From srs.rocks10 at gmail.com Mon Jan 5 05:32:04 2009 From: srs.rocks10 at gmail.com (john henry) Date: Mon, 5 Jan 2009 02:32:04 -0800 Subject: [Mp4-tech] H.264 SVC [JSVM] [FGS] Message-ID: Dear experts, Can anybody tell me when FGS would be proposed to introduced again in JSVM software? Can I use cfg files of older JSVM versions while using JSVM14 as there are no cfg files provided (data folder) with the software? Thirdly, is this good to work on DP and FMO in JSVM? There are some cfg files for FMO but I never heard any research on FMO or DP in JSVM. Please answer any or all of the questions, I would highly appreciate your help. Thanks in advance. ~John -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090105/cdc9704d/attachment.html From singer at apple.com Mon Jan 5 18:16:44 2009 From: singer at apple.com (Dave Singer) Date: Mon, 5 Jan 2009 15:16:44 -0800 Subject: [Mp4-tech] Differences between mp41 and mp42 brands In-Reply-To: References: Message-ID: At 20:49 +0000 15/12/08, Mark RISON wrote: >Hello, > >I understand that the mp42 brand indicates conformance to ISO 14496-14 >(and therefore ISO 14496-12 on which it is based), while the mp41 >brand (or absence of an ftyp box) indicates conformance to ISO 14496-1 >section 13. > >Is a comparison of these two available somewhere? That is, a list of >the substantive differences between a file which conforms to the mp41 >brand but not the mp42 brand, and vice-versa. > >Mark Does annex E in the (freely available) ISO/IEC 14496-12:2005/Amd.2:2008 at help? (Sorry, your question got close to the holiday break)... -- David Singer Multimedia Standards, Apple Inc. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090105/ed3e08df/attachment.html From singer at apple.com Mon Jan 5 18:59:02 2009 From: singer at apple.com (Dave Singer) Date: Mon, 5 Jan 2009 15:59:02 -0800 Subject: [Mp4-tech] Differences between mp41 and mp42 brands In-Reply-To: References: Message-ID: At 23:51 +0000 5/01/09, Mark RISON wrote: > >> Is a comparison of these two available somewhere? That is, a list of >>> the substantive differences between a file which conforms to the mp41 >>> brand but not the mp42 brand, and vice-versa. >> Does annex E in the (freely available) ISO/IEC 14496-12:2005/Amd.2:2008 at >> help? > >Thanks for the pointer, but this seems to cover only the isom, avc1, >iso2, mp71 and iso3 brands. Nothing about the mp41 and mp42 brands. Doh! mp41 and mp42 are covered, rather briefly, in 14496-14, of course The brand 'mp41' is defined as identifying version 1 of this specification (ISO/IEC 14496-1:2001), and the brand 'mp42' identifies this version of the specification; at least one of these brands shall appear in the compatible-brands list in the file-type box, in all files conforming to this specification. > >Mark > >-- >CPC/IP - A TCP/IP stack for Amstrad CPCs >-- http://www.cepece.info/cpcip/ >"Z88 vs CPC? Christ. How did we miss that platform war?" >-- http://www.ntk.net/index.cgi?b=02000-01-28&l=111#l >_________________________________________________________________ >Get a bird's eye view of the world with Multimap >http://clk.atdmt.com/GBL/go/115454059/direct/01/ -- David Singer Multimedia Standards, Apple Inc. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090105/ab042a2c/attachment.html From Shevach.Riabtsev at zoran.com Tue Jan 6 02:31:35 2009 From: Shevach.Riabtsev at zoran.com (Shevach Riabtsev) Date: Tue, 6 Jan 2009 09:31:35 +0200 Subject: [Mp4-tech] [Video][H.264] Question on MBAFF mode and frame height In-Reply-To: <91C4B16D6D58C541BF2F841D27F82536F88D29@zmlex.zml.zoran.com> References: <91C4B16D6D58C541BF2F841D27F82536F88CBA@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CC6@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CCE@zmlex.zml.zoran.com><244oh4lk912k9tocf4q1vct0bqt2a1fd6v@4ax.com><91C4B16D6D58C541BF2F841D27F82536F88CCF@zmlex.zml.zoran.com> <2BAAC5E4AF2518459F0AB5D92794204701B8850B@bur-exch-01.dolby.net> <91C4B16D6D58C541BF2F841D27F82536F88CD6@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88CE1@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D29@zmlex.zml.zoran.com> Message-ID: <91C4B16D6D58C541BF2F841D27F82536F88D6D@zmlex.zml.zoran.com> Dear Expert As far as I understand H.264 standard in order to encode a frame in MBAFF mode the frame height in MBs (FrameHeightInMbs in H.264 wording) should be even number otherwise the last row of MBs can't be paired. So, if frame height in MB units is odd number then we can't utilize MBAFF mode at all, although input source can be originated from interlaced video and MBAFF mode would give a positive gain. Unlike MPEG2 where MBAFF mode (i.e. field motion type and DCT field) can be applied for any resolution, in H.264 MBAFF mode is restricted to even frame heights. Is there any trick in H.264 except adding dummy MBs row to enable MBAFF for video with odd frame height? Regards Shevach Riabtsev Zoran -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090106/d0447c0c/attachment.html From jc-work at uphall.net Tue Jan 6 06:51:31 2009 From: jc-work at uphall.net (John Cox) Date: Tue, 06 Jan 2009 11:51:31 +0000 Subject: [Mp4-tech] [Video][H.264] Question on MBAFF mode and frame height In-Reply-To: <91C4B16D6D58C541BF2F841D27F82536F88D6D@zmlex.zml.zoran.com> References: <244oh4lk912k9tocf4q1vct0bqt2a1fd6v@4ax.com><91C4B16D6D58C541BF2F841D27F82536F88CCF@zmlex.zml.zoran.com> <2BAAC5E4AF2518459F0AB5D92794204701B8850B@bur-exch-01.dolby.net> <91C4B16D6D58C541BF2F841D27F82536F88CD6@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88CE1@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D29@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D6D@zmlex.zml.zoran.com> Message-ID: Hi You indeed have to code unused data, but you then use the frame_crop_* SPS parameters to restrict the display area to that desired. John Cox > >Dear Expert > >As far as I understand H.264 standard in order to encode a frame in >MBAFF mode the frame height in MBs (FrameHeightInMbs in H.264 wording) >should be even number otherwise the last row of MBs can't be paired. > >So, if frame height in MB units is odd number then we can't utilize >MBAFF mode at all, although input source can be originated from >interlaced video and MBAFF mode would give a positive gain. >Unlike MPEG2 where MBAFF mode (i.e. field motion type and DCT field) can >be applied for any resolution, in H.264 MBAFF mode is restricted to even >frame heights. > >Is there any trick in H.264 except adding dummy MBs row to enable MBAFF >for video with odd frame height? > >Regards >Shevach Riabtsev >Zoran From ksuehring at web.de Tue Jan 6 09:14:30 2009 From: ksuehring at web.de (Karsten Suehring) Date: Tue, 06 Jan 2009 15:14:30 +0100 Subject: [Mp4-tech] [Video][H.264][CABAC] Why I_PCM samples are notarithmetically coded In-Reply-To: References: <91C4B16D6D58C541BF2F841D27F82536F88CC6@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CCE@zmlex.zml.zoran.com><244oh4lk912k9tocf4q1vct0bqt2a1fd6v@4ax.com><91C4B16D6D58C541BF2F841D27F82536F88CCF@zmlex.zml.zoran.com> <2BAAC5E4AF2518459F0AB5D92794204701B8850B@bur-exch-01.dolby.net> <2BAAC5E4AF2518459F0AB5D92794204701B88606@bur-exch-01.dolby.net> Message-ID: <49636746.8070608@web.de> Hi, maybe it's a bit late for this discussion, but let me add that it's not possible to code all video data unmodified as I_PCM in Baseline, Main and Extended profiles: > When profile_idc is not equal to 66, 77, or 88, pcm_sample_luma[ i ] > shall not be equal to 0. (Semantics of pcm_sample_luma) This constraint was added to avoid start code emulations. Best regards, Karsten John Cox wrote: > Alexis > > No - I'm still confused - but I guess I'd better refine my statement to > something like: > > I do not see why I_PCM data should differ from the source data assuming that > the encoders intent is to render to source material faithfully and there is no > need to resize or reposition either the chroma or luma or in any other way > convert the data in order for it to fit into the picture description in the > SPS. > > If the encoder feels a need to "improve" the source then of course all bets are > off. I could also believe that there are cases where surrounding MBs are known > to be poorly encoded where a sharply encoded MB might be jarring but it seems > unlikely to me that you would want I_PCM in that case. > > Awaiting enlightenment... > > Thanks > > John Cox > > >> That is an interesting question indeed. Will let you think about it more >> carefully before I give an answer. You may wish to ask though what you >> consider as "original" and what I_PCM or even think carefully of what >> the so called lossless modes try to do. >> >> Best regards, >> >> Alexis >> >> -----Original Message----- >> From: John Cox [mailto:jc-work at uphall.net] >> Sent: Friday, November 14, 2008 2:53 AM >> To: Tourapis, Alexis >> Cc: Shevach Riabtsev; mp4-tech at lists.mpegif.org >> Subject: Re: [Mp4-tech] [Video][H.264][CABAC] Why I_PCM samples are >> notarithmetically coded >> >> Alexis >> >>> wish to code data without any alterations (note I_PCM does not >>> guarantee that the data you encode will be the original data, if you >>> understand what I mean) could satisfy the max bits per MB constraints? >> No - I don't understand what you mean. Why wouldn't the data you you >> encode be the original data in an I_PCM MB? (Assuming that we haven't >> got the rare QP conditions that allow us to deblock the samples in an >> I_PCM MB) >> >> Ta >> >> John Cox >> >> ----------------------------------------- >> This message (including any attachments) may contain confidential >> information intended for a specific individual and purpose. If you >> are not the intended recipient, delete this message. If you are >> not the intended recipient, disclosing, copying, distributing, or >> taking any action based on this message is strictly prohibited. > > _______________________________________________ > NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropriate identifier to indicate the type of question you have. > > Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php From jc-work at uphall.net Tue Jan 6 09:23:02 2009 From: jc-work at uphall.net (John Cox) Date: Tue, 06 Jan 2009 14:23:02 +0000 Subject: [Mp4-tech] [Video][H.264][CABAC] Why I_PCM samples are notarithmetically coded In-Reply-To: <49636746.8070608@web.de> References: <91C4B16D6D58C541BF2F841D27F82536F88CCE@zmlex.zml.zoran.com><244oh4lk912k9tocf4q1vct0bqt2a1fd6v@4ax.com><91C4B16D6D58C541BF2F841D27F82536F88CCF@zmlex.zml.zoran.com> <2BAAC5E4AF2518459F0AB5D92794204701B8850B@bur-exch-01.dolby.net> <2BAAC5E4AF2518459F0AB5D92794204701B88606@bur-exch-01.dolby.net> <49636746.8070608@web.de> Message-ID: Hi again OK - It may be need to do some serious rereading of the spec but I thought Annex B was meant to take care of all that. Is the rest of the standard really immune from start code emulation without Annex B (I'm pretty sure it isn't)? If not then why special case the I_PCM data? JC >Hi, > >maybe it's a bit late for this discussion, but let me add that it's not >possible to code all video data unmodified as I_PCM in Baseline, Main >and Extended profiles: > >> When profile_idc is not equal to 66, 77, or 88, pcm_sample_luma[ i ] >> shall not be equal to 0. > >(Semantics of pcm_sample_luma) > >This constraint was added to avoid start code emulations. > >Best regards, >Karsten > >John Cox wrote: >> Alexis >> >> No - I'm still confused - but I guess I'd better refine my statement to >> something like: >> >> I do not see why I_PCM data should differ from the source data assuming that >> the encoders intent is to render to source material faithfully and there is no >> need to resize or reposition either the chroma or luma or in any other way >> convert the data in order for it to fit into the picture description in the >> SPS. >> >> If the encoder feels a need to "improve" the source then of course all bets are >> off. I could also believe that there are cases where surrounding MBs are known >> to be poorly encoded where a sharply encoded MB might be jarring but it seems >> unlikely to me that you would want I_PCM in that case. >> >> Awaiting enlightenment... >> >> Thanks >> >> John Cox >> >> >>> That is an interesting question indeed. Will let you think about it more >>> carefully before I give an answer. You may wish to ask though what you >>> consider as "original" and what I_PCM or even think carefully of what >>> the so called lossless modes try to do. >>> >>> Best regards, >>> >>> Alexis >>> >>> -----Original Message----- >>> From: John Cox [mailto:jc-work at uphall.net] >>> Sent: Friday, November 14, 2008 2:53 AM >>> To: Tourapis, Alexis >>> Cc: Shevach Riabtsev; mp4-tech at lists.mpegif.org >>> Subject: Re: [Mp4-tech] [Video][H.264][CABAC] Why I_PCM samples are >>> notarithmetically coded >>> >>> Alexis >>> >>>> wish to code data without any alterations (note I_PCM does not >>>> guarantee that the data you encode will be the original data, if you >>>> understand what I mean) could satisfy the max bits per MB constraints? >>> No - I don't understand what you mean. Why wouldn't the data you you >>> encode be the original data in an I_PCM MB? (Assuming that we haven't >>> got the rare QP conditions that allow us to deblock the samples in an >>> I_PCM MB) >>> >>> Ta >>> >>> John Cox >>> >>> ----------------------------------------- >>> This message (including any attachments) may contain confidential >>> information intended for a specific individual and purpose. If you >>> are not the intended recipient, delete this message. If you are >>> not the intended recipient, disclosing, copying, distributing, or >>> taking any action based on this message is strictly prohibited. >> >> _______________________________________________ >> NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropriate identifier to indicate the type of question you have. >> >> Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php From mrison at hotmail.com Mon Jan 5 18:51:37 2009 From: mrison at hotmail.com (Mark RISON) Date: Mon, 5 Jan 2009 23:51:37 +0000 Subject: [Mp4-tech] Differences between mp41 and mp42 brands In-Reply-To: References: Message-ID: >> Is a comparison of these two available somewhere? That is, a list of >> the substantive differences between a file which conforms to the mp41 >> brand but not the mp42 brand, and vice-versa. > Does annex E in the (freely available) ISO/IEC 14496-12:2005/Amd.2:2008 at > help? Thanks for the pointer, but this seems to cover only the isom, avc1, iso2, mp71 and iso3 brands. Nothing about the mp41 and mp42 brands. Mark -- CPC/IP - A TCP/IP stack for Amstrad CPCs -- http://www.cepece.info/cpcip/ "Z88 vs CPC? Christ. How did we miss that platform war?" -- http://www.ntk.net/index.cgi?b=02000-01-28&l=111#l _________________________________________________________________ Get a bird?s eye view of the world with Multimap http://clk.atdmt.com/GBL/go/115454059/direct/01/ From mrison at hotmail.com Mon Jan 5 19:14:46 2009 From: mrison at hotmail.com (Mark RISON) Date: Tue, 6 Jan 2009 00:14:46 +0000 Subject: [Mp4-tech] Differences between mp41 and mp42 brands In-Reply-To: References: Message-ID: >>>> I understand that the mp42 brand indicates conformance to ISO >>>> 14496-14 (and therefore ISO 14496-12 on which it is based), >>>> while the mp41 brand (or absence of an ftyp box) indicates >>>> conformance to ISO 14496-1 section 13. >>>> Is a comparison of these two available somewhere? That is, a >>>> list of the substantive differences between a file which >>>> conforms to the mp41 brand but not the mp42 brand, and >>>> vice-versa. >>> Does annex E in the (freely available) ISO/IEC >>> 14496-12:2005/Amd.2:2008 help? >> Thanks for the pointer, but this seems to cover only the isom, >> avc1, iso2, mp71 and iso3 brands. Nothing about the mp41 and mp42 >> brands. > Doh! mp41 and mp42 are covered, rather briefly, in 14496-14, of > course > The brand 'mp41' is defined as identifying version 1 of this > specification (ISO/IEC 14496-1:2001), and the brand 'mp42' > identifies this version of the specification; at least one of these > brands shall appear in the compatible-brands list in the file-type > box, in all files conforming to this specification. Err, right, that's where I started: is there a list of the substantive differences between these two specifications? Mark -- CPC/IP - A TCP/IP stack for Amstrad CPCs -- http://www.cepece.info/cpcip/ "Z88 vs CPC? Christ. How did we miss that platform war?" -- http://www.ntk.net/index.cgi?b=02000-01-28&l=111#l _________________________________________________________________ Get a bird?s eye view of the world with Multimap http://clk.atdmt.com/GBL/go/115454059/direct/01/ From hanlongzhe at gmail.com Mon Jan 5 21:33:39 2009 From: hanlongzhe at gmail.com (Longzhe Han) Date: Tue, 6 Jan 2009 11:33:39 +0900 Subject: [Mp4-tech] About JSVM Decoder Message-ID: <705b8b2b0901051833y78d0f4eeo7cd5b4fc02d58f7c@mail.gmail.com> Hi, I am doing H.264/SVC streaming in NS-2. I would like to simulate SVC streaming in wireless networks. There is a function of JSVM which can generate packet trace file and use modified trace file to extract a bit-stream. I found that if any packet with Qid 0 and discardable has been deleted from trace file, the modified SVC bit-stream can be generated, however, the modified bit-stream can not be decoded fully. So the PSNR of two bit-stream can not be compared. I am not sure it is a problem of JSVM decoder or not. Is there other method to compare PSNR? Best regards, Longzhe Han From charlie.hubbard at gmail.com Tue Jan 6 10:38:33 2009 From: charlie.hubbard at gmail.com (Charlie Hubbard) Date: Tue, 6 Jan 2009 10:38:33 -0500 Subject: [Mp4-tech] Help parsing audio stream information from Mp4a (bit rate, sample rate, etc) Message-ID: I'm fairly new the mp4 format, and I've been successful at parsing most of the format so far, but I can't figure out where information related to the bit rate of the file, sample rate, channel modes (stereo, mono, joint stereo, etc) are contained? How do you detect variable bitrates in mp4 files? I'm mainly interested in audio files. Thanks Charlie From mp4-tech-owner at lists.mpegif.org Wed Jan 7 06:43:04 2009 From: mp4-tech-owner at lists.mpegif.org (MPEGIF List Admins) Date: Wed, 7 Jan 2009 12:43:04 +0100 Subject: FW: [Mp4-tech] [Video][H.264][CABAC] Why I_PCM samples arenotarithmetically coded Message-ID: <008d01c970bd$1b75bac0$52613040$@mpegif.org> 'nother one -----Original Message----- From: Barry Haskell [mailto:bhaskell at apple.com] Sent: Tuesday, 06 January 2009 20:29 To: Karsten Suehring Cc: John Cox; mp4-tech at lists.mpegif.org Subject: Re: [Mp4-tech] [Video][H.264][CABAC] Why I_PCM samples arenotarithmetically coded On Jan 6, 2009, at 11:21 AM, Barry Haskell wrote: > > On Jan 6, 2009, at 6:14 AM, Karsten Suehring wrote: > >> Hi, >> >> maybe it's a bit late for this discussion, but let me add that it's >> not >> possible to code all video data unmodified as I_PCM in Baseline, Main >> and Extended profiles: >> >>> When profile_idc is not equal to 66, 77, or 88, >>> pcm_sample_luma[ i ] >>> shall not be equal to 0. >> > > Hmmm. The text I found was "When profile_idc is not equal to 100, > 110, 122, or 144, pcm_sample_luma[ i ] shall not be equal to 0." > I see the corregendum changed this to "When profile_idc is equal to 66, 77, or 88, pcm_sample_luma[ i ] shall not be equal to 0." Barry G. Haskell tel +1 408 974 6333 Apple Inc. 2 Infinite Loop MS: 302-3KS Cupertino, CA 95014 bhaskell at apple.com (also B.Haskell at ieee.org) From mp4-tech-owner at lists.mpegif.org Wed Jan 7 06:43:04 2009 From: mp4-tech-owner at lists.mpegif.org (MPEGIF List Admins) Date: Wed, 7 Jan 2009 12:43:04 +0100 Subject: FW: [Mp4-tech] [Video][H.264][CABAC] Why I_PCM samples arenotarithmetically coded Message-ID: <008e01c970bd$1c6b3e50$5541baf0$@mpegif.org> Forwarding auto-discarded message from non-subscribed address. -----Original Message----- From: Barry Haskell [mailto:bhaskell at apple.com] Sent: Tuesday, 06 January 2009 20:21 To: Karsten Suehring Cc: John Cox; mp4-tech at lists.mpegif.org Subject: Re: [Mp4-tech] [Video][H.264][CABAC] Why I_PCM samples arenotarithmetically coded On Jan 6, 2009, at 6:14 AM, Karsten Suehring wrote: > Hi, > > maybe it's a bit late for this discussion, but let me add that it's > not > possible to code all video data unmodified as I_PCM in Baseline, Main > and Extended profiles: > >> When profile_idc is not equal to 66, 77, or 88, >> pcm_sample_luma[ i ] >> shall not be equal to 0. > Hmmm. The text I found was "When profile_idc is not equal to 100, 110, 122, or 144, pcm_sample_luma[ i ] shall not be equal to 0." Barry G. Haskell tel +1 408 974 6333 Apple Inc. 2 Infinite Loop MS: 302-3KS Cupertino, CA 95014 bhaskell at apple.com (also B.Haskell at ieee.org) From charlie.hubbard at gmail.com Wed Jan 7 16:10:13 2009 From: charlie.hubbard at gmail.com (Charlie Hubbard) Date: Wed, 7 Jan 2009 16:10:13 -0500 Subject: [Mp4-tech] Help parsing audio stream information from Mp4a (bit rate, sample rate, etc). Message-ID: I'm fairly new the mp4 format, and I've been successful at parsing most of the format so far, but I can't figure out where information related to the bit rate of the file, sample rate, channel modes (stereo, mono, joint stereo, etc) are contained? How do you detect variable bitrates in mp4 files? I'm mainly interested in audio files. Thanks Charlie From singer at apple.com Wed Jan 7 19:45:31 2009 From: singer at apple.com (Dave Singer) Date: Wed, 7 Jan 2009 16:45:31 -0800 Subject: [Mp4-tech] Help parsing audio stream information from Mp4a (bit rate, sample rate, etc). In-Reply-To: References: Message-ID: At 16:10 -0500 7/01/09, Charlie Hubbard wrote: >I'm fairly new the mp4 format, and I've been successful at parsing >most of the format so far, but I can't figure out where information >related to the bit rate of the file, sample rate, channel modes >(stereo, mono, joint stereo, etc) are contained? in the sample description atom mostly >How do you detect >variable bitrates in mp4 files? I'm mainly interested in audio files. um, you look at the sample sizes and sample durations, and perceive that their division is not constant? it sounds like you are going at it without documentation, which is unlikely to succeed. Merely finding AAC frames and ignoring the moov atom may seem to work some or even much of the time, but trust me, it ain't right apologies if this isn't where you are... > >Thanks >Charlie >_______________________________________________ >NOTE: Please use clear subject lines for your posts. Include [audio, >[video], [systems], [general] or another apppropriate identifier to >indicate the type of question you have. > >Note: Conduct on the mailing list is subject to the Antitrust >guidelines found at >http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php -- David Singer Multimedia Standards, Apple Inc. From charlie.hubbard at gmail.com Wed Jan 7 23:27:51 2009 From: charlie.hubbard at gmail.com (Charlie Hubbard) Date: Wed, 7 Jan 2009 23:27:51 -0500 Subject: [Mp4-tech] Help parsing audio stream information from Mp4a (bit rate, sample rate, etc). In-Reply-To: References: Message-ID: Dave, Thanks for the info. I've made a lot of progress today with this. I do have some of the documentation with ISO 14496-12. Unfortunately that doesn't have all the definitions for the boxes I need. I found the sample rate in the AudioSampleEntry in the SampleDescriptionBox both of which are in part 12. However, esds box, which looks like where most of the info I was looking for is, is not discussed in part 12. Somewhere I read part 1 discussed this, but I haven't found that spec yet. I have found example source code for parsing those, but I'd rather have real official docs. It just hard to find all of the docs needed. When you are referring to looking at the sample sizes and durations are you referring to looking at the values in 'stsz' and 'stz2' box? I had considered comparing the maxBitRate and averageBitRate from 'esds' to see if they were different in order to say if something was vbr or not. Is that not going to work? Thanks, Charlie On Wed, Jan 7, 2009 at 7:45 PM, Dave Singer wrote: > At 16:10 -0500 7/01/09, Charlie Hubbard wrote: >> >> I'm fairly new the mp4 format, and I've been successful at parsing >> most of the format so far, but I can't figure out where information >> related to the bit rate of the file, sample rate, channel modes >> (stereo, mono, joint stereo, etc) are contained? > > in the sample description atom mostly > >> How do you detect >> variable bitrates in mp4 files? I'm mainly interested in audio files. > > um, you look at the sample sizes and sample durations, and perceive that > their division is not constant? > > it sounds like you are going at it without documentation, which is unlikely > to succeed. Merely finding AAC frames and ignoring the moov atom may seem > to work some or even much of the time, but trust me, it ain't right > > apologies if this isn't where you are... > >> >> Thanks >> Charlie >> _______________________________________________ >> NOTE: Please use clear subject lines for your posts. Include [audio, >> [video], [systems], [general] or another apppropriate identifier to indicate >> the type of question you have. >> >> Note: Conduct on the mailing list is subject to the Antitrust guidelines >> found at >> http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php > > > -- > David Singer > Multimedia Standards, Apple Inc. > From singer at apple.com Thu Jan 8 13:52:27 2009 From: singer at apple.com (Dave Singer) Date: Thu, 8 Jan 2009 10:52:27 -0800 Subject: [Mp4-tech] Help parsing audio stream information from Mp4a (bit rate, sample rate, etc). In-Reply-To: References: Message-ID: At 23:27 -0500 7/01/09, Charlie Hubbard wrote: >Dave, > >Thanks for the info. I've made a lot of progress today with this. I >do have some of the documentation with ISO 14496-12. Unfortunately >that doesn't have all the definitions for the boxes I need. I found >the sample rate in the AudioSampleEntry in the SampleDescriptionBox >both of which are in part 12. However, esds box, which looks like >where most of the info I was looking for is, is not discussed in part >12. Somewhere I read part 1 discussed this, but I haven't found that >spec yet. I have found example source code for parsing those, but I'd >rather have real official docs. It just hard to find all of the docs >needed. the esds box itself is described in 14496-14; it contains an elementary stream descriptor as described in 14496-1 (sorry). > >When you are referring to looking at the sample sizes and durations >are you referring to looking at the values in 'stsz' and 'stz2' box? sample sizes in stsz or stz2, yes; sample durations in the time-to-sample (stts). >I had considered comparing the maxBitRate and averageBitRate from >'esds' to see if they were different in order to say if something was >vbr or not. Is that not going to work? if they are filled in right, that'll work. The sample durations and sizes is an easy check, though, since for most audio the frames have constant duration (e.g. AAC at 1024 samples/frame) and if the size is constant, the sample size table is also compact. > >Thanks, >Charlie > >On Wed, Jan 7, 2009 at 7:45 PM, Dave Singer wrote: >> At 16:10 -0500 7/01/09, Charlie Hubbard wrote: >>> >>> I'm fairly new the mp4 format, and I've been successful at parsing >>> most of the format so far, but I can't figure out where information >>> related to the bit rate of the file, sample rate, channel modes >>> (stereo, mono, joint stereo, etc) are contained? >> >> in the sample description atom mostly >> >>> How do you detect >>> variable bitrates in mp4 files? I'm mainly interested in audio files. >> >> um, you look at the sample sizes and sample durations, and perceive that >> their division is not constant? >> >> it sounds like you are going at it without documentation, which is unlikely >> to succeed. Merely finding AAC frames and ignoring the moov atom may seem >> to work some or even much of the time, but trust me, it ain't right >> >> apologies if this isn't where you are... >> >>> >>> Thanks >>> Charlie >>> _______________________________________________ >>> NOTE: Please use clear subject lines for your posts. Include [audio, >>> [video], [systems], [general] or another apppropriate identifier >>>to indicate >>> the type of question you have. >>> >>> Note: Conduct on the mailing list is subject to the Antitrust guidelines >>> found at >>> http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php >> >> >> -- >> David Singer >> Multimedia Standards, Apple Inc. >> -- David Singer Multimedia Standards, Apple Inc. From srs.rocks10 at gmail.com Thu Jan 8 14:50:53 2009 From: srs.rocks10 at gmail.com (john henry) Date: Thu, 8 Jan 2009 11:50:53 -0800 Subject: [Mp4-tech] H.264 SVC [JSVM] Message-ID: Dear experts, Can anybody tell me when FGS would be proposed to introduced again in JSVM software? Is this good to work on DP and FMO in JSVM? There are some cfg files for FMO but I never heard any research on FMO or DP in JSVM. I would highly appreciate your help. Thanks in advance. ~John -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090108/c49177fd/attachment.html From singer at apple.com Thu Jan 8 19:07:51 2009 From: singer at apple.com (Dave Singer) Date: Thu, 8 Jan 2009 16:07:51 -0800 Subject: [Mp4-tech] Re: avgBitrate (was: Help parsing audio stream information from Mp4a (bitrate, sample rate, etc.)) In-Reply-To: References: Message-ID: At 22:42 +0000 8/01/09, Mark RISON wrote: > >> I had considered comparing the maxBitRate and averageBitRate from >>> 'esds' to see if they were different in order to say if something was >>> vbr or not. Is that not going to work? >> if they are filled in right, that'll work. > >OK, there's something I've never understood: > >Somewhere in one of the 14496 specs -- something like the >DecoderConfigDescriptor description in 14496-1 -- it says that >avgBitrate is zero for variable-rate data. Isn't it precisely for >variable-rate data that you want to know the average rate, since for >constant-rate data the average bitrate is the same as the maximum >bitrate (maxBitrate just next to it)? yes, decoderconfig: maxBitrate - is the maximum bitrate in bits per second of this elementary stream in any time window of one second duration. avgBitrate - is the average bitrate in bits per second of this elementary stream. For streams with variable bitrate this value shall be set to zero. Before my time, alas. I hope someone from the old systems team can explain! I think the question was "average over what interval"... -- David Singer Multimedia Standards, Apple Inc. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090108/77f60525/attachment.html From ghughes at motorola.com Thu Jan 8 11:24:11 2009 From: ghughes at motorola.com (Hughes Gary-DJWV76) Date: Thu, 8 Jan 2009 11:24:11 -0500 Subject: [Mp4-tech] Differences between mp41 and mp42 brands In-Reply-To: References: Message-ID: The MP4 registration authority (http://www.mp4ra.org/index.html) is the, uh, authoritative place to look for this information. Brand mp41 (MP4v1) is described in 14496-1 ch 13. Brand mp42 (MP4v2) is described in 14496-14 (which cites 14496-1 and -12, so you need -1 either way) gary Gary Hughes Distinguished Member, Technical Staff Advanced Engineering Motorola On Demand Video, MA34 80 Central St. Boxborough, MA 01719 Email: ghughes at motorola.com Office: 978 266 7269 Mobile: 978 339 3615 Fax: 978 264 9108 > -----Original Message----- > From: mp4-tech-bounces at lists.mpegif.org > [mailto:mp4-tech-bounces at lists.mpegif.org] On Behalf Of Mark RISON > Sent: Monday, January 05, 2009 6:52 PM > To: singer at apple.com; mp4-tech at lists.mpegif.org > Subject: RE: [Mp4-tech] Differences between mp41 and mp42 brands > > > >> Is a comparison of these two available somewhere? That is, > a list of > >> the substantive differences between a file which conforms > to the mp41 > >> brand but not the mp42 brand, and vice-versa. > > Does annex E in the (freely available) ISO/IEC > > 14496-12:2005/Amd.2:2008 at help? > > Thanks for the pointer, but this seems to cover only the > isom, avc1, iso2, mp71 and iso3 brands. Nothing about the > mp41 and mp42 brands. > > Mark > > -- > CPC/IP - A TCP/IP stack for Amstrad CPCs > -- http://www.cepece.info/cpcip/ > "Z88 vs CPC? Christ. How did we miss that platform war?" > -- http://www.ntk.net/index.cgi?b=02000-01-28&l=111#l > _________________________________________________________________ > Get a bird's eye view of the world with Multimap > http://clk.atdmt.com/GBL/go/115454059/direct/01/ > _______________________________________________ > NOTE: Please use clear subject lines for your posts. Include > [audio, [video], [systems], [general] or another apppropriate > identifier to indicate the type of question you have. > > Note: Conduct on the mailing list is subject to the Antitrust > guidelines found at > http://www.mpegif.org/public/documents/vault/mp-out-30042-Anti > trust.php > From mrison at hotmail.com Thu Jan 8 17:42:13 2009 From: mrison at hotmail.com (Mark RISON) Date: Thu, 8 Jan 2009 22:42:13 +0000 Subject: [Mp4-tech] avgBitrate (was: Help parsing audio stream information from Mp4a (bitrate, sample rate, etc.)) In-Reply-To: References: Message-ID: >> I had considered comparing the maxBitRate and averageBitRate from >> 'esds' to see if they were different in order to say if something was >> vbr or not. Is that not going to work? > if they are filled in right, that'll work. OK, there's something I've never understood: Somewhere in one of the 14496 specs -- something like the DecoderConfigDescriptor description in 14496-1 -- it says that avgBitrate is zero for variable-rate data. Isn't it precisely for variable-rate data that you want to know the average rate, since for constant-rate data the average bitrate is the same as the maximum bitrate (maxBitrate just next to it)? Mark -- CPC/IP - A TCP/IP stack for Amstrad CPCs -- http://www.cepece.info/cpcip/ "Z88 vs CPC? Christ. How did we miss that platform war?" -- http://www.ntk.net/index.cgi?b=02000-01-28&l=111#l _________________________________________________________________ Cut through the jargon: find a PC for your needs. http://clk.atdmt.com/UKM/go/130777504/direct/01/ From arjun_majumdar at rediffmail.com Sat Jan 10 11:34:45 2009 From: arjun_majumdar at rediffmail.com (arjun majumdar) Date: Sat, 10 Jan 2009 16:34:45 -0000 Subject: [Mp4-tech] FMO/ASO query Message-ID: <20090110164401.20755.qmail@f6mail-145-186.rediffmail.com> Hi Garry Sir, We are implementing a H.264 Baseline Decoder in hardware. But we have some issues regarding the FMO mode. As per our understanding, the Macroblocks are coming in a paticular Slice Group order. So it means we have to first store all the Macroblock level information in normal raster scan order and then read them back in normal raster order to start the decoding of each macroblock there onwards. Is our understanding correct?? Again, if you could please tell us some formulas to find out the NextMbAddress depending upon the Slice Group Map type. What we are implenting is that we are storing the Slice Group Number of each Macroblock in a memory.But it is really becoming non feasable to find out the NextMbAddress in Dispersed and the Explicit Slice Group Map Type, because in that case, in the worst we have to traverse the whole memory in order to find out the NextMbAddress with the same Slice Group Number.This is very time consuming and a lot of power is consumed in hardware. Could you please suggest some feasable means to implement the FMO mode in hardware?? Hope to hear from you soon. Regards, Arjun -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090110/0a55f2c8/attachment.html From garysull at windows.microsoft.com Sun Jan 11 18:53:14 2009 From: garysull at windows.microsoft.com (Gary Sullivan) Date: Sun, 11 Jan 2009 15:53:14 -0800 Subject: [Mp4-tech] FMO/ASO query In-Reply-To: <20090110164401.20755.qmail@f6mail-145-186.rediffmail.com> References: <20090110164401.20755.qmail@f6mail-145-186.rediffmail.com> Message-ID: <406D6FEFCBB98643A34170C59E1ABEF31F7356B653@NA-EXMSG-W601.wingroup.windeploy.ntdev.microsoft.com> Arjun Majumdar et al, Actually, there are lots of other people besides me on this list -- I just tend to be excessively responsive. It is not necessary to store macroblock information and re-order it prior to decoding. Each slice can be decoded independently of any other slice in the (same) picture. And each slice (and each macroblock within the slice) can be decoded as soon as it arrives. The macroblocks of each slice are in raster scan order within the slice group that contains the slice (and all macroblocks of the slice are in the same slice group). For determining the next macroblock address after decoding a macroblock -- it is simply the next macroblock in raster scan order that is in the same slice group as the current macroblock. There are various ways to determine that, but I ultimately think it is up to you to figure out what is the best to do in your implementation environment. One possibility is, before decoding the picture, to build a map that identifies the next macroblock address for each macroblock. You can build such a map very quickly using num_slice_groups_minus1 + 1 pointers to track the preceding macroblock location in each slice group while scanning once through the map-unit-to-slice-group map in raster order. After building the map once, it can be used for decoding every slice in the picture. But you can do things any way you want to, as long as the decoded result is the same. Best Regards, Gary Sullivan ________________________________ From: mp4-tech-bounces at lists.mpegif.org [mailto:mp4-tech-bounces at lists.mpegif.org] On Behalf Of arjun majumdar Sent: Saturday, January 10, 2009 8:44 AM To: mp4tech Subject: [Mp4-tech] FMO/ASO query Hi Garry Sir, We are implementing a H.264 Baseline Decoder in hardware. But we have some issues regarding the FMO mode. As per our understanding, the Macroblocks are coming in a paticular Slice Group order. So it means we have to first store all the Macroblock level information in normal raster scan order and then read them back in normal raster order to start the decoding of each macroblock there onwards. Is our understanding correct?? Again, if you could please tell us some formulas to find out the NextMbAddress depending upon the Slice Group Map type. What we are implenting is that we are storing the Slice Group Number of each Macroblock in a memory.But it is really becoming non feasable to find out the NextMbAddress in Dispersed and the Explicit Slice Group Map Type, because in that case, in the worst we have to traverse the whole memory in order to find out the NextMbAddress with the same Slice Group Number.This is very time consuming and a lot of power is consumed in hardware. Could you please suggest some feasable means to implement the FMO mode in hardware?? Hope to hear from you soon. Regards, Arjun -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090111/2f484af9/attachment.html From dmitriy at graphics.cs.msu.ru Sun Jan 11 22:24:39 2009 From: dmitriy at graphics.cs.msu.ru (Dmitriy Vatolin) Date: Mon, 12 Jan 2009 06:24:39 +0300 Subject: [Mp4-tech] x264 analysis and updated metric tool In-Reply-To: <406D6FEFCBB98643A34170C59E1ABEF31F7356B653@NA-EXMSG-W601.wingroup.windeploy.ntdev.microsoft.com> References: <20090110164401.20755.qmail@f6mail-145-186.rediffmail.com> <406D6FEFCBB98643A34170C59E1ABEF31F7356B653@NA-EXMSG-W601.wingroup.windeploy.ntdev.microsoft.com> Message-ID: <346399763.20090112062439@graphics.cs.msu.ru> Dear experts! Several news from us: 1. We have analyzed x264 - following options: * Macroblocks partitions selection (--partitions) * Number of B-franes (--bframes) * Number of reference franes (--ref) * Motion Estimation method (--me) * Algorithm of subpixel compensation usage (--subme) * Mixing of predictions from differnt frames (--mixed-refs) * Weighted predictions in B-frames (--weightb) Document present novel type of options efficiency analysis, that can answer on following questions: * What set of options is optimal (in terms of optimal speed/quality) for exact class of video? * What option implementation is more efficient? * Is results options preset stable (in terms of optimal speed/quality balance) from sequence to sequence? Results can be found here: http://compression.ru/video/codec_comparison/x264_options_analysis_08_en.html 2. Also we published MSU Video Quality Measurement Tool 2.0 beta! New metrics release has following new features: * Simpler *.MOV, *.VOB, *.WMV, *.MP4, *.MPG, *.MKV, *.FLV and other media formats support * Bigger number of metrics (all types of PSNR) and etc. * 10, 14 and 16 bits color depth raw files support (in PRO). This is good tool for modest research in area of video compression or video filtering (for professional work we recommend our PRO tool). Program can be found here: http://www.compression.ru/video/quality_measure/video_measurement_tool_en.html 3. For next regular H.264 codecs comparison we need some help. It's reasonable to include as more codecs as possible to make comparison appropriate, but many companies stop H.264 codecs R&D now, so they are not interested in next comparisons participation. So it's necessary to start DirectShow or GUI codecs (like Apple or LEAD) for batch running with different parameters. Any help to solve this problem is welcome. Thank you! Yours, Dr. Vatolin From engrsajidanwar at yahoo.com Mon Jan 12 00:18:02 2009 From: engrsajidanwar at yahoo.com (Sajid Anwar SoEE SNU) Date: Sun, 11 Jan 2009 21:18:02 -0800 (PST) Subject: [Mp4-tech] mpeg 4 sp decoder on arm926ej-s platform Message-ID: <513653.3625.qm@web34804.mail.mud.yahoo.com> ? Dear Experts, ? Have any body experienced designing an mpeg 4 sp decoder for arm 926ejs platform. I want to achieve a frame rate of about 30 fps with VGA Format. I have optimized IDCT, VLD, Inverse Quantization and currently my frame rate is about 25 fps. ? I need some general idea about optimization apart from this thats why i am not giving any profiling information. Kindly if you have experienced such a case, then kindly let me know. Thanks for your time. Kind Regards, Sajid Anwar -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090111/497c41a1/attachment.html From Shevach.Riabtsev at zoran.com Mon Jan 12 08:35:36 2009 From: Shevach.Riabtsev at zoran.com (Shevach Riabtsev) Date: Mon, 12 Jan 2009 15:35:36 +0200 Subject: [Mp4-tech] [Video][H.264] Question on MBAFF mode and frame height In-Reply-To: <91C4B16D6D58C541BF2F841D27F82536F88D6D@zmlex.zml.zoran.com> References: <91C4B16D6D58C541BF2F841D27F82536F88CBA@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CC6@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CCE@zmlex.zml.zoran.com><244oh4lk912k9tocf4q1vct0bqt2a1fd6v@4ax.com><91C4B16D6D58C541BF2F841D27F82536F88CCF@zmlex.zml.zoran.com> <2BAAC5E4AF2518459F0AB5D92794204701B8850B@bur-exch-01.dolby.net> <91C4B16D6D58C541BF2F841D27F82536F88CD6@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88CE1@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D29@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D6D@zmlex.zml.zoran.com> Message-ID: <91C4B16D6D58C541BF2F841D27F82536F88D7E@zmlex.zml.zoran.com> Dear Experts Can a picture which was marked as "unused for reference" become later "used for reference" by any memory_management_control_operation command (of course provided that the picture is still in DPB)? Regards Shevach Riabtsev Zoran -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090112/56c44b2f/attachment.html From Shevach.Riabtsev at zoran.com Mon Jan 12 08:37:31 2009 From: Shevach.Riabtsev at zoran.com (Shevach Riabtsev) Date: Mon, 12 Jan 2009 15:37:31 +0200 Subject: [Mp4-tech] [Video][H.264] Can a picture which was marked as "unused for reference" become later "used for reference" In-Reply-To: <91C4B16D6D58C541BF2F841D27F82536F88D7E@zmlex.zml.zoran.com> References: <91C4B16D6D58C541BF2F841D27F82536F88CBA@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CC6@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CCE@zmlex.zml.zoran.com><244oh4lk912k9tocf4q1vct0bqt2a1fd6v@4ax.com><91C4B16D6D58C541BF2F841D27F82536F88CCF@zmlex.zml.zoran.com> <2BAAC5E4AF2518459F0AB5D92794204701B8850B@bur-exch-01.dolby.net> <91C4B16D6D58C541BF2F841D27F82536F88CD6@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88CE1@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D29@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D6D@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D7E@zmlex.zml.zoran.com> Message-ID: <91C4B16D6D58C541BF2F841D27F82536F88D7F@zmlex.zml.zoran.com> Dear Experts Can a picture which was marked as "unused for reference" become later "used for reference" by any memory_management_control_operation command (of course provided that the picture is still in DPB)? Regards Shevach Riabtsev Zoran -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090112/ae2e802f/attachment.html From garysull at windows.microsoft.com Mon Jan 12 10:02:30 2009 From: garysull at windows.microsoft.com (Gary Sullivan) Date: Mon, 12 Jan 2009 07:02:30 -0800 Subject: [Mp4-tech] [Video][H.264] Question on MBAFF mode and frame height In-Reply-To: <91C4B16D6D58C541BF2F841D27F82536F88D7E@zmlex.zml.zoran.com> References: <91C4B16D6D58C541BF2F841D27F82536F88CBA@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CC6@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CCE@zmlex.zml.zoran.com><244oh4lk912k9tocf4q1vct0bqt2a1fd6v@4ax.com><91C4B16D6D58C541BF2F841D27F82536F88CCF@zmlex.zml.zoran.com> <2BAAC5E4AF2518459F0AB5D92794204701B8850B@bur-exch-01.dolby.net> <91C4B16D6D58C541BF2F841D27F82536F88CD6@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88CE1@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D29@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D6D@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D7E@zmlex.zml.zoran.com> Message-ID: <406D6FEFCBB98643A34170C59E1ABEF31F7356B6BD@NA-EXMSG-W601.wingroup.windeploy.ntdev.microsoft.com> Shevach, No. Best Regards, Gary Sullivan ________________________________ From: mp4-tech-bounces at lists.mpegif.org [mailto:mp4-tech-bounces at lists.mpegif.org] On Behalf Of Shevach Riabtsev Sent: Monday, January 12, 2009 5:36 AM To: mp4-tech at lists.mpegif.org Subject: [Mp4-tech] [Video][H.264] Question on MBAFF mode and frame height Dear Experts Can a picture which was marked as ?unused for reference? become later ?used for reference? by any memory_management_control_operation command (of course provided that the picture is still in DPB)? Regards Shevach Riabtsev Zoran -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090112/6a912651/attachment.html From arjun_majumdar at rediffmail.com Mon Jan 12 11:15:54 2009 From: arjun_majumdar at rediffmail.com (arjun majumdar) Date: Mon, 12 Jan 2009 16:15:54 -0000 Subject: [Mp4-tech] FMO/ASO query Message-ID: <20090112162509.62409.qmail@f6mail-145-188.rediffmail.com> Hi Garry Sir, Thanks for your quick response. I have now understood the concept of FMO. I had just one query regarding FMO/ASO. If my understanding is correct,in scattered and dispersed slice group map type, there is a reduction in coding efficiency, wherever spatial redundancy is exploited eg. Intra prediction, Motion Vector Prediction etc. I just request a confirmation from your side on my understanding of the above issue. Thank you again. On Mon, 12 Jan 2009 Gary Sullivan wrote : >Arjun Majumdar et al, > >Actually, there are lots of other people besides me on this list -- I just tend to be excessively responsive. It is not necessary to store macroblock information and re-order it prior to decoding. > >Each slice can be decoded independently of any other slice in the (same) picture. And each slice (and each macroblock within the slice) can be decoded as soon as it arrives. > >The macroblocks of each slice are in raster scan order within the slice group that contains the slice (and all macroblocks of the slice are in the same slice group). > >For determining the next macroblock address after decoding a macroblock -- it is simply the next macroblock in raster scan order that is in the same slice group as the current macroblock. There are various ways to determine that, but I ultimately think it is up to you to figure out what is the best to do in your implementation environment. > >One possibility is, before decoding the picture, to build a map that identifies the next macroblock address for each macroblock. You can build such a map very quickly using num_slice_groups_minus1 + 1 pointers to track the preceding macroblock location in each slice group while scanning once through the map-unit-to-slice-group map in raster order. After building the map once, it can be used for decoding every slice in the picture. > >But you can do things any way you want to, as long as the decoded result is the same. > >Best Regards, > >Gary Sullivan > >________________________________ > From: mp4-tech-bounces at lists.mpegif.org [mailto:mp4-tech-bounces at lists.mpegif.org] On Behalf Of arjun majumdar >Sent: Saturday, January 10, 2009 8:44 AM >To: mp4tech >Subject: [Mp4-tech] FMO/ASO query > > >Hi Garry Sir, > >We are implementing a H.264 Baseline Decoder in hardware. But we have some issues regarding the FMO mode. > >As per our understanding, the Macroblocks are coming in a paticular Slice Group order. So it means we have to first store all the Macroblock level information in normal raster scan order and then read them back in normal raster order to start the decoding of each macroblock there onwards. > >Is our understanding correct?? > >Again, if you could please tell us some formulas to find out the NextMbAddress depending upon the Slice Group Map type. What we are implenting is that we are storing the Slice Group Number of each Macroblock in a memory.But it is really becoming non feasable to find out the NextMbAddress in Dispersed and the Explicit Slice Group Map Type, because in that case, in the worst we have to traverse the whole memory in order to find out the NextMbAddress with the same Slice Group Number.This is very time consuming and a lot of power is consumed in hardware. > >Could you please suggest some feasable means to implement the FMO mode in hardware?? > >Hope to hear from you soon. > >Regards, > >Arjun > > >_______________________________________________ >NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropriate identifier to indicate the type of question you have. > >Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090112/14a8c11e/attachment.html From garysull at windows.microsoft.com Mon Jan 12 14:27:33 2009 From: garysull at windows.microsoft.com (Gary Sullivan) Date: Mon, 12 Jan 2009 11:27:33 -0800 Subject: [Mp4-tech] FMO/ASO query In-Reply-To: <20090112162509.62409.qmail@f6mail-145-188.rediffmail.com> References: <20090112162509.62409.qmail@f6mail-145-188.rediffmail.com> Message-ID: <406D6FEFCBB98643A34170C59E1ABEF31F736C5A17@NA-EXMSG-W601.wingroup.windeploy.ntdev.microsoft.com> Arjun Majumdar et al, Yes - Whenever spatially-neighboring macroblocks are not available to use for prediction of a particular macroblock (due to being outside the slice that contains the particular macroblock), there is likely to be some reduction of compression capability. For maximum compression capability purposes, it is typically best to make the slices be large and to make the slices cover spatially contiguous regions -- e.g., to use rectangular slices. (Aside from region-of-interest coding tricks, perhaps.) Best Regards, Gary Sullivan ________________________________ From: arjun majumdar [mailto:arjun_majumdar at rediffmail.com] Sent: Monday, January 12, 2009 8:25 AM To: Gary Sullivan Cc: mp4tech Subject: Re: RE: [Mp4-tech] FMO/ASO query Hi Garry Sir, Thanks for your quick response. I have now understood the concept of FMO. I had just one query regarding FMO/ASO. If my understanding is correct,in scattered and dispersed slice group map type, there is a reduction in coding efficiency, wherever spatial redundancy is exploited eg. Intra prediction, Motion Vector Prediction etc. I just request a confirmation from your side on my understanding of the above issue. Thank you again. On Mon, 12 Jan 2009 Gary Sullivan wrote : >Arjun Majumdar et al, > >Actually, there are lots of other people besides me on this list -- I just tend to be excessively responsive. It is not necessary to store macroblock information and re-order it prior to decoding. > >Each slice can be decoded independently of any other slice in the (same) picture. And each slice (and each macroblock within the slice) can be decoded as soon as it arrives. > >The macroblocks of each slice are in raster scan order within the slice group that contains the slice (and all macroblocks of the slice are in the same slice group). > >For determining the next macroblock address after decoding a macroblock -- it is simply the next macroblock in raster scan order that is in the same slice group as the current macroblock. There are various ways to determine that, but I ultimately think it is up to you to figure out what is the best to do in your implementation environment. > >One possibility is, before decoding the picture, to build a map that identifies the next macroblock address for each macroblock. You can build such a map very quickly using num_slice_groups_minus1 + 1 pointers to track the preceding macroblock location in each slice group while scanning once through the map-unit-to-slice-group map in raster order. After building the map once, it can be used for decoding every slice in the picture. > >But you can do things any way you want to, as long as the decoded result is the same. > >Best Regards, > >Gary Sullivan > >________________________________ > From: mp4-tech-bounces at lists.mpegif.org [mailto:mp4-tech-bounces at lists.mpegif.org] On Behalf Of arjun majumdar >Sent: Saturday, January 10, 2009 8:44 AM >To: mp4tech >Subject: [Mp4-tech] FMO/ASO query > > >Hi Garry Sir, > >We are implementing a H.264 Baseline Decoder in hardware. But we have some issues regarding the FMO mode. > >As per our understanding, the Macroblocks are coming in a paticular Slice Group order. So it means we have to first store all the Macroblock level information in normal raster scan order and then read them back in normal raster order to start the decoding of each macroblock there onwards. > >Is our understanding correct?? > >Again, if you could please tell us some formulas to find out the NextMbAddress depending upon the Slice Group Map type. What we are implenting is that we are storing the Slice Group Number of each Macroblock in a memory.But it is really becoming non feasable to find out the NextMbAddress in Dispersed and the Explicit Slice Group Map Type, because in that case, in the worst we have to traverse the whole memory in order to find out the NextMbAddress with the same Slice Group Number.This is very time consuming and a lot of power is consumed in hardware. > >Could you please suggest some feasable means to implement the FMO mode in hardware?? > >Hope to hear from you soon. > >Regards, > >Arjun > > >_______________________________________________ >NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropriate identifier to indicate the type of question you have. > >Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090112/059191df/attachment.html From lbrtlee at gmail.com Tue Jan 13 04:00:55 2009 From: lbrtlee at gmail.com (Albert Lee) Date: Tue, 13 Jan 2009 18:00:55 +0900 Subject: [Mp4-tech] [Video][H.264] Question on MBAFF mode and frame height In-Reply-To: <406D6FEFCBB98643A34170C59E1ABEF31F7356B6BD@NA-EXMSG-W601.wingroup.windeploy.ntdev.microsoft.com> References: <91C4B16D6D58C541BF2F841D27F82536F88CBA@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88CCF@zmlex.zml.zoran.com> <2BAAC5E4AF2518459F0AB5D92794204701B8850B@bur-exch-01.dolby.net> <91C4B16D6D58C541BF2F841D27F82536F88CD6@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88CE1@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D29@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D6D@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D7E@zmlex.zml.zoran.com> <406D6FEFCBB98643A34170C59E1ABEF31F7356B6BD@NA-EXMSG-W601.wingroup.windeploy.ntdev.microsoft.com> Message-ID: I searched some explanations to this issue. According to 7,4,3,3, the marking of a reference picture can be "unused", "short", "long", but only one among these three. And marking process is applied only to short-term, long-term or current picture. So if a picture is marked to "unused", marking process will never be invoked for this picture again. Regards Albert On Tue, Jan 13, 2009 at 12:02 AM, Gary Sullivan < garysull at windows.microsoft.com> wrote: > Shevach, > > No. > > Best Regards, > > Gary Sullivan > > ------------------------------ > *From:* mp4-tech-bounces at lists.mpegif.org [mailto: > mp4-tech-bounces at lists.mpegif.org] *On Behalf Of *Shevach Riabtsev > *Sent:* Monday, January 12, 2009 5:36 AM > *To:* mp4-tech at lists.mpegif.org > *Subject:* [Mp4-tech] [Video][H.264] Question on MBAFF mode and frame > height > > Dear Experts > > Can a picture which was marked as "unused for reference" become later "used > for reference" by any memory_management_control_operation command (of > course provided that the picture is still in DPB)? > > Regards > > Shevach Riabtsev > > Zoran > > > _______________________________________________ > NOTE: Please use clear subject lines for your posts. Include [audio, > [video], [systems], [general] or another apppropriate identifier to indicate > the type of question you have. > > Note: Conduct on the mailing list is subject to the Antitrust guidelines > found at > http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php > -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090113/68706c8a/attachment.html From Shevach.Riabtsev at zoran.com Tue Jan 13 06:43:06 2009 From: Shevach.Riabtsev at zoran.com (Shevach Riabtsev) Date: Tue, 13 Jan 2009 13:43:06 +0200 Subject: [Mp4-tech] [Video][H.264] Can a picture which was marked as "unused for reference" become later "used for reference" In-Reply-To: References: <91C4B16D6D58C541BF2F841D27F82536F88CBA@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88CCF@zmlex.zml.zoran.com> <2BAAC5E4AF2518459F0AB5D92794204701B8850B@bur-exch-01.dolby.net> <91C4B16D6D58C541BF2F841D27F82536F88CD6@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88CE1@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D29@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D6D@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D7E@zmlex.zml.zoran.com> <406D6FEFCBB98643A34170C59E1ABEF31F7356B6BD@NA-EXMSG-W601.wingroup.windeploy.ntdev.microsoft.com> Message-ID: <91C4B16D6D58C541BF2F841D27F82536F88D84@zmlex.zml.zoran.com> Albert Thanks for your comments. So, I see that the only way to exclude temporarily a reference picture from consideration can be done by reordering commands. Indeed, by corresponding reordering command(s) we can move a selected picture at the bottom of reference list then by setting num_ref_idx_active_override_flag=1 and decreasing num_ref_idx_lx_active_minus1 by 1 we actually make the selected picture temporarily "unused". Regards ________________________________ From: Albert Lee [mailto:lbrtlee at gmail.com] Sent: Tuesday, January 13, 2009 11:01 AM To: Shevach Riabtsev Cc: mp4-tech at lists.mpegif.org Subject: Re: [Mp4-tech] [Video][H.264] Question on MBAFF mode and frame height I searched some explanations to this issue. According to 7,4,3,3, the marking of a reference picture can be "unused", "short", "long", but only one among these three. And marking process is applied only to short-term, long-term or current picture. So if a picture is marked to "unused", marking process will never be invoked for this picture again. Regards Albert On Tue, Jan 13, 2009 at 12:02 AM, Gary Sullivan wrote: Shevach, No. Best Regards, Gary Sullivan ________________________________ From: mp4-tech-bounces at lists.mpegif.org [mailto:mp4-tech-bounces at lists.mpegif.org] On Behalf Of Shevach Riabtsev Sent: Monday, January 12, 2009 5:36 AM To: mp4-tech at lists.mpegif.org Subject: [Mp4-tech] [Video][H.264] Question on MBAFF mode and frame height Dear Experts Can a picture which was marked as "unused for reference" become later "used for reference" by any memory_management_control_operation command (of course provided that the picture is still in DPB)? Regards Shevach Riabtsev Zoran _______________________________________________ NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropriate identifier to indicate the type of question you have. Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090113/652558ae/attachment.html From Shevach.Riabtsev at zoran.com Tue Jan 13 07:05:16 2009 From: Shevach.Riabtsev at zoran.com (Shevach Riabtsev) Date: Tue, 13 Jan 2009 14:05:16 +0200 Subject: [Mp4-tech] [Video][H.264] Can one field in complementary field pair be "used for reference" while the second one is marked "unused for reference" In-Reply-To: <91C4B16D6D58C541BF2F841D27F82536F88D7F@zmlex.zml.zoran.com> References: <91C4B16D6D58C541BF2F841D27F82536F88CBA@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CC6@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CCE@zmlex.zml.zoran.com><244oh4lk912k9tocf4q1vct0bqt2a1fd6v@4ax.com><91C4B16D6D58C541BF2F841D27F82536F88CCF@zmlex.zml.zoran.com> <2BAAC5E4AF2518459F0AB5D92794204701B8850B@bur-exch-01.dolby.net> <91C4B16D6D58C541BF2F841D27F82536F88CD6@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88CE1@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D29@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D6D@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D7E@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D7F@zmlex.zml.zoran.com> Message-ID: <91C4B16D6D58C541BF2F841D27F82536F88D85@zmlex.zml.zoran.com> Dear Experts As far as I understand memory_management_control_operation commands we can mark the first field in a complementary field pair as "used for reference" while the second field we can mark as "unused for reference". If I am right then the following scenario looks very vague: Assume that in a complementary field pair the first field is marked as "used for reference" and the second one is marked as "unused for reference". Assume also the above-mentioned complementary field pair is followed by a frame picture. How the frame picture can use the above-mentioned complementary field pair for reference if one of fields is excluded from reference list? Regards Shevach Riabtsev Zoran -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090113/e36d5ad4/attachment.html From garysull at windows.microsoft.com Tue Jan 13 10:02:31 2009 From: garysull at windows.microsoft.com (Gary Sullivan) Date: Tue, 13 Jan 2009 07:02:31 -0800 Subject: [Mp4-tech] [Video][H.264] Can one field in complementary field pair be "used for reference" while the second one is marked "unused for reference" In-Reply-To: <91C4B16D6D58C541BF2F841D27F82536F88D85@zmlex.zml.zoran.com> References: <91C4B16D6D58C541BF2F841D27F82536F88CBA@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CC6@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CCE@zmlex.zml.zoran.com><244oh4lk912k9tocf4q1vct0bqt2a1fd6v@4ax.com><91C4B16D6D58C541BF2F841D27F82536F88CCF@zmlex.zml.zoran.com> <2BAAC5E4AF2518459F0AB5D92794204701B8850B@bur-exch-01.dolby.net> <91C4B16D6D58C541BF2F841D27F82536F88CD6@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88CE1@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D29@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D6D@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D7E@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D7F@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D85@zmlex.zml.zoran.com> Message-ID: <406D6FEFCBB98643A34170C59E1ABEF31F736C5E10@NA-EXMSG-W601.wingroup.windeploy.ntdev.microsoft.com> Shevach et al, A frame cannot use a complementary reference field pair for reference if only one of its fields is currently marked as "used for reference". So if there are no other reference pictures in the DPB, the next frame would need to contain only I slices. Best Regards, Gary Sullivan ________________________________ From: mp4-tech-bounces at lists.mpegif.org [mailto:mp4-tech-bounces at lists.mpegif.org] On Behalf Of Shevach Riabtsev Sent: Tuesday, January 13, 2009 4:05 AM To: mp4-tech at lists.mpegif.org Subject: [Mp4-tech] [Video][H.264] Can one field in complementary field pair be "used for reference" while the second one is marked "unused for reference" Dear Experts As far as I understand memory_management_control_operation commands we can mark the first field in a complementary field pair as ?used for reference? while the second field we can mark as ?unused for reference?. If I am right then the following scenario looks very vague: Assume that in a complementary field pair the first field is marked as ?used for reference? and the second one is marked as ?unused for reference?. Assume also the above-mentioned complementary field pair is followed by a frame picture. How the frame picture can use the above-mentioned complementary field pair for reference if one of fields is excluded from reference list? Regards Shevach Riabtsev Zoran -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090113/34304f69/attachment.html From alexis.tourapis at dolby.com Tue Jan 13 20:19:13 2009 From: alexis.tourapis at dolby.com (Tourapis, Alexis) Date: Tue, 13 Jan 2009 17:19:13 -0800 Subject: [Mp4-tech] [Video][H.264] Can a picture which was marked as"unused for reference" become later "used for reference" In-Reply-To: <91C4B16D6D58C541BF2F841D27F82536F88D84@zmlex.zml.zoran.com> References: <91C4B16D6D58C541BF2F841D27F82536F88CBA@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CCF@zmlex.zml.zoran.com><2BAAC5E4AF2518459F0AB5D92794204701B8850B@bur-exch-01.dolby.net><91C4B16D6D58C541BF2F841D27F82536F88CD6@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CE1@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88D29@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88D6D@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88D7E@zmlex.zml.zoran.com><406D6FEFCBB98643A34170C59E1ABEF31F7356B6BD@NA-EXMSG-W601.wingroup.windeploy.ntdev.microsoft.com> <91C4B16D6D58C541BF2F841D27F82536F88D84@zmlex.zml.zoran.com> Message-ID: <2BAAC5E4AF2518459F0AB5D92794204701D33046@bur-exch-01.dolby.net> Dear Shevach, If your desire is to just not use a reference for motion compensation purposes, although reordering is one way, another way is to just add a condition in your encoder that disallows that reference to be used during motion estimation. Even though the reference is available, you just tell the encoder not to use it. Obviously special handling needs to be considered for skip and direct modes if those modes require the usage of that particular reference (i.e. you need to disallow also the usage of those modes for the cases that that reference is required by them). Reordering may perform overall a bit better in terms of performance (depends on the scenario of course), but I would not call it as being the only way. Of course, such mechanisms depend on what you are also trying to accomplish (i.e. with reordering the picture was removed from the reference list, even though temporarily, with the above restriction, it is in the list but never used. This may have implications for some cases such as when you wish to consider error concealment for example). Best regards, Alexis ________________________________ From: mp4-tech-bounces at lists.mpegif.org [mailto:mp4-tech-bounces at lists.mpegif.org] On Behalf Of Shevach Riabtsev Sent: Tuesday, January 13, 2009 3:43 AM To: Albert Lee Cc: mp4-tech at lists.mpegif.org Subject: RE: [Mp4-tech] [Video][H.264] Can a picture which was marked as"unused for reference" become later "used for reference" Albert Thanks for your comments. So, I see that the only way to exclude temporarily a reference picture from consideration can be done by reordering commands. Indeed, by corresponding reordering command(s) we can move a selected picture at the bottom of reference list then by setting num_ref_idx_active_override_flag=1 and decreasing num_ref_idx_lx_active_minus1 by 1 we actually make the selected picture temporarily "unused". Regards ________________________________ From: Albert Lee [mailto:lbrtlee at gmail.com] Sent: Tuesday, January 13, 2009 11:01 AM To: Shevach Riabtsev Cc: mp4-tech at lists.mpegif.org Subject: Re: [Mp4-tech] [Video][H.264] Question on MBAFF mode and frame height I searched some explanations to this issue. According to 7,4,3,3, the marking of a reference picture can be "unused", "short", "long", but only one among these three. And marking process is applied only to short-term, long-term or current picture. So if a picture is marked to "unused", marking process will never be invoked for this picture again. Regards Albert On Tue, Jan 13, 2009 at 12:02 AM, Gary Sullivan wrote: Shevach, No. Best Regards, Gary Sullivan ________________________________ From: mp4-tech-bounces at lists.mpegif.org [mailto:mp4-tech-bounces at lists.mpegif.org] On Behalf Of Shevach Riabtsev Sent: Monday, January 12, 2009 5:36 AM To: mp4-tech at lists.mpegif.org Subject: [Mp4-tech] [Video][H.264] Question on MBAFF mode and frame height Dear Experts Can a picture which was marked as "unused for reference" become later "used for reference" by any memory_management_control_operation command (of course provided that the picture is still in DPB)? Regards Shevach Riabtsev Zoran _______________________________________________ NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropriate identifier to indicate the type of question you have. Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php ----------------------------------------- This message (including any attachments) may contain confidential information intended for a specific individual and purpose. If you are not the intended recipient, delete this message. If you are not the intended recipient, disclosing, copying, distributing, or taking any action based on this message is strictly prohibited. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090113/29db1416/attachment.html From ruchi.sharma at patni.com Wed Jan 14 00:30:10 2009 From: ruchi.sharma at patni.com (Ruchi Sharma) Date: Wed, 14 Jan 2009 11:00:10 +0530 Subject: [Mp4-tech] Re: [audio] Regarding conformance testing of HE-AAC Message-ID: <006401c97609$2b83f020$ab04a8c0@patni.com> Hello All, Can Any one provide me Raw ER_AAC_LC bitstream ??? If any one knows , plz let me know the link. Thanks in advance. _____________________________________________________________________ This e-mail message may contain proprietary, confidential or legally privileged information for the sole use of the person or entity to whom this message was originally addressed. Any review, e-transmission dissemination or other use of or taking of any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you have received this e-mail in error kindly delete this e-mail from your records. If it appears that this mail has been forwarded to you without proper authority, please notify us immediately at netadmin at patni.com and delete this mail. _____________________________________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090114/cbfeab61/attachment.html From ralph.sperschneider at iis.fraunhofer.de Wed Jan 14 06:30:04 2009 From: ralph.sperschneider at iis.fraunhofer.de (Ralph Sperschneider) Date: Wed, 14 Jan 2009 12:30:04 +0100 Subject: [Mp4-tech] Re: [audio] Regarding conformance testing of HE-AAC In-Reply-To: <006401c97609$2b83f020$ab04a8c0@patni.com> References: <006401c97609$2b83f020$ab04a8c0@patni.com> Message-ID: <496DCCBC.90101@iis.fraunhofer.de> Ruchi Sharma wrote: > > Hello All, > > > > Can Any one provide me Raw ER_AAC_LC bitstream ??? > > If any one knows , plz let me know the link. > > Thanks in advance. > > > You may use any adif file you can find here: ftp://mpaudconf:adif2mp4 at ftp.iis.fraunhofer.de/mpeg2aac-conformance/compressedAdif/ as a basis. By simply removing the adif header you will get the raw bitstream you need. Best regards, Ralph > > > > > > > > > > > > _____________________________________________________________________ > > This e-mail message may contain proprietary, confidential or legally > privileged information for the sole use of the person or entity to > whom this message was originally addressed. Any review, e-transmission > dissemination or other use of or taking of any action in reliance upon > this information by persons or entities other than the intended > recipient is prohibited. If you have received this e-mail in error > kindly delete this e-mail from your records. If it appears that this > mail has been forwarded to you without proper authority, please notify > us immediately at netadmin at patni.com and delete this mail. > _____________________________________________________________________ > ------------------------------------------------------------------------ > > _______________________________________________ > NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropriate identifier to indicate the type of question you have. > > Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php -- Dipl.-Ing. Ralph Sperschneider | Phone: +49 9131 776 6167 Fraunhofer IIS | Fax: +49 9131 776 6099 Am Wolfsmantel 33 | mailto:ralph.sperschneider at iis.fraunhofer.de D 91058 Erlangen | http://www.iis.fraunhofer.de/amm/ From ralph.sperschneider at iis.fraunhofer.de Wed Jan 14 08:04:15 2009 From: ralph.sperschneider at iis.fraunhofer.de (Ralph Sperschneider) Date: Wed, 14 Jan 2009 14:04:15 +0100 Subject: [Mp4-tech] Re: [Audio] AAC CCE dpcm_gain_element sign bit extraction order In-Reply-To: <482E79A9.7000402@improvsys.com> References: <482E79A9.7000402@improvsys.com> Message-ID: <496DE2CF.20503@iis.fraunhofer.de> Alex Converse wrote: > Dear Experts, > > ISO/IEC 14496-3:2005(E) states that in the case when gain_element_sign > is set then the cc_sign is extracted and removed from each > dpcm_gain_element before the dpcm_gain_elements are summed (Section > 4.6.8.3.3 in the couple_channel() sample function) . However the > reference decoder (ISO/IEC 14496-5:2001/Amd.10:2007 > > cc_hufffac() function in mp4mcDec/src/coupling.c) sums the > dpcm_gain_elements then strips the cc_sign bit. > > These operations don't appear to be commutable. For example (taken > from al_07_44, block 167 starting at dpcm_gain_element[1][0][20]) take > this sequence of dpcm_gain_elements: {21, -10, 17, ...}. According to > the way the specification seems to be written, these should be decoded > to cc_sign = {-1, 1, -1, ...} and gain_element = {10, 5, 13}. > According to the reference decoder, these should be cc_sign = {-1, -1, > 1, ...} and gain_element = {10, 5, 14}. > > Which behavior is correct? Or did I misinterpret something? > > Regards, > Alex Converse > _______________________________________________ > NOTE: Please use clear subject lines for your posts. Include [audio, > [video], [systems], [general] or another apppropriate identifier to > indicate the type of question you have. > > Note: Conduct on the mailing list is subject to the Antitrust > guidelines found at > http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php Dear Alex, not sure anybody has sent a response to you yet - at least I just cannot see any. So let me step in - although its already some time age when you sent this request. Meanwhile this issue was evaluated within MPEG and we found that your observation is correct - thank you very much for reporting it. Accordingly, we drafted a corrigendum (ISO/IEC 14496-3:2005/COR 6) to adopt the standard text to the reference software. This Cor will be published soon (see http://www.itscj.ipsj.or.jp/sc29/29w42911.htm#MPEG-4 for its status). Best regards, Ralph -- Dipl.-Ing. Ralph Sperschneider | Phone: +49 9131 776 6167 Fraunhofer IIS | Fax: +49 9131 776 6099 Am Wolfsmantel 33 | mailto:ralph.sperschneider at iis.fraunhofer.de D 91058 Erlangen | http://www.iis.fraunhofer.de/amm/ From Shevach.Riabtsev at zoran.com Wed Jan 14 08:18:17 2009 From: Shevach.Riabtsev at zoran.com (Shevach Riabtsev) Date: Wed, 14 Jan 2009 15:18:17 +0200 Subject: [Mp4-tech] [Video][H.264] Can one field in complementary field pair be "used for reference" while the second one is marked "unused for reference" In-Reply-To: <406D6FEFCBB98643A34170C59E1ABEF31F736C5E10@NA-EXMSG-W601.wingroup.windeploy.ntdev.microsoft.com> References: <91C4B16D6D58C541BF2F841D27F82536F88CBA@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CC6@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CCE@zmlex.zml.zoran.com><244oh4lk912k9tocf4q1vct0bqt2a1fd6v@4ax.com><91C4B16D6D58C541BF2F841D27F82536F88CCF@zmlex.zml.zoran.com><2BAAC5E4AF2518459F0AB5D92794204701B8850B@bur-exch-01.dolby.net><91C4B16D6D58C541BF2F841D27F82536F88CD6@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CE1@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88D29@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88D6D@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88D7E@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88D7F@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D85@zmlex.zml.zoran.com> <406D6FEFCBB98643A34170C59E1ABEF31F736C5E10@NA-EXMSG-W601.wingroup.windeploy.ntdev.microsoft.com> Message-ID: <91C4B16D6D58C541BF2F841D27F82536F88D8B@zmlex.zml.zoran.com> Gary I assume that the same restrictions on an encoder are also valid for dangling field case. As for dangling field case it is not clear how to display the dangling field since only one field is available. How to compose a frame picture in such case? Regards ________________________________ From: Gary Sullivan [mailto:garysull at windows.microsoft.com] Sent: Tuesday, January 13, 2009 5:03 PM To: Shevach Riabtsev; mp4-tech at lists.mpegif.org Subject: RE: [Mp4-tech] [Video][H.264] Can one field in complementary field pair be "used for reference" while the second one is marked "unused for reference" Shevach et al, A frame cannot use a complementary reference field pair for reference if only one of its fields is currently marked as "used for reference". So if there are no other reference pictures in the DPB, the next frame would need to contain only I slices. Best Regards, Gary Sullivan ________________________________ From: mp4-tech-bounces at lists.mpegif.org [mailto:mp4-tech-bounces at lists.mpegif.org] On Behalf Of Shevach Riabtsev Sent: Tuesday, January 13, 2009 4:05 AM To: mp4-tech at lists.mpegif.org Subject: [Mp4-tech] [Video][H.264] Can one field in complementary field pair be "used for reference" while the second one is marked "unused for reference" Dear Experts As far as I understand memory_management_control_operation commands we can mark the first field in a complementary field pair as "used for reference" while the second field we can mark as "unused for reference". If I am right then the following scenario looks very vague: Assume that in a complementary field pair the first field is marked as "used for reference" and the second one is marked as "unused for reference". Assume also the above-mentioned complementary field pair is followed by a frame picture. How the frame picture can use the above-mentioned complementary field pair for reference if one of fields is excluded from reference list? Regards Shevach Riabtsev Zoran -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090114/9afd2ff9/attachment.html From garysull at windows.microsoft.com Wed Jan 14 13:56:47 2009 From: garysull at windows.microsoft.com (Gary Sullivan) Date: Wed, 14 Jan 2009 10:56:47 -0800 Subject: [Mp4-tech] [Video][H.264] Can one field in complementary field pair be "used for reference" while the second one is marked "unused for reference" In-Reply-To: <91C4B16D6D58C541BF2F841D27F82536F88D8B@zmlex.zml.zoran.com> References: <91C4B16D6D58C541BF2F841D27F82536F88CBA@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CC6@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CCE@zmlex.zml.zoran.com><244oh4lk912k9tocf4q1vct0bqt2a1fd6v@4ax.com><91C4B16D6D58C541BF2F841D27F82536F88CCF@zmlex.zml.zoran.com><2BAAC5E4AF2518459F0AB5D92794204701B8850B@bur-exch-01.dolby.net><91C4B16D6D58C541BF2F841D27F82536F88CD6@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CE1@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88D29@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88D6D@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88D7E@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88D7F@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D85@zmlex.zml.zoran.com> <406D6FEFCBB98643A34170C59E1ABEF31F736C5E10@NA-EXMSG-W601.wingroup.windeploy.ntdev.microsoft.com> <91C4B16D6D58C541BF2F841D27F82536F88D8B@zmlex.zml.zoran.com> Message-ID: <406D6FEFCBB98643A34170C59E1ABEF31F736C63F1@NA-EXMSG-W601.wingroup.windeploy.ntdev.microsoft.com> Shevach et al, The concepts relating to field pairing in the coding standard are not really for display purposes. For example, it is certainly possible for a field to be considered "non-paired" in the coding standard despite having no discontinuity in the top-bottom field ordering and timing of the output process. The pairing in the standard is primarily only for purposes of determining how pairs of fields can be referenced by subsequent coded frames and how to model the memory consumption for DPB capacity measurements relating to profile and level constraints. People should consider the display issues separately. From the perspective of the coding standard, display is completely outside the scope. Also, these days many displays are not actually interlaced and there is a great diversity of display resolutions and display refresh rate capabilities, so there is a lot that needs to happen between decoding the video and displaying it. My advice in that regard would be to simply study all the issues very carefully and do what you think is best. It's hard to get more precise than that without extensive study of the exact circumstances of the particular target system design environments for a given product. Best Regards, Gary Sullivan ________________________________ From: Shevach Riabtsev [mailto:Shevach.Riabtsev at zoran.com] Sent: Wednesday, January 14, 2009 5:18 AM To: Gary Sullivan; mp4-tech at lists.mpegif.org Subject: RE: [Mp4-tech] [Video][H.264] Can one field in complementary field pair be "used for reference" while the second one is marked "unused for reference" Gary I assume that the same restrictions on an encoder are also valid for dangling field case. As for dangling field case it is not clear how to display the dangling field since only one field is available. How to compose a frame picture in such case? Regards ________________________________ From: Gary Sullivan [mailto:garysull at windows.microsoft.com] Sent: Tuesday, January 13, 2009 5:03 PM To: Shevach Riabtsev; mp4-tech at lists.mpegif.org Subject: RE: [Mp4-tech] [Video][H.264] Can one field in complementary field pair be "used for reference" while the second one is marked "unused for reference" Shevach et al, A frame cannot use a complementary reference field pair for reference if only one of its fields is currently marked as "used for reference". So if there are no other reference pictures in the DPB, the next frame would need to contain only I slices. Best Regards, Gary Sullivan ________________________________ From: mp4-tech-bounces at lists.mpegif.org [mailto:mp4-tech-bounces at lists.mpegif.org] On Behalf Of Shevach Riabtsev Sent: Tuesday, January 13, 2009 4:05 AM To: mp4-tech at lists.mpegif.org Subject: [Mp4-tech] [Video][H.264] Can one field in complementary field pair be "used for reference" while the second one is marked "unused for reference" Dear Experts As far as I understand memory_management_control_operation commands we can mark the first field in a complementary field pair as ?used for reference? while the second field we can mark as ?unused for reference?. If I am right then the following scenario looks very vague: Assume that in a complementary field pair the first field is marked as ?used for reference? and the second one is marked as ?unused for reference?. Assume also the above-mentioned complementary field pair is followed by a frame picture. How the frame picture can use the above-mentioned complementary field pair for reference if one of fields is excluded from reference list? Regards Shevach Riabtsev Zoran -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090114/a31c0c07/attachment.html From mp4-tech-owner at lists.mpegif.org Thu Jan 29 16:15:50 2009 From: mp4-tech-owner at lists.mpegif.org (mpegif list admins) Date: Thu, 29 Jan 2009 22:15:50 +0100 Subject: [Mp4-tech] FW: What is x264_param_t b_aud? Message-ID: <000001c98256$c42076b0$4c616410$@mpegif.org> Forwarding message that got lost From: Hadi Hadizadeh [mailto:hadi.hadizadeh at hotmail.com] Sent: Thursday, 15 January, 2009 01:04 To: mp4-tech at lists.mpegif.org Subject: What is x264_param_t b_aud? Hi, Do you know what is x264_param_t param.b_aud? Thanks! _____ -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090129/571ae4c1/attachment.html From koumaras at iit.demokritos.gr Wed Jan 14 08:00:12 2009 From: koumaras at iit.demokritos.gr (Harilaos G. Koumaras) Date: Wed, 14 Jan 2009 15:00:12 +0200 Subject: [Mp4-tech] CFP: Special Issue on the Quality of Experience issues in Multimedia Provision Message-ID: <496DE1DC.9000804@iit.demokritos.gr> Dear colleagues, The attached call for papers may be of interest to you. Please feel free to forward it. Thank you -- ----------------------------------------------------------------------- Dr. Harilaos G. Koumaras Associate Researcher http://koumaras.gr.eu.org Institute of Informatics and Telecommunications, N.C.S.R "Demokritos", Digital Communications Lab. 15310, Agia Paraskevi, Greece phone: + (30) 210 650 31 07 fax: + (30) 210 653 21 75 ------------------------------------------------------------------------ -------------- next part -------------- A non-text attachment was scrubbed... Name: Telecommunication Systems Special Issue CFP short.pdf Type: application/pdf Size: 11081 bytes Desc: not available Url : /pipermail/mp4-tech/attachments/20090114/33ccaad4/attachment.pdf From hadi.hadizadeh at hotmail.com Wed Jan 14 19:04:10 2009 From: hadi.hadizadeh at hotmail.com (Hadi Hadizadeh) Date: Wed, 14 Jan 2009 16:04:10 -0800 Subject: [Mp4-tech] What is x264_param_t b_aud? Message-ID: Hi, Do you know what is x264_param_t param.b_aud? Thanks! _________________________________________________________________ Keep in touch and up to date with friends and family. Make the connection now. http://www.microsoft.com/windows/windowslive/ -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090114/fe767987/attachment.html From hadi.hadizadeh at hotmail.com Wed Jan 14 19:54:09 2009 From: hadi.hadizadeh at hotmail.com (Hadi Hadizadeh) Date: Wed, 14 Jan 2009 16:54:09 -0800 Subject: [Mp4-tech] X.264, NAL->p_payload Message-ID: Hi, Suppose that we have: x264_t h;Now, what is the difference between the following experssions: h->out.p_bitstreamh->out.NAL[0].p_payload Actually, I want to access to the final encoded data in NAL units that will be written into the output file or be transmitted. Any help? Thanks! _________________________________________________________________ Drag n? drop?Get easy photo sharing with Windows Live? Photos. http://www.microsoft.com/windows/windowslive/photos.aspx -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090114/a75a465d/attachment.html From ruchi.sharma at patni.com Wed Jan 14 23:54:05 2009 From: ruchi.sharma at patni.com (Ruchi Sharma) Date: Thu, 15 Jan 2009 10:24:05 +0530 Subject: [Mp4-tech] RE: [audio] Regarding conformance testing of HE-AAC In-Reply-To: <496DCCBC.90101@iis.fraunhofer.de> References: <006401c97609$2b83f020$ab04a8c0@patni.com> <496DCCBC.90101@iis.fraunhofer.de> Message-ID: <002e01c976cd$4bdd0900$ab04a8c0@patni.com> Hi Ralph, Thanks for your reply. I am downloading the bit-stream from the ftp server you had mentioned. Does those bit-stream are only AAC_LC and HE-AAC raw stream?? I need ER_AAC_LC raw bitstream.....can you tell me the name of those bitstream those are ER_AAC_LC raw stream?? Thanks. Regards, Ruchi -----Original Message----- From: Ralph Sperschneider [mailto:ralph.sperschneider at iis.fraunhofer.de] Sent: Wednesday, January 14, 2009 5:00 PM To: ruchi.sharma at patni.com Cc: mp4-tech at lists.mpegif.org Subject: Re: [audio] Regarding conformance testing of HE-AAC Ruchi Sharma wrote: > > Hello All, > > > > Can Any one provide me Raw ER_AAC_LC bitstream ??? > > If any one knows , plz let me know the link. > > Thanks in advance. > > > You may use any adif file you can find here: ftp://mpaudconf:adif2mp4 at ftp.iis.fraunhofer.de/mpeg2aac-conformance/compress edAdif/ as a basis. By simply removing the adif header you will get the raw bitstream you need. Best regards, Ralph > > > > > > > > > > > > _____________________________________________________________________ > > This e-mail message may contain proprietary, confidential or legally > privileged information for the sole use of the person or entity to > whom this message was originally addressed. Any review, e-transmission > dissemination or other use of or taking of any action in reliance upon > this information by persons or entities other than the intended > recipient is prohibited. If you have received this e-mail in error > kindly delete this e-mail from your records. If it appears that this > mail has been forwarded to you without proper authority, please notify > us immediately at netadmin at patni.com and delete this mail. > _____________________________________________________________________ > ------------------------------------------------------------------------ > > _______________________________________________ > NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropriate identifier to indicate the type of question you have. > > Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php -- Dipl.-Ing. Ralph Sperschneider | Phone: +49 9131 776 6167 Fraunhofer IIS | Fax: +49 9131 776 6099 Am Wolfsmantel 33 | mailto:ralph.sperschneider at iis.fraunhofer.de D 91058 Erlangen | http://www.iis.fraunhofer.de/amm/ _____________________________________________________________________ This e-mail message may contain proprietary, confidential or legally privileged information for the sole use of the person or entity to whom this message was originally addressed. Any review, e-transmission dissemination or other use of or taking of any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you have received this e-mail in error kindly delete this e-mail from your records. If it appears that this mail has been forwarded to you without proper authority, please notify us immediately at netadmin at patni.com and delete this mail. _____________________________________________________________________ From ralph.sperschneider at iis.fraunhofer.de Thu Jan 15 05:38:49 2009 From: ralph.sperschneider at iis.fraunhofer.de (Ralph Sperschneider) Date: Thu, 15 Jan 2009 11:38:49 +0100 Subject: [Mp4-tech] Re: [audio] Regarding conformance testing of HE-AAC In-Reply-To: <002e01c976cd$4bdd0900$ab04a8c0@patni.com> References: <006401c97609$2b83f020$ab04a8c0@patni.com> <496DCCBC.90101@iis.fraunhofer.de> <002e01c976cd$4bdd0900$ab04a8c0@patni.com> Message-ID: <496F1239.7080908@iis.fraunhofer.de> Ruchi Sharma wrote: > Hi Ralph, > > Thanks for your reply. > I am downloading the bit-stream from the ftp server you had mentioned. > Does those bit-stream are only AAC_LC and HE-AAC raw stream?? > I need ER_AAC_LC raw bitstream.....can you tell me the name of those > bitstream those are ER_AAC_LC raw stream?? > Thanks. > > Regards, > Ruchi > > > Hi Ruchi, it seems I have overlooked the ER in your first e-mail - sorry for that. To my knowledge there is no raw bitstream syntax defined for any ER AOT. The ER AAC LC sequences on the server are stored in the mp4ff container. You find them here: mpeg4audio-conformance/compressedMp4/er_al* If really needed, you could write a little program which parses those mp4 files and then dumps just the AU payloads - at least for epConfig = 0|3 those bitstreams might still be decodable as long as all relevant meta data is available. Best regards, Ralph > > > -----Original Message----- > From: Ralph Sperschneider [mailto:ralph.sperschneider at iis.fraunhofer.de] > Sent: Wednesday, January 14, 2009 5:00 PM > To: ruchi.sharma at patni.com > Cc: mp4-tech at lists.mpegif.org > Subject: Re: [audio] Regarding conformance testing of HE-AAC > > Ruchi Sharma wrote: > >> Hello All, >> >> >> >> Can Any one provide me Raw ER_AAC_LC bitstream ??? >> >> If any one knows , plz let me know the link. >> >> Thanks in advance. >> >> >> >> > > You may use any adif file you can find here: > > ftp://mpaudconf:adif2mp4 at ftp.iis.fraunhofer.de/mpeg2aac-conformance/compress > edAdif/ > as a basis. By simply removing the adif header you will get the raw > bitstream you need. > > Best regards, > Ralph > >> >> >> >> >> >> >> >> >> >> >> _____________________________________________________________________ >> >> This e-mail message may contain proprietary, confidential or legally >> privileged information for the sole use of the person or entity to >> whom this message was originally addressed. Any review, e-transmission >> dissemination or other use of or taking of any action in reliance upon >> this information by persons or entities other than the intended >> recipient is prohibited. If you have received this e-mail in error >> kindly delete this e-mail from your records. If it appears that this >> mail has been forwarded to you without proper authority, please notify >> us immediately at netadmin at patni.com and delete this mail. >> _____________________________________________________________________ >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> NOTE: Please use clear subject lines for your posts. Include [audio, >> > [video], [systems], [general] or another apppropriate identifier to indicate > the type of question you have. > >> Note: Conduct on the mailing list is subject to the Antitrust guidelines >> > found at > http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php > > > -- Dipl.-Ing. Ralph Sperschneider | Phone: +49 9131 776 6167 Fraunhofer IIS | Fax: +49 9131 776 6099 Am Wolfsmantel 33 | mailto:ralph.sperschneider at iis.fraunhofer.de D 91058 Erlangen | http://www.iis.fraunhofer.de/amm/ From video.codec.help at gmail.com Thu Jan 15 07:34:08 2009 From: video.codec.help at gmail.com (J. Miles) Date: Thu, 15 Jan 2009 13:34:08 +0100 Subject: [Mp4-tech] [MP4 container] License issue with MP4 container format? Message-ID: <14f06c5d0901150434y76fb9873j7daf38c18be5db16@mail.gmail.com> Hello everyone, We are looking into supporting the MP4 container (file) format as specified in MPEG-4 Part 14 (formally ISO/IEC 14496-14). This specification is an extension of MPEG-4 Part 12 on the ISO base file format, which again builds upon the principles of MPEG-4 Part 1 (Systems). Both Part 12 and Part 14 contain patent claims, as does Part 1. We are aware of a license provided by MPEG-LA for MPEG-4 Part 1 Systems, although that has been "closed down" according to their homepage. Our question is thus: - Does MP4 usage require licensing? - If so, who provides such a license? - And do we also need to license Part 1? (and then who provides that, if not MPEG-LA?) - Is there any difference between reading/accessing an MP4 file and writing/creating MP4 files? Any help and comments will be appreciated. Thanks. - Jay -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090115/9152f721/attachment.html From hadi.hadizadeh at hotmail.com Sat Jan 17 19:10:07 2009 From: hadi.hadizadeh at hotmail.com (Hadi Hadizadeh) Date: Sat, 17 Jan 2009 16:10:07 -0800 Subject: [Mp4-tech] X.264 doesn't support multiple slices per frame!!! Message-ID: Hi, Someone says that X.264 doesn't support multiple slices per frame. But is it true? I think it is strange because if there was only one slice per frame, then by dropping a NAL unit (due to channel errors), then the whole frame would be lost! But what is your opinion? Suppose that we want to deal with very high resolutions, then it would be better to use multiple slices per frame. But really how can we use different slices per frame in X.264? Thank you in advance! Best Regards,Hadi _________________________________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090117/15ed91ad/attachment.html From praveen.sisodia at gmail.com Mon Jan 19 04:35:35 2009 From: praveen.sisodia at gmail.com (praveen sisodia) Date: Mon, 19 Jan 2009 15:05:35 +0530 Subject: [Mp4-tech] Need a 3GPP timed text clip Message-ID: Hi,I need a 3g2/3gp clip with audio,video and text stream present in it. Can some one help? Thanks Praveen -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090119/d7d01db9/attachment.html From Shevach.Riabtsev at zoran.com Tue Jan 20 06:12:37 2009 From: Shevach.Riabtsev at zoran.com (Shevach Riabtsev) Date: Tue, 20 Jan 2009 13:12:37 +0200 Subject: [Mp4-tech] [Video][H.264][Deblock] Question on FilterOffsetA In-Reply-To: <91C4B16D6D58C541BF2F841D27F82536F88D85@zmlex.zml.zoran.com> References: <91C4B16D6D58C541BF2F841D27F82536F88CBA@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CC6@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CCE@zmlex.zml.zoran.com><244oh4lk912k9tocf4q1vct0bqt2a1fd6v@4ax.com><91C4B16D6D58C541BF2F841D27F82536F88CCF@zmlex.zml.zoran.com> <2BAAC5E4AF2518459F0AB5D92794204701B8850B@bur-exch-01.dolby.net> <91C4B16D6D58C541BF2F841D27F82536F88CD6@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88CE1@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D29@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D6D@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D7E@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D7F@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D85@zmlex.zml.zoran.com> Message-ID: <91C4B16D6D58C541BF2F841D27F82536F88D9F@zmlex.zml.zoran.com> Dear Experts In the syntax element slice_alpha_c0_offset_div2 the suffix div2 sounds suspicious. Indeed, the syntax element slice_alpha_c0_offset_div2 is used for calculation of FilterOffsetA as follows FilterOffsetA = slice_alpha_c0_offset_div2 << 1 So, in 2's complement arithmetic FilterOffsetA is always even number. Why FilterOffsetA is restricted to be always even number? What's reasoning behind? The same question can be re-formulated for FilterOffsetB. Regards Shevach Riabtsev Zoran -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090120/c91d2616/attachment.html From Shevach.Riabtsev at zoran.com Tue Jan 20 08:57:47 2009 From: Shevach.Riabtsev at zoran.com (Shevach Riabtsev) Date: Tue, 20 Jan 2009 15:57:47 +0200 Subject: [Mp4-tech] [Video][H.264] On complementary reference field pair In-Reply-To: <91C4B16D6D58C541BF2F841D27F82536F88D9F@zmlex.zml.zoran.com> References: <91C4B16D6D58C541BF2F841D27F82536F88CBA@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CC6@zmlex.zml.zoran.com><91C4B16D6D58C541BF2F841D27F82536F88CCE@zmlex.zml.zoran.com><244oh4lk912k9tocf4q1vct0bqt2a1fd6v@4ax.com><91C4B16D6D58C541BF2F841D27F82536F88CCF@zmlex.zml.zoran.com> <2BAAC5E4AF2518459F0AB5D92794204701B8850B@bur-exch-01.dolby.net> <91C4B16D6D58C541BF2F841D27F82536F88CD6@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88CE1@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D29@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D6D@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D7E@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D7F@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D85@zmlex.zml.zoran.com> <91C4B16D6D58C541BF2F841D27F82536F88D9F@zmlex.zml.zoran.com> Message-ID: <91C4B16D6D58C541BF2F841D27F82536F88DA0@zmlex.zml.zoran.com> Dear Experts The standard defines complementary reference field pair as follow (see 3.34): Two reference fields that are in consecutive access units in decoding order as two coded fields and share the same value of the frame_num syntax element, where the second field in decoding order is not an IDR picture and does not include a memory_management_control_operation syntax element equal to 5. In the above definition is absent a condition on field parity. This leads me to think that the standard permits two top reference fields compose a single complementary field pair. Does the standard allow complimentary field pair consisting of two reference fields with the same polarity? Regards Shevach Riabtsev Zoran -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090120/286fbbce/attachment.html From hadi.hadizadeh at hotmail.com Wed Jan 21 13:22:41 2009 From: hadi.hadizadeh at hotmail.com (Hadi Hadizadeh) Date: Wed, 21 Jan 2009 10:22:41 -0800 Subject: [Mp4-tech] JM decoder crashes with a lost NAL unit... (erc_do_p.c) Message-ID: Hi, I have encoded a YUV file with an H.264 encoder (X.264). Then to simulate the channel error, I drop some of the NAL units of this encoded file (from the half of the file (frame #156) to the end of it (frame #300)). But unfortunately, when I try to decode it with JM decoder I got the following errors: > ldecod.exe decoder.cfg case #1: in decoder.cfg we set Err Concealment mode = 0in this case I got this error on frame #156 : An unintentional loss of pictures occures! Exit case #2: in decoder.cfg we set Err Concealment mode = 1 or 2in this case at the begining of the decoding process I got the following error:0000(I) 0 0 10 4:2:0 140Assertion failed: conceal_from_picture != NULL, file c:\jm\lencode\src\erc_do_p.c, line 1725 Please note that I can play this encoded file (which some of its NAL units have been dropped) with VLC player with no error. But what is the problem? Could you please help me? Thank you in advance! Best Regards,Hadi _________________________________________________________________ Show them the way! Add maps and directions to your party invites. http://www.microsoft.com/windows/windowslive/events.aspx -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090121/3076a710/attachment.html From ruchi.sharma at patni.com Thu Jan 22 06:39:08 2009 From: ruchi.sharma at patni.com (Ruchi Sharma) Date: Thu, 22 Jan 2009 17:09:08 +0530 Subject: [Mp4-tech] RE: [audio] Regarding conformance testing of HE-AAC In-Reply-To: <496F1239.7080908@iis.fraunhofer.de> References: <006401c97609$2b83f020$ab04a8c0@patni.com> <496DCCBC.90101@iis.fraunhofer.de> <002e01c976cd$4bdd0900$ab04a8c0@patni.com> <496F1239.7080908@iis.fraunhofer.de> Message-ID: <006801c97c86$09f7b3a0$ab04a8c0@patni.com> Hello Ralph, Need some help again!! I had dumped the AU payloads for er_aac files having epflag = 1, suitabletracks = 3; I had dumped the data after appending all the layers in an AU unit. But in the ISO code, there is a function "BsPrepareEp1Stream" which required layer information also. I had dumped AAC files also after appending all the layers in an AU unit. They are decoding well. So, is the er_aac raw file with epflag greater than 1 contains the data in layer form???? Regards, Ruchi Sharma Sr. Software Engineer Product Enginnering Services Patni Computers Systems Ltd. Desk no.-43, Unit-19, SDF-7 SEEPZ, Andheri (E) Mumbai-400096 Ext. N0. -5810 Link Line No. - 601-7115 -----Original Message----- From: Ralph Sperschneider [mailto:ralph.sperschneider at iis.fraunhofer.de] Sent: Thursday, January 15, 2009 4:09 PM To: ruchi.sharma at patni.com Cc: mp4-tech at lists.mpegif.org Subject: Re: [audio] Regarding conformance testing of HE-AAC Ruchi Sharma wrote: > Hi Ralph, > > Thanks for your reply. > I am downloading the bit-stream from the ftp server you had mentioned. > Does those bit-stream are only AAC_LC and HE-AAC raw stream?? > I need ER_AAC_LC raw bitstream.....can you tell me the name of those > bitstream those are ER_AAC_LC raw stream?? > Thanks. > > Regards, > Ruchi > > > Hi Ruchi, it seems I have overlooked the ER in your first e-mail - sorry for that. To my knowledge there is no raw bitstream syntax defined for any ER AOT. The ER AAC LC sequences on the server are stored in the mp4ff container. You find them here: mpeg4audio-conformance/compressedMp4/er_al* If really needed, you could write a little program which parses those mp4 files and then dumps just the AU payloads - at least for epConfig = 0|3 those bitstreams might still be decodable as long as all relevant meta data is available. Best regards, Ralph > > > -----Original Message----- > From: Ralph Sperschneider [mailto:ralph.sperschneider at iis.fraunhofer.de] > Sent: Wednesday, January 14, 2009 5:00 PM > To: ruchi.sharma at patni.com > Cc: mp4-tech at lists.mpegif.org > Subject: Re: [audio] Regarding conformance testing of HE-AAC > > Ruchi Sharma wrote: > >> Hello All, >> >> >> >> Can Any one provide me Raw ER_AAC_LC bitstream ??? >> >> If any one knows , plz let me know the link. >> >> Thanks in advance. >> >> >> >> > > You may use any adif file you can find here: > > ftp://mpaudconf:adif2mp4 at ftp.iis.fraunhofer.de/mpeg2aac-conformance/compress > edAdif/ > as a basis. By simply removing the adif header you will get the raw > bitstream you need. > > Best regards, > Ralph > >> >> >> >> >> >> >> >> >> >> >> _____________________________________________________________________ >> >> This e-mail message may contain proprietary, confidential or legally >> privileged information for the sole use of the person or entity to >> whom this message was originally addressed. Any review, e-transmission >> dissemination or other use of or taking of any action in reliance upon >> this information by persons or entities other than the intended >> recipient is prohibited. If you have received this e-mail in error >> kindly delete this e-mail from your records. If it appears that this >> mail has been forwarded to you without proper authority, please notify >> us immediately at netadmin at patni.com and delete this mail. >> _____________________________________________________________________ >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> NOTE: Please use clear subject lines for your posts. Include [audio, >> > [video], [systems], [general] or another apppropriate identifier to indicate > the type of question you have. > >> Note: Conduct on the mailing list is subject to the Antitrust guidelines >> > found at > http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php > > > -- Dipl.-Ing. Ralph Sperschneider | Phone: +49 9131 776 6167 Fraunhofer IIS | Fax: +49 9131 776 6099 Am Wolfsmantel 33 | mailto:ralph.sperschneider at iis.fraunhofer.de D 91058 Erlangen | http://www.iis.fraunhofer.de/amm/ _____________________________________________________________________ This e-mail message may contain proprietary, confidential or legally privileged information for the sole use of the person or entity to whom this message was originally addressed. Any review, e-transmission dissemination or other use of or taking of any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you have received this e-mail in error kindly delete this e-mail from your records. If it appears that this mail has been forwarded to you without proper authority, please notify us immediately at netadmin at patni.com and delete this mail. _____________________________________________________________________ From ramya.desai at gmail.com Thu Jan 22 09:35:06 2009 From: ramya.desai at gmail.com (Ramya Desai) Date: Thu, 22 Jan 2009 20:05:06 +0530 Subject: [Mp4-tech] Video quality verification parameters Message-ID: <3e7aae30901220635g248a669bu8191115f9a0c01bc@mail.gmail.com> Dear Experts, I am a newbie to video and image encoding. I have been assigned to find the quality of the video that has been encoded. I am getting the decoded video as my input. Now I need to find the parameters that will define the quality of my video. The parameters that I need to calculate are.. 1)Blockiness 2)Black Frames 3)Freeze Frames 4)Blurriness (RGB) 5)Average Picture Level Can any one guide me where can I get the algorithms to calculate these parameters?. If any sample code available to do so, please let me know. Thanks in advance, Regards, RD From ralph.sperschneider at iis.fraunhofer.de Wed Jan 28 13:13:11 2009 From: ralph.sperschneider at iis.fraunhofer.de (Ralph Sperschneider) Date: Wed, 28 Jan 2009 19:13:11 +0100 Subject: [Mp4-tech] Re: [audio] Regarding conformance testing of HE-AAC In-Reply-To: <006801c97c86$09f7b3a0$ab04a8c0@patni.com> References: <006401c97609$2b83f020$ab04a8c0@patni.com> <496DCCBC.90101@iis.fraunhofer.de> <002e01c976cd$4bdd0900$ab04a8c0@patni.com> <496F1239.7080908@iis.fraunhofer.de> <006801c97c86$09f7b3a0$ab04a8c0@patni.com> Message-ID: <4980A037.5080707@iis.fraunhofer.de> An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090128/6147bbd0/attachment.html From mrison at hotmail.com Wed Jan 28 19:31:44 2009 From: mrison at hotmail.com (Mark RISON) Date: Thu, 29 Jan 2009 00:31:44 +0000 Subject: [Mp4-tech] Differences between mp41 and mp42 brands In-Reply-To: References: Message-ID: >>>>> I understand that the mp42 brand indicates conformance to ISO >>>>> 14496-14 (and therefore ISO 14496-12 on which it is based), >>>>> while the mp41 brand (or absence of an ftyp box) indicates >>>>> conformance to ISO 14496-1 section 13. >>>>> Is a comparison of these two available somewhere? That is, a >>>>> list of the substantive differences between a file which >>>>> conforms to the mp41 brand but not the mp42 brand, and >>>>> vice-versa. >>>> Does annex E in the (freely available) ISO/IEC >>>> 14496-12:2005/Amd.2:2008 help? >>> Thanks for the pointer, but this seems to cover only the isom, >>> avc1, iso2, mp71 and iso3 brands. Nothing about the mp41 and mp42 >>> brands. >> Doh! mp41 and mp42 are covered, rather briefly, in 14496-14, of >> course >> The brand 'mp41' is defined as identifying version 1 of this >> specification (ISO/IEC 14496-1:2001), and the brand 'mp42' >> identifies this version of the specification; at least one of these >> brands shall appear in the compatible-brands list in the file-type >> box, in all files conforming to this specification. > Err, right, that's where I started: is there a list of the substantive > differences between these two specifications? OK, I've analysed the relevant specs' box structures in a little more detail, and this is what I've found. Only in mp41 (Part 1 s13): - iods (this is in Part 14, of course) - more than one udta at a given level Only in mp42 (Part 12): - ftyp - stz2 - co64 - padb - sdtp, sbgp, sgpd, subs - pdin - meta and its children - mvex and its children - moof, mfra and their children - ipmc Other points of note: - Part 1 s13.2.3.17 seems to suggest that the mp4[vas] box is held immediately within the stbl atom and that there can only be one. Is this really the case, or is this an error and it's like Part 12/14, where the mp4[vas] box is held within an stsd atom within the stbl atom, and there can be more than one of them? There's additional confusion in that Part 1 *does* show a stsd box child of the stbl in Table 1. - udta is shown at top level in Table 55 in Part 1 but as a child of skip in Table 1 of Part 12 (but documented in both 13.2.3.27 and 8.27.1 respectively as being in moov or trak!). - as for Part 12, the cardinality is sometimes inconsistent with the mandatoryness; specifically ctts, stss, stsh, stdp, edts, elst are all (I assume) actually zero or one, not exactly one. Is this right? Or have I missed something? Why did Part 12 introduce a limitation on the udta cardinality? But it seems to me that this might only be part of the story. Maybe the blah box is different in Part 1 and Part 12, e.g. it has a new version, or a new restriction on allowed content or something. Has really no-one produced a (publicly-available) complete list of the differences -- e.g. the MPEG working group? Mark -- CPC/IP - A TCP/IP stack for Amstrad CPCs -- http://www.cepece.info/cpcip/ "Z88 vs CPC? Christ. How did we miss that platform war?" -- http://www.ntk.net/index.cgi?b=02000-01-28&l=111#l _________________________________________________________________ Hotmail, Messenger, Photos and more - all with the new Windows Live. Get started! http://www.download.live.com/ From singer at apple.com Thu Jan 29 09:12:30 2009 From: singer at apple.com (David Singer) Date: Thu, 29 Jan 2009 15:12:30 +0100 Subject: [Mp4-tech] Differences between mp41 and mp42 brands In-Reply-To: References: Message-ID: At 0:31 +0000 29/01/09, Mark RISON wrote: >Other points of note: > >- Part 1 s13.2.3.17 seems to suggest that the mp4[vas] box is held >immediately within the stbl atom and that there can only be one. Is >this really the case, or is this an error and it's like Part 12/14, >where the mp4[vas] box is held within an stsd atom within the stbl >atom, and there can be more than one of them? There's additional >confusion in that Part 1 *does* show a stsd box child of the stbl in >Table 1. There was no change in structure. > >- udta is shown at top level in Table 55 in Part 1 but as a child of >skip in Table 1 of Part 12 (but documented in both 13.2.3.27 and >8.27.1 respectively as being in moov or trak!). It was shown at top-level as a failed attempt to show it could be in multiple places. > >- as for Part 12, the cardinality is sometimes inconsistent with the >mandatoryness; specifically ctts, stss, stsh, stdp, edts, elst are all >(I assume) actually zero or one, not exactly one. Long been a problem, we've been cleaning it up... > >Is this right? Or have I missed something? Why did Part 12 introduce >a limitation on the udta cardinality? I do't recall, alas. > >But it seems to me that this might only be part of the story. Maybe >the blah box is different in Part 1 and Part 12, e.g. it has a new >version, or a new restriction on allowed content or something. Has >really no-one produced a (publicly-available) complete list of the >differences -- e.g. the MPEG working group? Surprisingly, no...very few people work from the original part 1 any more... > >Mark > >-- >CPC/IP - A TCP/IP stack for Amstrad CPCs >-- http://www.cepece.info/cpcip/ >"Z88 vs CPC? Christ. How did we miss that platform war?" >-- http://www.ntk.net/index.cgi?b=02000-01-28&l=111#l > > >_________________________________________________________________ > >Hotmail, Messenger, Photos and more - all with the new Windows >Live. Get started! >http://www.download.live.com/ -- David Singer Multimedia Standards, Apple Inc. From mp4-tech-owner at lists.mpegif.org Fri Jan 30 10:18:04 2009 From: mp4-tech-owner at lists.mpegif.org (mpegif list admins) Date: Fri, 30 Jan 2009 16:18:04 +0100 Subject: [Mp4-tech] List unstuck Message-ID: <002901c982ed$f317e7e0$d947b7a0$@mpegif.org> Folks, As you may have noticed and/or read on the News list, the mp4-tech list has been stuck for over two weeks. For a variety of reasons we only found out a week ago, and for other reasons it has taken until now to resolve the problem. The failure started on 14 January, and it looks like all messages sent after the list broke got lost. I will forward the messages that I received as moderator, but that doesn't cover all the messages that were sent (I don't get to see all messages for moderation). Please see if your message appears on the list, and if it doesn't, then please resend. We apologize for the inconvenience, and hope that everything is back to normal now. Kind Regards, Rob -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090130/49146369/attachment-0001.html From mp4-tech-owner at lists.mpegif.org Fri Jan 30 10:15:06 2009 From: mp4-tech-owner at lists.mpegif.org (mpegif list admins) Date: Fri, 30 Jan 2009 16:15:06 +0100 Subject: [Mp4-tech] RE: [audio] Regarding conformance testing of HE-AAC Message-ID: <002301c982ed$88d6ac40$9a8404c0$@mpegif.org> Another forward. _______________________________________ Ruchi Sharma wrote: > Hi Ralph, > > Thanks for your reply. > I am downloading the bit-stream from the ftp server you had mentioned. > Does those bit-stream are only AAC_LC and HE-AAC raw stream?? > I need ER_AAC_LC raw bitstream.....can you tell me the name of those > bitstream those are ER_AAC_LC raw stream?? > Thanks. > > Regards, > Ruchi > > > Hi Ruchi, it seems I have overlooked the ER in your first e-mail - sorry for that. To my knowledge there is no raw bitstream syntax defined for any ER AOT. The ER AAC LC sequences on the server are stored in the mp4ff container. You find them here: mpeg4audio-conformance/compressedMp4/er_al* If really needed, you could write a little program which parses those mp4 files and then dumps just the AU payloads - at least for epConfig = 0|3 those bitstreams might still be decodable as long as all relevant meta data is available. Best regards, Ralph > > > -----Original Message----- > From: Ralph Sperschneider [mailto:ralph.sperschneider at iis.fraunhofer.de ] > Sent: Wednesday, January 14, 2009 5:00 PM > To: ruchi.sharma at patni.com > Cc: mp4-tech at lists.mpegif.org > Subject: Re: [audio] Regarding conformance testing of HE-AAC > > Ruchi Sharma wrote: > >> Hello All, >> >> >> >> Can Any one provide me Raw ER_AAC_LC bitstream ??? >> >> If any one knows , plz let me know the link. >> >> Thanks in advance. >> >> >> >> > > You may use any adif file you can find here: > > ftp://mpaudconf:adif2mp4 at ftp.iis.fraunhofer.de/mpeg2aac-conformance/compress > edAdif/ > as a basis. By simply removing the adif header you will get the raw > bitstream you need. > > Best regards, > Ralph -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090130/826db7e8/attachment-0001.html From mp4-tech-owner at lists.mpegif.org Fri Jan 30 10:15:06 2009 From: mp4-tech-owner at lists.mpegif.org (mpegif list admins) Date: Fri, 30 Jan 2009 16:15:06 +0100 Subject: [Mp4-tech] FW: [audio] Regarding conformance testing of HE-AAC Message-ID: <002801c982ed$89a14f40$9ce3edc0$@mpegif.org> Another one -----Original Message----- From: Ruchi Sharma [mailto:ruchi.sharma at patni.com] Sent: Thursday, 15 January, 2009 05:54 To: 'Ralph Sperschneider' Cc: mp4-tech at lists.mpegif.org Subject: RE: [audio] Regarding conformance testing of HE-AAC Hi Ralph, Thanks for your reply. I am downloading the bit-stream from the ftp server you had mentioned. Does those bit-stream are only AAC_LC and HE-AAC raw stream?? I need ER_AAC_LC raw bitstream.....can you tell me the name of those bitstream those are ER_AAC_LC raw stream?? Thanks. Regards, Ruchi -----Original Message----- From: Ralph Sperschneider [mailto:ralph.sperschneider at iis.fraunhofer.de] Sent: Wednesday, January 14, 2009 5:00 PM To: ruchi.sharma at patni.com Cc: mp4-tech at lists.mpegif.org Subject: Re: [audio] Regarding conformance testing of HE-AAC Ruchi Sharma wrote: > > Hello All, > > > > Can Any one provide me Raw ER_AAC_LC bitstream ??? > > If any one knows , plz let me know the link. > > Thanks in advance. > > > You may use any adif file you can find here: ftp://mpaudconf:adif2mp4 at ftp.iis.fraunhofer.de/mpeg2aac-conformance/compress edAdif/ as a basis. By simply removing the adif header you will get the raw bitstream you need. Best regards, Ralph > > > > > > > > > > > > _____________________________________________________________________ > > This e-mail message may contain proprietary, confidential or legally > privileged information for the sole use of the person or entity to > whom this message was originally addressed. Any review, e-transmission > dissemination or other use of or taking of any action in reliance upon > this information by persons or entities other than the intended > recipient is prohibited. If you have received this e-mail in error > kindly delete this e-mail from your records. If it appears that this > mail has been forwarded to you without proper authority, please notify > us immediately at netadmin at patni.com and delete this mail. > _____________________________________________________________________ > ------------------------------------------------------------------------ > > _______________________________________________ > NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropriate identifier to indicate the type of question you have. > > Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php -- Dipl.-Ing. Ralph Sperschneider | Phone: +49 9131 776 6167 Fraunhofer IIS | Fax: +49 9131 776 6099 Am Wolfsmantel 33 | mailto:ralph.sperschneider at iis.fraunhofer.de D 91058 Erlangen | http://www.iis.fraunhofer.de/amm/ _____________________________________________________________________ This e-mail message may contain proprietary, confidential or legally privileged information for the sole use of the person or entity to whom this message was originally addressed. Any review, e-transmission dissemination or other use of or taking of any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you have received this e-mail in error kindly delete this e-mail from your records. If it appears that this mail has been forwarded to you without proper authority, please notify us immediately at netadmin at patni.com and delete this mail. _____________________________________________________________________ From mp4-tech-owner at lists.mpegif.org Fri Jan 30 10:18:04 2009 From: mp4-tech-owner at lists.mpegif.org (mpegif list admins) Date: Fri, 30 Jan 2009 16:18:04 +0100 Subject: [Mp4-tech] FW: X.264 doesn't support multiple slices per frame!!! Message-ID: <003301c982ed$f35d0730$da171590$@mpegif.org> forwarding From: Hadi Hadizadeh [mailto:hadi.hadizadeh at hotmail.com] Sent: Sunday, 18 January, 2009 01:10 To: mp4-tech at lists.mpegif.org Subject: X.264 doesn't support multiple slices per frame!!! Hi, Someone says that X.264 doesn't support multiple slices per frame. But is it true? I think it is strange because if there was only one slice per frame, then by dropping a NAL unit (due to channel errors), then the whole frame would be lost! But what is your opinion? Suppose that we want to deal with very high resolutions, then it would be better to use multiple slices per frame. But really how can we use different slices per frame in X.264? Thank you in advance! Best Regards, Hadi _____ _____ _____ Easily add maps and directions to your online party invites. Click to learn how. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090130/fea3a1f2/attachment-0001.html From mp4-tech-owner at lists.mpegif.org Fri Jan 30 10:18:04 2009 From: mp4-tech-owner at lists.mpegif.org (mpegif list admins) Date: Fri, 30 Jan 2009 16:18:04 +0100 Subject: [Mp4-tech] FW: JM decoder crashes with a lost NAL unit... (erc_do_p.c) Message-ID: <002e01c982ed$f33226a0$d99673e0$@mpegif.org> Forwarding From: Hadi Hadizadeh [mailto:hadi.hadizadeh at hotmail.com] Sent: Wednesday, 21 January, 2009 19:23 To: mp4-tech at lists.mpegif.org Subject: JM decoder crashes with a lost NAL unit... (erc_do_p.c) Hi, I have encoded a YUV file with an H.264 encoder (X.264). Then to simulate the channel error, I drop some of the NAL units of this encoded file (from the half of the file (frame #156) to the end of it (frame #300)). But unfortunately, when I try to decode it with JM decoder I got the following errors: > ldecod.exe decoder.cfg case #1: in decoder.cfg we set Err Concealment mode = 0 in this case I got this error on frame #156 : An unintentional loss of pictures occures! Exit case #2: in decoder.cfg we set Err Concealment mode = 1 or 2 in this case at the begining of the decoding process I got the following error: 0000(I) 0 0 10 4:2:0 140 Assertion failed: conceal_from_picture != NULL, file c:\jm\lencode\src\erc_do_p.c, line 1725 Please note that I can play this encoded file (which some of its NAL units have been dropped) with VLC player with no error. But what is the problem? Could you please help me? Thank you in advance! Best Regards, Hadi _____ -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090130/f8c0327e/attachment-0001.html From mp4-tech-owner at lists.mpegif.org Fri Jan 30 10:31:26 2009 From: mp4-tech-owner at lists.mpegif.org (mpegif list admins) Date: Fri, 30 Jan 2009 16:31:26 +0100 Subject: [Mp4-tech] FW: Differences between mp41 and mp42 brands Message-ID: <003801c982ef$d0eae760$72c0b620$@mpegif.org> Moe still -----Original Message----- From: Mark RISON [mailto:mrison at hotmail.com] Sent: Thursday, 29 January, 2009 01:32 To: singer at apple.com; mp4-tech at lists.mpegif.org Subject: RE: [Mp4-tech] Differences between mp41 and mp42 brands >> The brand 'mp41' is defined as identifying version 1 of this >> specification (ISO/IEC 14496-1:2001), and the brand 'mp42' >> identifies this version of the specification; at least one of these >> brands shall appear in the compatible-brands list in the file-type >> box, in all files conforming to this specification. > Err, right, that's where I started: is there a list of the substantive > differences between these two specifications? OK, I've analysed the relevant specs' box structures in a little more detail, and this is what I've found. Only in mp41 (Part 1 s13): - iods (this is in Part 14, of course) - more than one udta at a given level Only in mp42 (Part 12): - ftyp - stz2 - co64 - padb - sdtp, sbgp, sgpd, subs - pdin - meta and its children - mvex and its children - moof, mfra and their children - ipmc Other points of note: - Part 1 s13.2.3.17 seems to suggest that the mp4[vas] box is held immediately within the stbl atom and that there can only be one. Is this really the case, or is this an error and it's like Part 12/14, where the mp4[vas] box is held within an stsd atom within the stbl atom, and there can be more than one of them? There's additional confusion in that Part 1 *does* show a stsd box child of the stbl in Table 1. - udta is shown at top level in Table 55 in Part 1 but as a child of skip in Table 1 of Part 12 (but documented in both 13.2.3.27 and 8.27.1 respectively as being in moov or trak!). - as for Part 12, the cardinality is sometimes inconsistent with the mandatoryness; specifically ctts, stss, stsh, stdp, edts, elst are all (I assume) actually zero or one, not exactly one. Is this right? Or have I missed something? Why did Part 12 introduce a limitation on the udta cardinality? But it seems to me that this might only be part of the story. Maybe the blah box is different in Part 1 and Part 12, e.g. it has a new version, or a new restriction on allowed content or something. Has really no-one produced a (publicly-available) complete list of the differences -- e.g. the MPEG working group? Mark -- CPC/IP - A TCP/IP stack for Amstrad CPCs -- http://www.cepece.info/cpcip/ "Z88 vs CPC? Christ. How did we miss that platform war?" -- http://www.ntk.net/index.cgi?b=02000-01-28&l=111#l _________________________________________________________________ Hotmail, Messenger, Photos and more - all with the new Windows Live. Get started! http://www.download.live.com/ From mp4-tech-owner at lists.mpegif.org Fri Jan 30 10:31:26 2009 From: mp4-tech-owner at lists.mpegif.org (mpegif list admins) Date: Fri, 30 Jan 2009 16:31:26 +0100 Subject: [Mp4-tech] FW: [audio] Regarding conformance testing of HE-AAC Message-ID: <003e01c982ef$d11f64e0$735e2ea0$@mpegif.org> more -----Original Message----- From: Ruchi Sharma [mailto:ruchi.sharma at patni.com] Sent: Thursday, 22 January, 2009 12:39 To: 'Ralph Sperschneider' Cc: mp4-tech at lists.mpegif.org Subject: RE: [audio] Regarding conformance testing of HE-AAC Hello Ralph, Need some help again!! I had dumped the AU payloads for er_aac files having epflag = 1, suitabletracks = 3; I had dumped the data after appending all the layers in an AU unit. But in the ISO code, there is a function "BsPrepareEp1Stream" which required layer information also. I had dumped AAC files also after appending all the layers in an AU unit. They are decoding well. So, is the er_aac raw file with epflag greater than 1 contains the data in layer form???? Regards, Ruchi Sharma Sr. Software Engineer Product Enginnering Services Patni Computers Systems Ltd. Desk no.-43, Unit-19, SDF-7 SEEPZ, Andheri (E) Mumbai-400096 Ext. N0. -5810 Link Line No. - 601-7115 -----Original Message----- From: Ralph Sperschneider [mailto:ralph.sperschneider at iis.fraunhofer.de] Sent: Thursday, January 15, 2009 4:09 PM To: ruchi.sharma at patni.com Cc: mp4-tech at lists.mpegif.org Subject: Re: [audio] Regarding conformance testing of HE-AAC Ruchi Sharma wrote: > Hi Ralph, > > Thanks for your reply. > I am downloading the bit-stream from the ftp server you had mentioned. > Does those bit-stream are only AAC_LC and HE-AAC raw stream?? > I need ER_AAC_LC raw bitstream.....can you tell me the name of those > bitstream those are ER_AAC_LC raw stream?? > Thanks. > > Regards, > Ruchi > > > Hi Ruchi, it seems I have overlooked the ER in your first e-mail - sorry for that. To my knowledge there is no raw bitstream syntax defined for any ER AOT. The ER AAC LC sequences on the server are stored in the mp4ff container. You find them here: mpeg4audio-conformance/compressedMp4/er_al* If really needed, you could write a little program which parses those mp4 files and then dumps just the AU payloads - at least for epConfig = 0|3 those bitstreams might still be decodable as long as all relevant meta data is available. Best regards, Ralph > > > -----Original Message----- > From: Ralph Sperschneider [mailto:ralph.sperschneider at iis.fraunhofer.de] > Sent: Wednesday, January 14, 2009 5:00 PM > To: ruchi.sharma at patni.com > Cc: mp4-tech at lists.mpegif.org > Subject: Re: [audio] Regarding conformance testing of HE-AAC > > Ruchi Sharma wrote: > >> Hello All, >> >> >> >> Can Any one provide me Raw ER_AAC_LC bitstream ??? >> >> If any one knows , plz let me know the link. >> >> Thanks in advance. >> >> >> >> > > You may use any adif file you can find here: > > ftp://mpaudconf:adif2mp4 at ftp.iis.fraunhofer.de/mpeg2aac-conformance/compress > edAdif/ > as a basis. By simply removing the adif header you will get the raw > bitstream you need. > > Best regards, > Ralph > >> >> >> >> >> >> >> >> >> >> >> _____________________________________________________________________ >> >> This e-mail message may contain proprietary, confidential or legally >> privileged information for the sole use of the person or entity to >> whom this message was originally addressed. Any review, e-transmission >> dissemination or other use of or taking of any action in reliance upon >> this information by persons or entities other than the intended >> recipient is prohibited. If you have received this e-mail in error >> kindly delete this e-mail from your records. If it appears that this >> mail has been forwarded to you without proper authority, please notify >> us immediately at netadmin at patni.com and delete this mail. >> _____________________________________________________________________ >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> NOTE: Please use clear subject lines for your posts. Include [audio, >> > [video], [systems], [general] or another apppropriate identifier to indicate > the type of question you have. > >> Note: Conduct on the mailing list is subject to the Antitrust guidelines >> > found at > http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php > > > -- Dipl.-Ing. Ralph Sperschneider | Phone: +49 9131 776 6167 Fraunhofer IIS | Fax: +49 9131 776 6099 Am Wolfsmantel 33 | mailto:ralph.sperschneider at iis.fraunhofer.de D 91058 Erlangen | http://www.iis.fraunhofer.de/amm/ _____________________________________________________________________ This e-mail message may contain proprietary, confidential or legally privileged information for the sole use of the person or entity to whom this message was originally addressed. Any review, e-transmission dissemination or other use of or taking of any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you have received this e-mail in error kindly delete this e-mail from your records. If it appears that this mail has been forwarded to you without proper authority, please notify us immediately at netadmin at patni.com and delete this mail. _____________________________________________________________________ From mp4-tech-owner at lists.mpegif.org Fri Jan 30 10:31:26 2009 From: mp4-tech-owner at lists.mpegif.org (mpegif list admins) Date: Fri, 30 Jan 2009 16:31:26 +0100 Subject: [Mp4-tech] FW: AAC sample rate Message-ID: <003901c982ef$d102b520$73081f60$@mpegif.org> Yet more From: adamc7 at inter.net.il [mailto:adamc7 at inter.net.il] Sent: Tuesday, 27 January, 2009 18:47 To: mp4-tech at lists.mpegif.org Subject: AAC sample rate Dear experts, I'm new in AAC-LC standart, so my question may be basic. I wonder if the audio sample rate can be changed in the middle of the stream. >From one hand, ADTS header includes the sample rate, but from the other hand I've an analyzer telling me it's an error. Will the answer be different if the stream has SBR? Thanks, Adam -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090130/97203056/attachment.html From mrison at hotmail.com Fri Jan 30 13:28:23 2009 From: mrison at hotmail.com (Mark RISON) Date: Fri, 30 Jan 2009 18:28:23 +0000 Subject: [Mp4-tech] ADIF -- what's the correct spec? Message-ID: I'm having trouble working out what an ADIF header looks like. If I look at Annex 1.A of ISO 14496-3:2001 I find it's: adif_id[32] copyright_id_present[1] if (copyright_id_present) copyright_id[72] orig_copy[1] home[1] bitstream_type[1] bitrate[23] num_pce[4] for (i = 0; i < num_pce; ++i) { if (bitstream_type == 0) adif_buffer_fullness[20] program_config_element() } However, if I look at Annex 1.A of ISO 14496-3:2005 I find it's: adif_id[32] copyright_id_present[1] if (copyright_id_present) copyright_id[72] orig_copy[1] home[1] bitstream_type[1] bitrate[23] num_pce[4] if (bitstream_type == 0) adif_buffer_fullness[20] for (i = 0; i < num_pce; ++i) { program_config_element() } So, which is correct? Is there, if the bitstream_type is 0, one adif_buffer_fullness for each PCE as in 2001, or exactly one for the whole ADIF sequence as in 2005? Similarly there seems to be a difference in the raw data stream, after the header. In 2001: while (data()) { raw_data_block() byte_alignment() } But in 2005: byte_alignment() while (data()) { raw_data_block() } So, aligned after every raw data block, or just once before the first one? If 2005 didn't supersede 2001, or if it did but there might be some 2001-flavour files kicking around, how does one tell the two apart? Mark -- CPC/IP - A TCP/IP stack for Amstrad CPCs -- http://www.cepece.info/cpcip/ "Z88 vs CPC? Christ. How did we miss that platform war?" -- http://www.ntk.net/index.cgi?b=02000-01-28&l=111#l _________________________________________________________________ Twice the fun?Share photos while you chat with Windows Live Messenger. Learn more. http://www.microsoft.com/uk/windows/windowslive/products/messenger.aspx From alex.converse at gmail.com Fri Jan 30 16:17:12 2009 From: alex.converse at gmail.com (Alex Converse) Date: Fri, 30 Jan 2009 16:17:12 -0500 Subject: [Mp4-tech] FW: AAC sample rate In-Reply-To: <003901c982ef$d102b520$73081f60$@mpegif.org> References: <003901c982ef$d102b520$73081f60$@mpegif.org> Message-ID: On Fri, Jan 30, 2009 at 10:31 AM, mpegif list admins wrote: > Yet more > > > > From: adamc7 at inter.net.il [mailto:adamc7 at inter.net.il] > Sent: Tuesday, 27 January, 2009 18:47 > To: mp4-tech at lists.mpegif.org > Subject: AAC sample rate > > > > Dear experts, > I'm new in AAC-LC standart, so my question may be basic. > I wonder if the audio sample rate can be changed in the middle of the > stream. The sample rate is part of the ADTS "fixed header." Contents of the "fixed header" are not allowed to change from frame to frame" > >From one hand, ADTS header includes the sample rate, but from the other > hand I've an analyzer telling me it's an error. It is an error. > Will the answer be different if the stream has SBR? SBR does not allow you to change this. Regards, Alex Converse From krista.joe at gmail.com Fri Jan 30 20:39:07 2009 From: krista.joe at gmail.com (Krista Joe) Date: Fri, 30 Jan 2009 17:39:07 -0800 Subject: [Mp4-tech] Question on Maximum NAL Size Message-ID: Dear Experts, I am new to H.264 JM Software, I am working on transmitting H.264 encoded video packets over Wireless networks. I am using JM12.1 version, extended Profile with DP and FMO enabled.I use slice mode option 2 and currently give slice_argument as 400 minimum. In Slice.c while assigining the buffer size it takes, buffer_size = imax(2 * input->slice_argument, 764); I want to be able to generate even smaller packets with slice argument of 100,200,300 bytes. Is there any reason why is this 764 bytes given and can we reduce it more, because I made some changes and my Heap memory got corrupted. Any help will be very much appreciated, Thanks, Krista. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20090130/13544f76/attachment.html From mrison at hotmail.com Sat Jan 31 09:24:18 2009 From: mrison at hotmail.com (Mark RISON) Date: Sat, 31 Jan 2009 14:24:18 +0000 Subject: [Mp4-tech] Differences between mp41 and mp42 brands In-Reply-To: References: Message-ID: >> - Part 1 s13.2.3.17 seems to suggest that the mp4[vas] box is held >> immediately within the stbl atom and that there can only be one. Is >> this really the case, or is this an error and it's like Part 12/14, >> where the mp4[vas] box is held within an stsd atom within the stbl >> atom, and there can be more than one of them? There's additional >> confusion in that Part 1 *does* show a stsd box child of the stbl in >> Table 1. > There was no change in structure. Ah, OK. I certainly find s13.2.3.17 rather confusing, then (a bit like the url and urn boxes in Part 12 s8.13.1). >> - udta is shown at top level in Table 55 in Part 1 but as a child of >> skip in Table 1 of Part 12 (but documented in both 13.2.3.27 and >> 8.27.1 respectively as being in moov or trak!). > It was shown at top-level as a failed attempt to show it could be in > multiple places. Ah, OK. >> - as for Part 12, the cardinality is sometimes inconsistent with the >> mandatoryness; specifically ctts, stss, stsh, stdp, edts, elst are all >> (I assume) actually zero or one, not exactly one. > Long been a problem, we've been cleaning it up... OK. Is there a plan to maintain Part 1 s13, or is it due to be removed/marked as informative only in future editions? >> But it seems to me that this might only be part of the story. Maybe >> the blah box is different in Part 1 and Part 12, e.g. it has a new >> version, or a new restriction on allowed content or something. Has >> really no-one produced a (publicly-available) complete list of the >> differences -- e.g. the MPEG working group? > Surprisingly, no...very few people work from the original part 1 any more... I see. Am I still going to come across files conforming to mp41 but not mp42, and if so, where can I find some examples? Mark -- CPC/IP - A TCP/IP stack for Amstrad CPCs -- http://www.cepece.info/cpcip/ "Z88 vs CPC? Christ. How did we miss that platform war?" -- http://www.ntk.net/index.cgi?b=02000-01-28&l=111#l _________________________________________________________________ Check out the new and improved services from Windows Live. Learn more! http://clk.atdmt.com/UKM/go/132630768/direct/01/ From singer at apple.com Sat Jan 31 11:05:09 2009 From: singer at apple.com (David Singer) Date: Sat, 31 Jan 2009 17:05:09 +0100 Subject: [Mp4-tech] FW: AAC sample rate In-Reply-To: References: <003901c982ef$d102b520$73081f60$@mpegif.org> Message-ID: At 16:17 -0500 30/01/09, Alex Converse wrote: > > Will the answer be different if the stream has SBR? > >SBR does not allow you to change this. well, careful. if you use the euphemistically-called 'implicit signalling' for SBR (actually no signal at all that SBR might be in use), SBR frames can, I think, turn up after you start decoding, whereupon suddenly double the sample-rate output would be needed. which is why implicit signalling is, um, not always desirable. -- David Singer Multimedia Standards, Apple Inc. From alex.converse at gmail.com Sat Jan 31 11:41:00 2009 From: alex.converse at gmail.com (Alex Converse) Date: Sat, 31 Jan 2009 11:41:00 -0500 Subject: [Mp4-tech] FW: AAC sample rate In-Reply-To: References: <003901c982ef$d102b520$73081f60$@mpegif.org> Message-ID: On Sat, Jan 31, 2009 at 11:05 AM, David Singer wrote: > At 16:17 -0500 30/01/09, Alex Converse wrote: >> >> > Will the answer be different if the stream has SBR? >> >> SBR does not allow you to change this. > > well, careful. if you use the euphemistically-called 'implicit signalling' > for SBR (actually no signal at all that SBR might be in use), SBR frames > can, I think, turn up after you start decoding, whereupon suddenly double > the sample-rate output would be needed. > SBR does change the sample rate but the ADTS header still advertises the core coder (AAC) frequency. As far as implicit SBR just popping up on you in the middle of a stream, I always took Andreas's explanation here to imply that if you got through the first frame without finding it, you were safe the beauty and horror that is SBR. [...] --Alex From ralph.sperschneider at iis.fraunhofer.de Sat Jan 31 17:04:29 2009 From: ralph.sperschneider at iis.fraunhofer.de (Ralph Sperschneider) Date: Sat, 31 Jan 2009 23:04:29 +0100 Subject: [Mp4-tech] AAC sample rate In-Reply-To: References: <003901c982ef$d102b520$73081f60$@mpegif.org> Message-ID: <4984CAED.9060901@iis.fraunhofer.de> Alex Converse wrote: > On Fri, Jan 30, 2009 at 10:31 AM, mpegif list admins > wrote: >> Yet more >> >> >> >> From: adamc7 at inter.net.il [mailto:adamc7 at inter.net.il] >> Sent: Tuesday, 27 January, 2009 18:47 >> To: mp4-tech at lists.mpegif.org >> Subject: AAC sample rate >> >> >> >> Dear experts, >> I'm new in AAC-LC standart, so my question may be basic. >> I wonder if the audio sample rate can be changed in the middle of the >> stream. > > The sample rate is part of the ADTS "fixed header." Contents of the > "fixed header" are not allowed to change from frame to frame" That is correct, even for MPEG-2 AAC. In ISO/IEC 13818-4:2004 it can be read in subclause 8.5.2.4 adts_fixed_header(): "All data elements of the adts_fixed_header shall not change from frame to frame." Best regards, Ralph -- Dipl.-Ing. Ralph Sperschneider | Phone: +49 9131 776 6167 Fraunhofer IIS | Fax: +49 9131 776 6099 Am Wolfsmantel 33 | mailto:ralph.sperschneider at iis.fraunhofer.de D 91058 Erlangen | http://www.iis.fraunhofer.de/amm/