| Time |
Nick |
Message |
| 00:02 |
|
Sharpman joined #minetest-dev |
| 00:55 |
|
ShadowBot joined #minetest-dev |
| 01:02 |
|
MTDiscord joined #minetest-dev |
| 02:14 |
|
SFENCE joined #minetest-dev |
| 02:31 |
|
jonadab joined #minetest-dev |
| 05:00 |
|
MTDiscord joined #minetest-dev |
| 05:31 |
|
SFENCE joined #minetest-dev |
| 07:12 |
|
v-rob joined #minetest-dev |
| 07:30 |
|
calcul0n joined #minetest-dev |
| 08:05 |
|
vampirefrog joined #minetest-dev |
| 09:14 |
|
r00t_ joined #minetest-dev |
| 09:15 |
chaosesqueteam |
you fucking faggots are going to be physically killed |
| 09:15 |
chaosesqueteam |
you banned me for "toxic behavior" |
| 09:15 |
MTDiscord |
<greenxenith> Self-awareness is not his strong suit I see |
| 09:15 |
chaosesqueteam |
for telling you fucking faggots about extending the map format to 32 bit |
| 09:15 |
chaosesqueteam |
FUCK YOU |
| 09:15 |
chaosesqueteam |
it's called freedom of speech |
| 09:15 |
chaosesqueteam |
I extended the netcode |
| 09:16 |
chaosesqueteam |
for Q1 to above 16 bit |
| 09:16 |
chaosesqueteam |
so I was telling you FAGS |
| 09:16 |
chaosesqueteam |
that it is NOT HARD |
| 09:16 |
MTDiscord |
<greenxenith> What free speech, this is a privately operated space :] |
| 09:16 |
chaosesqueteam |
that I DID IT in another C application |
| 09:16 |
chaosesqueteam |
freedom of speech is a concept |
| 09:16 |
chaosesqueteam |
and it means I am going to murder you for banning me |
| 09:16 |
chaosesqueteam |
and for being a pro-trans rights, anti-loli bride FAGGOT |
| 09:16 |
MTDiscord |
<greenxenith> Ok |
| 09:17 |
chaosesqueteam |
fucking scum faggots |
| 09:17 |
MTDiscord |
<greenxenith> Such insecurities |
| 09:17 |
MTDiscord |
<greenxenith> I hope you recover |
| 09:17 |
chaosesqueteam |
bet you support women's rights too |
| 09:17 |
chaosesqueteam |
im a licensed attorney |
| 09:17 |
chaosesqueteam |
what the fuck are you |
| 09:17 |
chaosesqueteam |
faggot |
| 09:17 |
rubenwardy |
doubt |
| 09:17 |
chaosesqueteam |
yes i am |
| 09:17 |
|
chaosesqueteam was kicked by rubenwardy: kick |
| 09:17 |
MTDiscord |
<greenxenith> I definitely believe you |
| 09:17 |
MTDiscord |
<greenxenith> I patiently await being murdered by a licensed attorney |
| 09:18 |
MTDiscord |
<greenxenith> Looking forward to his legal defense |
| 09:18 |
rubenwardy |
death threats in a public place too |
| 09:19 |
|
Bahhumbug joined #minetest-dev |
| 09:21 |
ROllerozxa |
oh they're in #minetest-docs too now |
| 09:21 |
rubenwardy |
I don't have op there |
| 09:23 |
Bahhumbug |
I've dealt with it for the moment. |
| 09:23 |
Bahhumbug |
That behavior is explicitly against our TOS. |
| 09:23 |
ROllerozxa |
thanks |
| 09:24 |
MTDiscord |
<greenxenith> Is that a network op? 😄 |
| 09:24 |
Bahhumbug |
o/ |
| 09:24 |
Bahhumbug |
Hi :) |
| 09:24 |
ROllerozxa |
libera staff, it seems like |
| 09:26 |
MTDiscord |
<greenxenith> Guess ill have to keep an eye out now on this side 🙄 |
| 09:27 |
MTDiscord |
<greenxenith> Maybe we'll be lucky and they'll consider it beneath themselves to use Discord .... though their bar is challenging Challenger Deep |
| 09:32 |
rubenwardy |
I don't know how these people have this much time |
| 09:32 |
MTDiscord |
<greenxenith> Lack of employment, parents with a spare basement |
| 09:33 |
MTDiscord |
<greenxenith> The removed forum posts from them make it even funnier, when he was upset about people allegedly telling him it isnt possible to expand the map size when ... there are multiple PRs that did and do just that |
| 10:46 |
MTDiscord |
<herowl> Didn't he say "I am not a lawyer" in one of his forum posts? :thonkhammer: |
| 12:27 |
|
proller joined #minetest-dev |
| 12:54 |
|
chmodsayshello joined #minetest-dev |
| 12:55 |
chmodsayshello |
I hope this not too offtopic for #minetest-dev, but why are the filenames of all media (images and sound) hashed, and not the original filename as it is on the server? Also, is it just the sha256 of the name? |
| 13:04 |
|
chmodsayshello joined #minetest-dev |
| 13:05 |
sfan5 |
I'm not aware of any part where Minetest hashes filenames |
| 13:06 |
celeron55_ |
are you asking why they aren't stored by their original filename? that's because then the name wouldn't necessarily match the content, and the hash would have to be stored in a separate database |
| 13:06 |
celeron55_ |
much handier to just use the hash as the filename, to directly find the correct content by the hash |
| 13:07 |
celeron55_ |
the hash that is used as the filename is the hash of the content of the file, not the hash of the original filename (that wouldn't make any sense) |
| 13:07 |
celeron55_ |
the minetest cache is essentially a content-addressed database |
| 13:07 |
celeron55_ |
or storage |
| 13:08 |
chmodsayshello |
if it's not the hash of the original filename though, how would the client know which textures from a texturepack to use though instead of the ones sent by the server? |
| 13:09 |
celeron55_ |
(or, really, it's probably the purest form of content-addressed storage that you can come up with, the textbook example) |
| 13:09 |
rubenwardy |
chmodsayshello: the client knows both the filename and the hash |
| 13:09 |
celeron55_ |
the server sends a list of files which contains, for each file, the original name of the file, and the hash of the content of the file |
| 13:10 |
celeron55_ |
afte that the client checks its cache for the actual contents for each file, and for the ones it didn't have, it asks from the server, and stores them into the cache by the content hash |
| 13:10 |
celeron55_ |
+r |
| 13:12 |
celeron55_ |
the client needs to know the filename not just for texture packs but for being able to render stuff. the server refers to textures by their original name, when it tells the client what everything should look like |
| 13:13 |
chmodsayshello |
aah, i see, makes sense now, thanks! Unrelated to my original question: if it checks for the contents of the file (comparing hashes), wouldn't the client go ahead and request files that it never needs due to having a different texture from a texturepack for example? |
| 13:13 |
celeron55_ |
i would assume it doesn't take into account the texture packs it has available, at that point. it's true that it could do that |
| 13:48 |
MTDiscord |
<warr1024> Oh wait, does using a texturepack prevent the client from downloading at all the original texture from the server?  If that's really the case, that's a pretty significant factor in coming up with strategies to reduce server join times... 🤔 |
| 13:54 |
sfan5 |
the easiest way to reduce join times is join once and then have the cache filled for the second time ;) |
| 14:10 |
|
appguru joined #minetest-dev |
| 14:40 |
|
thelounge656 joined #minetest-dev |
| 14:40 |
|
MisterE123 joined #minetest-dev |
| 14:40 |
|
hlqkj joined #minetest-dev |
| 14:40 |
|
basxto joined #minetest-dev |
| 14:40 |
|
calcul0n joined #minetest-dev |
| 14:41 |
|
ShadowBot joined #minetest-dev |
| 14:41 |
|
ivanbu joined #minetest-dev |
| 14:43 |
|
ShadowNinja joined #minetest-dev |
| 14:49 |
celeron55_ |
alternatively you can join half-way two times, after which the third time will take only one fourth of the time |
| 14:51 |
|
izzyb joined #minetest-dev |
| 15:08 |
|
appguru joined #minetest-dev |
| 15:15 |
MTDiscord |
<warr1024> The easiest way to reduce join times is to have players use a custom client build that has the cache prepopulated. Â I'm not interested in hearing which single strategy is the "best" for reducing join times, I'm interested in hearing what strategies exist... |
| 15:16 |
sfan5 |
it seems the IPO changes in combination with ccache or LLD broke compiling |
| 15:16 |
sfan5 |
just says "ld.lld: error: undefined symbol: main" for me |
| 15:18 |
MTDiscord |
<grorp> maybe you're using gcc + lld? that doesn't seem to work for me with LTO |
| 15:20 |
sfan5 |
I am |
| 15:26 |
|
Lupercus joined #minetest-dev |
| 16:59 |
|
Desour joined #minetest-dev |
| 18:25 |
|
v-rob joined #minetest-dev |
| 18:45 |
|
appguru joined #minetest-dev |
| 18:58 |
|
v-rob joined #minetest-dev |
| 19:03 |
|
proller__ joined #minetest-dev |
| 19:38 |
|
calcul0n joined #minetest-dev |
| 20:39 |
|
appguru joined #minetest-dev |
| 21:46 |
|
v-rob joined #minetest-dev |
| 21:56 |
MTDiscord |
<luatic> I would like to merge / push #14366, #14351 #13789 and https://gist.github.com/appgurueu/a2e5378946f960c8761c03bfa60f80d4 in 20m |
| 21:56 |
ShadowBot |
https://github.com/minetest/minetest/issues/14366 -- Allow VBO to be used for meshes all the way down to 4 vertices by paradust7 |
| 21:56 |
ShadowBot |
https://github.com/minetest/minetest/issues/14351 -- Rename MINETEST_SUBGAME_PATH to MINETEST_GAME_PATH by cx384 |
| 21:56 |
ShadowBot |
https://github.com/minetest/minetest/issues/13789 -- Fix liquid falling if in "float" group, fixes #13782 by kromka-chleba |
| 21:56 |
MTDiscord |
<luatic> The error message patch basically just throws with the proper errmsg instead of giving the very useful |
| 21:56 |
MTDiscord |
<luatic> ServerError: AsyncErr: Lua: Runtime error from mod 'builtin' in callback player_event(): "Runtime error from mod '??' in callback ??(): <no description>" |
| 22:11 |
jonadab |
That may be the _longest_ completely information-free error message I have ever seen coming from software not published by Microsoft. |
| 22:12 |
jonadab |
(The absolute top of the list, of course, is the error message some versions of MSIE give for DNS resolution failure, unreachable servers, and http 404, because users weren't already confused enough without all those things being conflated.) |
| 22:25 |
MTDiscord |
<luatic> done |
| 22:31 |
ROllerozxa |
jonadab: oh my god windows error codes. microsoft is still at it to this day with completely opaque and nondescriptive error codes whenever anything goes wrong in windows |
| 22:32 |
ROllerozxa |
even in windows 10 if say, windows update fails and pulls some error code like 0xDEADBEEF or 43. it cheekily tells you to search online for the issue but the only results you get are microsoft support threads where people give the standard cookie cutter "`sfc /scannow` or reinstall" advice |
| 22:32 |
ROllerozxa |
microsoft must have intentionally coded in the codes to occur when something happens, so they obviously must know what has happened when any given error code is raised, why can't they just tell you?? "the code is the documentation" only works when it's open source, maybe they intentionally want you to be helpless because it's in their interest to make windows as opaque as possible |
| 22:32 |
ROllerozxa |
(ok sorry for ranting about windows in #minetest-dev haha, I should bottle it up and save for a blog post or something) |
| 23:32 |
|
panwolfram joined #minetest-dev |
| 23:58 |
|
pmp-p joined #minetest-dev |
