Some of that code look familiar.
As you’re using the tweak the texture coordinates and use a single texture read approach you should provide a .glslp file so people will use linear texture filtering.
Some of that code look familiar.
As you’re using the tweak the texture coordinates and use a single texture read approach you should provide a .glslp file so people will use linear texture filtering.
Yeah I am using the filter from CRT-Pi, I like how sharp it looks. Yeah you are right about the linear filter.
A complete rework based on the same idea, due to initital not scaling well with integer more than 4x. I believe it should be really fast and good looking for what it needs to run.
Parts of this shader come from:
https://mega.nz/file/m6gUwKiZ#dUQXPSDwM9KP7Dn0vWlGpCjZ3hM0Sp_OHSEWlncZFY4
I’m planning on checking this out when I get a minute. You’re making me want to not be a potato
New version here for anyone who cares lol. Some unneeded things removed, added color temperature from guest.r-dr.venom. Don’t expect any miracles on overall brightness.
https://mega.nz/file/375wDDBS#gJSHAgZbPrGVuj1uY4S-bgLzGoIBTdCyDEjXJ393HsM
Imho, I’d have preferred if you had gutted grades white point/sat/gamma code instead of using guests white point code tbh
Nothing wrong with guests wp code, just prefer grades overall control over color.
Regardless, I love the progress and I’m hyped you continue to update it.
You can always use Grade as 1st pass, and yes it’s White point works better, plus many more options.
shaders = "2"
feedback_pass = "0"
shader0 = "shaders_glsl/misc/grade.glsl"
filter_linear0 = "false"
wrap_mode0 = "clamp_to_border"
mipmap_input0 = "false"
alias0 = ""
float_framebuffer0 = "false"
srgb_framebuffer0 = "false"
scale_type_x0 = "source"
scale_x0 = "1.000000"
scale_type_y0 = "source"
scale_y0 = "1.000000"
shader1 = "shaders_glsl/crt/shaders/CRT-Beam.glsl"
filter_linear1 = "true"
wrap_mode1 = "clamp_to_border"
mipmap_input1 = "false"
alias1 = ""
float_framebuffer1 = "false"
srgb_framebuffer1 = "false"
Ye ik I can just add grade, it just made my sense to me to bake it’s code into the shader vs using the other code that did less.
Still cool regardless, just throwing ideas at the wall ya know.
Thing is that is a huge shader and imo it would be better just add it as 1st pass instead of just gut it out and add it to a shader. I just checked to add “sega luminance fix” and that would require quite some job while you can have it in 5 seconds just by adding it as 1st pass. But some interesting things from other shaders could be added if they are light and straight-forward.
You don’t need the whole thing, give me a couple hours I have to run some errands.
When I get home I’ll gut the pertinent code and see what you think about the size and such.
It does a lot more than Guest.r’s code that’s why it looks better too trust me
(tons of options that affect each other)
I’ve worked/altered code in grade before so I’m aware
A small update with contrast added and off switches to speedup things if it’s needed.
https://mega.nz/file/e7xESTJJ#RVfrKqFdl34F2M6CjcH1LhxcmCxWZnqVEDRl0sBYpQM
As a bonus, included a modified old guest.r shader with things backported from this shader (and some back to their natural enviroment, like color temperature)
Holy shit that guest revision is a blast from the past.
I can’t remember the last time his shader codebase was that simple.
Yeah really simple but still good looking. I think that comes from 2018, and I removed some things too, like saturation function that wasn’t working properly. That runs 70-75 fps on my old HTC One with an 80 gflops GPU.
That old dog can’t do much now, but it’s slim with stereo speakers and 1080p, used entirely as a RetroArch machine
A small update, trying to save some brightness for normal monitors
https://mega.nz/file/3vJjnYqL#mEMwDXNE6OE_F6sv_4G82FaGKgUFhNF0kEETuUeVAL8
And a different profile for Guest.r shader to somehow match my Trinitron in scanlines, colors and brightnessAdded a new mask on both shaders (CGWG Slot)
https://mega.nz/file/SmJSgDYb#j7TsvtdaovzDcighgGRXCJ1gpgMKcRH0qlt38Zvs_ag