VirtualDub Conversion unintentionally also changes video speed

[hwvd7]

Hi everyone,

I have encountered a small issue with video speeds when using Virtualdub to convert files from .mpg to AVI and was wondering if anyone can help.

Does anyone possibly know why the duration of an event in a converted video might be faster than when compared to the original file? And what role the frame rate might have (even if the audio and video have the same duration)?

The process involved: an .mpg file being converted to uncompressed AVI form, with the "Video" > "Frame rate" > "Source rate adjustment"> "Change so video and audio durations match" selection checked.

I have attached “properties” page of the videos in the image below (Columns 2 and 3). Sometimes the converted videos did not change speed (row 2) but most did (row 1) and the only difference I noticed was that the converted AVI file had fewer frames per second. To amend for this, I had changed the frame rate selection to “no change”. The video converted from this maintained the frame rate but was then sped up to 18 s, compared to the 46 s of the original video.

I have included an extra column (first column) which shows the properties of the original video in case that is of any use - that had a much longer duration than the .mpg file as only a certain segment of interest was used. The original video was converted from .wmv to .mpg using Freemake.


[Attachment 43222 - Click to enlarge]

Sorry for my ignorance. Any help will be greatly appreciated. Thanks!

[jagabo]

Is the audio in sync in the MPEG file you made?

WMV is inherently variable frame rate. Some programs will give audio sync errors when converting from WMV.

[hwvd7]

I'm not able to tell if the audio is in sync - as I believe the microphone was turned off. There is no sound even if I turn the volume up to 100%.

Does saying "WMV has an inherently variable frame rate" mean that the frame rate between videos recorded by the same device changes randomly?

[sneaker]

Not only between different videos but also within a single video. Often, we will convert such videos to a constant framerate by duplicating/deleting frames as required to increase compatibility for editing or playback. Many editors do not really handle variable framerates (this includes VirtualDub).

[hwvd7]

[Blank]

I can't seem to delete the post so I've put this here instead.

[hwvd7]

Not only between different videos but also within a single video. Often, we will convert such videos to a constant framerate by duplicating/deleting frames as required to increase compatibility for editing or playback. Many editors do not really handle variable framerates (this includes VirtualDub).

Ah that's interesting to know. The variable frame rates may be one source of the problem then! Would that mean there's no good fix to find the duration of that event anymore? Apart from measuring it in real life? What would be some good video file types to record in so the frame rate is constant?

Though, I'm still not sure why the .mpg and .avi also present the event at different speeds (differences ~ 3 seconds).

[edit] I am guessing that the .mpg may also have had a variable frame rate. I will update everyone when I have access to the videos again after next Tuesday.

[jagabo]

When the action gets too high for the requested bitrate the WMV encoder reduces the frame rate (fewer frames per second require less bits per second). So a 30 fps video may drop to 15 fps during high action shots.

Cell phones usually drop the frame rate during very dark shots to allow for longer exposure times and hence higher image quality.

Variable frame rate video includes time codes that indicate how long each frame is to be displayed. Many programs don't understand VFR time codes and simply assume a constant frame rate.

Your MPG and AVI files all show a running time of 46 seconds. What's the problem?

[hwvd7]

When the action gets too high for the requested bitrate the WMV encoder reduces the frame rate (fewer frames per second require less bits per second). So a 30 fps video may drop to 15 fps during high action shots.

Cell phones usually drop the frame rate during very dark shots to allow for longer exposure times and hence higher image quality.

Variable frame rate video includes time codes that indicate how long each frame is to be displayed. Many programs don't understand VFR time codes and simply assume a constant frame rate.

Your MPG and AVI files all show a running time of 46 seconds. What's the problem?

Yes the overall running times are the same (strangely). The problem is that the duration of an event in the MPG file is different to in the AVI file - and timing is important as it is a variable that I am measuring.

Though, now that you mention time codes, I'm guessing that perhaps the MPG file also had a variable frame rate which was removed by Virtualdub during the conversion to AVI. And that may have caused the time difference. I shall check the next time I have access to the videos again (which will be next Tuesday), so I'll update everyone then. If it is the case with the variable frame rate, are there any good alternatives to Virtualdub that works with variable frame rates (preferably free)?

[jagabo]

If you can, supply a sample MPG and AVI file that shows the problem. MPG files you make yourself usually don't have variable frame rates. At least give a much more detailed explanation of how the timing varies. Also use MediaInfo to get more details about the files. GSpot is also useful for examining AVI and MPG files.

[hwvd7]

If you can, supply a sample MPG and AVI file that shows the problem. MPG files you make yourself usually don't have variable frame rates. At least give a much more detailed explanation of how the timing varies. Also use MediaInfo to get more details about the files. GSpot is also useful for examining AVI and MPG files.

I see, I shall check the MPG file just in case. I'm not sure if I'd be able to supply a sample of the videos but I'll definitely use MediaInfo to obtain some more details about the files when I can and post them here.

As for a more detailed explanation of the timing problem - I am measuring how long an event goes on for in the AVI videos (in ImageJ, had to set the frame rate manually) but when I measure the duration of the same event in the MPG files (in Windows Media Player), it is usually ~3 seconds longer. Though oddly, some videos (only a small percentage) did not have this problem and these are the videos in second row of the image. I'm not sure if that explanation is detailed enough, was there something specific you were wondering about?

[Edit] Added the names of the programs that I have been using to measure the duration of the event in case that might also be a source of the problem.

[hwvd7]

If you can, supply a sample MPG and AVI file that shows the problem. MPG files you make yourself usually don't have variable frame rates. At least give a much more detailed explanation of how the timing varies. Also use MediaInfo to get more details about the files. GSpot is also useful for examining AVI and MPG files.

I see, I shall check the MPG file just in case. I'm not sure if I'd be able to supply a sample of the videos but I'll definitely use MediaInfo to obtain some more details about the files when I can and post them here.

As for a more detailed explanation of the timing problem - I am measuring how long an event goes on for in the AVI videos (in ImageJ, had to set the frame rate manually) but when I measure the duration of the same event in the MPG files (in Windows Media Player), it is usually ~3 seconds longer. Though oddly, some videos (only a small percentage) did not have this problem and these are the videos in second row of the image. I'm not sure if that explanation is detailed enough, was there something specific you were wondering about?

[Edit] Added the names of the programs that I have been using to measure the duration of the event in case that might also be a source of the problem.

Hi Jagabo and sorry for the delay. These are the information obtained from MediaInfo and I have put them below. Do you or possibly anyone have any insight after looking at them?

Number 1 is the video that when converted, did not change the length of the event of interest whilst Number 9 is the video that was converted oddly. I have included the information from the original .wmv, the .mpg (converted by freemake) and the final uncompressed .avi file.

The MPEG files do not have a "Frame rate mode" entry so I am guessing it does have a constant frame rate and isn't a cause of the problem.

One other guess is, would having a video watermark affect the video speeds in any way? When Freemake was used to convert the wmv file to MPEG, it added 3 s watermarks before and after the video. Before (I think last year), when it didn't add watermarks, the videos converted did not have the problem that I am having.

Firstly, Number 9, which had the conversion error:
General
Complete name : E:\Video Error\Justin 8-3-17 no. 9.wmv
Format : Windows Media
File size : 72.0 MiB
Duration : 9 min 23 s
Overall bit rate mode : Constant
Overall bit rate : 1 072 kb/s
Maximum Overall bit rate : 2 661 kb/s
Encoded date : UTC 2017-03-08 05:32:11.553

Video
ID : 2
Format : VC-1
Format profile : MP@HL
Codec ID : WMV3
Codec ID/Info : Windows Media Video 9
Codec ID/Hint : WMV3
Description of the codec : Windows Media Video 9 - Professional
Bit rate mode : Constant
Bit rate : 2 477 kb/s
Width : 960 pixels
Height : 544 pixels
Display aspect ratio : 16:9
Frame rate mode : Variable
Nominal frame rate : 30.000 FPS
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Compression mode : Lossy
Language : English (AU)

Audio
ID : 1
Format : WMA
Format version : Version 2
Codec ID : 161
Codec ID/Info : Windows Media Audio
Description of the codec : Windows Media Audio 9.2 - 160 kbps, 44 kHz, stereo (A/V) 1-pass CBR
Duration : 9 min 23 s
Bit rate mode : Constant
Bit rate : 160 kb/s
Channel(s) : 2 channels
Sampling rate : 44.1 kHz
Bit depth : 16 bits
Stream size : 10.8 MiB (15%)
Language : English (AU)

----------------------------------------------------------------------------------------------------------------------------------------------
General
Complete name : Number 9 9-02-2017\Justin 8-3-17 no. 9 Freemake.mpg
Format : MPEG-PS
File size : 6.42 MiB
Duration : 46 s 640 ms
Overall bit rate mode : Variable
Overall bit rate : 1 155 kb/s

Video
ID : 224 (0xE0)
Format : MPEG Video
Format version : Version 2
Format profile : Main@High 1440
Format settings : BVOP
Format settings, BVOP : Yes
Format settings, Matrix : Default
Duration : 46 s 640 ms
Bit rate mode : Variable
Bit rate : 876 kb/s
Width : 960 pixels
Height : 544 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 : Progressive
Compression mode : Lossy
Bits/(Pixel*Frame) : 0.067
Time code of first frame : 00:00:00:00
Time code source : Group of pictures header
GOP, Open/Closed : Open
Stream size : 4.87 MiB (76%)

Audio
ID : 192 (0xC0)
Format : MPEG Audio
Format version : Version 1
Format profile : Layer 2
Duration : 46 s 629 ms
Bit rate mode : Constant
Bit rate : 256 kb/s
Channel(s) : 2 channels
Sampling rate : 44.1 kHz
Compression mode : Lossy
Stream size : 1.42 MiB (22%)

----------------------------------------------------------------------------------------------------------------------------------------------
General
Complete name : no._9 Vir.avi
Format : AVI
Format/Info : Audio Video Interleave
File size : 686 MiB
Duration : 46 s 629 ms
Overall bit rate : 123 Mb/s
Writing library : VirtualDub build 32618/release

Video
ID : 0
Format : RGB
Codec ID : 0x00000000
Codec ID/Info : Basic Windows bitmap format. 1, 4 and 8 bpp versions are palettised. 16, 24 and 32bpp contain raw RGB samples
Duration : 46 s 629 ms
Bit rate : 122 Mb/s
Width : 960 pixels
Height : 544 pixels
Display aspect ratio : 16:9
Frame rate : 9.737 FPS
Bit depth : 8 bits
Bits/(Pixel*Frame) : 24.000
Stream size : 678 MiB (99%)

Audio
ID : 1
Format : PCM
Format settings : Little / Signed
Format settings, Endianness : Little
Format settings, Sign : Signed
Codec ID : 1
Duration : 46 s 629 ms
Bit rate mode : Constant
Bit rate : 1 411.2 kb/s
Channel(s) : 2 channels
Sampling rate : 44.1 kHz
Bit depth : 16 bits
Stream size : 7.84 MiB (1%)
Alignment : Aligned on interleaves
Interleave, duration : 103 ms (1.01 video frame)
Interleave, preload duration : 500 ms

----------------------------------------------------------------------------------------------------------------------------------------------
Now Number 1, which did not have the conversion error.


General
Complete name : 1 (2) wmv.wmv
Format : Windows Media
File size : 80.4 MiB
Duration : 10 min 0 s
Overall bit rate mode : Constant
Overall bit rate : 1 124 kb/s
Maximum Overall bit rate : 2 661 kb/s
Encoded date : UTC 2017-04-28 03:53:07.176

Video
ID : 2
Format : VC-1
Format profile : MP@HL
Codec ID : WMV3
Codec ID/Info : Windows Media Video 9
Codec ID/Hint : WMV3
Description of the codec : Windows Media Video 9 - Professional
Bit rate mode : Constant
Bit rate : 2 477 kb/s
Width : 960 pixels
Height : 544 pixels
Display aspect ratio : 16:9
Frame rate mode : Variable
Nominal frame rate : 30.000 FPS
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Compression mode : Lossy
Language : English (AU)

Audio
ID : 1
Format : WMA
Format version : Version 2
Codec ID : 161
Codec ID/Info : Windows Media Audio
Description of the codec : Windows Media Audio 9.2 - 160 kbps, 44 kHz, stereo (A/V) 1-pass CBR
Duration : 10 min 0 s
Bit rate mode : Constant
Bit rate : 160 kb/s
Channel(s) : 2 channels
Sampling rate : 44.1 kHz
Bit depth : 16 bits
Stream size : 11.4 MiB (14%)
Language : English (AU)

----------------------------------------------------------------------------------------------------------------------------------------------

General
Complete name : no. 1 Freemake.mpg
Format : MPEG-PS
File size : 8.09 MiB
Duration : 46 s 942 ms
Overall bit rate mode : Variable
Overall bit rate : 1 446 kb/s

Video
ID : 224 (0xE0)
Format : MPEG Video
Format version : Version 2
Format profile : Main@High 1440
Format settings : BVOP
Format settings, BVOP : Yes
Format settings, Matrix : Default
Format settings, GOP : Variable
Duration : 46 s 922 ms
Bit rate mode : Variable
Bit rate : 1 161 kb/s
Width : 960 pixels
Height : 544 pixels
Display aspect ratio : 16:9
Frame rate : 23.976 (24000/1001) FPS
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Compression mode : Lossy
Bits/(Pixel*Frame) : 0.093
Time code of first frame : 00:00:00:00
Time code source : Group of pictures header
Stream size : 6.50 MiB (80%)

Audio
ID : 192 (0xC0)
Format : MPEG Audio
Format version : Version 1
Format profile : Layer 2
Duration : 46 s 942 ms
Bit rate mode : Constant
Bit rate : 256 kb/s
Channel(s) : 2 channels
Sampling rate : 44.1 kHz
Compression mode : Lossy
Stream size : 1.43 MiB (18%)

----------------------------------------------------------------------------------------------------------------------------------------------
General
Complete name : no._1 VD.avi
Format : AVI
Format/Info : Audio Video Interleave
File size : 1.65 GiB
Duration : 46 s 942 ms
Overall bit rate : 302 Mb/s
Writing library : VirtualDub build 32618/release

Video
ID : 0
Format : RGB
Codec ID : 0x00000000
Codec ID/Info : Basic Windows bitmap format. 1, 4 and 8 bpp versions are palettised. 16, 24 and 32bpp contain raw RGB samples
Duration : 46 s 942 ms
Bit rate : 300 Mb/s
Width : 960 pixels
Height : 544 pixels
Display aspect ratio : 16:9
Frame rate : 23.944 FPS
Bit depth : 8 bits
Bits/(Pixel*Frame) : 24.000
Stream size : 1.64 GiB (100%)

Audio
ID : 1
Format : PCM
Format settings : Little / Signed
Format settings, Endianness : Little
Format settings, Sign : Signed
Codec ID : 1
Duration : 46 s 942 ms
Bit rate mode : Constant
Bit rate : 1 411.2 kb/s
Channel(s) : 2 channels
Sampling rate : 44.1 kHz
Bit depth : 16 bits
Stream size : 7.90 MiB (0%)
Alignment : Aligned on interleaves
Interleave, duration : 42 ms (1.01 video frame)
Interleave, preload duration : 500 ms



Thank you and anyone else who is trying to help.

[hwvd7]

Not only between different videos but also within a single video. Often, we will convert such videos to a constant framerate by duplicating/deleting frames as required to increase compatibility for editing or playback. Many editors do not really handle variable framerates (this includes VirtualDub).

Ah that's interesting to know. The variable frame rates may be one source of the problem then! Would that mean there's no good fix to find the duration of that event anymore? Apart from measuring it in real life? What would be some good video file types to record in so the frame rate is constant?

Though, I'm still not sure why the .mpg and .avi also present the event at different speeds (differences ~ 3 seconds).

[edit] I am guessing that the .mpg may also have had a variable frame rate. I will update everyone when I have access to the videos again after next Tuesday.

From the information obtained with MediaInfo, I think the MPEG files don't have a variable frame rate as there is no entry for it, unlike for the .wmv file.