@Zibelas "Adoption of a parasitic mode of life has occurred repeatedly and independently more than once in many groups" (source). I'd argue the same thing happens in tech and perhaps more generally in society even.
@DMGregory regarding your suggestion thanks for the vote of confidence - I'll see what I can do. I haven't used RPG Maker, so I was reluctant to try to field a full answer. I'll look it over again & see if I can pull something together.
Regarding the follow question in the comments about limiting the area of influence on a front line: is there enough going on there to be its own question? I kinda think so, but I wanted to hear what others think.
I've often heard people I respect say LLM output should be treated sort of like undergrad interns: it can do a lot of ground work for you, but you shouldn't treat it like a finished product - you've accumulated expertise that needs to be used to vet the work.
Just ran into the above error again when I tried to go to meta. Errored out a few times & then it worked. According to this it's a generic message from the server side Tag Engine which may or may not be normal.
But so far, things seem to be tilting the desired direction. I wasn't sure if a) pushing things from the CPU to GPU would pay off in terms of speed or b) I should be using shaders or something more general like OpenCL.
So then last night I'm prodding at it when I finally realize: the shaders were expecting the textures as uniforms, but the CPU code never sent them as such. Like at all. I have no idea why I was able to sample from the first texture.
By my second half day, I get the nagging suspicion that maybe things that seem like they're working are not, that my some of my incremental success was accidental.
I thought maybe I was using the wrong part of the FBO, or not binding correctly & spend a half day of trying stuff, changing code block orders, dumping buffers to bitmaps and other debugging.
I'm fumbling my way through modern OpenGL / GLSL via OpenTK. I thought I had a handle on how to sample from a texture. But when I try to sample from two texture, my shader code seemed unable to pull anything from the 2nd texture.
@Zibelas Yeah, it's not really relevant to me either. Some of my friends have been hearing their higher ups tilting toward it. Their profits & metrics are up though, so it's hard not to see it as something other than a management / C suite flex.
> Use what is easiest for you, but be aware that any choice is arbitrary and that you may need to convert between them and use multiple coordinate systems within one program.
lol - true. I did think that was one of the silver linings to the whole mess. I've been hearing of 405 type regression errors with that feature though :/
Some parts of it feel opaque. I'm trying to figure out how SHADERed works + compute shaders + GLSL at the same time, so it could also just be that I'm lacking in mental scaffolding.
My OpenGL know how was from the era of glBegin & glEnd & as far as shaders are concerned, right now I'm frequently saying "dər", so I guess I'll go with that last one.
I'm using Silk.net which as far as I can tell defines shaders as strings where they don't seem to benefit from any of the usual quality of life perks from VS - ShaderEd is (so far) filling some of that gap.
Might be premature - we're only a couple of days into using it, but I wanna toss out a mention to SHADERed - it's kinda like VS but for shaders. I like the dev tools & the ability to hammer out drop in ready code.
If multiple other users flagged it as spam, I'll accept that I the consensus is was spam. If it just had one flag & multiple down votes for being a poor question, I'm less accepting of the outcome. In retrospect, I should have popped a comment sharing my "this is weak, but not spam" take before hammering out an answer.
