Time Nick Message 00:00 luatic this is actually not as far off as you think. (i think you are mistaken.) 00:02 luatic !tell erle AFAICT, there is no magic "if you have this GPU, set this near plane". check logs 00:02 MinetestBot luatic: I'll pass that on when erle is around 00:03 luatic while of course the range of reasonable values depends on the GPU, which may be worse on android (or is there another rationale?), the current solution is of course a crutch. 00:03 luatic a proper solution for a game engine should however let game devs control the near and far plane of the perspective camera view frustum. because where you want to leverage your depth buffers precision depends on the game. is rendering things close to the player important or not? 00:04 luatic (or by contradiction: if it solely depended on the GPU, OGL would not have made it configurable.) 00:36 [MatrxMT] 😏 15:01 rubenwardy cdblitelite: generates a static site for some packages, compatible with the CDB APK 15:02 rubenwardy *API 16:33 repetitivestrain sfan5: regarding https://github.com/luanti-org/luanti/pull/16397: 16:33 repetitivestrain i'll adjust points 1 2 and 4 as you suggest. i suppose i can't object to 3, i'll fix that also. as regards the final two points you raise, however, i simply copied the existing default vm_area, and hence i'm a bit confused since that test would only be incorrect if l_generate_decoration's default pmin and pmax has always been one block wider than the vm's emerged area, and lastly oldvm is always initialized before it is accessed by 16:33 repetitivestrain reason of the "oldvm = NULL" statement in the !mgp branch, which is always taken if oldvm has not already been initialized from mgp->vm 16:35 sfan5 what 16:35 sfan5 what is "the existing default vm_area"? 16:39 sfan5 nevermind, I misread the code 16:46 sfan5 added a different comment 23:26 cheapie Messing with people who went AFK next to a PDP-13 with a programming terminal on it is too much fun sometimes :P