| Time |
Nick |
Message |
| 00:21 |
|
scr267a joined #minetest-hub |
| 01:31 |
|
ANAND joined #minetest-hub |
| 01:43 |
|
paramat joined #minetest-hub |
| 03:49 |
|
GreenDimond joined #minetest-hub |
| 03:56 |
|
paramat joined #minetest-hub |
| 05:18 |
|
diemartin joined #minetest-hub |
| 08:52 |
|
CWz joined #minetest-hub |
| 10:52 |
|
Fixer joined #minetest-hub |
| 11:27 |
Fixer |
FUCK NVIDIA |
| 11:46 |
|
IhrFussel joined #minetest-hub |
| 11:46 |
IhrFussel |
IS THIS DUDE CRAZY? https://forum.minetest.net/viewtopic.php?p=344639#p344639 |
| 11:46 |
IhrFussel |
'dedicated_server_step = 0.01' |
| 11:46 |
IhrFussel |
No wonder that his server lags and ABMs have problem running really |
| 11:47 |
IhrFussel |
MT is not made for 100 ticks per second... almost no game is made for that! |
| 11:51 |
IhrFussel |
Even shooters like Call of Duty only have 64 ticks/sec by default I think...or at least the older titles had that |
| 12:00 |
IhrFussel |
I don't wanna tell him something wrong so... while ABMs run the whole server is 'stalled' correct? |
| 12:01 |
IhrFussel |
So if the log mentions for example 'took longer than 200ms' it means there happened a lag spike of at least 0.2+ secs |
| 12:01 |
VanessaE |
IhrFussel: no, doesn't work like that. |
| 12:01 |
VanessaE |
ABMs are timed in seconds, not server ticks. |
| 12:02 |
IhrFussel |
But while they run the rest of the server is paused AFAIK |
| 12:02 |
VanessaE |
server step time will mostly only affect globalstep stuff |
| 12:02 |
VanessaE |
ABMs will take just as much time to run whether server step is 0.1 or 0.0001. |
| 12:03 |
IhrFussel |
I know that but his lag in general is high which means globalsteps run 100 times per second (if that is even possible) compared to default which would be a bit more than 10 times per second |
| 12:03 |
VanessaE |
no. |
| 12:03 |
VanessaE |
my servers have globalstep at 0.02 I think. better latency that way, no effect at all on ABMs |
| 12:03 |
IhrFussel |
Not ABMs |
| 12:03 |
sfan5 |
there might be badly coded mods that do some things at every globalstep |
| 12:03 |
IhrFussel |
But every mod that uses globalsteps now runs 100 times per second or 50 times in your case |
| 12:04 |
VanessaE |
sure, both of those are true |
| 12:04 |
VanessaE |
but that guy's ABMs will still take 202, 204, 262, 212... milliseconds regardless of the server step setting. |
| 12:04 |
IhrFussel |
Many mods don't use a timer in globalstep cause they assume the server owner won't change it (much) |
| 12:05 |
IhrFussel |
Yes that is why I told him to profile his ABMs and looks like moretrees is the culprit (who would've guessed) |
| 12:06 |
sfan5 |
VanessaE: well if you reduce the server step more ABMs might have to run at once, thus taking more time |
| 12:06 |
sfan5 |
that's not the case here, though |
| 12:07 |
IhrFussel |
But I still think I'm right when I say 'if ABMs take longer than [server tick setting] the server will hang until they are done/cancelled' |
| 12:07 |
sfan5 |
eh, by reduce I mean making server steps less frequent |
| 12:08 |
sfan5 |
wait, nevermind this doesn't make sense, ignore what I said |
| 12:08 |
VanessaE |
sfan5: well that may still be true, but any one ABM still takes the same amount of time when it does fire, regardless of server step |
| 12:08 |
IhrFussel |
And he says it happens every 30 secs so the ABMs do lag his server for 200+ms every 30 secs |
| 12:08 |
VanessaE |
and sure, it makes sense, since globalstep code is also blocking. |
| 12:09 |
IhrFussel |
'playeranim' for example has no timer in gloalstep...imagine that mod trying to update all player animations 100 times per second...goodbye server :P |
| 12:09 |
VanessaE |
IhrFussel: sure, if an ABM interval is 30s, and the chance is low enough or the code is slow enough, yeah it'll hang the machine every 30s. regardless of globalstep. |
| 12:10 |
VanessaE |
and yeah, playeranim probably ought to check dtime :) |
| 12:10 |
VanessaE |
all my mods that use globalstep do that afaik. |
| 12:11 |
IhrFussel |
Or...I wonder what happens when you try to update a player HUD 100 times per second...I could imagine that the client even starts freaking out |
| 12:12 |
VanessaE |
well if the client isn't cacheing such things, that'll be a problem yeah |
| 12:13 |
IhrFussel |
Here is the deadly code xP https://github.com/minetest-mods/playeranim/blob/master/init.lua#L297 |
| 12:17 |
VanessaE |
ouch. |
| 12:18 |
VanessaE |
pretty simple to rate-limit it, but that needs rather more optimization than that. |
| 12:18 |
|
kilbith joined #minetest-hub |
| 12:20 |
IhrFussel |
But doesn't the engine also use the server step for internal stuff? |
| 12:20 |
IhrFussel |
Which would mean in any case it requires somewhat more preoccessing power |
| 12:21 |
IhrFussel |
processing* |
| 12:49 |
|
_Xenon joined #minetest-hub |
| 12:52 |
nerzhul |
i think we should have a registration for stepping over lua or player |
| 12:52 |
nerzhul |
it can permit to prevent many duplicated calls like this |
| 13:05 |
|
Hillbilly joined #minetest-hub |
| 13:13 |
IhrFussel |
nerzhul, I have at least 15 mods with globalstep callbacks that loop through players every few secs so that sounds like a good idea |
| 13:13 |
nerzhul |
remember me to open an issue when the 5.1 dev cycle is open |
| 13:13 |
ChimneySwift |
playeranim is ridiculously network heavy too |
| 13:14 |
nerzhul |
yes, because the animation code is very low level, we talk in frames not in animations |
| 13:14 |
nerzhul |
it's very bad, but historical |
| 13:14 |
ChimneySwift |
yeah, not a great implementation |
| 13:16 |
IhrFussel |
It's currently the only solution available if your server needs head movement |
| 13:17 |
VanessaE |
why not fix that before 5.0.0 goes out? |
| 13:20 |
VanessaE |
seems to me it would only take two bytes to represent the head pitch and yaw, could just be sent from the client, as a normal packet, any time the head moves more than some amount (like half a degree maybe) |
| 13:20 |
VanessaE |
then the rest could be handled on the server with some callback |
| 13:21 |
VanessaE |
i.e. in lua. |
| 13:21 |
ChimneySwift |
yeah, that would be a worthwhile implementation, playeranim works but it's usually not worth it |
| 13:21 |
VanessaE |
register_on_move_head(player,pitch,yaw).... |
| 13:21 |
ChimneySwift |
ez |
| 13:36 |
|
kaeza joined #minetest-hub |
| 13:40 |
nerzhul |
VanessaE we are in freeze and very close to release |
| 13:40 |
nerzhul |
https://github.com/minetest/minetest/milestone/9 |
| 13:41 |
VanessaE |
I thought you were just in translation freeze? |
| 13:42 |
jas_ |
:) |
| 13:42 |
nerzhul |
no everything is frozen |
| 13:42 |
VanessaE |
oh |
| 13:43 |
jas_ |
FULL FREEZE ?! |
| 13:43 |
* jas_ |
steps away in utter amazement |
| 13:43 |
VanessaE |
I guess that's why there are icicles everywhere :P |
| 13:43 |
IhrFussel |
Make sure to include https://github.com/minetest/minetest/pull/8249 in 5.0.0 ... don't allow/encourage cheating |
| 13:44 |
VanessaE |
probably too late, IhrFussel |
| 13:44 |
IhrFussel |
You have no clue how many mods depend on player movement keys in order to function properly |
| 13:44 |
nerzhul |
the freeze has been announced on the forum quite time ago guys, hurry up :) |
| 13:45 |
nerzhul |
too late for 8249 |
| 13:45 |
IhrFussel |
It's a BUGFIX |
| 13:45 |
IhrFussel |
You said you will include BUGFIXES |
| 13:45 |
nerzhul |
https://forum.minetest.net/viewtopic.php?f=18&t=21733 |
| 13:45 |
nerzhul |
it's not a bug fix it's a behaviour change |
| 13:45 |
nerzhul |
and it's too risky |
| 13:45 |
nerzhul |
we are in RC2 |
| 13:46 |
IhrFussel |
No behavior change...auto foward simply forgot to send movement keys to the server |
| 13:46 |
nerzhul |
did you just read the PR ? |
| 13:46 |
IhrFussel |
Even movement in MTG looks awkward without it |
| 13:47 |
IhrFussel |
Then you basically don't care if VANILLA clients can cheat on servers |
| 13:47 |
VanessaE |
nerzhul: he's right. that needs added. |
| 13:47 |
IhrFussel |
My server CANNOT reduce saturation with auto forward |
| 13:48 |
nerzhul |
guys minetest is not finished we will have a 5.1.0 after 5.0.0 which should not be released in 3 years but 3-4 months if possible |
| 13:48 |
VanessaE |
not acceptable. |
| 13:49 |
IhrFussel |
You have to include the BUGFIXES if they exist |
| 13:49 |
VanessaE |
you know full well how long we tend to go between releases. |
| 13:49 |
nerzhul |
if you wanted this fix before just read the forum and permit to have it in the roadmap before 5.0.0 -rc 2 |
| 13:49 |
nerzhul |
only blockers are allowed now. It's not a blocker minetest work and had worked since ages without it. |
| 13:49 |
nerzhul |
VanessaE: i know 5.0.0 was too long |
| 13:49 |
nerzhul |
but now it's time for it to come. |
| 13:50 |
nerzhul |
and go ahead with 5.1 and release more and more often |
| 13:50 |
IhrFussel |
It's 14 LINES of code... you can review this in a matter of 1 minute probably |
| 13:50 |
VanessaE |
nerzhul: I'm not talking about 0.4.x -> 5.0.0, but 0.4.x -> 0.4.x, tends to be 6 months or so at a time. |
| 13:50 |
nerzhul |
yeah because it wasn't so major |
| 13:50 |
nerzhul |
and i have insisted on the release in december to trigger this freeze in january |
| 13:50 |
nerzhul |
and i wanted a release in february, we are now in march and it's not released |
| 13:51 |
VanessaE |
yeah, thta just means there was a lot of work to do. |
| 13:51 |
VanessaE |
and there's still more, if only the blockers. |
| 13:51 |
IhrFussel |
If you don't include this I will have to find a way to kick players with auto forward enabled |
| 13:52 |
IhrFussel |
I want all players to have EQUAL hunger/saturation... and there are more mods that depend on movement sent by players such as animation |
| 13:52 |
nerzhul |
VanessaE: the problem was not sufficient testing in january |
| 13:52 |
nerzhul |
i want to be rich but it's not for this release, i just need to wait :) |
| 13:52 |
nerzhul |
just wait we cannot fix ALL things, else we won't have a 5.0.0 before next century |
| 13:53 |
* VanessaE |
shrugs |
| 13:53 |
IhrFussel |
Youz can fix the things that are KNOWN and have a PR ready |
| 13:53 |
nerzhul |
but promise i add it to 5.1.0 roadmap and we we will look at this and merge if the fix is correct |
| 13:53 |
nerzhul |
for me the fix is not really correct |
| 13:53 |
VanessaE |
maybe see what the other core devs have to say? |
| 13:53 |
nerzhul |
why each time i come here i battle with some community members because roadmap is never sufficient ? |
| 13:54 |
IhrFussel |
This move by you shows server owners again how little you care about their server functionality |
| 13:54 |
nerzhul |
we can wait, but for me we merge the latest PR blocker and we release. |
| 13:54 |
VanessaE |
nerzhul: do you really want me to answer that? |
| 13:55 |
IhrFussel |
People waited for almost 2 years now... a few more days to include a very important bugfix wouldn't matter at all |
| 13:56 |
IhrFussel |
You should release when no simple bugfixes are left to include else your product is buggy at RELEASE |
| 13:56 |
nerzhul |
VanessaE: maybe because i answer to those community members and they know i answer ? :p |
| 13:57 |
nerzhul |
ihrFussel sorry but this bugfix is not easy to review |
| 13:57 |
nerzhul |
it's a behaviour change |
| 13:58 |
nerzhul |
you can't only look at the PR line numbers here |
| 13:58 |
IhrFussel |
It simply continuously sends up key packets AFAIK ... how is this complex? |
| 13:58 |
nerzhul |
and as i said, it's a bug fix but not a blocker bugfix |
| 13:58 |
nerzhul |
the PR doesn't send more packets |
| 13:58 |
nerzhul |
you didn't understand it |
| 13:59 |
nerzhul |
it modify the controls sent to server |
| 14:00 |
nerzhul |
it's why i don't want to take risk |
| 14:00 |
VanessaE |
nerzhul: my answer will not be very polite.. |
| 14:01 |
IhrFussel |
The 'superspeed' bit was removed I think cause servers cannot detect it... if you have auto foward enabled and fast move and press W you suddenly move faster BUT the server has no way of checking this |
| 14:01 |
IhrFussel |
Other than maybe polling velocity |
| 14:02 |
IhrFussel |
forward* |
| 14:02 |
VanessaE |
nerzhul: you guys (core devs) have a long history, very obviously, of caring more about the code and what it does, than about how it behaves to the end user. i.e. behind-the-scenes is hugely more important to you than presentation. As my late husband used to say, you guys are coders, not programmers. |
| 14:02 |
nerzhul |
if 2 other coredev want to merge it before release go ahead, for me it's not blocker and the behaviour change is a little bit tricky and needs more and more testing and we are near release |
| 14:02 |
VanessaE |
no offense, but it is what it is, I call it as I see it. |
| 14:03 |
nerzhul |
i don't want 1 month of blocker because of a major fix (major, not critical, not blocker) |
| 14:04 |
IhrFussel |
You should focus on bugfixes that directly affect mods/servers in a way that it might allow exploiting of gameplay |
| 14:05 |
IhrFussel |
A camera bug is not part of that for example |
| 14:06 |
nerzhul |
we should focus on bug which blocks minetest from working. |
| 14:06 |
nerzhul |
this behaviour is historical, for me it's just a major, it has never making MT not working |
| 14:07 |
VanessaE |
nerzhul: in other words, it would be, in a game like Quake, the difference between printing "you died. 10 shots fired, 5 frags." and displaying a bunch of gore, sound effects, displaying scores, and so on in a nice graphical summary. |
| 14:07 |
nerzhul |
the latest blocker i own is particle spawner breakage in 5.0. |
| 14:07 |
nerzhul |
sorry we will never add gore option to minetest, in base game :) |
| 14:07 |
VanessaE |
I"m not saying to. |
| 14:07 |
VanessaE |
but you see the difference, right? |
| 14:07 |
nerzhul |
now sell your PR to other coredev if you want. For me it's not blocker. |
| 14:07 |
nerzhul |
i see the difference |
| 14:07 |
IhrFussel |
I only see 2 max 3 PRs that fall into the category I mentioned above |
| 14:08 |
nerzhul |
but i want that fucking 5.0.0 to be released with no real blocker |
| 14:08 |
nerzhul |
5.1.0 is here to integrate all changes you want we can implement in 4 months |
| 14:08 |
VanessaE |
nerzhul: I'm just saying, a simple print "works", the other looks proper for that game type and is what a player would expect. |
| 14:08 |
nerzhul |
i didn't finished the changelog in the dev wiki i have 2 or 3 months commits to add |
| 14:08 |
nerzhul |
it's the biggest we ever had in mt |
| 14:09 |
VanessaE |
in minetest, I guess the equivalent would be placing a torch adds a flat, yellowish light to its surroundings in a diamond pattern, versus placing a torch adds bloom, haze, proper spherical light spreading, light color, etc. |
| 14:09 |
|
NathanS21 joined #minetest-hub |
| 14:09 |
VanessaE |
one works, the other is what players would expect (given time and effort of course) |
| 14:10 |
nerzhul |
i wanted those things too, but i don't want to block the release for them, we can add it in next release |
| 14:10 |
ANAND |
nerzhul: Everyone is looking forward to 5.0.0 as much as you. But since this is a major release, it'd be great if it achieves a certain level of perfection. And I'm sure you understand that has nothing to do with ticking off items from a list of blockers. :) |
| 14:10 |
VanessaE |
I'm not saying to add them now, or anytime soon. I"m making comparisons. |
| 14:10 |
nerzhul |
you think about this "simple" PR, but look at the 5.1.0 milestone, there is a bunch of PR it's wow |
| 14:11 |
nerzhul |
ANAND: i want to polish, but i also know we will have a release after, and i will try to wake up the team to have a 4 month based release |
| 14:11 |
IhrFussel |
Sometimes I don't think you understand what gameplay exploits can actually do to servers... they can destroy the balance, they can make the whole game that runs on the server pointless ... and that isn't important enough for you to fix it now instead of in a few months |
| 14:11 |
nerzhul |
less features, but features which are released |
| 14:11 |
nerzhul |
what is the game exploit there |
| 14:12 |
VanessaE |
nerzhul: ok, here's a more concrete example... how long have players and modders been begging for colorized light sources? as in a torch is yellow-orange, a blue lightston is blue, sun is yellow-white, etc. just a rough guess? |
| 14:12 |
IhrFussel |
I already told you that many mods listen to the movement keys of players and it's exploiting the system if 'auto forward' acts like NOTHING is pressed |
| 14:12 |
VanessaE |
+e |
| 14:13 |
|
calcul0n joined #minetest-hub |
| 14:13 |
nerzhul |
VanessaE: i know, i also wanted more features, but at a point there are compromised. And this PR was published too late after freeze for me |
| 14:13 |
VanessaE |
nerzhul: just answer the question. |
| 14:13 |
nerzhul |
IhrFussel, and why do they do this ? |
| 14:13 |
nerzhul |
for me it's a mod bug |
| 14:13 |
nerzhul |
i don't know how many time they wait |
| 14:13 |
IhrFussel |
Your VERY OWN MTG does it! |
| 14:14 |
ANAND |
lol |
| 14:14 |
VanessaE |
nerzhul: a few years at least. |
| 14:14 |
VanessaE |
nerzhul: so call it 3. what version were we at 3 years ago? |
| 14:14 |
IhrFussel |
MTG NEEDS the up key packets to play the walking animation of the player |
| 14:15 |
ANAND |
The engine not working as it should isn't a "mod bug", lol :) |
| 14:16 |
IhrFussel |
But there are way worse cases where the missing up key allows actual exploits ... such a hunger/stamina mods ... I could RIGHT NOW go to CTF probably with auto forward and NO STAMINA will be reduced when I run (unless rubenwardy added a velocity check too) |
| 14:16 |
nerzhul |
VanessaE: maybe find a nice dev and pay it can be the solution, but we are not ready for a such thing |
| 14:16 |
VanessaE |
nerzhul: and actually, that was requested almost 4 years ago, I just looked it up |
| 14:16 |
ANAND |
True, and we can't blame sprint for that |
| 14:17 |
nerzhul |
IhrFussel: i don't know about mtg i never looked at the game code itself. Only think i know about it is mobs must be added to it to be attractive for me |
| 14:17 |
nerzhul |
VanessaE: i don't know about the core part you talked about sorry :) |
| 14:17 |
nerzhul |
i cannot help on that part |
| 14:18 |
VanessaE |
nerzhul: it's not about paying for dev time to get a feature. it's about resources being aimed wrong. if you guys are gonna spend time coding *anyway*, why not work on stuff people actually asked for? |
| 14:18 |
VanessaE |
I actually don't care about colored lights, btw. it's just one thing that came to mind |
| 14:19 |
IhrFussel |
If you add all sorts of things people never wanted...they will not really be happy cause it#s basically the same as if you don't work on MT at all...that is the truth |
| 14:20 |
VanessaE |
as for this movement issue, it's simple: client initiates a movement, but doesn't send the keypresses that go with it. how's a mod supposed to detect that in a sane and fast manner? |
| 14:20 |
IhrFussel |
Not trying to sound disrespectful but the end result is exactly that... you worked on MT but people still don't get what they asked for |
| 14:21 |
nerzhul |
VanessaE: i understand the point. But you know about autoforward or not ? |
| 14:21 |
VanessaE |
nerzhul: I know only a little about it -- but i don't need to know ANything about it to know that movement without a corresponding keypress, outside of inertia or falling, is a bug. |
| 14:22 |
IhrFussel |
VanessaE, there is a way but it could return lots of false positives: Velocity ... you cannot assume at all times however that the player must be walking just cause velocity is 4 |
| 14:23 |
VanessaE |
IhrFussel: hence my "sane" requirement. |
| 14:23 |
nerzhul |
IhrFussel: i'm generally working on the MT core code, the non feature part because it needs to be polished more and more to permit those features to be coded properly, i understand you point about the feature. I hope all 5.0.0 will be used properly by modders because they will have many work to use the new feature set :) |
| 14:23 |
nerzhul |
VanessaE: the problem is autofoward is not keypress |
| 14:23 |
nerzhul |
it's why this PR is hack for me |
| 14:23 |
nerzhul |
and not a bugfix |
| 14:24 |
VanessaE |
nerzhul: right, but autoforward initiates a movement, and a movement NEEDS a keypress to be handled properly by a mod. whether that PR fixes it properly is a matter for debate. but a fix is needed. |
| 14:24 |
IhrFussel |
How else you want to fix this? Remove auto forward? |
| 14:24 |
VanessaE |
ergo, there is a bug to be fixed here. |
| 14:25 |
IhrFussel |
Auto Forward HAS TO tell the server that the player is moving right now |
| 14:25 |
IhrFussel |
And sending infinite up key packets sounds right cause those can be recognized by mods |
| 14:26 |
VanessaE |
I guess the correct solution would be to send a single keypress event on the start of automove, and send the key-up event on stop |
| 14:26 |
IhrFussel |
infinite until you turn off Auto Forward* |
| 14:26 |
VanessaE |
so that the server perceives such a move as "W" being held down, or whatever. |
| 14:26 |
nerzhul |
IhrFussel: i don't want to remove autoforward, but if we send the autoforward toggle is this sufficient ? |
| 14:27 |
IhrFussel |
Then all mods need to add yet another check |
| 14:27 |
nerzhul |
but it's the proper way to do it |
| 14:27 |
IhrFussel |
You could set another controls table key to true |
| 14:28 |
IhrFussel |
Like 'if controls.up or controls.auto_forward then' |
| 14:28 |
VanessaE |
that would work. |
| 14:28 |
IhrFussel |
But then the server needs to keep track of it |
| 14:29 |
VanessaE |
nerzhul: see? a "battle" with us has benefits :P |
| 14:29 |
IhrFussel |
Pretty sure if 'forgets' all other key presses next step unless the client still presses them |
| 14:29 |
IhrFussel |
it* |
| 14:30 |
nerzhul |
VanessaE: maybe but it's tiring, you are never happy with me :D |
| 14:30 |
IhrFussel |
Cause continuously sending 'auto_foward = true' is the same as sending 'up' |
| 14:30 |
VanessaE |
nerzhul: I think it comes down players/admins want X, but you don't want X, you want Y... |
| 14:31 |
VanessaE |
(when perhaps most of the time, both X and Y could be added without a problem) |
| 14:32 |
IhrFussel |
Why do I keep typing 'foward' even though I know how to spell it... forward* |
| 14:32 |
VanessaE |
IhrFussel: well you wouldn't need a continuous stream of "auto-forward = true", you could just toggle it server-side in response to start and stop packets. |
| 14:32 |
|
aerozoic joined #minetest-hub |
| 14:33 |
IhrFussel |
Yep that's why I meant the server needs to remember the flag...and it doesn't currently remember any other movement flags from clients (correct me if I'm wrong) |
| 14:36 |
nerzhul |
we can add it to movement packet |
| 14:36 |
VanessaE |
nerzhul: the reason you perceive as no one being happy with you is because of the core of this argument: devs focusing on background stuff and "X must be coded now so that we can implement Y that won't happen for 5 years anyway" instead of "users want Y, so let's code a basic Y now, with hardcoded defaults and settings, and work on making X later and putting Y on top of X." |
| 14:37 |
nerzhul |
VanessaE: yeah, remember MT performance before i come to the project, now we have have 0.1 step, before it was 0.5 :) |
| 14:37 |
VanessaE |
(which is how MT used to be done early on, and how I do my mods) |
| 14:37 |
IhrFussel |
nerzhul, but if you send it as movement packet isn't it the same as sending 'up' forever? I see no difference in that |
| 14:38 |
nerzhul |
one way is crappy other is proper :) |
| 14:38 |
VanessaE |
nerzhul: yes, I remember. I also remember when we had a texture atlas and it was simply removed instead of being made dynamic to handle more texture data. |
| 14:38 |
IhrFussel |
Only that the 'proper' way then requires all mods to add another flag for no reason |
| 14:39 |
VanessaE |
(not saying that's your doing, just an example of something that was just torn out instead of being properly fixed) |
| 14:40 |
IhrFussel |
I thought you meant it like this: Send 1 packet to server which tells 'i enabled auto forward now' and server remembers the flag for the client and just adds it to get_player_control() until the client tells server 'i disabled it now' and server sets the flag to false |
| 14:41 |
IhrFussel |
But that would require reliable packets of course |
| 14:41 |
IhrFussel |
Nothing can get lost |
| 14:54 |
nerzhul |
VanessaE: texture atlas should be very old, i never heard about it |
| 14:55 |
nerzhul |
IhrFussel i'm more in favor to send autoforward in the movement packet |
| 14:56 |
VanessaE |
nerzhul: it's quite old, removed in early 0.4.x because people were starting to use more texture data than the static atlas could hold. |
| 14:56 |
|
DI3HARD139 joined #minetest-hub |
| 14:56 |
VanessaE |
nerzhul: c55 had a variant of it in his sorta-recent farmap code, I think (I mean the LoD thing, not the old "farmesh") |
| 14:59 |
|
entuland joined #minetest-hub |
| 15:13 |
|
benrob0329 joined #minetest-hub |
| 15:57 |
|
twoelk joined #minetest-hub |
| 16:03 |
|
Krock joined #minetest-hub |
| 16:03 |
Krock |
o/ |
| 16:05 |
ANAND |
Greetings :) |
| 16:06 |
|
RichardTheTurd joined #minetest-hub |
| 16:13 |
twoelk |
o/ |
| 16:15 |
|
scr267 joined #minetest-hub |
| 16:50 |
|
IhrFussel joined #minetest-hub |
| 17:12 |
IhrFussel |
Does inv:get_list(listname) return a table? |
| 17:13 |
IhrFussel |
I need to loop through the detached armor inventory |
| 17:16 |
Calinou |
what are some well-written mods to you? I'd just like to take a look at the code used in modern mods :) |
| 17:16 |
Krock |
IhrFussel: yes it does |
| 17:16 |
Krock |
a table of itemstacks |
| 17:17 |
Krock |
Calinou: all recent ones of mine ofc :OP |
| 17:26 |
|
FrostRanger joined #minetest-hub |
| 17:28 |
VanessaE |
Calinou: and all of mine ;) |
| 17:30 |
IhrFussel |
Looks like 3d_armor got a bug (at least in my version) where you can wear any item by first putting a valid armor type then taking it with your mouse and swapping with something else on the fly |
| 17:31 |
IhrFussel |
Then your player turns into 'invalid texture' and the chat spits errors |
| 17:39 |
Calinou |
nerzhul: the server step was never 0.5, it was 0.2 before 0.4 IIRC |
| 17:42 |
nerzhul |
Calinou if i remember it was 0.5 in 0.4.13 |
| 17:44 |
Calinou |
https://github.com/minetest/minetest/blame/8432efa308e0d6c56ad61df189981ed03b0af463/minetest.conf.example#L1244 says otherwise :) |
| 17:44 |
Calinou |
(that's before you reduced it to 0.09 in https://github.com/minetest/minetest/pull/6252) |
| 17:47 |
nerzhul |
right |
| 17:53 |
IhrFussel |
Can I use dump(table) if I just want to check if a table contains a certain string anywhere? |
| 17:54 |
|
entuland_ joined #minetest-hub |
| 17:54 |
IhrFussel |
Like 'if string.find(dump(def.groups),'armor') then' |
| 17:54 |
|
GreenDimond joined #minetest-hub |
| 17:55 |
sfan5 |
you can |
| 17:55 |
sfan5 |
but it's terrible programming |
| 17:56 |
Krock |
IhrFussel: why don't you just use table.contains ? |
| 17:56 |
Krock |
or table.indexof which is in the API? |
| 17:58 |
Krock |
it's documented in 5.0.0-dev but exists for a long time already |
| 17:58 |
GreenDimond |
THATS A THING?! |
| 17:58 |
sfan5 |
don't armor tables just use keys? so you'd have to do def.groups.armor ~= nil |
| 17:58 |
Krock |
oh right. that function compares values only |
| 17:59 |
Krock |
well, but even for key comparison it's a very short helper function to add to your mod |
| 17:59 |
Krock |
that's a sane and faster solution than dump() and find() |
| 18:00 |
GreenDimond |
So should I be using table.indexof instead of looping through tables to find matching values? |
| 18:01 |
sfan5 |
table.indexof does just that |
| 18:02 |
sfan5 |
but you should be using it as there is no need to reimplement that yourself |
| 18:03 |
GreenDimond |
I totally would have had I known it existed |
| 18:06 |
Krock |
GreenDimond: exists for ~4 years, but the documentation is fresh |
| 18:07 |
Krock |
as in: added a few commits ago |
| 18:09 |
|
aerozoic joined #minetest-hub |
| 18:11 |
xerox123_ |
hello o/ |
| 18:11 |
Krock |
o/ xerox123_ |
| 18:20 |
IhrFussel |
I just decided to do it another way... 'if string.find(sname,"helmet") or string.find(sname,"chestplate").......' |
| 18:20 |
IhrFussel |
Checking armor groups is kinda complex |
| 18:21 |
IhrFussel |
Cause some armor types have groups like armor_water=1 or armor_fire=1 and other just have armor_helmet=1 x.x |
| 18:21 |
|
entuland joined #minetest-hub |
| 18:30 |
IhrFussel |
Now players can still use this bug to put any items into their armor inventory, but the visuals are not upgraded -> no errors in chat |
| 18:30 |
IhrFussel |
updated* |
| 18:30 |
|
entuland_ joined #minetest-hub |
| 18:30 |
|
aerozoic joined #minetest-hub |
| 18:32 |
|
entuland_ joined #minetest-hub |
| 18:32 |
|
paramat joined #minetest-hub |
| 18:37 |
|
entuland joined #minetest-hub |
| 18:38 |
paramat |
nerzhul is right and i agree that 8249 isn't a critical enough a bug to merge now, MT has many other non-critical bugs which will have to wait otherwise 5.0.0 will never be released. 5.1.0 is intended to be soon |
| 18:40 |
paramat |
besides, the PR has issues, is not tested and review has barely begun |
| 18:41 |
|
entuland joined #minetest-hub |
| 18:46 |
|
tenplus1 joined #minetest-hub |
| 18:46 |
tenplus1 |
hi folks :) |
| 18:46 |
tenplus1 |
hey Krock |
| 18:46 |
|
entuland joined #minetest-hub |
| 18:46 |
tenplus1 |
hi entuland |
| 18:48 |
Krock |
hi tenplus1 |
| 18:48 |
tenplus1 |
o/ how's things ? |
| 18:50 |
Krock |
but apart from that I'm fine. earlier weekend this time :D |
| 18:50 |
|
entuland joined #minetest-hub |
| 18:50 |
tenplus1 |
heh, went to see Alita: Battle Angel last night... was amazing :D |
| 18:53 |
tenplus1 |
Stamina mod (my fork) has been updated, drunk effect working well now :D |
| 18:54 |
Krock |
heh. did you tweak the player physics a little? |
| 18:54 |
tenplus1 |
not physics, player yaw :) it's funny trying to walk around while drunk and you're tweaking in all directions |
| 18:55 |
tenplus1 |
players on server are getting each other drunk on purpose now :) especially playing Spleef! |
| 18:57 |
tenplus1 |
turns out Wuzzy has made a player physics api as well :) think we're totalling 5 now :D ehehe |
| 18:57 |
|
Sokomine joined #minetest-hub |
| 18:58 |
lumberJ |
hey tenplus1 |
| 18:58 |
tenplus1 |
hi lumberJ o/ |
| 18:58 |
lumberJ |
what are the other 4 physics apis? |
| 18:59 |
tenplus1 |
Pova, player_monoids, late, physicsh and wuzzy's player_physics now |
| 18:59 |
tenplus1 |
looks ok but is overly simple with no fallback |
| 18:59 |
Krock |
time for a 6th standard |
| 18:59 |
tenplus1 |
https://forum.minetest.net/viewtopic.php?f=11&t=22172 |
| 18:59 |
Krock |
!title |
| 18:59 |
MinetestBot |
Krock: [Mod] Player Physics API [1.0.0] [playerphysics] - Minetest Forums |
| 19:01 |
tenplus1 |
3d_armor, playerplus and sprint support pova and player_monoids |
| 19:01 |
lumberJ |
i see. its hard for me to keep track of so many mods |
| 19:01 |
tenplus1 |
yeah, there was a pull request on git to include this kinda thing in minetest itself... never got included |
| 19:02 |
lumberJ |
i was thinking of player effects. had it confused |
| 19:03 |
tenplus1 |
the potions ? |
| 19:04 |
lumberJ |
i think it supports those but it is a framework of sorts for managing effects |
| 19:05 |
tenplus1 |
yeah, handles timers and the like to apply effects |
| 19:06 |
tenplus1 |
I use pova for simplicity and ease of use :P been running on xanadu for over 10 months now without issue |
| 19:14 |
tenplus1 |
how's your game coming along lumberJ? |
| 19:14 |
tenplus1 |
o/ Green |
| 19:15 |
lumberJ |
slow and steady |
| 19:15 |
GreenDimond |
Oh hello |
| 19:15 |
tenplus1 |
:P how's you? |
| 19:15 |
lumberJ |
not a lot of coding time when you have 1 year old around |
| 19:16 |
tenplus1 |
very true, tech the kid how to code and make it a joint effort :)))))) |
| 19:16 |
tenplus1 |
*teach |
| 19:18 |
lumberJ |
when he gets a bit older, i will for sure. |
| 19:18 |
lumberJ |
he already gets too much screen time as it is for his age though |
| 19:19 |
lumberJ |
he and I both |
| 19:19 |
tenplus1 |
yeah, kids and ipads/phones etc. should be limited |
| 19:20 |
lumberJ |
waiting for the rain to let up a bit to get out of the house with him now |
| 19:21 |
tenplus1 |
kewl, it's cold here again, the heatwave didnt last long :D |
| 19:21 |
lumberJ |
yeah, its not bad today but supposed to get fairly cold over the next week |
| 19:22 |
tenplus1 |
they say a storm is coming, brr |
| 19:23 |
tenplus1 |
speaking of which, I saw the wind and storm mod on minecraft, damn did it look really well done... if only |
| 19:23 |
scr267 |
Hi tenplus1 |
| 19:23 |
tenplus1 |
hey scr267 :) hows u ? |
| 19:23 |
scr267 |
Hi lumberJ |
| 19:23 |
scr267 |
not bad |
| 19:23 |
scr267 |
yourself? |
| 19:23 |
lumberJ |
hey scr267 |
| 19:24 |
tenplus1 |
good ta, watching anime and mod tinkering as always :D |
| 19:24 |
lumberJ |
i'm going to head out. the rain is slowing up so, better seize the chance. later all |
| 19:24 |
tenplus1 |
cya dude, enjoy o/ |
| 19:27 |
scr267 |
:) |
| 19:41 |
|
entuland_ joined #minetest-hub |
| 19:41 |
tenplus1 |
wb entu :P |
| 19:46 |
tenplus1 |
time for noms :) laters all o// |
| 19:46 |
|
tenplus1 left #minetest-hub |
| 20:30 |
luk3yx |
Can someone review (or merge) https://github.com/minetest-mods/quartz/pull/10? |
| 20:43 |
sofar |
sure |
| 20:46 |
sofar |
don't see major issues with it |
| 20:52 |
|
kaeza joined #minetest-hub |
| 20:57 |
luk3yx |
Fixed |
| 21:15 |
|
Fixer joined #minetest-hub |
| 22:37 |
|
scr267a joined #minetest-hub |
| 22:43 |
|
T^4im joined #minetest-hub |
| 22:56 |
|
T4im joined #minetest-hub |
| 23:03 |
|
kaeza joined #minetest-hub |
| 23:10 |
|
paramat joined #minetest-hub |
| 23:34 |
|
IcyDiamond joined #minetest-hub |
