| Time |
Nick |
Message |
| 03:32 |
|
ivanbu joined #luanti-dev |
| 03:32 |
Noisytoot |
I object to blocking older versions, although I care more about this for clients than servers. I am still using 5.10 (as the last version to support OpenGL without shaders and therefore have not-terrible performance on Intel GMA4500MHD) and will likely continue to do so and backport security fixes if necessary |
| 04:00 |
|
MTDiscord joined #luanti-dev |
| 05:11 |
sofar |
Noisytoot: that hardware isn't supported by an open source linux driver. It's bad enough it ever made it out in product hardware. Please consider getting rid of it. |
| 05:12 |
sofar |
wait, I could be confused, but, in any case, it's really really ancient, and pre-x86_64-v2 as well. |
| 05:29 |
MTDiscord |
<jordan4ibanez> That thing must run xfce at 30 fps holy moly |
| 08:22 |
|
Warr1024 joined #luanti-dev |
| 08:47 |
|
Warr1024 joined #luanti-dev |
| 09:00 |
|
Farooq joined #luanti-dev |
| 09:33 |
|
Farooq joined #luanti-dev |
| 09:36 |
|
Farooq joined #luanti-dev |
| 09:55 |
|
Farooq joined #luanti-dev |
| 10:02 |
|
Farooq joined #luanti-dev |
| 11:11 |
Juri |
GPU so old it can legally drink in most places on earth |
| 12:30 |
MTDiscord |
<sheriff_u3> Give it some steroids, that might give it some more power. |
| 13:38 |
sfan5 |
I think you can basically get a better laptop for 10$ at a yard sale |
| 13:59 |
MTDiscord |
<rollerozxa> that 10$ yard sale laptop is obviously not usable to anyone with good tastes because it has the Intel Malware Engine embedded in the CPU |
| 14:43 |
|
Farooq joined #luanti-dev |
| 14:57 |
|
Farooq joined #luanti-dev |
| 15:00 |
|
Farooq joined #luanti-dev |
| 15:25 |
|
Farooq joined #luanti-dev |
| 15:44 |
Noisytoot |
sofar: it's supported by the i915 kernel driver and crocus (previously i965) mesa driver |
| 15:46 |
Noisytoot |
it is pre-x86_64-v2 but only just (it supports SSE4.1 but not SSE4.2 and POPCNT) |
| 15:47 |
cheapie |
Noisytoot: Huh, so if you run RVController in Luanti there, the emulated processor ends up having at least one feature that the host doesn't :P |
| 15:48 |
cheapie |
(RVController supports the B extension, which includes the "cpop" population count instruction) |
| 15:51 |
|
Panquesito7 joined #luanti-dev |
| 17:14 |
rubenwardy |
The app is live on the play store again |
| 17:14 |
sfan5 |
nice |
| 17:20 |
Sheriff_U3 |
finally |
| 17:24 |
pgimeno |
nice, is the blog entry being updated? |
| 17:25 |
MTDiscord |
<wsor4035> betting pool on how long till the next time tracer.ai does this again? |
| 17:26 |
Sheriff_U3 |
6 months |
| 17:27 |
|
jstein joined #luanti-dev |
| 17:28 |
rubenwardy |
Comms are working on it |
| 17:28 |
rubenwardy |
I'd like to confirm why it was restored before the 10 days mandated by the dmca |
| 18:21 |
sofar |
no popcnt? throw it out XD |
| 18:55 |
|
[MatrxMT] joined #luanti-dev |
| 19:25 |
rubenwardy |
edited the post. We'll do a new post when we get further information |
| 19:28 |
pgimeno |
great |
| 19:33 |
rubenwardy |
confirmed that Microsoft rescinded the notice |
| 19:35 |
celeron55 |
i wonder if i'll ever get any official email about it |
| 19:38 |
Sheriff_U3 |
I guess they decided that they still needed their test bed lol |
| 19:49 |
|
YuGiOhJCJ joined #luanti-dev |
| 20:02 |
rubenwardy |
yeah it's weird how the play console just unlocked and we were never told |
| 20:03 |
rubenwardy |
I wonder if this happened back in 2023 and explains the 46 day thing |
| 20:06 |
celeron55 |
well, it's possible. but also seems like i'd have went in and poked around earlier than that |
| 20:06 |
celeron55 |
+likely |
| 21:36 |
sfan5 |
@luatic if you have some time I'd appreciate your opinion on https://github.com/Uberi/Minetest-WorldEdit/issues/271 |
| 21:36 |
sfan5 |
(off-topic, sorry) |
| 22:36 |
MTDiscord |
<wsor4035> cc @grilled cheesus as iirc you had an unreleased better solution to the one i commented |
| 22:38 |
|
panwolfram joined #luanti-dev |
| 23:10 |
MTDiscord |
<kifie> that’s super old but it does incorporate most of those boxes and can probably handle all with minor modification. that being said i don’t think it’s ideal in the slightest, it worked well enough to prove a point and maybe thats all they need tbh |
| 23:14 |
MTDiscord |
<kifie> spitballing but a good schematic in my eyes should derive from a suitable spatial data structure like an octree or something. Most algos fall apart on high entropy data so any schematic and algorithms should work well enough for mapgen and builds. it also isnt a bad assumption to store metadata stuff in an entirely separate data structure, since more than likely there’ll be relatively little nodes with custom stuff |
| 23:16 |
MTDiscord |
<kifie> I’ve been detached from luanti development for several years so idk the proper terms for stuff like mapblocks and chunks and whatnot, but iirc there’s an efficient way to load bulk data in chunks or whatever and if your algos and data structures can leverage that, you can make something crazy fast and memory friendly |
| 23:24 |
MTDiscord |
<kifie> Please re-ping me if anything is needed, I never check this channel |
