Page 18 of 44 FirstFirst ... 8161718192028 ... LastLast
Results 256 to 270 of 654

Thread: tyremux a GPLed TyStream demuxer and to be remuxer

  1. #256
    Join Date
    Jan 2003
    Posts
    32
    Yes, that's correct. Your DVD would not be compliant at 480x480. However, you may not need to reencode. You can use software called DVDPatcher to 'fool' your player into thinking the video is compliant. The only prob is not all players are able to accept this. Go to vcdhelp.com to see a list of players that accept this type of video.

  2. #257
    Join Date
    Nov 2002
    Posts
    17
    thought DTiVos only did 480x480 and DVDs have to be 720x480 or 352x480.
    He's using a SA tho....

  3. #258
    Join Date
    Aug 2002
    Posts
    319
    I'm quite sure this is not a bug in tydemux triggering a bug in TMPG. Beside it can actually be TMPG tiggering a TMPG bug.

    I also very sure that the mpeg that is come out of tydemux is very valid. The mpeg video stream is more or less trippel checked by tydemux.

    Cheers Olaf

    Originally posted by racingclub
    Hi All,

    I've now tested a few more UK streams - and they seem to cause TMPG to fail (crash).

    I'm trying to re-encode the .m2v and .m2a as SVCD's using the simple TMPG project wizard with no 'frills'. I seem to be able to encode small sections with no problems.

    I can use the TMPG MPEG tools to mux the 2 files together without any problems - but when I run the resulting .mpg through the TMPG SVCD wizard I get a crash in the same place.

    I've also tried authoring a DVD with IfoEdit then running the VOBs through DVD2SVCD (which uses CCE) - this works fine and gives me a 'good' SVCD with proper sync.

    Is anyone else (maybe just in the UK) trying SVCD creation?

  4. #259
    Join Date
    Jan 2003
    Posts
    25
    I doubt its a tydemux bug, but is it possible its a TIVO encoder bug, (given that the TIVO encoder was put together before MPEG2 spec was complete?) and that although the mpeg structure data, of GOPs etc is valid, that the mpeg data within the frames in the GOPs may be what is causing the problem.

    I assume this as muxing/demuxing isnt going to care too much about the content of each bit of mpeg data, it seems pretty much the payload is just transferred as-is from one stream to the next. But if the payload itself contained errors, then when tmpgenc came to decode it, it falls over in a heap.
    Other players [powerdvd/dvd2avi/set-top] might not care about the correctness of the data.

    Ive just completed my first frameserve approach with output split using tydemux, fed thru DVD2AVI, loading the d2v from DVD2AVI and the m2a from tydemux. This produced a good SVCD mpeg [as others indicated it would].

    Both using the frameserving approach, and using the MPEG2 plugin for tmpgenc from vcdhelp.com allow tmpgenc to process the data without invoking its own mpeg2 decoding routines. So it does look like its got a problem decoding the tivo data. The fault could lie either with the data, or the decoder..

    To find out which, is there any way to 'validate' the payload data within an mpeg stream? [Im assuming here that the frames that are deltas are some form of modification instruction for the preceeding frames, and that this modification could be invalid. (I dimly recall hearing something in the past of something about motion vectors initating/terminating beyond the edge of the frame causing problems on some other project, I cant recall which tho)]

    Cheers,
    Nugget

  5. #260
    Join Date
    Jan 2003
    Location
    Hampshire, UK, Timezone: GMT
    Posts
    158
    Originally posted by Nugget
    To find out which, is there any way to 'validate' the payload data within an mpeg stream? [Im assuming here that the frames that are deltas are some form of modification instruction for the preceeding frames, and that this modification could be invalid. (I dimly recall hearing something in the past of something about motion vectors initating/terminating beyond the edge of the frame causing problems on some other project, I cant recall which tho)]

    Cheers,
    Nugget
    In the last couple of days I saw a download link to an official MPEG stream validation program released by the MPEG group. I'm sorry but I cannot remember where I saw it but that may be what you are looking for.

  6. #261
    Join Date
    Aug 2002
    Posts
    319

    Re: My first DVD that is in sync

    Originally posted by MrBassMan
    Olaf: There have been some requests here to accept multiple .ty input files. I think multiple output files would also be useful to make editing easier. i.e. Output until the video file reaches roughly 1GB, close the files and create new ones with _2 appended to the filenames etc.
    Yes, that one would be nice to implement - it would be instyle with what e.g. mkisofs uses when producing dvd images on systems not supporting 2GB and bigger files.

    There is also no reason to why we should not be able to make DVD's under a OS that doesn't support large files. Since the maximum size of a vob file is just 1024MB.


    Have you detected any major changes in streams when the adverts start? If so, this could also be a cue to start new files.
    There are a few but that is mostly on DirecTV mpeg streams, nothing solid in SA/UK streams. It's a very interesting subject although and something that is definitly worth investigating. However my main point in adv detection is that has to be done in the compressed space - decompression of the mpeg stream in order to detect adv is a big no no in my view. Why? It simply takes to much time to process even on a fast computer.

    Cheers Olaf

  7. #262
    Join Date
    Aug 2002
    Posts
    319
    That one is intersting, I didn't check what audio it was when I did tested UK Tystreams.

    Cheers Olaf

    Originally posted by MrBassMan
    Opening the IFO file produced by IFOEdit shows a line that says:

    Audio 1: English (Mpeg-1 ) 2ch 48Kbs 16bps (ID: 0xc0)

    So the answer would APPEAR to be yes it is at 48KHz

  8. #263
    Join Date
    Jan 2003
    Posts
    25
    My UK SA Tivo streams have 32khz/192kbps audio, at least the ones I've checked have. Didn't think it was known how to tell Tivo to change it to another value. However, the SVCD transcoded output from tmpgenc had 44.1khz/224kpbs output.

    I just tried using MPEG2VCR's 'GOP Fix' feature on a tydemux'd m2v, and then passing the 'fixed' result to tmpgenc. Still crashes. Think this points nicely to whatever the problem is, not being tydemux.

    You want the real fun test? Feed an mpeg2 CREATED by tmpgenc (I used the output from my frameserved approach to create an svcd mpeg) back to tmpgenc and ask it to transcode it. You would expect the mpeg2 it created would be valid. Still crashes.

    I guess tmpgenc's builtin mpeg2 decoder has some 'issues'.

    Guess I'm stuck with the frameserving approach. In which case, anyone know if DVD2AVI supports commandline options to create the d2v file? I found some pages mentioning /preview and /saveavi but no obvious docs on /makemead2vfilefromthis ;-)

    Cheers
    Nugget

  9. #264
    Join Date
    Jan 2003
    Posts
    25
    Swiftly answering my own question.

    A commandline variant of DVD2AVI can be found over at DVD2SVCD Project Homepage
    Where its available on the bottom left of the page, complete with source.

    Can use that version within a batchfile with tydemux to create the d2v to feed to tmpgenc.

    I read somewhere tmpgenc can be started with an argument to make it process a batch list, which you can generate beforehand. If I can get this working then can have a windows batch file to convert .ty to svcd / vcd compatible mpeg streams.

    I worry about audio offsets though ? maybe its better to let besweet convert the audio, tmpgenc just convert video, and then use mplex to mux with the offset. Anyone tried this?

    Cheers,
    Nugget

  10. #265
    Join Date
    Aug 2002
    Posts
    319
    Part 5 of ISO-13818 standard includes the software simulation of ISO-13818 part 1, 2 and 3 (system, video and audio). I have attached it in my post.

    However note that some of this is outdated and newer versions of the software (parts) is avalible at www.mpeg.org under mssg.

    Cheers Olaf

    Originally posted by MrBassMan
    In the last couple of days I saw a download link to an official MPEG stream validation program released by the MPEG group. I'm sorry but I cannot remember where I saw it but that may be what you are looking for.

  11. #266
    Join Date
    Jan 2002
    Posts
    63

    Re: Re: My first DVD that is in sync

    Originally posted by olaf_sc
    [re: allow multiple input files] Yes, that one would be nice to implement - it would be instyle with what e.g. mkisofs uses when producing dvd images on systems not supporting 2GB and bigger files.
    i was going to wait and look at the code to see if this is feasible but since you're considering playing around with the input routines, another option would be to allow cutoff points as arguments for the input streams in much the same way that bassman's tychopper does now. there is less total output and more importantly, we can ditch any of the stream we don't want right off the bat, rather than having to keep all the stream and cut it out with editing tools down the road.

    problem i see with this is that i'm not sure how easy it is to detect the chunk:time ratio with enough accuracy to be able to take time offsets (from begining of first input ty) and cut up accordingly. obviously wouldn't need to be exact, but being able to cut with +/- 1 minute accuracy would be necessary to make it worthwhile. obviously taking chunks as a parameter would be easy but we don't really have the chunk handy while we're watching our recordings at hit a 15-minute commercial marathon

    ps- i chopped that stream with +/- 50 chunks around the bad spot and ran the 20MB file through tydemux040w32 again... same result. filename is tungsten2k01.ty - have fun with it

  12. #267
    Join Date
    Aug 2002
    Posts
    319

    Mpeg analyzer

    Hello Folks

    Found a nice little app "Moonlight MPEG Stream Explorer" it's a demo but it worked fine for me on a small random mpeg file I had at my disposal at work.

    Use this url to download it http://www.elecard.com/ftp/pub/mpeg/...MSExplorer.zip.

    Cheers Olaf

    PS: Yes it's not much of a analyzer but it lets your step trough the stream in both decoder and display order and it shows all frames etc...

  13. #268
    Join Date
    Jan 2003
    Location
    Hampshire, UK, Timezone: GMT
    Posts
    158

    Re: Re: Re: My first DVD that is in sync

    Originally posted by tungsten2k
    i was going to wait and look at the code to see if this is feasible but since you're considering playing around with the input routines, another option would be to allow cutoff points as arguments for the input streams in much the same way that bassman's tychopper does now. there is less total output and more importantly, we can ditch any of the stream we don't want right off the bat, rather than having to keep all the stream and cut it out with editing tools down the road.

    problem i see with this is that i'm not sure how easy it is to detect the chunk:time ratio with enough accuracy to be able to take time offsets (from begining of first input ty) and cut up accordingly. obviously wouldn't need to be exact, but being able to cut with +/- 1 minute accuracy would be necessary to make it worthwhile. obviously taking chunks as a parameter would be easy but we don't really have the chunk handy while we're watching our recordings at hit a 15-minute commercial marathon
    In theory you can do this now with tychopper. If you can work out how long a chunk plays (prabably very variable due to compression) you could calculate the chunks to chop out and keep. It's more like using a blunt axe than a surgeons scaple but it would sort of do what you want.
    Last edited by MrBassMan; 01-15-2003 at 05:38 PM.

  14. #269
    Join Date
    Aug 2002
    Posts
    319

    Re: Re: Re: Re: My first DVD that is in sync

    Choping chunks is not such a good idea, there are several limitations to it.

    Both video and audio data can spaw over more than one chunk hence you will end up with half frames (video/audio). Tydemux naturally detects that but that is not a guarntie that things will work.

    There are also times when a chunk is mainly video or audio hence cuting out a chunk can reder in a long loss of audio or video (audio loss is definitly bad).

    One way to do it would be to have a app to display I frames (we have to cut at gop boundary anyways which is tied to the I frame). We could very easily scan the whole tystream and and build up a list of I frames or build a list of where the iframes are located in the stream.

    Now when playing back and forth we could extract those I frames. The user can review them and add/delet cut points.
    Tydemux is numbering every seq/gop with a sequence number. This sequence number is also attached to the I frame (which is in the same container as the gop and seq header). A cut start will therefore be tied to one seq number and the stop to another. When extracting the data from the stream we can simply skip those seq/gops we don't want. We will naturally need to skip the same amount of audio data.

    Anyideas?

    Cheers Olaf



    Originally posted by MrBassMan
    In theory you can do this now with tychopper. If you can work out how long a chunk plays (prabably very variable due to compression) you could calculate the chunks to chop out and keep. It's more like using a blunt axe than a surgeons scaple but it would sort of do what you want.

  15. #270
    Join Date
    Sep 2001
    Location
    Delaware
    Posts
    237
    Originally posted by MrBassMan
    Opening the IFO file produced by IFOEdit shows a line that says:

    Audio 1: English (Mpeg-1 ) 2ch 48Kbs 16bps (ID: 0xc0)

    So the answer would APPEAR to be yes it is at 48KHz
    MrBassMan -

    I pulled a medium stream from my Philps SA and verified that the stream was 480x480 and the audio was 32Kbps. After I authored a DVD in IfoEdit, I checked the IFO info and it showed me:


    Video: MPEG-2 352x240 (NTSC) blah blah blah...
    Audio 1: English (Mpeg-1) 2ch 48Kbps 16pbs (ID:0xc0)


    So, I'm guessing Derrow didn't put in any stream checks in this tool, to make sure it would be DVD compliant.

    When I record a stream on my Philips SA @ 720x480 (Setting Zero in resource editor), resample the audio to 48Kbps and author with IfoEdit, it plays perfect on all my DVD players.

    -mp

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •