Who needs HDR?
Credit : https://m.youtube.com/@JimMusilPainter/shorts

@troy_s fantastic video as always but unfortunately he never elaborate the statement in the thumbnail.

@yiningkarlli "rendering at 2400fps" 🤦‍♂️

@demofox @bitinn Can you imagine the brain rot required to use the same kind of software that upscale images to make critical funding & military decisions?

@bitinn separate pass, but only if they can explain what is color.

@troy_s 0-1 considered limited.

@bsrgb @troy_s FYI you can stretch pre-formation too, it's just a bit hidden. in the "Pre" section, play around with pre_scale_a and b.
The fun thing here is that you can pick any axis to do this transformation by changing the pre_neutral_axis. The a/b scale control will squash the plane perpendicular to that axis.

@ScriptFanix @troy_s let me know if I can help. That tool's UI is a bit of a mess 😆

To get started correctly, go to the Generator section and pick "image" as the generator. You will then be able to use the "image" drop down to pick between three source images.

You can ignore the pre and post sections for this experiment. In the AgX section, you can play around with the rotation and compression triplets.

So I just remembered our family CRT monitor used to have two little bullet shaped plastic things on the top left and bottom right corner. They were held by velcro.

I think my dad told me they were like privacy screens but better, but I don't think I ever believed him.

Does anyone know what I'm talking about?

When in doubt, just cull by distance.

@petrikas @apq people that retire because you retire 😆

@apq @petrikas I thought about FIRE a lot and it's a good target. I often hope to be poor/rich enough to do this.

But I read a post the other day that made me question the idea : all of your friends will be working while you are "free", and people that are retired are not in your age group (e.g. 65+). You'll have to find a new social life, or work.

Which is fine! But not something that comes to mind when first imagining this lifestyle.

@troy_s @TonyVladusich OK I think I see two things that went wrong here :
- I say "stimuli" to sometimes mean a single pixel, but often as the entirety of what hits the retina. I don't know how to express those two concepts distinctly. Stimuli field?
- Utility of colorimetry as stimuli encoding vs. perceptual model.

Any perceptual model will have to be based on the stimuli field (i.e. above). Practically, the starting point of this model will most likely be colorimetry-encoded data.

Yes?

@TonyVladusich @troy_s I totally agree.
Although in reality, I believe colorimetry to be slightly more useful than a broken clock. As stimuli encoding, it seems fine?

@TonyVladusich @troy_s You already know I will buy a copy of that book.

My argument here is not "derpy derp color is outside".
My argument is "shit model sometimes right at midnight on a full moon is better than absurd shit model that is always wrong".

@troy_s if I pick a paint tube, and it's labeled red, and I open it and it looks red, was the label a good predictor? Is the label nonsense?

@troy_s the point is that the colorimetric model has some non-zero value as a predictor of color.

Otherwise, if it was *complete* nonsense like using the number of poppy seeds on my bagel as a predictor of color, then folks would have a hard time doing anything with the model.

Folks open Photoshop, picks (1,0,0) a draw a line, and it looks red. And the next day they do the same, and it still looks red. That's useful and absolutely not nonsense.

@troy_s it's really not. Only color fuckers like us argue about this crap. Muggles just go (1,0,0) = red = apple and go about their day.

@troy_s As color is (in part!) constructed from stimuli, any perceptual predictor based on stimuli will land somewhere on the accuracy dartboard.

Is (1,0,0) red? Sometimes. Quite often I would even say. Not always!

But the most damning thing is, if we want a better model, it will have to be constructed from the stimuli data, as we don't have anything else to manipulate and inspect.

@troy_s Fine! I'll pick the number of distinct letters in my alphabet soup bowl.