Never encode while capturing, Always capture lossless and encode later, H.264 is way better than MPEG-2.

For Dolby D and DTS, there is a better way of doing it without having to go thru HDMI route, Just get a multichannel audio capture card and while capturing select video from the video capture device and audio from the multichannel audio capture card.

Another way without having to use a separate audio capture card is to get a SDI audio serializer/injector box that can be inserted in the SDI stream (it has SDI in + a separate audio in, and SDI out with audio embedded in video).

Capturing multi channel audio can be a pain and I think only major NLE software support it, I don't think Vdub captures videos with multichannel audio but I hope I'm wrong.

H.264 isn't better than MPEG-2, it's just different.

I can see why LaserDisc would need a TBC. Its still an analog format and motors are far from perfect. I know some units have dropout compensation, so a TBC wouldn't be out of the ordinary, particularly when they dropped in cost in the 90s. There would be less time-base error on a LD then a tape based format though since you aren't dealing with a tape transport with far more slop in it vs. a spinning disc. The tape itself is also subject to stretching and deforming, further needing time-base correcting.

Yes they can be both at the same level of quality but MPEG-2 will be 10 times or more larger in size, That's what I meant to say. The compression algorthims of newer codecs are much more sophisticated and efficient than the ancient MPEG-2 which makes them yield a better picture quality for the same bitrate.

It's not just about bitrate, or "efficiency" (which is honestly a meaningless term).

One of the main differences, aside from file size, is how the video will artifact. MPEG will have mosquito noise and blocks, while H.264 just smears it. Which is best becomes subjective.

Smaller bitrate also isn't necessarily important. There are valid uses for both, just as there are valid uses for uncompressed, lossless, H.265/HEVC, etc.

Video is scenario based. Workflows, end goals -- those matter, and determine formats.

It's not a simple matter of "newest = best, we shall use that". Age is really inconsequential for many uses.