Time |
Nick |
Message |
00:50 |
|
Fixer joined #minetest-dev |
01:26 |
|
hlqkj1 joined #minetest-dev |
02:52 |
|
calcul0n__ joined #minetest-dev |
03:34 |
Sokomine |
ups: that is new: i965: Failed to submit batchbuffer: Bad address |
03:34 |
Sokomine |
but in general it performed quite well up to this point. the world seemed to have loaded far better and possible view range better |
05:00 |
|
MTDiscord joined #minetest-dev |
06:19 |
|
flux__ joined #minetest-dev |
07:27 |
nerzhul |
it's why i think we really need to link MT with irrlicht statically, the conflict will happen now :D |
08:00 |
|
ShadowNinja joined #minetest-dev |
08:31 |
|
hlqkj joined #minetest-dev |
09:37 |
|
proller joined #minetest-dev |
10:00 |
|
tech_exorcist joined #minetest-dev |
10:03 |
|
tech_exorcist joined #minetest-dev |
10:08 |
|
tech_exorcist joined #minetest-dev |
10:10 |
|
tech_exorcist joined #minetest-dev |
10:10 |
|
tech_exorcist joined #minetest-dev |
10:11 |
|
tech_exorcist joined #minetest-dev |
10:22 |
|
Fixer joined #minetest-dev |
10:33 |
|
absurb joined #minetest-dev |
10:39 |
|
numzero joined #minetest-dev |
11:30 |
|
ivanb joined #minetest-dev |
11:34 |
|
Fixer_ joined #minetest-dev |
11:45 |
|
tech_exorcist joined #minetest-dev |
12:35 |
celeron55 |
i'm still thinking renaming the library would allow distributions to package MT's irrlicht separately if they insist on dynamic linking and separate packages |
12:36 |
celeron55 |
i guess they can patch it themselves if they want to, but just something to keep in mind if we want to be nice |
12:38 |
celeron55 |
it has been one of MTs advantages that it's easy to build and the build breaks very rarely on its own |
12:40 |
celeron55 |
so either rename the library and build it dynamically, or make it easy to rename the library and build it dynamically |
12:40 |
|
z812_ joined #minetest-dev |
12:49 |
|
z812_ joined #minetest-dev |
13:07 |
|
proller joined #minetest-dev |
13:26 |
|
hlqkj1_ joined #minetest-dev |
13:28 |
sfan5 |
hmm yea that'd be good |
15:18 |
|
z812_ joined #minetest-dev |
15:26 |
|
Fixer joined #minetest-dev |
15:31 |
|
z812__ joined #minetest-dev |
15:56 |
|
proller joined #minetest-dev |
16:03 |
|
Wuzzy joined #minetest-dev |
16:03 |
|
I_am_6r1d joined #minetest-dev |
16:11 |
|
tech_exorcist joined #minetest-dev |
16:19 |
Wuzzy |
Out of curiosity: Did the act of forking Irrlicht fix any major or longstanding bugs in Minetest? Would be interesting to know. ? |
16:21 |
sfan5 |
some input issues probably |
16:21 |
Krock |
also macos rendering |
16:21 |
|
DS-minetest joined #minetest-dev |
16:22 |
Wuzzy |
input issues? like the M key on macOS? |
16:22 |
sfan5 |
like shift on macos |
16:22 |
Wuzzy |
ooff. okay |
16:23 |
sfan5 |
there's likely many other issues fixed but we don't know yet until users test it and say "oh hey <thing> works in 5.5.0-dev" |
16:23 |
Wuzzy |
is there a GitHub label like "broken because of Irrlicht"? ? |
16:24 |
sfan5 |
we call that "Upstream issue" |
16:24 |
Wuzzy |
ah |
16:24 |
sfan5 |
(of course that includes more than just irrlicht issues) |
16:24 |
Wuzzy |
such as? |
16:25 |
sfan5 |
the luajit memory issue, broken graphics drivers, ... |
16:25 |
Wuzzy |
i just tried to compile minetest with the new irrlicht fork, and it worked. but i needed to fool around with ccmake first, i think the readme could be still a bit friendlier on compile instructions |
16:25 |
sfan5 |
by "the readme" you're referring to which? |
16:25 |
Wuzzy |
minetest |
16:26 |
Wuzzy |
but i think the minetest irrlicht readme also needs some love in genral, its very short and vague |
16:26 |
Wuzzy |
for minetest irrlicht, i just guessed cmake, make and it worked ? |
16:27 |
sfan5 |
minetest's readme could use example instructions so people don't have to figure out IRRLICHT_LIBRARY and IRRLICHT_INCLUDE_DIR themselves |
16:27 |
sfan5 |
but irrlicht's readme is fully complete, it even includes an example |
16:27 |
Wuzzy |
oh right |
16:28 |
Wuzzy |
my memory is really broken sometimes. i even looked at the readme... |
16:28 |
MTDiscord |
<Jordach> Wuzzy: the only non functional key under M1 Mac is the apostrophe |
16:28 |
MTDiscord |
<Jordach> shift and everything else works fine |
16:28 |
MTDiscord |
<Jordach> i can use a quote " |
16:28 |
MTDiscord |
<Jordach> not an apostrophe ' |
16:29 |
Wuzzy |
was this tested in latest dev version as well (i.e. wth irrlicht fork)? sfan5 suspects many bugs were "accidentally" fixed ? |
16:29 |
MTDiscord |
<Jordach> i can test that one min |
16:31 |
MTDiscord |
<Jordach> recompiling library+mt |
16:31 |
Wuzzy |
oh i remember now. my actual problem with the minetest irrlicht readme was that it doesnt really explain well what this thing is and what it does and why you need it. i think just saying "this is a fork of irrlicht" is too vague IMHO |
16:31 |
Wuzzy |
I generally like readmes to answer the one crucial question: "WTF does this thing?" ? |
16:32 |
MTDiscord |
<Jordach> what was the uh -D command for irrlicht folder locaiton |
16:32 |
MTDiscord |
<Jordach> oh right there |
16:32 |
Wuzzy |
use ccmake ? |
16:33 |
MTDiscord |
<Jordach> lemme uh launch cmake gui |
16:33 |
pgimeno |
"contains features, customizations and fixes specifically for use in Minetest" is accurate enough, it can't name each customization and fix |
16:33 |
Wuzzy |
i think you need IRRLICHT_INCLUDE_DIR, IRRLICHT_LIBRARY_DIR. |
16:33 |
pgimeno |
(re https://github.com/minetest/irrlicht/blob/master/README.md ) |
16:33 |
Wuzzy |
IRRLICHT_LIBRARY... oops |
16:34 |
Wuzzy |
"This is a fork of X" is not a complete description IMHO. Does not say what it *does*. |
16:35 |
pgimeno |
so you expect it to contain a description of what Irrlicht is? |
16:35 |
sfan5 |
"The Irrlicht Engine is an open source realtime 3D engine written in C++." this text i guess |
16:35 |
Wuzzy |
let me think... |
16:36 |
MTDiscord |
<Jordach> ah fuck, there's some deprecated shit in irrlicht_changes that needs to go |
16:36 |
Wuzzy |
yeah it goes in the right direction, let me think a bit ... |
16:39 |
MTDiscord |
<Jordach> sfan5, https://gist.github.com/Jordach/70bf8b3f9ca671c211778aa4fc59f9c0 on latest commit from minetest/minetest with irrlicht branch |
16:40 |
sfan5 |
"irrlicht branch"? |
16:40 |
MTDiscord |
<Jordach> well it's using the fork |
16:40 |
sfan5 |
works here |
16:41 |
MTDiscord |
<Jordach> eh i'll redownload it |
16:41 |
MTDiscord |
<Jordach> the good old git way |
16:42 |
Wuzzy |
Suggested description: "This is a fork of the Irrlicht Engine, an open source realtime 3D engine written in C++. It was created specifically for the Minetest project for rendering. It contains features, customizations and fixes specifically for use in Minetest." |
16:46 |
Wuzzy |
By the way, the more i think about it, the more i think the irrlicht fork should be renamed. I have no idea how to install this lib in a non-awkward way because of the name "libIrrlicht.so", which clashes with the real Irrlicht, obviously. |
16:46 |
sfan5 |
the answer to that would be not to install it system-wide |
16:47 |
Wuzzy |
? |
16:47 |
MTDiscord |
<Jordach> there we go, compiled |
16:47 |
Wuzzy |
I think it gets even funnier when Linux distros have to deal with it which also need to figure out how to deal with the name-clash... |
16:47 |
MTDiscord |
<Jordach> the apostrophe doesn't work when using chat, but is recognised by formspecs |
16:48 |
DS-minetest |
shouldn't distros just bundle out irrlicht together with minetest? it's unlikely that any other software will use our fork |
16:48 |
sfan5 |
Wuzzy: you make a good point but I'm saying that you can avoid all trouble by linking it statically |
16:48 |
DS-minetest |
s/out/our/ |
16:49 |
Wuzzy |
true. i'm not saying i can't work around the awkwardness, but its still super annoying to have this name-clash to begin with |
16:50 |
MTDiscord |
<Jordach> just bundle it into minetest ;) |
16:50 |
Wuzzy |
heh, thats also an option ... |
16:50 |
MTDiscord |
<Jordach> like we already for for lua 5.1 |
16:51 |
Wuzzy |
hmmm that would simplify the build process as well |
16:51 |
Wuzzy |
since you dont have to download 2 repos |
16:52 |
Wuzzy |
do you approve of the suggested description above? |
16:52 |
celeron55 |
what arguments are there for keeping the same name? |
16:53 |
Wuzzy |
none so far ? |
16:53 |
celeron55 |
there are many arguments for changing it |
16:53 |
Wuzzy |
YES! |
16:53 |
celeron55 |
it needs to be changed if there's nothing to gain from keeping the name |
16:53 |
Wuzzy |
oh finally. i thought i was talking against a brick wall. finally some agreement |
16:53 |
pgimeno |
<celeron55> what arguments are there for keeping the same name? <-- that no one has offered to bell that cat yet? |
16:54 |
Wuzzy |
"MTIrrlicht" has been suggested. but i really dont care about the new name, as long its different. ? |
16:55 |
celeron55 |
i'd say IrrMT |
16:55 |
sfan5 |
Wuzzy: if you think the points I made are what a brick wall is like you must have misunderstood me |
16:55 |
celeron55 |
it might be a bad idea to keep the full "irrlicht" in the name, it can cause confusion and could even be a trademark problem |
16:56 |
Wuzzy |
it doesnt seem the name "Irrlicht" is trademarked tho |
16:56 |
celeron55 |
trademarks are implied always |
16:57 |
sfan5 |
did supertuxkart rename their fork? |
16:57 |
Wuzzy |
Antarctica |
16:57 |
DS-minetest |
call it ErrLight |
16:57 |
Wuzzy |
wait, STK did fork Irrlicht, too? |
16:57 |
DS-minetest |
Errlight* |
16:58 |
Wuzzy |
anyway, the engine behind STK is called "Antarctica" |
17:14 |
rubenwardy |
worth noting that STK includes Antarctica in their source code, not even statically or dynamically linked |
17:15 |
MTDiscord |
<Jordach> that is what i alluded to earlier |
17:39 |
|
numzero joined #minetest-dev |
17:53 |
|
Taoki joined #minetest-dev |
18:08 |
MTDiscord |
<IhrFussel> @Wuzzy Irrlicht was not specifically made for MT, but the description suggestion you gave earlier included that sentence |
18:08 |
Wuzzy |
"It was created for ..." referring to the fork, not Irrlicht |
18:09 |
Wuzzy |
but ok probably needs a rephrase then :/ |
18:10 |
MTDiscord |
<IhrFussel> No, I missed the "fork" part somehow...it's fine |
18:18 |
|
turtleman joined #minetest-dev |
18:26 |
|
tech_exorcist joined #minetest-dev |
19:46 |
MTDiscord |
<exe_virus> Well, minetest is a voxel engine, and for now irrlicht is our graphics and input backend so perhaps we name it more like a underlying backend name? |
19:46 |
MTDiscord |
<exe_virus> Dungeon? |
19:47 |
MTDiscord |
<exe_virus> Shaft? |
19:47 |
MTDiscord |
<exe_virus> bedrock |
20:39 |
Krock |
irrtest |
20:39 |
Krock |
naming convention solved |
20:39 |
Krock |
!next |
20:39 |
ShadowBot |
Another satisfied customer. Next! |
20:39 |
rubenwardy |
I don't think it's worth giving a marketting name to it |
20:40 |
DS-minetest |
What about "meinlicht", to keep it german? |
20:41 |
DS-minetest |
(speaks minelicht) |
20:41 |
rubenwardy |
apparently german for Mine is Mine |
20:42 |
rubenwardy |
according to https://dictionary.cambridge.org/dictionary/english-german/mine |
20:42 |
rubenwardy |
not sure what the die is |
20:42 |
rubenwardy |
die Mine! DIE |
20:42 |
rubenwardy |
I shall see myself out |
20:43 |
DS-minetest |
it's the article, like the the in english |
20:43 |
MTDiscord |
<appguru> ^ |
20:44 |
DS-minetest |
for german mine it should be minenlicht, btw. |
20:53 |
MTDiscord |
<appguru> Lichtmine :thonking: |
20:59 |
freshreplicant[m |
Minelight = Lantern or (Laterne in German). |
21:02 |
|
absurb joined #minetest-dev |
23:03 |
MTDiscord |
<exe_virus> Okay so if no marketing name, then it should be as least mention irrlicht. So just mtirrlicht |
23:03 |
MTDiscord |
<exe_virus> All lower case |
23:04 |
MTDiscord |
<exe_virus> Easy for programmers to remember and type |
23:04 |
sfan5 |
I was thinking IrrlichtMT so that the library would be libIrrlichtMt.so.1.9.0 |
23:05 |
MTDiscord |
<GreenXenith> That seems more reasonable |
23:36 |
|
hlqkj_ joined #minetest-dev |