[Mjpeg-users] mpeg2enc options
Conditions: My source is a cheap CCD, not too bad, pretty good detail, excellent chroma, some mild noise when in the dark but otherwise it's clean. The camera is amateurishly handled: quiet scenes interrupted by bursts of brutal motion. It stores the image in NTSC 480i DV format. I capture it with dvgrab over IEEE1394. Goals: I want to put my movies on DVDs with the highest quality that's achievable with the mpeg2enc encoder. Also i want to obtain the highest portability and compatibility especially w.r.t. standalone players. When i can choose between two settings that make no difference quality-wise, i want the one which provides faster encoding. I don't care at all about the compression factor or the resulting file size (i never make DVDs bigger than 60 minutes). Now the questions: How often is actually needed -s? What do i loose if i don't use it? Any side-effects of -c? Is it a good choice to use -g 6 -G 18? Any possible side-effects of doing that? In general, when would you recommend to tweak -g -G? Did anyone actually encounter a set-top DVD player that gets upset by -R 0? Any chances for that to happen in the real world? What are the default values of -4 and -2? (the equivalent of not using them at all) -E -20 corresponds to what kind of quality? VHS-like? Broadcast-like? What's the _visual_ difference (the result) between positive and negative values? When would you recommend either side? What -E value would you recommend to start experimenting with in my case? Could you please comment on this combination of parameters? Anything that doesn't make sense? Any contradictory parameters? -b 9800 -E -15 -q 6 -R 0 -s -c -I 1 -g 6 -G 18 -4 2 -2 1 -K tmpgenc -N 1 Whenever that makes sense (such as what are default values, etc.), please add the answers to these questions (if any) to the mpeg2enc man page. Thank you. -- Florin Andrei http://florin.myip.org/ --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: Malloc failed, Was: Re: [Mjpeg-users] MPEG2 encoding performance
Hi, On Monday 24 November 2003 18.16, you wrote: > On Mon, 24 Nov 2003, Michael Hanke wrote: > > Thanks a lot for all of your help. I had to upgrade libtool and autoconf. > > Then the build went smoothly. But when I tried yuvdenoise, I got an > > memory > In utils/cpu_accel.c you should see the function bufalloc() which > is where the malloc error is coming from. The only thing I can > think of to try is add a "fprintf(stderr, "size = %d\n", size);" > at the beginning to see if a bogus request size is being passed > in. I did it. The request size is moderate: 450560. So need to have a closer look at the problem. For tracing down memory allocations, is it possible to link against dmalloc or ElectricFence? Will this have an effect on posix_memalign? Another possibility: With reference to http://sources.redhat.com/ml/bug-glibc/2002-09/msg00037.html, http://sources.redhat.com/ml/bug-glibc/2002-06/msg00202.html maybe there is a bug in glibc?? Did you use posix_memalign in version 1.6.1? > > > [EMAIL PROTECTED]:/usr/local/xcdroast/MJPEG > lav2yuv final01.eli |yuvdenoise > > | y4mscaler -I active=692x564+8+4 -S option=cubicCR -O preset=svcd | > > yuvplay > > The vertical dimension/offset really must be a multiple of 4 for > interlaced material (multiple of 2 per field but there are 2 > fields). All input dimensions are a multiple of 4! The cause of the problem is probably the reallocation in the output stream. Here, I left all decisions to y4mscaler. Michael --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: [Mjpeg-users] mpeg2enc options
On 25 Nov 2003, Florin Andrei wrote: > Conditions: > My source is a cheap CCD, not too bad, pretty good detail, excellent > chroma, some mild noise when in the dark but otherwise it's clean. > > Goals: > I want to put my movies on DVDs with the highest quality that's > achievable with the mpeg2enc encoder. > Also i want to obtain the highest portability and compatibility > especially w.r.t. standalone players. > > Now the questions: > > How often is actually needed -s? What do i loose if i don't use it? DVDs require it - "-f 8" sets it automatically for you. > Any side-effects of -c? Minimal increase in filesize (few tenths of a percent) due to very slight decrease in compression efficiency. For editing purposes -c is very useful (and _required_ for multi angle DVDs). > Is it a good choice to use -g 6 -G 18? > Any possible side-effects of doing that? > In general, when would you recommend to tweak -g -G? High motion videos might benefit from '-g' being decreased - in fact you could set that to 1 (if you're using the cvs version of mpeg2enc or at least the current release candidate). Lowering -g will increase the size of the file because more I frames will be generated. -G is set correctly for the standards - don't raise it unless you're only playing back on a computer (or a *very* forgiving settop box). > Did anyone actually encounter a set-top DVD player that gets upset by -R > 0? Nope - and you won't. B frames are an additional feature of the higher levels of the MPEG standards - they're not required. > What are the default values of -4 and -2? (the equivalent of not using > them at all) I use -4 2 -2 1. Dropping -4 from 2 to 1 increases the encoding time by a lot (30 - 40%) for a nominal (couple percent at best) decrease in average bitrate. > -E -20 corresponds to what kind of quality? VHS-like? Broadcast-like? It's not that simple... But you knew that ;) -20 is moderately aggressive, and I've not tried that setting, but it most definitely would not degrade things to VHS levels. Since you're not overly concerned about bitrate/filesize I'd suggest a modest -E setting. > What's the _visual_ difference (the result) between positive and > negative values? When would you recommend either side? Don't know that - I just started with - values and left it there ;) > What -E value would you recommend to start experimenting with in my > case? -10 is a good non-aggressive/mild setting (i've seen good results with -5 to -8). > Could you please comment on this combination of parameters? Anything > that doesn't make sense? Any contradictory parameters? > > -b 9800 -E -15 -q 6 -R 0 -s -c -I 1 -g 6 -G 18 -4 2 -2 1 -K tmpgenc -N 1 9800 is too high. Some standalone players can not handle the maximum. For portability you'll really want to back off to something like 8500. Skip the '-s' - it's redundant. The "-I 1" is the default so there's no need for that but I can understand having explicit reminders in the command line. Just take the default for -G (mpeg2enc knows the NTSC and PAL limits for the max GOP size so why second guess it?) and leave that out. The minimum is up to you For maximum quality use just I frames "-g 1 -G 1" :-)The filesize will be large but the encoding will be very very fast. Steven Schultz --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: [Mjpeg-users] mpeg2enc options
On Wed, 2003-11-26 at 00:15, Steven M. Schultz wrote: > On 25 Nov 2003, Florin Andrei wrote: > > Is it a good choice to use -g 6 -G 18? > > Any possible side-effects of doing that? > > In general, when would you recommend to tweak -g -G? > > High motion videos might benefit from '-g' being decreased - in fact > you could set that to 1 (if you're using the cvs version of mpeg2enc > or at least the current release candidate). Issues with set-top-boxes in that case? (sorry if asking silly questions) > > What are the default values of -4 and -2? (the equivalent of not using > > them at all) > > I use -4 2 -2 1. Dropping -4 from 2 to 1 increases the encoding > time by a lot (30 - 40%) for a nominal (couple percent at best) > decrease in average bitrate. ...and no visible quality gain, right? > > -E -20 corresponds to what kind of quality? VHS-like? Broadcast-like? > > It's not that simple... But you knew that ;) yes :-( I thought i should try anyway. > > What -E value would you recommend to start experimenting with in my > > case? > > -10 is a good non-aggressive/mild setting (i've seen good results > with -5 to -8). What's the default? (the equivalent if i don't use -E at all) > > -b 9800 -E -15 -q 6 -R 0 -s -c -I 1 -g 6 -G 18 -4 2 -2 1 -K tmpgenc -N 1 > > 9800 is too high. Some standalone players can not handle the maximum. > For portability you'll really want to back off to something like 8500. But most of the professional DVDs i've seen are at or around 9800 kb/s -- Florin Andrei http://florin.myip.org/ --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: [Mjpeg-users] mjpegtools-1.6.2 release candidate 2
Hi, On Wed, 2003-11-26 at 05:22, Steven M. Schultz wrote: > Rather than have version numbers go backwards which would be > confusing (is version 1.6.x newer or older than 2.2.2?) I > changed it so that the version is 2.2.3 in both places and the > date changed to be the GMT time the version number was incremented. Isn't it easier to change everything to use the VERSION macro? That way, everything is incremented automatically. Ronald -- Ronald Bultje <[EMAIL PROTECTED]> Linux Video/Multimedia developer --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: [Mjpeg-users] mjpegtools-1.6.2 release candidate 2
On Tue, 2003-11-25 at 08:28, Ronald Bultje wrote: > > As of today, release candidate 2 of mjpegtools-1.6.2 (official version: > 1.6.1.91) is available on your local sourceforge mirror. For URLs, see mplex -h does not work according to manual. Instead of activating the --keep-hf, it prints the usage help. :-) -- Florin Andrei http://florin.myip.org/ --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: [Mjpeg-users] mjpegtools-1.6.2 release candidate 2
On Wed, 2003-11-26 at 01:52, Florin Andrei wrote: > On Tue, 2003-11-25 at 08:28, Ronald Bultje wrote: > > > > As of today, release candidate 2 of mjpegtools-1.6.2 (official version: > > 1.6.1.91) is available on your local sourceforge mirror. For URLs, see > > mplex -h does not work according to manual. Instead of activating the > --keep-hf, it prints the usage help. :-) Ah, ok, it should really be -H, and it seems to work fine, but in the manpage is specified as -h by mistake. -- Florin Andrei http://florin.myip.org/ --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
[Mjpeg-users] -H and -E
I'm not sure i understand how -E works, so... Is there any conflict if i use -H and -E simultaneously? If not, does it make sense from a logical p.o.v. to use them both? Does -H override -K, or the other way 'round? -- Florin Andrei http://florin.myip.org/ --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: [Mjpeg-users] Better MPEG decoder for WMP
Bernhard - thanks for the tip. Unfortunately, the application this is for is embedded within a Macromedia Director movie, and VLC doesn't support ActiveX (at least as far as I could tell), although this might be a good open-source contribution for my company to make. Thanks! Jon On Wed, 26 Nov 2003, Bernhard Praschinger wrote: > Hallo > > > Does anyone know of a better MPEG decoder for Windows Media Player - maybe > > one that will play 1024x768 videos? The Windows one won't work - it seems > > like DVD players install a better one, but I was looking for something I > > could redistribute with the video. Any ideas? > If you just want a mpeg decoder you should take a look at VLC > (http://www.videolan.org/) it exists for many OS. > > The Windows Media Player does only support MPEG1 videos. I think I read > a article where they described how to install a seperate packege of the > windows install CD, and than you have a very basic DVD player. But that > is much mor complicated than downloading and sterting VLC. > The typical DVD Player SW in Windows is able to play back MPEG1/2. > > BTW: You do not have to install (just copy them to a location) VLC on a > Windows system, so you still can play back videos with VLC. That > important if the sysadmin has removed the right to install SW on you PC. > ;) > > auf hoffentlich bald, > > Berni the Chaos of Woodquarter > > Email: [EMAIL PROTECTED] > www: http://www.lysator.liu.se/~gz/bernhard > --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: [Mjpeg-users] mpeg2enc options
On Tue, 2003-11-25 at 23:35, Florin Andrei wrote: > Whenever that makes sense (such as what are default values, etc.), > please add the answers to these questions (if any) to the mpeg2enc man > page. Thank you. Actually, it would be great if the man page would specify all default values (wherever there is a default). -- Florin Andrei http://florin.myip.org/ --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: [Mjpeg-users] Making mpeg stills
It seems to be down right now, but when it's up you should check out http://www.vcdimager.org/stillencode/ It will encode stills for you AND give you the commands it used to do so. Basically, you combine Xtopnm, pnmscale, ppmntsc, ppmtoy4m, and mpeg2enc into a pipe. Do it for both the large and small versions, then mplex them. Jon On Wed, 26 Nov 2003, Bernhard Praschinger wrote: > Hallo > > > There is a Windows tool for VCD authoring (www.vcdeasy.org) that has a nice tab > > for making mpg stills (for use on VCD menus/segment items) from JPG/PNG files. It > > is supposed to use regular Linux tools like netpgm, mjpeg, etc. > > > > It handles all the issues Michael mentions in > > http://www.vcdimager.org/guides/mpeg_still_images.html i.e. aspect ratio and > > overscan/cropping. > > > > Can someone please give me a commandline that does this? i.e. convert a JPG or PNG > > file to a VCD2.0 MPG1 still with 4:3 aspect ratio for display on a NTSC TV? > > In Unix you usally use that guide: http://www.mir.com/DMG/stills.html > And there you have all the tools you need in the mjpeg package. > (mjpeg.sf.net) > > To convert images to differnet formart use convert from ImageMagic > Package (http://www.imagemagick.org/). > > > auf hoffentlich bald, > > Berni the Chaos of Woodquarter > > Email: [EMAIL PROTECTED] > www: http://www.lysator.liu.se/~gz/bernhard > > > --- > This SF.net email is sponsored by: SF.net Giveback Program. > Does SourceForge.net help you be more productive? Does it > help you create better code? SHARE THE LOVE, and help us help > YOU! Click Here: http://sourceforge.net/donate/ > ___ > Mjpeg-users mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/mjpeg-users > --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: [Mjpeg-users] mpeg2enc options
Am Mittwoch, 26. November 2003 09:36 schrieb Florin Andrei: > On Wed, 2003-11-26 at 00:15, Steven M. Schultz wrote: > > > -b 9800 -E -15 -q 6 -R 0 -s -c -I 1 -g 6 -G 18 -4 2 -2 1 -K > > > tmpgenc -N 1 > > > > 9800 is too high. Some standalone players can not handle the > > maximum. For portability you'll really want to back off to > > something like 8500. > > But most of the professional DVDs i've seen are at or around 9800 > kb/s AFAIK 9800 is max overall bitrate => video+audio thus a max video bitrate of 8500 is safer. After all i don't think you have material that reaches that bitrate anyway. normal bitrates are 3500-5000 with some peaks sometimes. Steffen --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: [Mjpeg-users] mjpegtools-1.6.2 release candidate 2
On 26 Nov 2003, Florin Andrei wrote: > On Tue, 2003-11-25 at 08:28, Ronald Bultje wrote: > > > > As of today, release candidate 2 of mjpegtools-1.6.2 (official version: > > 1.6.1.91) is available on your local sourceforge mirror. For URLs, see > > mplex -h does not work according to manual. Instead of activating the > --keep-hf, it prints the usage help. :-) mplex has no --keep-hf option. mpeg2enc has a -H option which is the same as --keep-hf That was changed some time ago but apparently the manpage was overlooked. I normally just use the -h or usage() summary anyhow ;) Cheers, Steven Schultz --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: Malloc failed, Was: Re: [Mjpeg-users] MPEG2 encoding performance
On Wed, 26 Nov 2003, Michael Hanke wrote: Hi! > > In utils/cpu_accel.c you should see the function bufalloc() which > > is where the malloc error is coming from. The only thing I can > > think of to try is add a "fprintf(stderr, "size = %d\n", size);" > I did it. The request size is moderate: 450560. So need to have a closer look Quite reasonable size so there must be something else causing malloc() to return NULL. > at the problem. For tracing down memory allocations, is it possible to link > against dmalloc or ElectricFence? Will this have an effect on posix_memalign? It should be possible. The easiest way to do that I think is edit the generated Makefile to add the necessary libraries. > Another possibility: With reference to > http://sources.redhat.com/ml/bug-glibc/2002-09/msg00037.html, > http://sources.redhat.com/ml/bug-glibc/2002-06/msg00202.html > maybe there is a bug in glibc?? > Did you use posix_memalign in version 1.6.1? Use of posix_memalign is a recent addition to mjpegtools and was not used in earlier versions of mjpegtools. Ah, it could well be a glibc bug. What happens if you change posix_memalign to simply be malloc()? If that works then that would be evidence of a glibc bug. > > > | y4mscaler -I active=692x564+8+4 -S option=cubicCR -O preset=svcd | > > > All input dimensions are a multiple of 4! The cause of the problem is probably And so they are - my mentall 'bc' had a fault that night ;) Cheers, Steven Schultz --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: [Mjpeg-users] mpeg2enc options
On 26 Nov 2003, Florin Andrei wrote: > > High motion videos might benefit from '-g' being decreased - in fact > > you could set that to 1 (if you're using the cvs version of mpeg2enc > > Issues with set-top-boxes in that case? None. Smaller GOP sizes just increase the filesize but allow the encoder to handle scene changes better. Set -g to 1 for maximum flexibility. I have a fairly complete set of test results showing the effect of '-g 1 -G N' where N was varied from 1 to 15. The effect of longer GOP sizes starts dropping a lot at around 10 or so. > > I use -4 2 -2 1. Dropping -4 from 2 to 1 increases the encoding > > time by a lot (30 - 40%) for a nominal (couple percent at best) > > ...and no visible quality gain, right? I suppose if you have golden eyeballs you might see (or imagine you see) a difference ;) > > -10 is a good non-aggressive/mild setting (i've seen good results > > with -5 to -8). > > What's the default? (the equivalent if i don't use -E at all) 0. Meaning no effect - the single coefficient logic is not invoked at all. > > 9800 is too high. Some standalone players can not handle the maximum. > > For portability you'll really want to back off to something like 8500. > > But most of the professional DVDs i've seen are at or around 9800 kb/s Not really. Or rather how are you determining that? By taking the .m2v streamsize in bytes and dividing by the playing time? If you do that you'll find it's a rare DVD that goes over 7500. I've a couple very high quality DVDs (they used a dual layer disc for a 140 minute concert) that averaged around 6000-6500 (but they did include 3 audio tracks - PCM, AC3 5.1, and DTS). It's necessary to leave room for peaks or spikes in the bitrate - exceeding the legal limit even briefly will cause playback problems Cheers, Steven Schultz --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: [Mjpeg-users] -H and -E
On 26 Nov 2003, Florin Andrei wrote: > Is there any conflict if i use -H and -E simultaneously? No conflict at all. -E processing is done after the quantization tables have been used. > If not, does it make sense from a logical p.o.v. to use them both? Makes a lot of sense. -E goes thru the output of the quantization process to make it more efficient to encode. > Does -H override -K, or the other way 'round? Well, yes. "-K hi-res" is the same as "-H" though so I'm not sure why you'd want to use both :) You can, if you wish, create your own custom matrices and use them with "-K file=path_to_your_file". Thus you can take the intra (I frame) table from the hi-resolution side and combine that with the non-Intra (B/P frame) table of TMPGEnc. I've attached the custom file where I did just that. The I frames end up being bigger (but there are only on average 2 of those per second) but the P/B frames become smaller. Overall the image looked very good (but not having done double blind testing against a default encoding that's not saying a lot ;)). Cheers, Steven Schultz # High resolution INTRA table 8,16,18,20,24,25,26,30 16,16,20,23,25,26,30,30 18,20,22,24,26,28,29,31 20,21,23,24,26,28,31,31 21,23,24,25,28,30,30,33 23,24,25,28,30,30,33,36 24,25,26,29,29,31,34,38 25,26,28,29,31,34,38,42 # TMPEGEnc NON-INTRA table 16,17,18,19,20,21,22,23 17,18,19,20,21,22,23,24 18,19,20,21,22,23,24,25 19,20,21,22,23,24,26,27 20,21,22,23,25,26,27,28 21,22,23,24,26,27,28,30 22,23,24,26,27,28,30,31 23,24,25,27,28,30,31,33
Re: [Mjpeg-users] mpeg2enc options
On Wed, 2003-11-26 at 08:43, Steven M. Schultz wrote: > On 26 Nov 2003, Florin Andrei wrote: > > But most of the professional DVDs i've seen are at or around 9800 kb/s > > Not really. Or rather how are you determining that? By taking > the .m2v streamsize in bytes and dividing by the playing time? If No, because that would only compute the average. And the average seems to be, for DVDs in that class, between 3000-6000 I normally use 2 different methods: 1. Extract one title with tccat then probe it with tcprobe: tccat -i /dev/dvd -t dvd -P 1 > title1.vob tcprobe -H 100 -i title1.vob 2. extract .m2v and .ac3 then mux them with mplex and watch the messages Both methods provide consistent results across a large number of tests: the max m2v bitrate is always 9800 kbps with commercial DVDs. I actually thought the total max mandated by the DVD standard is 10.08 Mbps or something like that (and it's the sum of the actively played tracks, not of all the stuff in the VOB). So that would leave, above a 9800 kbps ceiling, some room for a 448 kbps audio track and still some more room to spare (subtitles). Is my information wrong? Where the heck can we find some better and more trustworthy information about the DVD standard? (no, i don't wanna pay money) > It's necessary to leave room for peaks or spikes in the bitrate - > exceeding the legal limit even briefly will cause playback problems Right, but as far as m2v and mpeg2enc, _isolated_ (disregarding DVD limits, other tracks, etc.), are concerned, -b sets a HARD limit, never to be exceeded, is that correct? I guess what i'm asking is, if you get all the numbers right, you should be safe on the m2v side, isn't it true? -- Florin Andrei http://florin.myip.org/ --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: [Mjpeg-users] mjpegtools-1.6.2 release candidate 2
On Wed, 2003-11-26 at 08:11, Steven M. Schultz wrote: > On 26 Nov 2003, Florin Andrei wrote: > > mplex -h does not work according to manual. Instead of activating the > > --keep-hf, it prints the usage help. :-) > > mplex has no --keep-hf option. > mpeg2enc has a -H option which is the same as --keep-hf I'm sorry, you're right, i was talking about mpeg2enc. The man page has -h where it should be, i think, -H. Well, it was 2:00 AM... :-/ -- Florin Andrei http://florin.myip.org/ --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: [Mjpeg-users] mpeg2enc options
On Wed, Nov 26, 2003 at 11:21:37AM -0800, Florin Andrei wrote: > > I actually thought the total max mandated by the DVD standard is > 10.08 Mbps or something like that (and it's the sum of the actively > played tracks, not of all the stuff in the VOB). So that would > leave, above a 9800 kbps ceiling, some room for a 448 kbps audio > track and still some more room to spare (subtitles). > Is my information wrong? I've seen that same 10.08Mbps maximum upper limit for DVD's as well. My understanding was that to properly comply with the DVD consortium's rules, a player had be able to sustain that level of read speed from the DVD platter no matter what. In any case, I've been encoding mpeg2 files for DVD mastering using mpeg2enc and a -b value of 9576 for about a year to year and a half now, and so far my very small set of various hardware players have not shown any problems from files encoded with that bit rate ceiling. Your mileage may vary of course, but all I can say is that value works for me. --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: [Mjpeg-users] feature request
Hi Andrei, > If i understand correctly, one of the reasons why mplex cannot multiplex > subtitles was the scarcity of tools to generate subtitle streams with > timestamp information. Thanks for the info - adding subtitle muxing itself once you can get the timing info is pretty easy. It is quite possible that one result of recent discovery that the Author of the Windows tool IFOedit has embedded my 'mplex' in his code that the extension made there will flow back into the main mplex source tree. Even if that doesn't happen I'll probably add it pretty soon. The 'formats' required are trivial: simply some kind of header coding the PTS of each subtitle appearing before the subtitle itself. Andrew --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: [Mjpeg-users] mpeg2enc options
On Wed, 26 Nov 2003, Richard Ellis wrote: > I've seen that same 10.08Mbps maximum upper limit for DVD's as well. I haven't seen any that come even close to the upper limit. Perhaps one of the "Superbits" titles would be worth a look, they leave out all the extras (additional audio tracks, special features, etc) to make room for video bits. > My understanding was that to properly comply with the DVD > consortium's rules, a player had be able to sustain that level of > read speed from the DVD platter no matter what. And of course all manufacturers adhere completely to the standards when they're trying to churn out $49 DVD players :-) Apple's documentatin for DVD Studio Pro has repeated cautions about setting the rate too high. In practice even with a rate limit of 8500 guess what? The DVDs usually end up around 6000 to 6400 - much depends on the material, the options used (turning off B frames, eliminating single coefficients, and so on all work to lower the bit rate to well under the maximum). Cheers, Steven Schultz --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
Re: [Mjpeg-users] mpeg2enc options
On 26 Nov 2003, Florin Andrei wrote: > > It's necessary to leave room for peaks or spikes in the bitrate - > > exceeding the legal limit even briefly will cause playback problems > > Right, but as far as m2v and mpeg2enc, _isolated_ (disregarding DVD > limits, other tracks, etc.), are concerned, -b sets a HARD limit, never > to be exceeded, is that correct? No. It's a very strong hint but you will (I have) see (very) brief spikes above that. It's not a brickwall limit... When creating SVCDs for example the limit's 2500 but I will see peaks (very short, infrequent) a few percent higher - 2600 or so. > I guess what i'm asking is, if you get all the numbers right, you should > be safe on the m2v side, isn't it true? I prefer to be cautious/safe and allow for a few percent overshoot on the peaks. Besides which employing a couple of the things mentioned earlier (-E -8, -R 0) will lower the rate enough that specifying a max of 9800 is meaningless because it won't get that high ;) Cheers, Steven Schultz --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
