MAMEWorld Forums - EmuChat - question in 2022 will be “how much storage capacity does your computer have? “ *edit*

> No doubt they will be better, but the "old" captures where extremely bad in all terms
> and considering what hardware where used 20 years ago, comparing to what can be used
> today. It should be mentioned, because it is easy to beat the "old" captures in terms
> of quality and everything else and easy to say that the current AVIs will look
> "significantly" better. I am curious how "significantly" better they will look on
> games, that look like shit already on the original source. That is interesting.

Why do you always have to roll into threads like these and be so negative about things that developers are doing? If you have a better solution, go on then.

Both the software and the underlying hardware designs are 100% open-source and visible. Go have a look: https://github.com/happycube/ld-decode

I've been saying for years at this point that having the raw RF-domain signal (note that this is an entirely different thing from "RF-modulator output" from an old player) off of the disc allows for everything from more-accurate dropout detection, to more stable time-base correction, to both NTSC and PAL color decoding that would not have been viable to implement in hardware during the commercial lifespan of the medium.

You're free to have a look at some actual, practical examples of the ld-decode pipeline in action, with the caveat that the video was uploaded to the site in October 2020, and was posted there about a year after the original upload to YouTube (which has since been removed): https://odysee.com/@Domesday86:7/Domesday-Duplicator-and-ld-decode-(rev4)-demonstration:c

That level of drop-out correction (the Garden Birds segment) represented the state of the ld-decode toolchain back in 2019, and with the advent of the ld-disc-stacker utility, multiple captures of the same master - but from different discs - can now be "stacked" in order to eliminate drop-outs that exist in one or the other (but not both), and results increase in quality with the overall number of discs.

Worth noting is that the majority of the remaining speckling in that British Garden Birds segment is due to the original film grain. When viewing the raw signal off of the disc, drop-outs tend to show in a much more obvious way - the signal tends to swing wildly above or below a valid range for representing 0 to 100 IRE. There's no possibility for a false-positive when it comes to drop-outs, but there can still be the possibility for the occasional false-negative.

The thing about drop-out detection (or drop-out correction) is that it's something that Laserdisc players had to contend with in real-time. The best you could usually hope for was that the player would continue replicating the values found to the left of the drop-out, or simply store the previous line in a delay line in order to duplicate it downward. If desired, the ld-decode toolchain can do a 3D search: Looking to the left or right on the same line, up or down one line in the field, or forward or back one field. Still, this is a half-measure, and disc-stacking is preferred in order to backfill known-good data rather than best-guess data.

> Will those lossless AVIs keep the interlaced content, or will
> it be progressive?

Remains to be seen. The likelihood of anyone providing an ffmpeg-grade deinterlacer that can run in real-time is unlikely, as evidenced by the fact that in the nearly 20 years since the first LD games were supported, nobody has done so. Keeping interlaced content would make the whole thing completely turn-key, but going with pre-deinterlaced content for visual quality purposes could be a better interim solution.

What really matters is that once the LDFs have been captured and backed up, and ideally enough LDFs from separate discs have been captured in order to stack up a dropout-free capture, that's that. The discs can rot into dust, there's no further signal information that can even be acquired, it's simply down to the progression of the software itself.

So, why so negative? Why so skeptical? It's really easy to talk smack, but you're not bringing anything to the table other than that.