This comment by @tonroosendaal has been in #blender codebase for 17 years: "Waste of cpu cycles... but the imbuf API has no other way to scale fast (ton)"
Was trying to figure out why laptop of mother-in-law was very slow. It is 7 years old, 4GB RAM, HDD, so yeah. But it used to be not slow! Constant updates of both Windows and Chrome make everything sluggish. Even bringing up task manager takes half a minute. *One* app there that is still snappy? Picasa. Because it was discontinued before the laptop was made, so it had *zero* updates that would make it slower.
This is one of "all the apps automagically update themselves" reality downsides: it used to be that software you have stays the same, on the same computer. Now, computer stays the same, but software (slowly but surely) gets slower. Needs more memory, no longer acceptably works on non-SSD, etc. etc.
So everyone is still doomscrolling, eh? Isn't it time to move onto quakescrolling at this point? Or is that full 3D scrolling is too confusing to people, so they'd rather keep on scrolling in 2.5D.
And no mastodon/"fediverse" is not a sustainable and scalable answer. You still need someone to run the servers (costs $$$) and do moderation (costs human work and sometimes human misery).
Like, this instance running at all is a combination of Someone paying a $700 bill each.and.every.month, and about 5 moderators looking at reports and user registrations each.and.every.day. *So far* this has been working. Will it keep on working? Who knows! Forever? Probably not :/
My guitar teacher was talking about a whole-tone scale, but I misheard it as "Halton scale". I was: oh cool, a low-discrepancy sampler, but for notes. Now I'm disappointed it is not actually called a Halton scale.
@zeux Not here! My impression is generally that the period “oh our toolchain is ancient and/or strange” was 10-15 years ago. MSVC was stuck in weird state, some consoles had “strange” toolchains too. These days everyone’s either on clang or MSVC (and some on gcc), and all of those keep up with times well.
Remember Collada? Yeah the format is pretty much dead at this point, but Blender still has it (because of Sketchup, Chief Architect, Bricklink Studio etc.). However it felt increasingly uncomfortable to continue using OpenCollada (that embeds old 3rd party libraries with several dozen known security issues). So here's a "cleanup fork" of OpenCollada that gets rid of the security issues part (but also gets rid of everything Blender does not need). https://github.com/aras-p/OpenCOLLADA
@BartWronski ah, I remember a year or two ago you were asking around about Unreal jittered sampling and other “strange” texture filtering approaches. Now we know where all that went! Really nice!
@BartWronski there’s at least 10x effort (and prestige?) difference between a blog post and a GDC/Siggraph talk, but yes even a blog post is 1000x better than nothing. From personal experience though, “hey I found a gross hack!” the first instinct is to *not* write about it :) But of course you have no idea if your “gross hack” is actually a sensible application of a theory that has not been formulated yet.
@BartWronski your paper is a perfect blend of “proper literature” and “documenting gamedev practices” by the way. The latter is very often not well documented or even understood (I’m sure you are aware of a million reasons why :)). But it is curious that production environment sometimes stumbles upon actually sound theory by accident, without realizing it.