That looks pretty neat! I’m sure the screenshot looks noticeably different from the live preset in motion due to the use of noise. Have you observed this as well?
2 Likes
Well, do keep in mind everything I’ve said thus far applies ONLY to VGA games and CGA/EGA games running in VGA compatibility mode. Most CGA and EGA games originally ran at 320x200@60Hz, and so were proper 15KHz standards (the exception being EGA’s hi-res 640x350 mode, which was 24KHz), and so were their respective monitors, so it is not at all incorrect or inaccurate to display those games with 240p shaders or custom modelines on real PC CRTs (they will look quite a bit sharper than through their original period-correct monitors, though, as those were often essentially glorified TVs with RGB input).
However, once we got into the VGA era, even though VGA hardware was backwards compatible with both CGA and EGA, due to VGA monitors no longer being able to handle 15KHz (outside uncommon triple-sync monitors that could handle 15KHz, 24KHz, and 31KHz), it was made so CGA and EGA modelines were line-doubled and displayed at 70Hz refresh rate, which may introduce judder, as you might imagine. It wasn’t accurate, but hey, it got the job done back then! This is likely how many (most?) people remember and played CGA and EGA games, as many were often packaged on those old shareware CDs that made the rounds throughout the 90’s. Complicating things further is the fact that many DOS games actually let you choose from a variety of display modes, including CGA, EGA, Hercules, VGA, MCGA, Tandy, and PCJr. Prince of Persia is one such game, so honestly, it’s pretty much fair game to display it at 200p if you are so inclined. 
2 Likes
Oh the still shot definitely doesn’t do it justice. Seeing your 70s RF presets in action on my home laptop (which runs it way better than my work one) gave me the inspiration to try similar things as a learning experience 
1 Like
Just to clarify, I’m perfectly aware of all that. None of those pictured games by me above are VGA games. The gaps in that second PoP are just so large, I never recall any game looking like that, and indeed it looks on my monitor more like the first pic. Might be just another sign of it’s last days, it certainly used to be sharper (and brighter, but part of this is due to the photo.)
If I recall correctly, that picture was off of an aperture grille 19" LG Flatron, which explains why the gaps are so large. It’s probably TOO sharp for these old games. I like the look of your monitor better, even if it’s seen better days.
That’s why I started posting video clips. I first capture to H.264 using Geforce Experience then transcode to H.265 using Handbrake. You have to be really careful with the bitrates because too low a bitrate can cause the noise to just look smeared.
My video game recording and encoding settings:
GeForce Experience
I started with the High Quality preset then clicked Custom and set bitrate to 95Mbps.
Handbrake
I started with the:
Web - Vimeo YouTube HQ 2160p60 4K preset
I changed the following:
Video Encoder to H.265 10-bit (x265)
Encoder Preset to Placebo
Encoder Tune to Grain
Quality to Constant Quality 27 RF
I also discard the audio when I’m transcoding.
1 Like
Those things deserve to be seen on a TV! Lol Do you ever hook your laptop up to your TV for gaming?
1 Like
It’s been quite a while since I have - before I got much into shaders and bezels and the like. I can possibly try this weekend, though.
1 Like
I have a couple of screenshots using @guest.r masks 7 and 8 along with the white pixel active. First I should say that screen looks brighter and whites look white instead of grey (wooh!). Mask 8 has visible RGB (and white?) while 7 just looks greyscale and bright. I like 7 with scanlines 2 and 8 with scanlines 0.
using cool color temp and tweaks borrowed from @Nesguy Also, I may have deconvergence on.
4 Likes
Yeeeah that phosphor shape and definition still leaves something to be desired IMO.
On an aperture grille CRT (careful not to say “should!”), you never have phosphors that are horizontal rectangles, they’re always vertical rectangles or tiny squares.
I’m not really seeing distinct RGB phosphors, either. Looks like subpixel artifacts to me.
I think mask 7 looks better in this situation but maybe needs more strength over bright colors so it doesn’t fade as much. Just my two cents.
Oh and deconvergence will definitely interfere with any RGB mask, especially the aperture grille ones. It’s better used with black and white masks.
3 Likes
Plain scanline.glsl with CGWG mask, should run on any potato/toaster you have. It even runs 70-80 fps at 768p on absolute garbage Intel GMA 3600.
https://mega.nz/file/Svg1QCzL#Cm9MucvV6Vj6AXzkvyRvbHWwWrp4RPBUbzr8QDSQP8c
5 Likes
What is the mask of this monitor?
Notice the presence of what looks like ringing artifacts in parts of the image. The ringing that occurs with subtractive sharpness 1.00 and horizontal sharpness 3.00 isn’t any worse, as far as I can see.
3 Likes
are you talking about the area near his eyes and mouth? And at the edge of the wall?
UPDATE: I was finally able to get in touch with @Hyllian and he was kind enough to send me the latest version of his crt-super-xbr shader and allow me to reupload it for anyone interested.
for screens see Hyllians previous posts:
9 Likes
It’s most noticeable at the right edge of the sconces and the rightmost edge of the character. The edges aren’t completely clean due to one or more beams being slightly msaligned.
2 Likes
The manual calls it an “anti-doming invar shadow mask”.
1 Like
Check the difference a simple shader (fake-crt-geom with image-adjustment.glsl as 1st pass) can make
10 Likes
Holy cow that is an amazing difference. Not to mention that is a very cool website.
8 Likes
We all need to be using that site, that’s some spicy fire lol.
6 Likes