I had some similar weirdness when I used the cineform codec. The resulting file was squished like yours but was only in red & green colors. I imported it into vegas and it looked the way I would have expected. I rendered the 24p timeline with the file to a 24p WMV and that file also looked correct. I haven't figured out exactly what the issue there is. I did not see that scenario when using some of the other encoders though, only the cineform ones..
--Steve
error
After I set up everything and tried to process the .m2t a window popped up that said "No data. Check your PIDS." Forgive me, but I have no idea what I'm doing wrong. Any help would be greatly appreciated.
drahx: Here are a couple links to some googled threads about that error message your getting (I've not come across it myself). Hopefully one of them will help you with your particular issue:
http://forum.doom9.org/showthread.php?t=107482
http://forum.videohelp.com/topic329677.html
Thanks SO much. Those links really helped. Now it's running perfectly.
i believe i need help
i am on the go.bat part
what prompt do i punch into the command line?
I have a new problem. It says:
Avisynth open failure:
Script error: Invalid arguments to funtion "Delay Audio"
(E:\ViedeoPulldown\CFHDCap_.avs, line 7)
Has anybody any clue what that is and how to fix it?
Thanks,
Matthias
note: For the command line to go.bat, you want to specify the directory that contains the clips you want to process.
krause: What does your _.avs file have for the DelayAudio(xxx) value?
--Steve
could you provide an example?
thank you very much
note235, the farnsworth script will process all video files in a particular directory. So if you created a directory called C:\Test\ and copied the video files you wanted converted into that directory then you would run:
go.cmd C:\Test
The batch from Farsnworth would then run DGIndex against those files to create the appropriate .avs files that you can later import into VirtualDubMod for processing.
As an alternative you could use my executable (also linked in this thread), which would allow you to process your video files through DGIndex as well as VirtualDubMod.
--Steve
thanks
will try this
i have aspect HD
would converting do the same thing?
Please visit my website.
alliance-crafts.cm
Along with cahrdr, I'm also ending up with garbled, gibberish images. I've run through everything I can think of and I just can't seem to find what the problem is. Anyone else having this problem, or know how to solve it?
I think it may have something to do with the DGIndex step. I checked out the input images in VDub while Vdub was doing it's thing and the input was garbled.
Last edited by BrownJT; 2007 May 29th at 09:16.
I'm getting garbled images as wellOriginally Posted by BrownJT
What exactly do you guy's mean by "garbled" images?
A couple other questions:
1. What application are you using to capture your original clips?
2. What file format are you video files being initially created with?
3. Which codecs are you using to perform the conversion?I (and many others) have had great success with the Lagarith and Uncompressed codecs. Huffyuv has also produced good quality as well.
If you have a small sample original clip (say about 50-100 megs) you could post that up somewhere and I can download and run it through on my end to see what I get. You can also try manually running the file through DGIndex and see what happens there.
--Steve
I am getting the same video as cahrdr (previous page). When I run DGIndex manually, the output video I get is the "garbled" video. The video looks like the orignal video repeated 5 or 6 times, squished vertically and horizontally, and stuck into the upper left corner of the frame(s) leaving the rest of the frame(s) black. Also, the colors are all messed up (grayscale with nonsensical splashes of random colors, similar to a fuzy TV channel).
I've imported both mpeg-2s and AVIs to DGIndex, with the same results. I've viewed the DGIndex outputs in both Windows Media Player and VDub, both garbled.
I really feel that something funky is happening when saving the d2v file from DGIndex, but I could be wrong.
Brown, could you post up a short sample pre-processed clip on one of the file transfer sites? I'd like to take a look at it and see whether I wind up with the same issue that your seeing.
If DGIndex is causing the problem, then that would be irregardless of the which codec VDub is being told to use. It would have to be something in your template.avs file, at least best that I could guess offhand.
--Steve
Steve,
I was able to do everything manually using both Lagarith and Uncompressed codecs without problem. It does look stanning!
Now. What I might be missing, when I try to do it with the .exe?
So you manually ran the DGIndex portion as well as the VirtualDub parts? Can you post up the .avs file that was created out of DGIndex using your manual steps and a copy of one that was created using the automated process?
All the exe does is start each of those two apps and feed in the automation scripts to get them to do the appropriate actions. The exe just coordinates and batches those processes. It doesn't do anything to the videos itself. The closest it comes to doing that is updating the .avs output from DGIndex with the appropriate DelayAudio setting.
--Steve
Sorry for the dumb question, but why when shooting with pal camera this doesn't apply? And what would be the workflow then? And, what's the workflow if shooting in HD, but want to work in SD (my pc can't handle HD, but i've heard you can improve footage by shooting HD and downsizing to SD).
Is there something like 25p SD?
Garbled screen potential solution...
To Cahrdr:
Using Steve Szudzik's nice little application, I got results much like yours. Turns out I forgot to install either the Xvid or DivX codec on my new PC (both of which, from what I read, can decompress YV12, which seems to be required during the process).
If you have the same problem as I did, when you try to open your m2t file in VirtualDubMod, you should get the warning message:
"Couldn't locate decompressor for format YV12 (unknown)"
In my case, simply installing Xvid solved my problem.
Hope this helps.
why is it that the outputted video, ends up squishing my video length-wise?
We're talking that over at dvinfo right now. Our guess is that it has to do with the pixel size being set to square pixels instead of 1.333. One solution would be to re-render out the large AVI files into a smaller file (.m2t etc) and make sure you setup the Vegas or Premier project to specify the pixel aspect ratio of 1.333.
I'm looking to see what I can do to help force this through DGIndex/VDub.
Yes! Installing XVID fixed my scrambled image issue! Being that I just built this system, is was completely bare. For whatever reason, it uses some aspect of the XVID codec, but it is fixed now.
However, it does have that aspect ratio issue when playing it through Media Player Classic. When loaded into Vegas, it indicates and loads fine as a 1.33 pixel aspect ratio video. In the windows property, it reflects the 1440 x 1080 resolution, so I'm fine with it for now.
Thanks,
cahrdr
That's great to hear. I'll make sure to update my site to reflect the installation of XVId (or similar codec).
As a side note to an earlier question, can create your own "compressor" template for the automation exe that uses XVId directly. Follow the instructions on my install page for creating your own .SYL script for VDub. If XVid is installed, it will be one of the "Save As" options from VDub and you can script out the job from in there.
--Steve
Yeah I've been following it, the temporary solution of resaving the avi through vdub does work indeed, but yeah, annoying having to do it twice.
Installing XVID fixed the garbled video problem, but now I have another. It seems like there is still a lot of ghosting in my clips after running the SSzudzik program. I've got everything in the right spots and everything linked correctly (as far as I know). The only difference I see in the output video is the rescaling of the image size and some jittery camera movements (on panning shots, looks like the camera is jumping slightly instead of a smooth pan). Some other movements look jittery, such as people's mouths when they talk (mouth open to mouth closed, with no half-open mouth in between).
Is removing pulldown on a video with too much movement, not covered by this method? Is it impossible? Or am I just not doing something right?
Thanks, and keep up the good work.
Posting Permissions
Reply With Quote