Here's the "workbench implementation" mentioned at the end (too much blur/bleed), nothing came of tge project it was intended for:

https://www.masswerk.at/rterm/VT/

See also this "fancy" variant: https://www.masswerk.at/rterm/VT/fancy.html

Some people just don't get the fact that CRT images were pretty sharp. LCDs were crap due to high persistence.

Old games look horrible on LCD monitors.

Yes, esp. monochrome CRTs can be actually too sharp. (E.g., I've a Sharp MZ-80, which I prefer to run with somewhat reduced focus.)

Something I found out lately, on LCD, you really have to model pixel "softness" in the time domain.

See for example here: https://masswerk.at/pet/?prg=back2pet&screen=green&persisten...

(This is a demo by shiru8bit, "Back to the PET", from last year, running in a "hot" demo mode with longer persistence, which is a bit reminiscent of a well known video showing this demo in a dark room with video and audio in overdrive. The problem with this is that persistence is still modelled linearly, instead of sinusoidally as it ought to be, for economic reasons, since it's running in a web browser.)

There's a shortcut to this and other recent demos: https://masswerk.at/pet/?demos

(2019)

Priors:

https://news.ycombinator.com/item?id=25523322 (18 comments)

https://news.ycombinator.com/item?id=19528618 (54 comments)

https://news.ycombinator.com/item?id=19207373 (0 comments)

I imagine there's some trim pots in there somewhere that can affect aspect ratio, too.

Obligatory Technology Connections video explaining how CRT pixels aren't really pixels as we understand them. Definitely worth the watch.

https://www.youtube.com/watch?v=Ea6tw-gulnQ