> > as well as video > > codecs capable of preserving the video output in a way that can be read, > manipulated, > > and displayed true to the original hardware. > > As soon as you start talking about capturing video then you've failed. The video > output of a laser disc player is generated by the player, using sub standard circuits > which haven't gotten better with age.
You're essentially talking about the difference between frames read directly from the disc and the picture output by the player. Yes, it's understood that those are two different things.
However, given that the end goal is to transfer the analogue frames on the disc to a usable digital medium, there will always be some form of alteration of the original out of necessity as well as as part of the analogue-to-digital conversion process.
Additionally, no digital format that I'm aware of has a concept of picture fields in the same way as NTSC (or PAL, or SECAM) does, so even taking a direct read of a frame from a disc without further analogue processing would still require combining those fields into a single frame for digital capture.
Again, my understanding of how NTSC (and other TV standards) work at a low level is somewhat limited. But unless I'm misunderstanding something in your reply, there's no way to do this without having to process the video between reading it from the disc and storing it digitally using {insert codec here}.
> It may not affect arcade games, but it's > possible for discs to have audio stored in multiple formats and capturing all of them > when thinking about video capture is not going to be easy.
Also understood. But arcade games are the focus here, not retail discs - and nothing is currently in place (that I'm aware of) to back up either type of disc except for player output. Now, if something is developed to back up arcade game discs that can later be extended out to the more unusual retail titles, great, but retail titles aren't MAME's focus.
> Capturing the signal from the disc, before it's been degraded by the laser disc > player is the therefore the only practical solution. It's been proven to be possible, > it is just going to take a while for the gears to get in motion.
Sure, it's possible; nobody's suggesting otherwise. But if we want to get really nitpicky about it, we should be reading the video from each disc using only the players originally used by each game.
The thing is that while we're both in agreement that post-read processing (passive or otherwise) doesn't create an ideal output signal, we don't have "a while" to wait for the ideal solution. This is one of the reasons why I'm advocating that we need to figure out what constitutes acceptable losses and back up to the best of our present ability.