640x480@30fps is quite a lot to handle if images are changing each time, the decompression *might* not be keeping up if the machine has a low spec...
Since you're talking about quality settings, you're only single pass encoding the source. It would certainly work better if you capture the raw video and then two-pass encode it. This will give best results especially for quake footage where each separate frame will be different from the last.
I might suggest getting the latest XviD which is 1.0.0RC2 or something similar.
Stuff to try / explore
- Set the maximum key frame interval to something much lower, like 30 frames maybe - which will be one every second for you. You could even try like every 15 frames.
- Set I frame percentage boost a bit higher (20%)?
- Set target bitrate to something pretty high like 1500 for the second pass. Higher still, if filesize isn't too much of an issue for you.
If you need more info on other settings let me know... when I get time, I'll write up a bit of a guide on XviD encoding. It might make sense if this coincided with the release of XviD version 1 final.
Regards
-Fruey |