Minetest logo

IRC log for #minetest-dev, 2023-09-02

| Channels | #minetest-dev index | Today | | Google Search | Plaintext

All times shown according to UTC.

Time Nick Message
04:00 MTDiscord joined #minetest-dev
04:49 wsor joined #minetest-dev
05:28 izzyb joined #minetest-dev
05:45 calcul0n joined #minetest-dev
06:58 tekakutli joined #minetest-dev
07:47 calcul0n_ joined #minetest-dev
08:21 appguru joined #minetest-dev
09:40 appguru joined #minetest-dev
10:02 appguru joined #minetest-dev
10:33 appguru joined #minetest-dev
10:44 appguru joined #minetest-dev
10:58 Fleckenstein joined #minetest-dev
11:06 Fleckenstein joined #minetest-dev
11:08 LizzyFleck joined #minetest-dev
11:10 appguru joined #minetest-dev
11:23 Fleckenstein joined #minetest-dev
11:39 appguru joined #minetest-dev
11:53 Fleckenstein joined #minetest-dev
11:58 appguru joined #minetest-dev
12:15 appguru joined #minetest-dev
12:33 appguru joined #minetest-dev
12:34 Fleckenstein joined #minetest-dev
13:24 erle joined #minetest-dev
13:27 erle i have published the work-in-progress state of my unicode renderer in pure lua. the README also contains some examples and explanations on how you can cut texture size in minetest by a lot by sidestepping the whole optipng thing. this is probably interesting to engine developers, i believe this is the first easily-accessible code that even a super-lazy dev could use for texture encoding benchmarks in terms of encoding time and
13:27 erle payload. https://git.minetest.land/erlehmann/tga_unicode_renderer
13:28 erle and yes, i was surprised by the results myself, especially for the checkerboard pattern thing
13:32 erle since my goal is to have unicode on signs i would appreciate it if anyone on the input side of things could look at this example rendering and tell me if there exist characters that can not be input
13:32 erle https://mister-muffin.de/p/lgrI.png
13:36 appguru joined #minetest-dev
13:37 erle if you see any engine issues regarding my goal of fully backwards-compatible unicode on signs, please tell. as far as i can tell, my envisioned solution will work at least as far back as minetest 5.3 and require no new engine features (see readme for details).
13:37 erle in particular, i would like to know if there are issues having big textures
13:38 erle or if there is anything i should do to not fill the clients memory with garbage
13:39 erle (the garbage issue is one of several reasons why xmaps does not have live-updating maps, i fear the client caching a texture for each change in the map)
13:43 erle by the way, if anyone feels like fixing minetest.encode_png() to be able to compete with tga_encoder() in terms of features, feel free – but know that this is most likely a waste of your time, as the space-time tradeoff will *never* be in your favor for optimized PNG files (assuming no player wants to wait 20 seconds for their super-optimized sign texture to appear).
13:46 erle oh yeah, i already wrote it in #minetest – but you can probably reduce the transfer size of almost all textures by about 30% or so if you support .tga.z as a format (with fallback to sending .tga to older clients). this is probably by far the easiest thing you can do in terms of slashing teture filesize.
13:47 erle texture
13:51 erle btw, the shibboleths for not having understood the technical details are a) disputing the benchmarks without running them yourself b) proposing to compress all TGA files automatically (there is a simple reason why this does not work and it is so obvious i have not written it down, the random noise thing is a red herring)
13:52 erle by which i mean, i'd appreciate input from people knowledgeful in the problem space
13:52 erle thank you for coming to my ted talk :)
15:55 Fleckenstein joined #minetest-dev
16:21 Fleckenstein joined #minetest-dev
16:43 appguru joined #minetest-dev
16:47 vampirefrog joined #minetest-dev
17:03 appguru joined #minetest-dev
17:31 appguru joined #minetest-dev
17:35 LizzyFleck joined #minetest-dev
17:39 LizzyFleck joined #minetest-dev
17:42 Fleckenstein joined #minetest-dev
17:58 Fleckenstein joined #minetest-dev
18:03 appguru joined #minetest-dev
18:04 pgimeno joined #minetest-dev
18:59 Fleckenstein joined #minetest-dev
19:01 appguru joined #minetest-dev
19:36 MTDiscord <grorp> Merging #13768 in 15 min
19:36 ShadowBot https://github.com/minetest/minetest/issues/13768 -- Rewrite the gamebar by grorp
19:40 erle grorp why did you keep the function name the same when you radically changed the signature?
19:44 MTDiscord <grorp> If you mean the one of buttonbar_create, it's an internal function so why would renaming it be necessary?
20:44 Fleckenstein joined #minetest-dev
20:46 MTDiscord <grorp> Now also merging #13774 in 10 min
20:46 ShadowBot https://github.com/minetest/minetest/issues/13774 -- send ParticleSpawners to all players when time = 0 by chmodsayshello
20:57 erle grorp if it is an internal function and you fix all callsites, i understand completely
20:58 erle grorp i am so disappointed, i thought maybe this was a new coordinate exploit
20:58 erle > Modify a client to log the addition of ParticleSpawners
20:59 erle grorp did you use coras cheatclient for that or what?
21:01 MTDiscord <grorp> No, I didn't. Did they write one? I only know the one written by Fleckenstein.
21:02 erle grorp it is called waspsaliva and last time i used it we also used it to detect particle exploits
21:02 erle i think it is a fork of lizzys dragonfire?
21:02 erle not sure
21:03 erle https://repo.or.cz/waspsaliva.git this seems a bit old, but look at it
21:03 erle a bunch of these APIs can help a lot with debugging
21:05 erle grorp cora pointed me to https://github.com/dragonfireclient/dragonfireclient so ig she contributes to upstream dragonfire now
21:09 MTDiscord <grorp> Both PRs merged now.
21:25 erle joined #minetest-dev
21:51 ShadowNinja joined #minetest-dev
22:34 panwolfram joined #minetest-dev
22:59 YuGiOhJCJ joined #minetest-dev

| Channels | #minetest-dev index | Today | | Google Search | Plaintext