I think the guess in the article is pretty close to the truth, I've seen stuff like that happen countless of times. You make a bet on a early technology (Unity DOTS + ECS in this case) which gives you a lot of benefits but also, it's immature enough that you get a bunch of additional work to do, and you barely have time to get everything in place before publisher forces you to follow the initial deadline.

100,000 vertices for pile of logs isn't really a bad bet on tech, though. That is just piling vastly more onto any tech stack than it can handle, with nobody having the time or the political okay to do a perf pass through the code and put all these ideas on a diet.

But that means that everything is solvable. There's no need in this game for 100,000 vertices for a logpile, so that should be a relatively straightforward task to fix. And someone can rip out all the teeth and put "Principal Tooth Extraction Engineer" on their resume.

> 100,000 vertices for pile of logs isn't really a bad bet on tech, though. That is just piling vastly more onto any tech stack than it can handle, with nobody having the time or the political okay to do a perf pass through the code and put all these ideas on a diet.

I can easily see this happening though.

Artist starts making assets, asks "What's my budget for each model" and engineering/managers reply with "Do whatever you want, we'll automatically create different LODs later" and the day gold master is being done, the LOD system still isn't in place so the call gets made to just ship what they have, otherwise publisher deadline will be missed.

That sounds like exactly what happened. I've been in that position many times in games I've worked on and seen it happen.

Anyone in management or engineering who tells an artist they have no texture or mesh budget at all gets exactly what's coming to them.

That's essentially what Epic is trying to tell devs these days. Don't know if it will ever truly live up to such lofty goals, though.