| Time |
Nick |
Message |
| 01:54 |
|
whitepaperkat joined #minetest-dev |
| 04:00 |
|
MTDiscord joined #minetest-dev |
| 06:03 |
|
GreenXenith joined #minetest-dev |
| 06:32 |
|
YuGiOhJCJ joined #minetest-dev |
| 07:47 |
sfan5 |
NB: "even if 16 or 32 bit floats get sent to the graphics card." this isn't actually happening, instead the graphics card internally expands the data to floats |
| 07:49 |
sfan5 |
https://github.com/minetest/minetest/blob/6303334cc2f39732697a1278532cbe66ed481da8/irr/src/OpenGL/Driver.cpp#L61-L69 |
| 09:34 |
MTDiscord |
<greenxenith> #5664 can probably be closed? |
| 09:34 |
ShadowBot |
https://github.com/minetest/minetest/issues/5664 -- Asynchronous logging |
| 09:35 |
MTDiscord |
<greenxenith> And on that note, is minetest.get_translated_string something that could be added to the async environment? |
| 10:05 |
sfan5 |
thread-safety probably needs some addressing |
| 10:05 |
sfan5 |
but sure |
| 10:06 |
sfan5 |
why not make a PR? :) |
| 13:48 |
|
Noisytoot joined #minetest-dev |
| 14:17 |
|
[MTMatrix] joined #minetest-dev |
| 17:02 |
|
imi joined #minetest-dev |
| 17:47 |
|
TheCoffeMaker joined #minetest-dev |
| 17:48 |
|
Eragon joined #minetest-dev |
| 17:52 |
|
TheCoffeMaker joined #minetest-dev |
| 17:54 |
|
Eragon joined #minetest-dev |
| 18:04 |
|
Noisytoot joined #minetest-dev |
| 18:19 |
|
Noisytoot joined #minetest-dev |
| 18:29 |
|
imi joined #minetest-dev |
| 19:06 |
|
imi joined #minetest-dev |
| 21:29 |
|
Sharpman joined #minetest-dev |
| 21:58 |
MTDiscord |
<exe_virus> https://github.com/minetest/minetest/pull/14643 The Pull request for the unordered_multimap for spatial indexing is ready for more thorough testing and review. It's now stable in all my testing & passing all CI tests. Currently re-running benchmarks for comparisons/tradeoffs, and considering making a special case algorithm for getObjectsInRadius. |
| 22:31 |
|
panwolfram joined #minetest-dev |
| 22:35 |
MTDiscord |
<exe_virus> Actually thinking about the algorithm more, I should probably do some smart AABB filtering where we check once for the whole mapblock, not always checking all the entities inside. That would help with all of the cases where the current naive algo is not optimal. Still a big improvement as is, but could do a good chunk better |
