DDraceNetwork - developer

This should be a Big message louis

1

06:28

06:29

Finally u stoooed

06:29

Fuck

can anyone confirm this? "while in a stopper, your velocity vector parallel is set to 0 every tick (and you are also given a doublejump if the stopper is vertical). Since stoppers don't actively push you out, the tick that you enter the stopper will determine your position, as you will be stopped immediately after. This is the reason why when walking or falling into a stopper, your position will fluctuate based on how you entered. This also leads to a special case when completely skipping a stopper in one tick - afterwards, your speed will be set to 0, and you will be given a doublejump if the stopper is vertical."

06:29

took a long time cus im typing on my phone justatest

1

@louis leak part

pollito part

louis

can anyone confirm this? "while in a stopper, your velocity vector parallel is set to 0 every tick (and you are also given a doublejump if the stopper is vertical). Since stoppers don't actively push you out, the tick that you enter the stopper will determine your position, as you will be stopped immediately after. This is the reason why when walking or falling into a stopper, your position will fluctuate based on how you entered. This also leads to a special case when completely skipping a stopper in one tick - afterwards, your speed will be set to 0, and you will be given a doublejump if the stopper is vertical."

"while in a stopper, your velocity vector parallel is set to 0 every tick (and you are also given a doublejump if the stopper is vertical)" yes. "Since stoppers don't actively push you out, the tick that you enter the stopper will determine your position, as you will be stopped immediately after." yes. "This is the reason why when walking or falling into a stopper, your position will fluctuate based on how you entered. This also leads to a special case when completely skipping a stopper in one tick - afterwards, your speed will be set to 0, and you will be given a doublejump if the stopper is vertical." yes.

07:40

get doublejump and speed y set to 0.

07:41

what about how you don't get dj if you are hooking someone

louis

can anyone confirm this? "while in a stopper, your velocity vector parallel is set to 0 every tick (and you are also given a doublejump if the stopper is vertical). Since stoppers don't actively push you out, the tick that you enter the stopper will determine your position, as you will be stopped immediately after. This is the reason why when walking or falling into a stopper, your position will fluctuate based on how you entered. This also leads to a special case when completely skipping a stopper in one tick - afterwards, your speed will be set to 0, and you will be given a doublejump if the stopper is vertical."

" velocity vector parallel to the stopper direction" maybe

plsplsplslol

what about how you don't get dj if you are hooking someone

it may because ur speed y is not bigger than 0.

i think it also does when you are hooking someone above you

07:45

also how would you get dj by standing on stoppers then

i think it also relate about speed calculated from hooking and gravity, but i am not clear about it.

07:48

u can see speed is calculated earlier than tile(stopper)

07:49

if u are interesting in, u can find this in src\game\server\entities\character.cpp void CCharacter::Tick()

ok i think i found it actually

08:16

https://github.com/ddnet/ddnet/blob/469136b5e7c7ae30b5ea22aff59e9f0845e5452a/src/game/gamecore.cpp#L427

ddnet/gamecore.cpp at 469136b5e7c7ae30b5ea22aff59e9f0845e5452a · dd...

DDraceNetwork, a cooperative racing mod of Teeworlds - ddnet/gamecore.cpp at 469136b5e7c7ae30b5ea22aff59e9f0845e5452a · ddnet/ddnet

08:16

when you are hooking someone it calculates stoppers earlier than normal

1

08:17

so m_Vel.y is 0 at // stopper

08:18

if you arent hooking it clamps velocity after it checks m_Vel.y > 0

@Patiga i was checking ur twmap code and i found this, actually the code on line 174 is unreachable, because ? on a Err always returns, thats why its always better to return errors using "return" when you creare ur own Err() like this

1

08:37

ill pr later

twmap masterrace

https://gitlab.com/Patiga/twmap/-/merge_requests/3/diffs

09:38

09:44

tis beautiful

09:44

Well almost all languages have a swap in their stdlib :D

@Patiga how does the py script work

ah I hoped it would explain itself

11:10

generate_external_constants.py 6 ~/usr/share/ddnet/data/mapres ddnet_external.rs and generate_external_constants.py 7 ~/usr/share/teeworlds/data/mapres teeworlds_external.rs

11:11

(assuming you are on arch and have ddnet and teeworlds 0.7 installed)

ok i pushed

@Ryozuki gitlab gives me the option to squash the commits, mind if I do that?

i dont mind

@Patiga u should run cargo fmt too

I'm scared

11:46

but ye

scared of?

the amount of lines changed

about 4k

11:47

iirc

11:48

in the future u can use this https://gist.github.com/hryniuk/e3f4ae34e37947d0ff55a3f63add90f8

Git pre-commit hook that checks Rust code style with cargo fmt

Git pre-commit hook that checks Rust code style with cargo fmt - pre-commit-cargo-fmt

@Patiga all the highlighted ones dont show any maps and Stop could probably be renamed to Stopper

thanks for notifying :), changes are already submitted

def-

[ddnet-libs:pr-sdl-2012] 1 new commit

9207113 Upgrade SDL2 2.0.18 -> 2.0.20 (except on Windows) - def-

11:58

def-

[ddnet/ddnet-libs] New branch created: pr-sdl-2012

11:58

def-

[ddnet/ddnet-libs] Pull request opened: #24 Upgrade SDL2 2.0.18 -> 2.0.20 (except on Windows)

12:01

def-

[ddnet/ddnet] Pull request opened: #4588 Upgrade SDL2 2.0.18 -> 2.0.20 (except on Windows)

Still works on macOS

Checklist

[x] Tested the change ingame
[ ] Provided screenshots if it is a visual change
[ ] Tested in combination with possibly related configuration options
[ ] Written a unit test if it works standalone, system.c especially
[ ] Considered possible null pointers and out of bounds array indexing
[ ] Changed no physics that affect existing maps
[ ] Tested the change with [ASan+UBSan or valgrind's memcheck](https://github.com/ddnet/ddnet/#using...

Windows still having performance issues on newer sdl?

@Ryozuki scanned a bit over the files, and I dislike the formatting it does to the generated files. it messes up the arrays also I saw a weird thing it did to src/lib.rs:

pub mod map_checks;  // comment1
pub mod map_edit;    // comment2
pub mod map_dir;     // comment3

was changed to

+pub mod map_checks; // comment1
+pub mod map_dir;
+pub mod map_edit; // comment2 // comment3

hmm weird

You dare question cargo fmt?

i never had problems with it

12:08

can u show me the array thing

You obviously wanted comment 2 and 3 merged

definitely more concise or smth

ok do whathever

@Ryozuki don't wanna paste the whole array here, too long

12:09

I'll send in pm

I've actually had no issues with cargo fmt either. Odd that it's messing up there

12:13

honestly this looks like a bug in cargo fmt

12:14

maybe if u can find a reduced code version of it u can report it

12:15

i can reproduce it

12:15

ill report it

yeah its really simple to reproduce o.o

on the arrays u can add the skip tag i told u

Learath2

Windows still having performance issues on newer sdl?

didn't check but there was no relevant change in changelog

Patiga

thanks for notifying :), changes are already submitted

So I should just rerun it?

yes, with the new changes

12:17

I also finally rebased to master ^^

https://github.com/rust-lang/rustfmt/issues/5168

12:18

i opened it

hm

12:22

next time I wanna open that issue myself :)

@Ryozuki do you have a fix for the long lines of tiles or should I make one?

make it

12:37

u should wrap all the tiles in a div with some max weidth

12:37

width

12:37

and display:flex the div probs

def-

[ddnet-scripts:master] 1 new commit

a8b5d70 SDL 2.0.18 -> 2.0.20 - def-

with flex-wrap: wrap

@deen the names and grouping I did might've been a bit confusing: HOOKTHROUGH_ALL is just new hookthrough (THROUGH is old hookthrough), PLASMAU is the unfreezing plasma turret. I mapped the freezing+exploding bullet into the freezing bullet

ok

13:08

so no plasmae?

as it is right now yes, should I separate it from the plasmaf?

ok, no. that's fine

1

def-

[ddnet-web:master] 1 new commit

7f9c087 Fix alignment with more tiles - def-

@Patiga weird some maps still report PLASMAE and some PLASMAF

ah I mixed up the names in my head.

14:03

there are PLASMA, PLASMAF, PLASMAE, PLASMAU. F -> freeze, e -> explosion, u -> unfreeze, PLASMA -> freeze + explosion and I mapped PLASMA onto PLASMAF

14:04

so PLASMAE should have the icon at index 220, PLASMAF -> 221, PLASMAU -> 223

14:05

and PLASMA is mapped onto 220

ok

Any idea why we have 2 refresh rate sliders in the settings now?

14:28

abit confusing

graphics thread and input thread

Yeah, but why

you can set both, they have different effects

there are a couple of settings which should be moved to different tabs e.g: Show name plates + name plates size, Show clan above nameplates, in General to HUD Show only chat messages from friends, In General to HUD > Chat Tab Refresh rate, In General to Graphics and rename both according to what they do Automatically record demos and the slider in, General to DDNet in the demo section or do we have not enough space? ^^

layer arrows unseeable under blocks fix it place

hussainx3

layer arrows unseeable under blocks fix it place

like here i can't know how he did it was holding A or some A and D or not move at all zzzz

14:47

i suggest putting layer arrows ,up

https://www.bleepingcomputer.com/news/security/dev-corrupts-npm-libs-colors-and-faker-breaking-thousands-of-apps/

Dev corrupts NPM libs 'colors' and 'faker' breaking thousands of apps

Users of popular open-source libraries 'colors' and 'faker' were left stunned after they saw their applications, using these libraries, printing gibberish data and breaking. Some surmised if the NPM libraries had been compromised, but it turns out there's more to the story.

14:59

How can project rely on other library being updated without their consent? This js bullshit is beyond me

15:00

But the bottom topic is probably of concern here too

@Chairn this has to do with how semantic versioning works