Posts: 6,485
Threads: 176
Joined: Jan 2012
Thanks: 131
Given 1074 thank(s) in 852 post(s)
Yeah, our software is a bit better, although we did have some of those google-security-research incidents, too. Not that ridiculous, though.
I don't think it's the interpreted/compiled gap, it must be new optimizations to the display routines. It also depends very much on the world and direction you're looking. On my old laptop I get like 10 times slowdown when looking forward, compared to looking straight down. And I once wrote a world generator in Cuberite that brought any client to its knees, because the world was unoptimizable.
Posts: 6,485
Threads: 176
Joined: Jan 2012
Thanks: 131
Given 1074 thank(s) in 852 post(s)
The code didn't make it to the repo, quite understandably. It was when I was testing out various terrain height generators and had them output some test values into the world. I managed to make a more-or-less "50% gray" - every other block was set to stone, the rest was air. There were so many polygons that the client cannot fit them into the available memory, let alone render. I suppose modern client could have some optimizations to prevent a crash in this situation, thanks to occlusion culling; for such a case, a similar pattern, but less dense, should still do the trick.
It should be rather easy to write such a generator as a plugin, if you really want to test that.
Posts: 6,485
Threads: 176
Joined: Jan 2012
Thanks: 131
Given 1074 thank(s) in 852 post(s)
Sister is getting married tomorrow, the frenetic chaos levels are rising...
Posts: 204
Threads: 6
Joined: Oct 2015
Thanks: 23
Given 17 thank(s) in 16 post(s)
01-16-2016, 04:30 PM
(This post was last modified: 01-16-2016, 04:31 PM by Schwertspize.)
Yesterday, first time ever, I got compilation issues and it seems to be gcc itself. I'm on Ubuntu and using gcc-5.1 (after trying 4.9)
On another machine it compiles just normal