i love that in block game i'm agonizing about giving a stick a few more triangles when meanwhile minecraft is over here like "we're going to render dropped items using a 16x16 grid of tiny cubes :)"

i dunno if this has changed but last i checked every item slot in minecraft's GUI was also rendering one of these 3D items except that they're rotated so they look like 2D images and as far as i know there's no caching going on

meanwhile block game just caches 2d images to use as an icon for every item in the game at startup because uhh, hello?? why wouldn't you do that???

before i started block game there were lots of things in minecraft that made me go "that's weird why wouldn't they do it this better way? i guess the microsoft funded multi billion dollar studio must have a good reason for it"

and now that i've been making block game i am starting to strongly suspect they do not, in fact, have a good reason for it

@eniko I think this is a common thing! Developing software at scale necessitates streamlining, and the best way to streamline is to commodify the software process: remove custom code, cleverness, optimisation tricks, etc. and just rely more on brute force / volume when it's even slightly cost effective to do so.

Most for-profit software is unimaginably bad in ways you can only begin to imagine that just wouldn't fly if developed by individuals or in community settings.

@jsbarretto @eniko sometimes it's like that, but sometimes it's more that you have a "cleverness budget", and if you spend your cleverness budget on something which doesn't improve outcomes, then it's kinda a waste. brute force is often *faster* than clever, due to realities of hardware and compilers. (which isn't to say that's the case here, my gut feeling is that on anything moderately new it's not going to make much of a difference, but idk how well old gma hw deals with dynamic viewports)
@dotstdy @eniko For sure! Modern GPUs are hella fast at dispatching small triangles and I'm sure there are vanishingly few workloads where the difference is even noticeable. You're entirely right that you've got to pick your battles. Not trying to besmirch the Minecraft team for this or anything: there're plenty of places in @veloren where we've shrugged our shoulders & have said 'good enough'. Just trying to provide an explanation for why this sort of thing happens even in teams of clever folk.
@jsbarretto @eniko Yeha, I think in minecraft's case they're also pretty constrained by mods in the java edition, and that nobody wants to use bedrock for various reasons. Somebody seems to have had the grand plan that they'd migrate over to bedrock and everything would be happy times, and then obviously enough that didn't work.
@dotstdy @eniko Once you've got that much technical, social, and community debt (mods), a migration is never going to work out well. Migrating just one of my open source libs to a rewrite with only a few hundred users at the time was a years-long task...
@jsbarretto @dotstdy especially when the version you want people to migrate to is a walled garden that didn't support mods and had non-deterministic redstone >_>
@dotstdy @jsbarretto @eniko I mean they're currently in the process of a big rewrite of rendering, breaking a bunch of mods (including mine :( ), so I don't really think mods are a big problem here (and I think modders will figure everything out anyways)
@BasiqueEvangelist @jsbarretto @eniko something which they have put off doing for roughly a decade, due to the impact. you're right that people will adapt eventually, but I think they made an attempt at a clean break in bedrock, and it failed at displacing java edition, and they've been faffing around without strong direction ever since. but i think they do care about mods, it would have been easy to be a *lot* more disruptive to that side of things while chasing performance / whatever else.