VideoHelp Forum




+ Reply to Thread
Results 1 to 23 of 23
  1. Member
    Join Date
    Aug 2008
    Location
    Europe
    Search Comp PM
    Hello everyone

    Newbie question to which I hope the answer is fairly straightforward.

    I'm using VirtualDud and XviD in two pass mode for changing some avi files to different file sizes (want to get all the files on one dvd). Finding this difficult as the final actual file size often doesn't anywhere near match the value that I enter for the 'final file size' in the XviD dialog box. So find have to redo all the files and try to work out what value is going to land me in the park.

    Following questions on this issue:

    1. Am I correct in thinking that the first pass file produced by XviD only contains information concerning the video stream for an avi?


    2. Provided I always use the same avi 'source files' is it possible for me to reuse the matching 'WhatEverName.pass' file for second, third and 'nth' attempts at producing a final (second pass) avi file that matches the file size that I actually need? (Really asking is it possible to do this to avoid having to do a new first pass for any given file?)


    3. Suppose I in some way altered the audio stream in the source avi files - would the earlier first pass '.pass' files for those files still be valid (even though the audio stream has now been altered in the source files)?
    Quote Quote  
  2. Always Watching guns1inger's Avatar
    Join Date
    Apr 2004
    Location
    Miskatonic U
    Search Comp PM
    If you use a bitrate calculator to work out what bitrate you need for the file size you desire you take any guess work out of the process.
    Read my blog here.
    Quote Quote  
  3. Originally Posted by Chakra
    1. Am I correct in thinking that the first pass file produced by XviD only contains information concerning the video stream for an avi?
    Yes.

    Originally Posted by Chakra
    2. Provided I always use the same avi 'source files' is it possible for me to reuse the matching 'WhatEverName.pass' file for second, third and 'nth' attempts at producing a final (second pass) avi file that matches the file size that I actually need? (Really asking is it possible to do this to avoid having to do a new first pass for any given file?)
    Yes, you can reuse the .pass file as long as you don't change the video.

    Originally Posted by Chakra
    3. Suppose I in some way altered the audio stream in the source avi files - would the earlier first pass '.pass' files for those files still be valid (even though the audio stream has now been altered in the source files)?
    Yes, the .pass file has nothing to do with the audio.
    Quote Quote  
  4. Member
    Join Date
    Aug 2008
    Location
    Europe
    Search Comp PM
    Many thanks to you both. (This is a very good forum.)

    Off now to search for how to use a bitrate calculator.

    Thanks again.
    Quote Quote  
  5. Always Watching guns1inger's Avatar
    Join Date
    Apr 2004
    Location
    Miskatonic U
    Search Comp PM
    I like Vcalc as a standalone BRC. You can set a custom size, multiple audio tracks etc. Very easy to use.
    Read my blog here.
    Quote Quote  
  6. Member
    Join Date
    Aug 2008
    Location
    Europe
    Search Comp PM
    Thanks for that. Have it now and trying to use it. Confused only by the 'Audio' setting. It asks for number of tracks and bitrate. As example here is audio inf from GSpot:

    0x2000 (Dolby AC3) AC3
    48000Hz 112 kb/s tot , stereo (2/0) [CBR]

    Not sure what the (2/0) bit above is indicating.

    My questions as follows:

    1. Audio here is stereo. Does that mean that I regard that as 'two tracks' for VCalc?


    2. If it isn't two tracks, what would constitute two (or more tracks) in an avi video file?
    Quote Quote  
  7. Originally Posted by Chakra
    1. Audio here is stereo. Does that mean that I regard that as 'two tracks' for VCalc?
    Yes.
    Quote Quote  
  8. Always Watching guns1inger's Avatar
    Join Date
    Apr 2004
    Location
    Miskatonic U
    Search Comp PM
    Originally Posted by jagabo
    Originally Posted by Chakra
    1. Audio here is stereo. Does that mean that I regard that as 'two tracks' for VCalc?
    Yes.
    Er . . No.

    Single track with a bitrate of 112 kbps.

    A track in Vcalc is an audio stream. For instance, if you have a main audio track and a director's commentary track, then you have two tracks. If you have multiple languages tracks, then you have more than one track. A single stream that is stereo is one track, as is a single stream that is 5.1. Just use the bitrate from g-spot
    Read my blog here.
    Quote Quote  
  9. Oops, sorry. Yes, since the stereo audio is being encoded as a single audio stream it is one track. Wasn't thinking straight!
    Quote Quote  
  10. Member
    Join Date
    Aug 2008
    Location
    Europe
    Search Comp PM
    Brilliant. Thanks guys. Track = a Stream. And using that VCalc has sure kicked my wild-guesses out of the field. What a difference.

    One last question in this area:

    1. When XviD makes a 1st pass file it also produces an associated avi file at the same time. Am I correct in thinking that to do subsequent 2nd passes I do not need those 1st pass associated avi files (so it's okay to delete them before doing a 2nd pass)? I only need the '.pass' files?
    Quote Quote  
  11. Originally Posted by Chakra
    When XviD makes a 1st pass file it also produces an associated avi file at the same time. Am I correct in thinking that to do subsequent 2nd passes I do not need those 1st pass associated avi files (so it's okay to delete them before doing a 2nd pass)? I only need the '.pass' files?
    That is correct. The AVI is there because VirtualDub doesn't know that Xvid isn't saving the video during the first pass. Normally that file contains a black image. If you disable the "discard first pass" option the file will contain video that was compressed with a constant quantizer of 2.
    Quote Quote  
  12. Member
    Join Date
    Aug 2008
    Location
    Europe
    Search Comp PM
    Many thanks, much appreciated. That's me for this one - over and out.
    Quote Quote  
  13. Banned
    Join Date
    Oct 2004
    Location
    Freedonia
    Search Comp PM
    Originally Posted by guns1inger
    Originally Posted by jagabo
    Originally Posted by Chakra
    1. Audio here is stereo. Does that mean that I regard that as 'two tracks' for VCalc?
    Yes.
    Er . . No.

    Single track with a bitrate of 112 kbps.
    Actually it's a single track with a bitrate of 224 Kbps. 112 is not a multiple of 32 but 224 is. For reasons that I do not understand, many tools take
    total bitrate / number of channels
    and report that as your bitrate, so here
    224 / 2 = 112

    It's as stupid and misleading as it can possibly be, but that's how many tools do it.
    Quote Quote  
  14. Member
    Join Date
    Aug 2008
    Location
    Europe
    Search Comp PM
    Oh, oh, it's stating to get complicated.

    1. Why does it have to be a multiple of 32?


    2. What does this mean by way of what values I would actually want to enter in VCalc for a file that has values (supplied by GSpot):

    0x2000 (Dolby AC3) AC3
    48000Hz 112 kb/s tot , stereo (2/0) [CBR]

    And VCalc is asking for:

    Tracks:
    BitRate:

    To be getting it correct, what would I enter?
    Quote Quote  
  15. Always Watching guns1inger's Avatar
    Join Date
    Apr 2004
    Location
    Miskatonic U
    Search Comp PM
    1 track
    112 kbps bitrate

    Why does what have to be a multiple of 32 ?
    Read my blog here.
    Quote Quote  
  16. Member
    Join Date
    Aug 2008
    Location
    Europe
    Search Comp PM
    jman seems to be saying that bitrates of audio tracks are in some way related to having to be multiples of 32.

    Way he is speaking it seems like if the bitrate isn't a multiple of 32 then that implies that the value give for the track cannot be a valid one.

    (Least I think that is what he is saying.)
    Quote Quote  
  17. Always Watching guns1inger's Avatar
    Join Date
    Apr 2004
    Location
    Miskatonic U
    Search Comp PM
    I would have to see the file or the g-spot screen to confirm his conclusion, but he could well be right. That said, I have seen some dumb things done when encoding files, and not everything out there is encoded in a compliant manner.
    Read my blog here.
    Quote Quote  
  18. Member
    Join Date
    Aug 2008
    Location
    Europe
    Search Comp PM
    Hi Gunslinger

    Sorry for delay. Had to work out how to get images displayed here. I thought to put two GSpot shots from the series of avi's I'm working on. Some of them have 'packed bitstream', some do not.

    I got the unpacked ones using the 'avidemux' program (as suggested above) to deal with 'fixes' to append to the 'main' segments. I could though just as soon create those files without okaying the 'unpack bitstream' option that avidemux presents me with (I think).

    I did some reading on packed bitstreams. Seems concensus opinion is:

    a. File size savings are miniscule.
    b. Can cause compatibility issues for playback with some players.

    For me that means don't use packed bitstreams, so all arrange the entire series that way. However, it may be the case that some sources I use will have packed bitstreams so really need to know if that makes any difference over what is being discussed here - what values to enter for 'Track' and 'Bitrate' settings in VCalc so that it can accurately calculate the bitstream value to use in XviD.


    Example of File - Packed Bitstream




    Example of File - No Packed Bitstream




    Hope this brings closer to a solution on this multiples of 32 thing.
    Quote Quote  
  19. Always Watching guns1inger's Avatar
    Join Date
    Apr 2004
    Location
    Miskatonic U
    Search Comp PM
    It makes no difference to the bitrate calculations.

    You can use Mpeg4Modifier to unpack the bitstream without re-encoding.
    Read my blog here.
    Quote Quote  
  20. Packed bitstream isn't for saving space. It is a kludge to work around a problem with the Video For Windows' one-frame-in-one-frame-out" model which can't handle backward predicted frames. Some players can handle packed bitstream, some can't.
    Quote Quote  
  21. Member
    Join Date
    Aug 2008
    Location
    Europe
    Search Comp PM
    Sorry still a little confused. For the purposes of VCalc what values to enter:

    1. Audio: 1 track 112 kbps (as per Gunslinger suggestion)
    OR
    2. Audio: 1 track 224 kbps (as per jman98 suggestion)

    Also:

    3. Why use that value and how does it relate to the point made by jman98 that audio bitrate values have to be a multiple of 32.
    Quote Quote  
  22. Always Watching guns1inger's Avatar
    Join Date
    Apr 2004
    Location
    Miskatonic U
    Search Comp PM
    G-spot says it is 112 kbps, so that is what you should put in the bitrate calculator. 112 is very low for DVD, but is the type of value often used for AVI files, along with 96 kbps. G-spot does not give you a per channel bitrate, it gives you a total track bitrate.

    I suspect that 32 kbps for audio are like 16 pixel blocks for video - an optimised segmentation for getting the best quality out of the compression. Generally for DVD you would use 192, 224 or 256 for stereo, and 384 or 448 for 5.1. However if you are downloading material you can end up with any old odds and sods, as a lot of it is done by people who make it up as they go along.
    Read my blog here.
    Quote Quote  
  23. Member
    Join Date
    Aug 2008
    Location
    Europe
    Search Comp PM
    Okay, that will do me for this..

    Many thanks, Gunslinger - and everyone else - much appreciated.
    Quote Quote  



Similar Threads

Visit our sponsor! Try DVDFab and backup Blu-rays!