I really enjoy the results I get using your templates, Kwag! But I've noticed the color fidelity isn't quite as good as I like. A few simple RGB adjustments, and maybe brightness/contrast controls would do the trick, what do you recommend (I would like a simple drag and drop interface, if possible, and I'll pick up any retail software you recommend)?

Not sure about that Reno. If your source is DVD, the KVCD you make has exactly the same color space/brightness/contrast. Every KVCD I've made using the x3 template, looks almost identical to the original DVD. Quality wise and color wise. If your source is other than DVD, that could be the cause for the difference.

-kwag

My main sources up till now have been Divx Avi's. Could that explain the color problem?

And now that I look back at my source files, the color info does appear to be the same. Any way to improve on that?

It does end up the same colourspace (YV12), but goes through numerous transformations on the way. Usually this is:

YV12 stored on DVD
MPEG2DEC reads YV12 from DVD, converts to YUY2
Avisynth processes in YUY2
VFAPI plugin reads YUY2 from Avisynth, converts to RGB
TMPGEnc reads RGB from VFAPI plugin, converts to YV12
YV12 stored on VCD

Each conversion introduces errors, except for the initial YV12->YUY2 if it's done right. This is one of the reasons people are excited about Avisynth 2.5, which has native support for YV12. With it, the process would look like:

YV12 stored on DVD
MPEG2DEC reads YV12 from DVD
Avisynth processes in YV12
VFAPI plugin reads YV12 from Avisynth, converts to RGB
TMPGEnc reads RGB from VFAPI plugin, converts to YV12
YV12 stored on VCD

Which of course is better but still not perfect. This is the ideal process, as used by most 2.5 testers right now:

YV12 stored on DVD
MPEG2DEC reads YV12 from DVD
Avisynth processes in YV12
VDub mod reads YV12 from Avisynth
XViD reads YV12 from VDub, compresses
YV12 stored in MPEG-4 file

Perfect colour fidelity. Time for us to petition for support for YV12 source material in TMPGEnc .

Thanks Sansgrip.....

Do you know of a way to correct after the conversion has taken place (other than adjusting the TV )??

The only way I've tested and know for a fact maintains correct colour ranges when using TMPGEnc is:

1) Use the ReadAVS VFAPI plugin to open the .avs file -- uncheck all other plugins in the TMPGEnc config;

2) Avoid as much as possible extra colourspace conversions within your script;

3) DO NOT CHECK "Output YUV data as Basic YCbCr not CCIR601"; and

4) Make sure Avisynth is outputing YUY2.

If you do this you will at least maintain the correct ranges (i.e. pure black will stay 16, pure white will stay 235) all the way through, and errors due to conversions will be minimized.

It's probably possible to achieve the same thing with a ConvertToRGB at the end of the script and by using the VFW-compatible VFAPI plugin, but I've not tested it to see whether "Output YUV data as Basic YCbCr not CCIR601" should be checked or not in those circumstances.

You know, you just triggered a memory in my mind, about this user over at "Another forum" where he insisted on encoding directly from the VOB's He always claimed that it was the best quality/color fidelity, etc., and the moderators there slammed him around all the time and told him to shut up because he didn't know what he was talking about. I guess he is 100% correct on the color space issue, but the main problem is that encoding directly from a VOB you can't take advantage of rff flags to encode at 23.976fps.
And all of these discussions were PRIOR to AviSynth 2.5. So he has a point!. Here, have a laugh!: http://www.vcdhelp.com/forum/viewtopic.php?t=123172

-kwag

I just posted a reply in that thread .

Oh I can't wait for the replies
That site should be called VCDhell, not help, because most discussions end up in loose-loose situation for the users. Some moderators there just have a mind narrower than a pencil point

-kwag