Needing alittle help with my avi capture enhancement.

[friendtoall]

I don't know what I'm doing wrong and need help understanding what to do with a file that has been capture with virtual dub. This is how it happen, ok I read all the how to capture video for virtaul dub and it went ok, the thing I did wrong was the resolution with made my capture long weight and short in high but it works and the file fisinhed at about 460 MB. Now I read the how to process video with all fliter needed and save as avi, the file came out to be 10 GB. All I want to do is to enhance file and transfer to DVD. I read the how to compression but it was for the VCD and I want DVD quality. What is the standard or easiest way to capture an avi file, enhance, and transfer to DVDR? Thank you for all members help.

[Hiro2k]

Well you should capture to an Uncompressed Video Format first. This file will be huge! Once you have that AVI, you can then add fitlers and encode it into an MPEG2. Once you have that, you can make your DVD from there. Also save your audio as Uncompressed WAV so can convert that to DVD audio later without losing quality.

[jggimi]

Hmm. You're capturing with Vdub, and hopefully using an appropriate codec such as MJPEG (can be configured to be nearly lossless) or HuffYUV (lossless), to end up with 460MB for a short capture.

But when you re-encode into AVI again, you have not chosen a video codec for compression, and are producing uncompressed AVI, which is why it reaches 10GB.

Lossless codecs such as HuffYUV do consume a great deal of disk space, but they will consume a lot less than completely uncompressed video, which is what I think you might have ended up with at 460MB -> 10GB

Unless lossless AVI codecs are used for both capture and editing, I would not recommend AVI->AVI->MPEG2, as you seem to want to do. That's because of the loss in quality taken at each encoding step. If you are limited in disk space, and need to use a lossy codec for capture, then I recommend AVI->AviSynth->MPEG2 instead, where the captured video is frameserved to your MPEG-2 encoder.

You've mentioned you've been following some "how to capture" instructions, but you don't mention which ones. Doom9's Analog guide (also called the Capture Guide) was written by a number of people who spend time in the Capture Forum, where you'll also find a Capture FAQ.

The Capture Guide does not describe how to create DVDs from captured information. However, one can use an AVI or AVS as video source in most MPEG-2 encoders. For information on AVI->DVD, there are two guides. Doom9 has an AVI->DVD guide, and bb has written a good one on DV->DVD which is a sticky in the DV forum. DV is usually managed as an AVI codec on PCs, which is why that guide may apply.

[friendtoall]

Thank you my friend, this is my first time use virtualdub and processing. I been using WinDVR all this time which been working very good for me. But I have been reading that in order to enhance a capture video you will have in rencode to avi and enhance then back to Mpeg2 and on to dvd. Windvr does a great job of capturing and I never had a problem but on some VHS Tape that I want to back are poor in quality and need some enhancement. I will read more about the dv to dvd from the dv forum. Agian thank you so much for all your help. I'm very thanksful to be a member here.

[friendtoall]

Man, I feel so stupid and my brain is on fire from all this reading. But still I'm not getting the point on how to do a High quality DVD capture and enhancement. I have look at the process of going from capture avi to avisynth and on to DVD. I'm lost, someone please help me find my way back home!!! I have a few detail question now since I have been reading the guide on capture in the guide area. First of all let start by saying Thank you. Ok, this what I want to do is to capture my old VSH Tape in high quality and transfer to DVD. Knowing that VHS has very poor picture quality and high in niose, is it better to capture video with all filters needed and then to tmpgenc or is it better to capture video first with no filters and then process with filter after the capture? Also I have been setting VirtualDub according to the guide and in capture mode at the location Video>Compression the guide said to select the huffty codec, but all I have was the mpg 1 to 3. If I'm missing the codec then way can I find the codec to install or is it becasue my Nvidia capture card don't support any other type of codec? Plus reading the guide for AVISynth was killer to me, it took me on and on from one website to another and still to this moment I don't even an ideal how to use it. Can't posprocess in avisynt so must use Virtualdubmod, load up avisource and filter and so on. But were is Avisynth, wasn't I post to load up avisynth to enhance video quality? I tell ya what, I little a really stupid newbie. I know how to backup DVD Movie very well but this capture, enhancement, and transfering got me out of wact.

[Hiro2k]

I don't suggest you use the filters while your capturing unless you have a godly PC, because your CPU won't be able to keep up and will start dropping frames.

You can find the huffy codec on doom9's download page under codecs.
http://www.doom9.org/Soft21/Codecs/huffyuv0.2.2.rar

[ultimatebilly]

Man, I feel so stupid and my brain is on fire from all this reading
I'm here quite a while, and I feel the same every few days :)
I'm no capture expert at all, only did it once...
So I can give you no specific advice for the avisynth-filters you should use...
But I can answer a few things, though.
You can get the Huffyuv-codec here:
http://neuron2.net/www.math.berkeley.edu/benrg/huffyuv.html
Get the pre-build dll of the actual version...
Note that you will need loads of harddisk-space, for 90 minutes more than 20 GB I think, but I can't tell you for sure...
Another choice is MJPEG, as jggimi already stated.
When choosing high-quality levels, it will still give you very good quality, but not as good as huffyuv, so use this if you've got the hd-space.
You should consider getting the huffyuv-update from doom9s download-page (click on show all codecs).
Copy the .dll over your existing .dll (rename it, so it overwrites the old one). If you can't find the .dll, you can use the search-function in windows.
is it better to capture video with all filters needed
No. This would put extreme loads on your machine, and you would risk dropping frames.
Additionally, it makes no sense, because you can just as well filter in avisynth.
I would, however, capture in full resolution: 768x576 for PAL, (640x480 for NTSC, not sure about that, read the capture guide for the resolution you should use...).
This will give you the best possible source to work with later on (the same reason it is better not to filter when capturing, you can't experiment with filters and can't do anything if you don't like the results of one/a combination).
Many people will say that this is useless, because the resolution of VHS is supposed to be no more than 352x288 resp. 352x240, but I think better capture more than there is than less.
Now to avisynth:
Avisynth is a scripting-language, you create a text-file with the functions you want to use, and it processes the clip you specified according to that.
You have to save the text-file with the .avs-extension.
You can open that in TMPEG or CCE, and the video will be processed and frameserved to TMPEG or CCE.
Frameserving means that the video will given frame by frame to the encoder, without needing to be compressed and recompressed.
This gives you much power, because you can process the video on the way, and do an unimaginable lot of things with it.
The syntax for opening an avi-file is this:

avisource("c:\mypath\myclip.avi")

If you put that in a .avs-file, and open that in VDub, you will see that the video is frameserved into VDub.
Now for DVD I would do the following (I will described it here for PAL, because I don't feel too sure with NTSC, learned everything about it, but forgot, because I never have to use it).

avisource("c:\mypath\myclip.000.avi")+avisource("c:\mypath\myclip.001.avi")+....
loadplugin("c:\avisynthplugins\PeachSmoother.dll")
SeparateFields()
PeachSmoother(NoiseReduction=40, Stability=20, Spatial=100)
Weave()
LanczosResize(704,576)
addborders(8,0,8,0)

Line by line walk-through: ;)

Avisource is used to load all the segments.
Note that this is probably the most inelegant way to load the multiple segments VDub created, because there is a special function for that purpose, but it works just fine and is easier to remember, for me at least.
Loadplugin is used to load external filters, which are not included in avisynth.
I can't give you much information about the filters to use, because, as I said, I'm in no way a specialist on capturing.
But Peachsmoother is a very good noise-remover, and has the advantage that it can be used on interlaced content (which is what you get from VHS).
Seperatefields lets Peachsmoother process the fields on their own, which is important for it to work correctly (otherwise it may misunderstand information from the other field as being noise).
Peachsmoother: These values are very close to the default values, can't give you better values, read more in capture-area, use the search-function, and read the filter documentation.
You should also look into other filters, but make sure they can handle interlaced content.
Weave is used get the processed fields back together.
LanczosResize is a very precise resizing method, probably overkill, you should probably use bicubicresize or bilinearresize here...
I usually always scale DVD's (PAL!!!) to 704x576, because I'm one of the people who believes they will loke more the way they are intended on TV's then (but that's a different story).
I use the addborders-function, to get a final resolution of 720x576, even if 704x576 is DVD-compliant, I feel safer this way.
You can just as well resize directly to 720x576, and omit the addborders-stuff.
For possible values to use, especially if you have to process NTSC and cannot use my values as a base, you can also orientate on the VDub- and TMPEG-profiles (and use the search function).

That's it from me for now, I hope you can get some useful information out of it, and don't get demoralized.
All this avisynth-stuff may seem very confusing, but every headache about it is worth it a thousand times.
You have the chance to get the best out of your captures/encodes, and probably learn more about video-encoding than you would in most universities.
Don't give up, this isn't easy stuff, but you will master it if you are willing to invest some time...

[friendtoall]

Thank you for all of your input and information. I'm very glad to be here as a member. I think I proibly jump the gun when I though that I know how to backup my DVD movies and though it was very easy to capture and rencode movies. May I ask if all you pro can list me the basic step in which I should read and learn about the basic of video and video editing so that would understand what to do at serent point of video editing. I know there is alot to learn but one can't enlighten if he is to jump from one to ten and leave the step in between. Ulead make it easy to capture to dvd but offer almost no power to enhance. There for I must learn how to edit video the long but sure way. Thank you you all for your help and sharing information.