VideoHelp Forum
+ Reply to Thread
Page 1 of 3
1 2 3 LastLast
Results 1 to 30 of 67
Thread
  1. Member
    Join Date
    Apr 2012
    Location
    Hungary
    Search PM
    I have just created a DVD from an mpeg2 file. The video was exported from Sony Vegas through Debugmode Frameserver with HCEnc. The sound was saved directly from Vegas as ac3. The video and sound was muxed by ImagoMPEG Muxer. Then the mpeg2 file was imported to Corel DVD Movie Factory 7 Pro, and written to DVD without conversion. The result is: the video and the sound seems to be in sync when played back on the computer DVD drive, but when I play it back on the desktop DVD player, the video and sound seems a little bit out of sync. Can it be because of the bitrate I set for the video? In HCEnc I set 9500 as maximum and 6000 as average. Here is the mpeg2 file info from Mediainfo:

    General
    Complete name : C:\Users\Bencuri\Downloads\Liberation.mpg
    Format : MPEG-PS
    File size : 747 MiB
    Duration : 16 min 33 s
    Overall bit rate mode : Variable
    Overall bit rate : 6 306 kb/s

    Video
    ID : 224 (0xE0)
    Format : MPEG Video
    Format version : Version 2
    Format profile : Main@Main
    Format settings : CustomMatrix / BVOP
    Format settings, BVOP : Yes
    Format settings, Matrix : Custom
    Format settings, GOP : M=3, N=15
    Duration : 16 min 33 s
    Bit rate mode : Variable
    Bit rate : 5 988 kb/s
    Maximum bit rate : 9 500 kb/s
    Width : 720 pixels
    Height : 576 pixels
    Display aspect ratio : 16:9
    Frame rate : 25.000 FPS
    Standard : PAL
    Color space : YUV
    Chroma subsampling : 4:2:0
    Bit depth : 8 bits
    Scan type : Progressive
    Compression mode : Lossy
    Bits/(Pixel*Frame) : 0.578
    Time code of first frame : 00:00:00:00
    Time code source : Group of pictures header
    GOP, Open/Closed : Open
    GOP, Open/Closed of first frame : Closed
    Stream size : 710 MiB (95%)

    Audio
    ID : 189 (0xBD)-128 (0x80)
    Format : AC-3
    Format/Info : Audio Coding 3
    Muxing mode : DVD-Video
    Duration : 16 min 33 s
    Bit rate mode : Constant
    Bit rate : 192 kb/s
    Channel(s) : 2 channels
    Channel positions : Front: L R
    Sampling rate : 48.0 kHz
    Frame rate : 31.250 FPS (1536 SPF)
    Bit depth : 16 bits
    Compression mode : Lossy
    Stream size : 22.8 MiB (3%)
    Service kind : Complete Main
    Quote Quote  
  2. Member hech54's Avatar
    Join Date
    Jul 2001
    Location
    Yank in Europe
    Search Comp PM
    Originally Posted by Bencuri View Post
    The video was exported from Sony Vegas through Debugmode Frameserver with HCEnc. The sound was saved directly from Vegas as ac3. The video and sound was muxed by ImagoMPEG Muxer. Then the mpeg2 file was imported to Corel DVD Movie Factory 7 Pro, and written to DVD without conversion.
    Why?
    Are you trying for a Rube Goldberg award?
    "A Rube Goldberg machine is a machine intentionally designed to perform a simple task in an indirect and overcomplicated fashion."
    https://en.wikipedia.org/wiki/Rube_Goldberg_machine
    Quote Quote  
  3. Member
    Join Date
    Apr 2012
    Location
    Hungary
    Search PM
    Which part do you have problem with? I used this method because this way the video is very clear and edges are not jagged. If I render directly from Sony Vegas, the result is crap compared to it.
    Quote Quote  
  4. Member
    Join Date
    Apr 2012
    Location
    Hungary
    Search PM
    It seems it's a common type of problem, others are complaining about is, too:
    https://www.techadvisor.co.uk/forum/helproom-1/video-files-constantly-burning-audio-ou...-help-4475746/
    Quote Quote  
  5. Member
    Join Date
    Mar 2011
    Location
    Nova Scotia, Canada
    Search Comp PM
    Originally Posted by Bencuri View Post
    Which part do you have problem with? I used this method because this way the video is very clear and edges are not jagged. If I render directly from Sony Vegas, the result is crap compared to it.
    He was referring to the ridiculous number of steps and programs used, which in itself makes it much harder to pinpoint the issue. Ever heard of AVStoDVD?
    Quote Quote  
  6. Member DB83's Avatar
    Join Date
    Jul 2007
    Location
    United Kingdom
    Search Comp PM
    What is this 31.250 FPS for the audio ?

    But I agree that there are too many programs involved. Just what is this jagged edges you refer to ? If it is a interlacing artefact then you need to reconsider your workflow. Check the source. Was this interlaced in the first place ? Your final video is progressive which itself is not good for dvd.

    To start, let us see the mediainfo file for the original video before it's tortuous path.
    Quote Quote  
  7. Member
    Join Date
    Apr 2012
    Location
    Hungary
    Search PM
    Yes, it is 31.250

    This is the original:

    General
    ID : 1 (0x1)
    Complete name : C:\Users\Bence\Videos\00121.ts
    Format : MPEG-TS
    File size : 1.88 GiB
    Duration : 11 min 31 s
    Overall bit rate mode : Variable
    Overall bit rate : 23.3 Mb/s
    Maximum Overall bit rate : 35.5 Mb/s

    Video
    ID : 4113 (0x1011)
    Menu ID : 1 (0x1)
    Format : AVC
    Format/Info : Advanced Video Codec
    Format profile : High@L4
    Format settings, CABAC : Yes
    Format settings, ReFrames : 2 frames
    Format settings, GOP : M=3, N=12
    Codec ID : 27
    Duration : 11 min 31 s
    Bit rate mode : Variable
    Bit rate : 21.9 Mb/s
    Maximum bit rate : 22.7 Mb/s
    Width : 1 920 pixels
    Height : 1 080 pixels
    Display aspect ratio : 16:9
    Frame rate : 25.000 FPS
    Color space : YUV
    Chroma subsampling : 4:2:0
    Bit depth : 8 bits
    Scan type : Interlaced
    Scan type, store method : Separated fields
    Scan order : Top Field First
    Bits/(Pixel*Frame) : 0.422
    Stream size : 1.76 GiB (94%)

    Audio
    ID : 4352 (0x1100)
    Menu ID : 1 (0x1)
    Format : AC-3
    Format/Info : Audio Coding 3
    Format settings, Endianness : Big
    Codec ID : 129
    Duration : 11 min 31 s
    Bit rate mode : Constant
    Bit rate : 256 kb/s
    Channel(s) : 2 channels
    Channel positions : Front: L R
    Sampling rate : 48.0 kHz
    Frame rate : 31.250 FPS (1536 spf)
    Bit depth : 16 bits
    Compression mode : Lossy
    Stream size : 21.1 MiB (1%)
    Service kind : Complete Main


    Here are the two mpeg files: one is a Sony Vegas render, the other is HcEnc. If I burn the Sony Vegas one on to DVD i plays fine on all devices, the HCEnc version plays out of sync on DVD player. The delay is minor, but as it is a tutorial DVD, where people will try to observe what's happening carefully, you can notice that the audio comes later than the video. This is like that in case if the desktop player. If you just play back the mpeg file, there is no problem.

    https://forum.videohelp.com/attachments/46371-1533477902/Liberation_HCEnc.mpg
    https://forum.videohelp.com/attachments/46372-1533477902/Liberation_sony.mpg
    Quote Quote  
  8. Member
    Join Date
    Apr 2012
    Location
    Hungary
    Search PM
    I have to admit though that in case it wasn't this kinf of movie, maybe I wouldn't even notice that delay. It is only because the nature of this stuff, that you are trying to watch very carefully, and then you notice the problems
    Quote Quote  
  9. Member DB83's Avatar
    Join Date
    Jul 2007
    Location
    United Kingdom
    Search Comp PM
    With that sample, ok I did not watch it all, how can one determine audio sync ?

    Again, why have you encoded this as progressive from an interlaced source ? Always keep interlaced as interlaced.

    That fps is probably a quirk of the original source but mixing an audio source from, effectively, a different video source could well have caused the sync issue.

    And I agree with my friend above. Juts throw the source in to avstodvd and let it do its magic. If it is still not in sync then atleast there is a good thread here to look at the issue.
    Quote Quote  
  10. Member
    Join Date
    Apr 2012
    Location
    Hungary
    Search PM
    If you are looking at his picking hand and watch carefully, you can determine whether it is in sync or not.

    Someone suggested to me to treat it as progressive. And indeed, in the end I got a finer image than keeping it interlaced.

    The problem is when rendered with the same project settings, the mpg that Sony Vegas creates makes no problem. It is also an interlaced to progressive conversion, same edit, same project properties. So that's why I see no problem from the video being progressive. The Sony version is also progressive, but makes no issues.
    Quote Quote  
  11. Member DB83's Avatar
    Join Date
    Jul 2007
    Location
    United Kingdom
    Search Comp PM
    Since you are resizing in an NLE, interlaced to progressive may be the best option. It certainly is not when frame size is the same.

    But the sync issue probably is the mix of sources. There are other, cleaner, ways of demuxing audio from video if you feel it necessary to go one that tortuous route particularily since you exported the audio at a lower bit rate. Not enough to make a difference but audio is surely important in this example. But it the Sony version works then why go there.
    Quote Quote  
  12. Member
    Join Date
    Apr 2012
    Location
    Hungary
    Search PM
    I tried to omit going with the Sony version, because the two files have significant difference in quality. The file that Sony outputs is way much lower in quality.
    Quote Quote  
  13. Member DB83's Avatar
    Join Date
    Jul 2007
    Location
    United Kingdom
    Search Comp PM
    Well I took a closer look at both files.

    Both have the same video and audio bitrate. The Sony one has a small left/right border but since I am not familiar with the original I can not say whether that is true and the HCenc is actually stretched or the Sony is slightly squashed.

    More importantly is that the HCenc version runs some 8 secs longer than the Sony one. Again, only you know the true running time but you do have the un-muxed source to compare.

    Personally, I find the Sony more visually appealing.
    Quote Quote  
  14. Member
    Join Date
    Apr 2012
    Location
    Hungary
    Search PM
    I did not notice it runs longer. Will check.

    My problem with the Sony version is that it has flicker artifacts and the edges are jagged, and also there is more image noise. And most importantly, when I play back the mpeg file on the computer, it stucks for a moment on playback here and there. It never happens with the HCEnc version. The. Mpeg file plays smooth and has no flicker.
    Quote Quote  
  15. Mainconcept MPEG2 (what sony uses) does produce more noise (in general) than HCEnc , but it also tends to retain more detail . HCenc tends to drop fine texture detail and tends to have problems with gradients. Pros/cons.

    Your videos are not really comparable, because you probably inadvertently deinterlaced the sony version (you probably forgot to interpret the file as progressive, that's why you are getting jaggies), and it's pillarboxed, and the levels are quite different between them.

    The HCenc version is washed out, elevated black level , low contrast and dynamic range . Maybe it's a desired effect, but non standard. I'm guessing more likely it was you forgot to adjust for the frameserver . Vegas works in studio RGB, and if you frameserve out, you need to adjust the levels either converting RGB to YUV with a PC matrix, or using a studio to RGB preset before the frameserver in vegas

    The sony version has similar levels to the source . But it probably needs some adjustment, as there are some illegal levels and you might want to make some subjective adjustments too


    As for the audio - there should be a difference. If you use the same muxing tool, there should be a slight difference in the initial audio delay between MP2 and AC3, it's expected. (I'm talking less than a frame, maybe 10-20ms, normally negligible - imperceptible difference by 99.9% human ears) . All compressed audio types have slightly different delay. e.g. MP3 or AAC is slightly different too (but not for DVD-video, obviously) . Only uncompressed wav has no additional initial audio delay . So if you got the same, or no difference, I would look again at your procedures - how exactly are you muxing these ? I would also look at another authoring tool - why not Sony DVDA since you use vegas ?
    Last edited by poisondeathray; 6th Aug 2018 at 00:11.
    Quote Quote  
  16. Member
    Join Date
    Apr 2012
    Location
    Hungary
    Search PM
    Regarding the quality difference: why I prefer the HCEnc version is because it is more comfortable to watch for the eye. This video will serve tutorial purposes, there are many fast movements, you have to watch carefully. The Sony version is more difficult to watch when there are movements. There are more arifatcs, flickers, etc. Regarding image colors my only concern is the irritation that the much white causes for the eye after a while. When the images is washed, its better to watch this longterm. It is due to -25% gain reduction settind in the avisynth script.

    In case of both files, Sony project settings were set to progressive, deinterlacing none, adjust source unticked, resampling disabled, progressive selected for every video part inserted on timeline.

    Muxing was done by ImagoMuxer and TmpgEnc. But when I compare the project audio to the muxed audio zoomed in in Sony Vegas, you can see no more than 5ms delay between the original and the muxed. That's why i assume this sync problem is a player mpeg intepretation error rather than imporperly muxed audio.

    I stuck to the Corel authoring tool because that seems to be the only one that let you burn the project without reencoding the imported mpeg files. Tried Wondershare though, the result is poor, choppy playback, etc.

    I found an article about that on authoring time code of the mpeg is lost and not saved to vob. This is a problem when editing files containing GOP. Maybe the fact that I edited the MTS directly without converting it to avi caused this problem? If yes, can you somehow correct errors before authoring?
    Quote Quote  
  17. Originally Posted by Bencuri View Post
    Muxing was done by ImagoMuxer and TmpgEnc. But when I compare the project audio to the muxed audio zoomed in in Sony Vegas, you can see no more than 5ms delay between the original and the muxed. That's why i assume this sync problem is a player mpeg intepretation error rather than imporperly muxed audio.
    There are differences between muxers too, and decoders. What you see in sony isn't necessarily what you get in another program or hardware DVD player

    I stuck to the Corel authoring tool because that seems to be the only one that let you burn the project without reencoding the imported mpeg files. Tried Wondershare though, the result is poor, choppy playback, etc.
    Only one??? If your assets are compliant, ALL authoring tools should not be re-encoding . That's the whole point. You should be using elementary streams to author too (ie. m2v, not .mpeg, or .mpg, or mpeg-ps) . The authoring too should be doing the muxing


    I found an article about that on authoring time code of the mpeg is lost and not saved to vob. This is a problem when editing files containing GOP. Maybe the fact that I edited the MTS directly without converting it to avi caused this problem? If yes, can you somehow correct errors before authoring?
    Do you have a link to the article, maybe something was lost in your paraphrasing

    It doesn't really matter anyway, because the timecodes re-written when it's authored to VOB. e.g. if you have several separate streams, they all start at "zero". But when you place them into a single contiguous timeline, they can't all start at zero, right ? The timecodes are adjusted.

    But if there is an error in decoding the MTS, or editing it - then there will be errors propogated everywhere downstream (i.e. entire project should exhibit same problem). You would expect the same error in dfms, hcenc too then as well as sony export, even exporting to MP4. So converting it to AVI or all I-frame beforehand would make no difference in your specific case . (But there are cases where editing long GOP can pose problems.)
    Quote Quote  
  18. Member
    Join Date
    Apr 2012
    Location
    Hungary
    Search PM
    This is that article: http://www.myvideoproblems.com/ProblemPages/OOS_AV_Mpeg2editing.htm

    I will try a new authoring software, like Cyberlink, but in case the Sony mpeg has no issues after authoring as I mentioned, only the HCEnc one, I doubt it is an authoring tool problem. It seems it is the file that is making the problems.
    Quote Quote  
  19. The article deals with editing mpeg2 and long GOP (temporal compression, i,b, p frames). Your source is AVC (you are editing AVC), not MPEG2 , but the concept still applies in terms of long GOP

    But if that was the cause, then BOTH HCEnc and the Sony export would be affected, because you exported from the same project. So it's not your problem in this case
    Quote Quote  
  20. Member
    Join Date
    Apr 2012
    Location
    Hungary
    Search PM
    Well. I will try that software: Womble for multiplexing. They mention it so many times, so it should do the job properly I guess. Then I burn another disc with Cyberlink, if it will be out of sync as well, I will drop the idea to put these videos on DVD and I will store it on memory card. At least, the HCEnc output mpeg plays fine itself, and the quality is pristine compared to the capabilities of the 9,5 Mbps world. I have FullHD videos that are not as clear as this.
    Quote Quote  
  21. Member DB83's Avatar
    Join Date
    Jul 2007
    Location
    United Kingdom
    Search Comp PM
    Do what you will.

    Yet you have already been told TWICE by different respondents which version is actually better. Bottom line is that no version, other than the original, can be perfect.

    Even so I would still prefer to see an interlaced version from the original source just to satisfy myself.

    Also I am still in the dark as to why the HCenc version is some 12 secs (just re-checked) longer than the Sony one whereas IIRC the original clip was the same length as the source. If such a discrepancy is possible then there really is an issue.

    Why have you just not tried avstodvd as suggested or do you just want to be bloody-minded and try your own solutions rather than attempt suggestions that might actually work.
    Quote Quote  
  22. Member
    Join Date
    Apr 2012
    Location
    Hungary
    Search PM
    I am not sure at all the two versions were converted from the same project clip length. I just uploaded something to show the difference how the two videos look. I am continously doing tests, it's difficult to track which file was rendered from which length. The same applies to advices. As we have no clue about what is happening here and as I am also experimenting on my side, I have no time to check everything, but I try to do my best.

    And interesting outcome however: No matter if I feed the Sony or HCEnc version to Cyberlink or Corel DVD creator, even when reencoding is disabled, the outcome is a DVD with a video that is not on the level of the source. Especially the HCEnc one. As I said I like that the HCEnc output file is crisp, have no jagged edges, little noise. When the DVD is ready, it plays with jagged edges again and some more noise. INTERESTINGLY!!! when I author the DVD with Womble, the video retains the quality of the HCEnc mpeg. This is very big leap forward, and its a shame regarding the pricy DVD authoring softwares. Unfortunately, even with Wobmle, the sound is late on the player....

    Anyway, in my case, it has no use thinking about using a source from Sony to author the DVD. To my mind, only the HC Enc version has a quality that makes it woth to make a DVD project from it. If I have to stick to the Sony output, I prefer starting the project over with an intention to do a BluRay project. Because then I just export a non resized BluRay file from Sony, that is more or less okay, and burn that. The problem however, that whatever Sony outputs, the blurs on the video, even if Full HD are worse then in the HCEnc mpeg output. But at least the BD file would have a similar, clear video, though not as pristine.

    Here is your interlaced file:
    Image Attached Files
    Quote Quote  
  23. Member DB83's Avatar
    Join Date
    Jul 2007
    Location
    United Kingdom
    Search Comp PM
    Thank you for the clip.

    This one is, reportedly, the shortest of all and it includes a title not in the others. Clearly you are not been consistent with the source which, itself, makes it hard to judge.

    I watched this in vlc with de-interlacing off (which is my default viewing). I could not detect any jaggies.

    With the greatest of respect, perception of quality rests in the eye of the beholder. It is what you think not what anyone else thinks. I do not put myself on a pedestal but simply state what I think looks better. My friend, who inevitably, has a better understanding of these things, has another opinion.

    But if you think the sync issue revolves around the bitrate of the video, which is on the higher point of what is permitted for dvd, then simply reduce that to 8,000 and take another look.
    Quote Quote  
  24. Member
    Join Date
    Apr 2012
    Location
    Hungary
    Search PM
    I do not say I don't tolerate your opinion. It is just from my view what I am writing. If I have to stick to the Sony version, for me it's better to do it on BluRay. I wouldn't mind doing the project with the Sony files if it would render a video with same image chararteristics as HCEnc, but so far I didn't manage to do it like that.

    Unfortunately the bitrate is not a problem. I tried even 7000 bps, does not work. One thing is different between the two softwares however. In Sony you can specify the minimum bitrate, in HCEnc, you cannot. That does differ, I am not sure if it can be considered as a possible source for problems.

    I tried AVStoDVD: same like at Womble: quality very good, but sync problems.
    Quote Quote  
  25. Member DB83's Avatar
    Join Date
    Jul 2007
    Location
    United Kingdom
    Search Comp PM
    With the version of avstodvd I use (2.3.2) you can actually select HCenc as the preferred encoder.

    My friend has more knowledge and he described above the issues with the various encoders and what you need to do in Vegas to achieve maximum quality. But if you are happy with the final result, however unacceptable it may be to others, then no one can force you change.

    What you could always try is just use HCenc on the source outside Vegas. Just requires a simple avisynth script to feed the frames. Then if you did not get what you expected you know something is not right ie it is not the encoder.

    Just one thought. Since I am not in the position to tell - not having perfect vision or hearing - was the interlaced version in sync ? Could you try that for both your preferred method and via Vegas ?
    Quote Quote  
  26. Member
    Join Date
    Apr 2012
    Location
    Hungary
    Search PM
    I tried just loading the camera raw file to HCEnc, and do the mpeg like that, omiting Vegas, and sadly, when writing it to the disc, there is sync problem, just like in the Debugmode Frameserver case. So this is HCEnc that is doing something that is not okay. On computer the files are fine but the DVD does problems.


    The Interlaced version is okay, but as usual, the Sony Vegas output files has no sync problem. But you see there are lots of jagginess.
    Quote Quote  
  27. Member
    Join Date
    Apr 2012
    Location
    Hungary
    Search PM
    I decided to try to do it differently, instead of considering the file progressive, I try deinterlacing it and see what happens that way. Can you help in that? What is the QTGMC code for that? At the moment, my code is this:

    Code:
    AviSource("C:\Users\Bence\Videos\Liberation.avi")
    AssumeFPS(25)
    ColorYUV(gain_y=-25)
    ColorMatrix(mode="rec.709->rec.601") 
    BilinearResize(720,576)
    ConvertToYV12()
    Quote Quote  
  28. Member DB83's Avatar
    Join Date
    Jul 2007
    Location
    United Kingdom
    Search Comp PM
    As I reported in reply #23 I could not detect jagginess
    Quote Quote  
  29. Originally Posted by Bencuri View Post
    I decided to try to do it differently, instead of considering the file progressive, I try deinterlacing it and see what happens that way. Can you help in that? What is the QTGMC code for that? At the moment, my code is this:

    Code:
    AviSource("C:\Users\Bence\Videos\Liberation.avi")
    AssumeFPS(25)
    ColorYUV(gain_y=-25)
    ColorMatrix(mode="rec.709->rec.601") 
    BilinearResize(720,576)
    ConvertToYV12()

    I think you're "barking up the wrong tree" - meaning this is the wrong line of thought

    Your original was already progressive . Deinterlacing it will just make it 25p content with duplicates for 50p, and degrade the quality slightly (or maybe make it better, depends on your opinion. Because QTGMC denoises slightly and sharpens). 50p framerate is incompatible with DVD, you have to make it interlaced 25i

    What you can do is treat it as interlaced. 25p content encoded as 25i . Effectively 2:2 pulldown. Many DVDs are done this way, including most retail PAL DVD's that have 25p content . All you do is take 25p and encode it interlaced. 99.999% of players will detect it's 2:2 and playback full 25p, full quality. But it shouldn't have any effect on "sync".

    In theory 25p , encoded as 25p would be the highest quality with no chance of any playback being deinterlaced. If it's deinterlaced - if you didn't have jaggies before, and the playback system deinterlaced, you will end up with jaggies.

    Most of the time this sort of sync problem is a muxing issue, but it might be a bad player too (all I did was check on computer and it seems ok). I don't know why the sony version is immune to the problem
    Quote Quote  
  30. Member DB83's Avatar
    Join Date
    Jul 2007
    Location
    United Kingdom
    Search Comp PM
    @pdr The original source, as reported in post #7, is interlaced. The OP shows the encode as progressive.
    Quote Quote  



Similar Threads