Hi,
I have read most everything I can find from all over the Web and have been experimenting for a while now. I eventually determined that Kwag's KVCD template (NTSC) does not work well on my Apex 1100W with the 352x480. The audio has brief cut-outs every few seconds. I played with audio encoding settings first, but finally just went down to 352x288 and I *think* that fixes it. So now I am ready to just start capturing, encoding, and burning.
Most of what I will be doing is off DishNetwork, capturing movies off of premium channels, etc. I need a little help understanding the effects of different choices in capture codec and other settings.
I am using an AIW Radeon. Let's assume that Huffyuv is not a practical option (disk space), MJPEG may be sometimes. I am mostly interested in MPEG1 and MPEG2 captures with ATI MMC. I can do both at decent bitrates without major frame droppage.
My system is Athlon 1.2Ghz with 256MB DDR, 2 (20 and 30GB) partitioned ATA-100 drives.
Assuming I will probably convert mostly to VCD (maybe play some with SVCD) is it
best to:
Capture at high (4 or more mbps) bitrate VBR MPEG1
OR
Capture at high (" ") bitrate VBR MPEG2?
I'm using something like motion estimate 98-100 also. The captures look good with either MPEG1 or MPEG2.
I had thought that maybe MPEG2 would be better as it can capture interlaced. Is this an issue, though, when downsizing to 352x288 for VCD? And do I need to deinterlace when I am downsizing, or would I need to if I went to 352x240 instead?
OK, now I'm running into ENCODING stuff, but...
In encoding, I'd use avisynth or dvd2avi as a frameserver, for speed. Would I need to deinterlace *and* IVTC? And how do I know what would be IVTCed? Would television shows be original FILM, only live broadcasts being video, or not telecined?
I probably have other questions about this, but I'm not at my home computer now.
Thanks!
Geoff