I’d like to see more shots of that if/when you get the time, some good close-ups would be nice!
I think I’ll call this one “Arcade19XX”
I’d like to see more shots of that if/when you get the time, some good close-ups would be nice!
I think I’ll call this one “Arcade19XX”
Monitor is on ‘standard’, which is the closest the monitor settings has to srgb. I also calibrated the monitor with windows built in settings.
The second shot looks perfectly fine on my monitor. I had to bring the white point down from 6500 to 6000, as it was too blue for me.
Idk, all your shots look way too yellow on every screen I’ve tested and too yellow compared to everyone else’s shots, so maybe the problem is with your display? Have you used a monitor calibration device?
I have absolutely no idea. The monitor is fine as is, even on factory settings your shader was way too bright.
The only settings my monitor has are basic brightness and contrast and separate RGB sliders (all at 100)
For reference, my monitor is a Dell E2220H
Have you used a monitor calibration device? There might be something going on with the display that’s throwing off the built-in profiles.
It’s just that your shots look too yellow compared to everything I’ve seen- not just my own shots but other people’s, the raw colors without shaders, and the real thing as seen on a CRT.
I don’t have one of those. I just used the built in windows one and the lagom monitor test pages.
Ive just used the default setting until today when I calibrated it
I’m guessing it’s your display. That’s a TN panel without full sRGB coverage that likely has the gamut way off so that yellow is oversaturated compared the other secondaries which means it will look unbalanced.
If that’s the way you’re used to seeing the image then it will look fine to you but yeah it definitely looks way off to me as well.
Ah, that’s why.
I really didn’t pay too much attention to the specifics when buying it. I bought it earlier this year because I needed a bigger monitor (my last was an 18in 768p monitor). All I really paid attention to was screen size and if it was 1080p at least.
So there really isn’t going to be a shader that is good for all monitor types.
Yeah TN panels have really bad image quality unfortunately. Definitely recommend upgrading to at least a decent IPS LCD if possible. OLED or maybe QLED would be even better. Plasma is good too but the last ones were manufactured in 2013
I’ll keep that in mind for the next monitor I get. At the time this was the best I could afford (got this one for $90), and it really didn’t matter to me personally. I just needed a monitor with a not terrible resolution.
It’s fine enough for my needs. Looks pretty good and is a modern resolution, plus I can finally do decent integer scales and actually use crt shaders (compared to my last one, where the best I could do was 720p/672p.)
Switched to scanline type 1 and setting scanline 1&2 saturation to 1.00 gives the colors a bit more punch. Original settings here
EDIT: I think I still prefer type 0 as the dark scanlines can look a bit unnatural with type 1 and this mask. Anyway, they both have pros/cons.
scanline type 0
scanline type 1, scanline saturation 1.00
Didn’t know where to put this, but this is an interesting video about studio monitor calibration in the US.
Not much detail about how to clean the innards, but believe it or not you can just hose all that stuff off as well.
I think 1.00 for dark pixel bright boost in guest-dr-venom is probably the no-op setting:
I read somewhere that when using crt-royale-kurozumi the most accurate mask triad size desired at 4K is a value of 3.
I’m using a 55” display at 2880 x 1620, which I believe is 3K.
Does anyone know what mask triad size desired value I should use?
nope. no change. I have several commits just sitting in the queue.
Lookin’ good! Are you reusing your xBR pixel search and just looking for different patterns?
No, it has nothing to do with xBR.
It’s based on my jinc shaders. I’m exploring some properties of that function and it is very well suited for getting rid of dithering and vertical lines (fake 16-bit transparencies).
The good thing: it’s very fast and for transparencies it’s only one pass at 1x as first shader. For only dithering, it’s 1 or 2 passes (a sharp version needs two passes).
oh nice. Yeah, I think there’s a thread somewhere here on the forum about using jinc2-sharp(?) with some specific parameter values to smooth out dithering very effectively.