DDraceNetwork - developer

DDraceNetwork

Development / developer

Development discussion. Logged to https://ddnet.tw/irclogs/ Connected with DDNet's IRC channel, Matrix room and GitHub repositories — IRC: #ddnet on Quakenet | Matrix: #ddnet-developer:matrix.org GitHub: https://github.com/ddnet

Between 2023-07-16 00:00:00Z and 2023-07-17 00:00:00Z

Hi there!! my name is Alica mattielo how are you doing today, and how is everything going over there and your family,I am from Liverpool UK, though the situation in here isn't friendly, we are dealing with tyrant governments. How is your year going off ?

justatest

6

Omg a real person

[ddnet-maps:master] 1 new commit

fc21c2f A sus'terism - ddnet-maps

The server wasn’t full there were always enough slots left (@Iza)

talking about blocking slots

07:11

Im used to GitHub. It works nice to sync across all my devices and I am already authenticated anyways. (@Voxel)

chiller why are you storing it on github

GitHub

Click to see attachment 🖼️

Embedded image

[ddnet/ddnet] Issue opened: #6869 Crash in CPlayers::RenderHand

Reported by Astramast on Discord on DDNet 17.1 win64 website version: ``` da89321e018fbfb27a33861b5724ae5f.RTP Module offset: 0x00007FF6825D0000 Image base: 0x140000000 0xffff800b032da30f: ?? ??:0 0xbce57ffdbda30000: ?? ??:0 0xffff800afd22e62f: ?? ??:0 0x00000001400c45d1: std::atan(float) at /usr/x86_64-w64-mingw32/include/c++/12.2.0/cmath:126 (inlined by) angle(vector2_base const&) at src/base/vmath.h:123 (inlined by) CPlayers::RenderHand(CTeeRenderInfo const*, vector2_base, vect...

omg im this close to downgrading to debian 11

09:08

using touchpad moves the in game cursor only 50% of the time in debian 12 any other ideas than downgrading?

buy mouse

that indeed is a good alternative to touchpad

09:10

but im touchpad gamer when on laptop :D

09:10

just so annoying to travel with mouse also often there is no space for a mouse anyways

09:11

i should switch to keyboard only gaming mouse always making trouble

09:11

can we add aim assist to ddnet for keyboard only players?

can we add move assist for mouse only players

:D

09:12

mouse only players lmao

09:13

thing is you can move way better with mouse than aim with keyboard

09:14

maybe i just need a few angle binds on my laptop its not like im doing ESL gaming from ma hotspot in some cafe

09:14

eventho i did pretty well in a gctf tourney once axaxax

09:15

jopsti wanna hear latest 0.7 dev progress?

sure, please post a very detailed list

09:15

atleast 20000 sites

currently the client whispers its timeout code to a random person on the server on connect

09:15

:D

sounds good enough

however i managed to bug that

09:16

i will fix it today because its 0.7 weekend

i wish you best of luck

:)

Embedded image

09:29

epic vitory royale

09:29

also fakin 0.7 bullying by ddnet server

Robyt3

Somehow git branch --merged no longer works after we started using GitHub merge queue. I previously used this command in a script to delete merged branches automatically. Now git branch -d PR-Branch also says error: The branch 'PR-Branch' is not fully merged. for branches that have been merged with the merge queue.

this is likely a side effect of enabling rebase-merge. I can turn it back to merge-merge

heinrich5991

this is likely a side effect of enabling rebase-merge. I can turn it back to merge-merge

Yeah, that would be easier, although the history isn't as readable in that case

@Robyt3 thanks! https://github.com/ddnet/ddnet/pull/6861

Use member instead of static variables in CGameClient by Robyt3 ·...

Checklist Tested the change ingame Provided screenshots if it is a visual change Tested in combination with possibly related configuration options Written a unit test (especially base/) or add...

Robyt3

Yeah, that would be easier, although the history isn't as readable in that case

changed it to merge-merge

Thanks, that should make my script work much easier again, the workaround (git range-diff --left-only PR-Branch...master for every individual branch) would have made it much slower

wot script

Uses git branch --merged to get all local merged branches and then deletes them automatically, and uses git remote prune origin to prune references to deleted remote branches.

10:09

And updates my fork's master to upstream's master

I have a weird liking for the rebase merge

10:18

not sure why the non-linear history bothers me ^^ but I see no actual benefit of the linear history, so whatever

[ddnet/ddnet] Pull request opened: #6870 Use lock file for dependencies

This makes breakages like in #6863 less likely.

Checklist

[ ] Tested the change ingame
[ ] Provided screenshots if it is a visual change
[ ] Tested in combination with possibly related configuration options
[ ] Written a unit test (especially base/) or added coverage to integration test
[ ] 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 memchec...

ChillerDragon

wot script

Bot speaks like a human

Does anyone have some advice on how to format the changes of the current branch? Previously I just formatted before every commit, but somehow my editor does not format only modified lines anymore.

marmare_314

Does anyone have some advice on how to format the changes of the current branch? Previously I just formatted before every commit, but somehow my editor does not format only modified lines anymore.

are you using the scripts/fix_style.py script?

10:57

that one should format it just like in CI

I tried but it formatted everything. Let me try it again

but in the rest of the code, it shouldn't format anything, as it already is formatted

The issue is if you have a newer clang-format version it will start using a different format for comments (adding one space after every //)

Oh was about to ask if that may cause problems, I was using clang-format-17

Are you on Windows using Visual Studio?

No debian

scripts/fix_style.py should verify the clang-format version

11:02

before executing it

Is there not an option to set the commet-style? Otherwise I will have to figure out where to get clang-format-10

heinrich5991

scripts/fix_style.py should verify the clang-format version

Yes it does, but I changed it

marmare_314

Yes it does, but I changed it

I mean you might wanna stop hitting yourself

11:03

https://github.com/muttleyxd/clang-tools-static-binaries

GitHub - muttleyxd/clang-tools-static-binaries: static-linked versi...

static-linked versions of clang tools for Linux, Mac OS X and Windows, available in AUR as clang-format-static-bin - GitHub - muttleyxd/clang-tools-static-binaries: static-linked versions of clang ...

11:03

you can get the right clang-format version here ^

[ddnet/ddnet] Pull request opened: #6871 Remove usages of CUI::LastActiveItem

Use CLineInput::GetActiveInput and CLineInput::IsActive instead for consistency. The last active item pointer is now only tracked internally in CUI to deactivate the active line input when it's no longer used in the UI.

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 (especially base/) or added coverage to integration test
[ ] C...

Robyt3

The issue is if you have a newer clang-format version it will start using a different format for comments (adding one space after every //)

There is SpacesInLineCommentPrefix, but only from version 13 on. Does it throw errors on unknown options?

At the moment it probably will, unless we are using this flag https://reviews.llvm.org/D86137

github-merge-queue[bot]

[ddnet:master] 2 new commits

a2dd529 Use lock file for dependencies - heinrich5991 0fcbac2 Merge pull request #6870 from heinrich5991/pr_ddnet_ci_locked - Chairn

11:17

[ddnet/ddnet] Pull request closed: #6870 Use lock file for dependencies

11:19

[ddnet/ddnet] Pull request opened: #6872 Multiple selection in envelope editor

This PR allows multiple selection in the envelope editor and (because I am bad at separating features) adds a context menu to envelope points. !env_context

Checklist

[ ] Tested the change ingame
[ ] Provided screenshots if it is a visual change
[ ] Tested in combination with possibly related configuration options
[ ] Written a unit test (especially base/) or added coverage to ...

When compiling DDnet.exe X64 Debug I am facing this error https://media.discordapp.net/attachments/995332467219251221/1130091627810140190/image.png https://media.discordapp.net/attachments/995332467219251221/1130091628183441518/image.png https://media.discordapp.net/attachments/995332467219251221/1130091628443480074/image.png?width=1273&height=625

Embedded image

Embedded image

Embedded image

windows moment

11:35

do you have rust installed?

ChillerDragon

do you have rust installed?

I don't think I have installed it

11:38

How should I install?

Can you believe tokio still doesn't have a condition variable?

mAdIM

How should I install?

https://rustup.rs/

rustup.rs - The Rust toolchain installer

The Rust toolchain installer

Imagine living in 2023 and not being able to use a condition variable

11:45

I don't even know what really replaces it but just polling on an atomic with exponential backoff

Learath2

Can you believe tokio still doesn't have a condition variable?

can you use a semaphore maybe?

11:46

not sure what exactly a condition variable is, sorry

heinrich5991

not sure what exactly a condition variable is, sorry

You don't know what a pthread_cond_t is?

correct

heinrich5991

can you use a semaphore maybe?

Yes, but it's a fairly delicate process coding one out of the other

condition variable supports the following? waiting to be signalled

http://birrell.org/andrew/papers/ImplementingCVs.pdf thankfully it's documented

can at most one thread that's waiting be signalled?

11:49

“Condition Variable” with operations Wait, Signal and Broadcast

[ddnet/ddnet] Pull request opened: #6873 Use lambdas for editor index modify and sort functions

For cleaner code with less global variables.

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 (especially base/) or added coverage to integration test
[ ] 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](...

what are these three operations?

Condition variables allow you to wait on a "condition", when you wake up you are guaranteed to be holding a mutex

[ddnet/ddnet] Pull request opened: #6874 Port JSON writer from upstream, improve testing

Port the CJsonWriter utility class from upstream, which makes outputting correct JSON easier. Add CJsonWriter as an abstract class that can write to different outputs. Two implementations CJsonFileWriter (writes to a file) and CJsonStringWriter (writes to an std::string) are added. Upstream CJsonWriter can only write to files. The same tests are added for both implementations. Duplicate code is avoided by using typed tests with two separate test fixtures.

Checklist

...

ah

The condition is over a shared variable that is protected by said mutex

heinrich5991

https://rustup.rs/

I have installed rust but I am still facing the previous error

it's basically a mutex with wait function?

heinrich5991

it's basically a mutex with wait function?

Mutex with conditional wait

mAdIM

I have installed rust but I am still facing the previous error

can you open a terminal an execute rustup default stable?

Learath2

Mutex with conditional wait

where does the "condition" come from?

11:51

the "condition" is the other thread signalling the mutex?

No it's like an actual predicate, when you modify the shared variable changing the predicated you drop the mutex and signal the condvar

heinrich5991

can you open a terminal an execute rustup default stable?

Yes

If there is anyone is waiting, they will be woken up and they will be guaranteed to have the mutex when they do

Learath2

No it's like an actual predicate, when you modify the shared variable changing the predicated you drop the mutex and signal the condvar

so you can additionally specify a condition that is checked whenever someone releases the mutex?

Learath2

No it's like an actual predicate, when you modify the shared variable changing the predicated you drop the mutex and signal the condvar

s/predicated/predicate/

mAdIM

Yes

what's the output of that? does it work afterwards?

heinrich5991

so you can additionally specify a condition that is checked whenever someone releases the mutex?

Whenever someone releases the mutex AND signals the condvar, so a change in predicate is not guaranteed to wake you up

Learath2

Whenever someone releases the mutex AND signals the condvar, so a change in predicate is not guaranteed to wake you up

wait, so someone releases the mutex AND signals the condvar AND your user-provided condition changed?

Anyway, the desired behaviour is to wake someone up on changes, you just need to follow a protocol

heinrich5991

wait, so someone releases the mutex AND signals the condvar AND your user-provided condition changed?

If someone is changing the condition, they need to first be holding the mutex, then when they are done changing, they drop the mutex and signal the condvar

I see

11:55

but the condition is only checked on the thread that's waiting?

heinrich5991

but the condition is only checked on the thread that's waiting?

Yes, (or threads in case of Broadcast)

11:56

There are also spurious wakeups, so you have to check the condition, even if the signaller had checked it before

11:56

It's not the most ideal but it's the only way I know of waiting on a change in a variable without polling

@Learath2 can you simply use a std::sync::Mutex and std::sync::Condvar?

heinrich5991

@Learath2 can you simply use a std::sync::Mutex and std::sync::Condvar?

I think that'd pause the entire thread then none of the other async stuff could progress

ah, true, condvar doesn't work

11:57

mb

Maybe I could steal a thread from tokio for this one process actually

(standard mutex would)

11:58

is there a reason you need this atomic being-signalled-and-taking the lock?

heinrich5991

(standard mutex would)

Does it? Wouldn't a standard std::sync::Mutex::wait() block the entire thread?

Learath2

Anyway, the desired behaviour is to wake someone up on changes, you just need to follow a protocol

cant u use channels

you could use a broadcast channel to signal change and take the mutex manually

XY problem?

Learath2

Does it? Wouldn't a standard std::sync::Mutex::wait() block the entire thread?

yes, it would, but it's kind of desired AFAIK: https://docs.rs/tokio/latest/tokio/sync/struct.Mutex.html (edited)

Mutex in tokio::sync - Rust

An asynchronous Mutex-like type.

11:59

Contrary to popular belief, it is ok and often preferred to use the ordinary Mutex from the standard library in asynchronous code.

heinrich5991

is there a reason you need this atomic being-signalled-and-taking the lock?

It's the only non-polling way of waiting on a condition I know that doesn't require kernel functionality, futex/WaitOnAddress

ye it is explained somewhere

Learath2

It's the only non-polling way of waiting on a condition I know that doesn't require kernel functionality, futex/WaitOnAddress

broadcast channel

Doesn't handle the case of a waiter missing the signal

what does that mean

have a receiver, take the lock, check the value, release the lock, then wait for the broadcast signal

Ryozuki

what does that mean

If I start waiting on a change message on the channel right after one went through I'm stuck forever

i.e. you need to have the receiver beforehand

I guess that is a reasonable compromise

I'd also be interested in what you want to achieve in a bigger picture (edited)

the issue with cond vars is task cancelattion iirc

heinrich5991

I'd also be interested in what you want to achieve in a bigger picture (edited)

Something absolutely trivial as always, I just want to properly terminate my program in a sane way

doesnt tokio provide that

12:03

https://docs.rs/tokio/latest/tokio/signal/index.html

tokio::signal - Rust

Asynchronous signal handling for Tokio.

Learath2

Something absolutely trivial as always, I just want to properly terminate my program in a sane way

depending on how many people need to listen, a broadcast channel or a oneshot channel seem appropriate

https://tokio.rs/tokio/topics/shutdown

Graceful Shutdown | Tokio - An asynchronous Rust runtime

Tokio is a runtime for writing reliable asynchronous applications with Rust. It provides async I/O, networking, scheduling, timers, and more.

12:04

https://docs.rs/tokio-util/latest/tokio_util/sync/struct.CancellationToken.html

CancellationToken in tokio_util::sync - Rust

A token which can be used to signal a cancellation request to one or more tasks.

Ryozuki

https://docs.rs/tokio/latest/tokio/signal/index.html

This is how my main thread is waiting on the ctrl_c

Ryozuki

https://docs.rs/tokio-util/latest/tokio_util/sync/struct.CancellationToken.html

This is what I was looking for, I guess in this case it replaces a condvar

@Learath2 look the blog post i sent too

12:05

it shows the cancel token

12:06

Image attachment

heinrich5991

depending on how many people need to listen, a broadcast channel or a oneshot channel seem appropriate

I was going for a broadcast channel, but I was having trouble making sure everyone properly got the signal. It seems CancellationToken has my desired behaviour of immediately returning when the task is already cancelled

[ddnet/ddnet] Issue opened: #6875 RFC: Remove Windows file ending treatment

The Windows default editor is able to work with pure \n endings nowadays. Maybe it's time to drop the \r\n code? I just saw it in #6874.

Ryozuki

the issue with cond vars is task cancelattion iirc

Spurious wakeups are allowed for condvars, that's a perfect way to work in cancellation

[ddnet/ddnet] Pull request opened: #6876 Support master server address format in client

Support master server address format in client This still works

    
$ ./DDNet "connect 127.0.0.1"

But now also this works ``` $ ./DDNet "connect tw-0.6+udp://ger10.ddnet.org:8303" 2023-07-16 14:07:50 I engine: running on unix-linux-amd64 2023-07-16 14:07:50 I client: starting... 2023-07-16 14:07:50 I client: version 17.1.1 on linux amd64 2023-07-16 14:07:50 I client: git revision hash: 7f100e262032b50d 2023-07-16 14:07:50 I client: connecting to 'tw-0.6+udp...

github-merge-queue[bot]

[ddnet:master] 2 new commits

1acb94f Use lambdas for editor index modify and sort functions - Robyt3 4a0c850 Merge pull request #6873 from Robyt3/Editor-Lambda-Refactoring - def-

12:46

[ddnet/ddnet] Pull request closed: #6873 Use lambdas for editor index modify and sort functions

Jupstar ✪

what did u do before it happened xd

Idk I just played with dummy on multieasy

12:54

Is there any fun based ddnet ported to 07?

ddnet is ported to 0.7

Smetanolub

Idk I just played with dummy on multieasy

the crashlogs might still be useful

github-merge-queue[bot]

[ddnet:master] 8 new commits

19d78b9 Made client/server check for correct map version on load - oy 9c4b0c0 Finished skip tile based map loading - oy a92ab45 Also support loading maps with tilemap skip in front layer - Robyt3 32f2fe9 Extract CMap::ExtractTiles with more validation - Robyt3 785f03e Report actual data sizes in CDataFileReader::GetDataSize - Robyt3 4ae0928 Support bezier envelope curves in maps and editor - Robyt3 54df98b Support loading CMapItemImage version 2 - Robyt3 8d12bce Merge pull request #5737 from Robyt3/Map-Format-Port - def-

12:55

[ddnet/ddnet] Pull request closed: #5737 Support bezier envelope curves in maps and editor, forward compatiblity for upstream map item changes

[ddnet/ddnet] Pull request opened: #6877 Don't support loading RGB images ingame

They're essentially unused and not supporting them ingame probably allows to keep that part of the map format unused. See https://github.com/ddnet/ddnet/pull/5737#issuecomment-1538284956 or https://github.com/teeworlds/teeworlds/issues/2812.

Checklist

[ ] Tested the change ingame
[ ] Provided screenshots if it is a visual change
[ ] Tested in combination with possibly related configuration options
[ ] Written a unit test (especially base/) or added coverage to integrati...

jopsti can i still get deals on netcup if i show them the competition is cheaper?

i guess so

13:26

they got quite expensive since the war :/

where can i do that?

13:26

do i send them a mail or what

call support

https://www.1blu.de/server/vserver/

Linux vServer mit SSD-Power: 1blu

1blu-vServer mit KVM-Virtualisierung mit umfangreicher RAM- und SSD-Ausstattung für Spitzen-Performancewerte

13:27

ya think itll work?

13:27

1 euro 4 core 8 ram 120 ssd

oh its a time based discount

but limited 12 month offer

13:27

ye rigged

13:27

worth the call?

13:28

maybe they make my vps cheaper for 12 months xd

well depends

13:28

what do u want to have for that in exchange?

anything

like a vPS or root server

i take either

root server is a german word

well try it out

jopsti ban

13:29

for using german

I think server hosters have abused it so much that it doesn't mean anything anymore

are there even 1euro root servers? :D

you might mean dedicated server, but hosters also use that word for VPS

i though root means dedicated and starts from 40 euros

well its not truely dedicated

13:30

they only guarantee the core is fully yours while u use it

ah you mean the guranteed power

13:30

a ye

https://de.wikipedia.org/wiki/Root-Server

Root-Server nennt man: Root-Nameserver, welche die Wurzel des Domain Name Systems bereitstellen Dedizierte oder virtuelle Server mit einem Root-Konto, siehe Server #Formen

13:30

Dedizierte oder virtuelle Server mit einem Root-Konto, siehe Server #Formen

deutschland

"dedicated or virtual servers with a root account"

so they didnt lie to me

13:31

i have root account

xd

13:32

so regular vps is without root rights

13:32

epic

no. it's just a made-up word

13:33

VPS always came with root access

13:33

I think it was used as the translation of "dedicated server" at some point

13:34

but not anymore

i dont care about the names. VPS also means nothing to me

13:34

minecraft VPS server

13:34

where u can only install minecraft mods

13:34

lmao

13:34

vps server

13:34

chillerdragon: if u unlucky they say smth like "but we give higher bandwidth" and call it a day

13:35

but defs tell me if it worked xD

fak

13:35

virtual private server server

"private"

xd

laughs in intel security problems

its similarly trol than saying "sending an sms"

13:36

which means "sending and short message service"

13:36

an*

xD

a* i guess

13:37

omg fuck engl

chillerdragon: what CPUs do they usually use in your hoster?

13:38

have u ever bought one there

what is my hoster

13:39

and how does one buy a cpu?

the one u posted

a

13:40

ye also curious

13:40

idk what cpu they have

i can also ask different: do they have okayish benchmarks?

but i dont want a 12mo contract anyways

my 1,99€ VPS sucks hard xDD

xd

13:40

i can give you a teeworlds vps

13:40

you can install teeworlds mods

xd

13:41

for teeworlds any vps is good enough

looking at fokkonauts 100% cpu usage moments

but i managed to kill the VPS with a teeworlds ban list i ofc blame teeworlds for using the worst possible implementation possible but still sad day xd

bans bad

13:43

we should remove bans

we should remove evilness

13:45

chillerdragon; when open a server hoster i'd like to have a VPS with very high single core power xd bcs java vm demands it

we should remove jumping. its unintuitive and you can acheive the same thing by hooking up

we should remove rendering, not all people can see and it gives an unfair advantage over them

we should remove ddnet and replace it with uhhhhhhhh

we should replace the logo with yours

13:46

when finished

idk what to do for logo

jopsti this enough for java? 13th Gen Intel i9-13900K (32) @ 5.500GHz

barely, but yeah

13:51

if i have exclusive boost power to the core xD

its my main pc .-.

13:51

also gotta turn it off during nite :D

13:51

quality vps

atleast it can run minecraft

bug? why, after saving, the tiles from the Switch layer are not drawn correctly (painted with a gradient so that it can be seen clearly)

invalid tiles for ddrace game mode

14:09

isnt there an options somewhere

didn't understand

Matodor

didn't understand

Image attachment

renders only specified tile numbers?

Jupstar ✪

Click to see attachment 🖼️

yes its enabled

Image attachment

entities mlife?

14:12

i dunno if these also change some internal mode, but i guess the editor drops the tiles as invalid?

Image attachment

@Robyt3

Jupstar ✪

i dunno if these also change some internal mode, but i guess the editor drops the tiles as invalid?

renders only specified tile numbers?

well apparently yes

14:13

but dunno if thats indented

14:13

intended

but its render switch number and delay index

Image attachment

14:13

without image from tileset

14:14

strange behavior)

ah that is what u mean i understand

14:14

the tiles are there but not rendered is your complain

1

and stops rendering after saving the map and loading

yeah weird

front layer work correctly

github-merge-queue[bot]

[ddnet:master] 2 new commits

9c0c755 Don't support loading RGB images ingame - heinrich5991 ce96dfc Merge pull request #6877 from heinrich5991/pr_ddnet_no_rgb - def-

14:16

[ddnet/ddnet] Pull request closed: #6877 Don't support loading RGB images ingame

lmao clickbait yter on ddnet frontage xd

14:18

didnt expect to ever see such thumbnails on a tw video xxxxxxxxxxD

it say its semi clickbait

14:21

https://youtu.be/typCY28F214?t=129 chillerdragon: here u can see the tiles are actually gone

Stronghold BUT Only the Blocks the World Record Touches...

14:21

physics wise

maybe clickbait is the wrong wording

14:23

the concept of removing tiles to mimic world record it self is mainstream youtube strat

14:23

i mean the font etc

14:23

the red arrow

14:24

the white outline on the tees

lmao

14:24

that is clickbait for u? XD

as i said clickbait might be the wrong word

14:24

more like mainstream youtube tryhard shit

14:24

only thing missing is him saying pls turn on the bell

yeah but i think i seen that before

in tw?

14:24

i didnt

14:25

i dont rember seeing it

well i couldnt tell for sure, but i guess there was

https://www.youtube.com/watch?v=gp4wXgS3uO4

Every Technique You Need To Know (Novice) - DDRaceNetwork

14:25

there is this

but that guy does it in every video xD

14:25

xDDDDDDd

14:25

i thought its 3d game

yea idk something about this style really triggers me

14:27

its used by all the shallow yters

https://www.youtube.com/watch?v=vQdhY0fFz70

Teeworlds Shyzo - DDraceNetwork

FIRST FINISH! | Corrupted | 5 Star Insane Hardest Map??? | Team Fin...

14:27

xDDDDDDD

that have zero content and passion they only produce to produce and to generate views not because they care about content

from shyzo, whom else

yea xd touche

14:28

but i gotta say Gazebr perfected it more

tru

shyzos looks way more like a tw yter

but its funny how thumnails differ

being inspired by watch yt

beautemps puts bloom on max xD

shyzo makes runs cuz he likes gores

Image attachment

then he decides to share those runs

14:29

and with Gazebr it feels more like he couldnt care less and just thought of a concept that will be clicked

crack does art contests in his thumbnails xD

Image attachment

14:30

rqza is lazy af

14:30

he doesnt even creates thumbnails

@Jupstar ✪ "allow unused" not working after map reloaded

14:30

i mean tiles not rendering)

Matodor

@Jupstar ✪ "allow unused" not working after map reloaded

yeah report it as bug

but ye its a nice video and Gazebr has good quality i should stop hating

14:31

the style is just posioned by some money farmers with ai content

chillerdragon lmao no need to have midlife crisis now

Jupstar ✪

crack does art contests in his thumbnails xD

why kog

14:31

‍

Ryozuki

why kog

is it my fault that only kog hosts gores xd

14:32

ddnet clearly isnt interested

@Jupstar ✪ lets make a network

14:32

RustNet

GGnetwork

14:32

ggoresnetwork

it's nice that we have a couple of independent networks. you making your own would be appreciated

i dont want to spent so muhc money

we have ddnet and unique which are open source, then kog which is closed

ddnet has good ecosystem

14:33

i'd prefer to use its

Jupstar ✪

i dont want to spent so muhc money

we can just ask for donations, in a clear transparent way

14:33

im sure many ppl would be interested in a kog alt

a few for sure

14:33

but most probably simply dont care

true

and creating a new community isnt ez

would be cool if something new appears

14:34

not gores i mean

14:34

tee smash network

xD

heinrich5991

it's nice that we have a couple of independent networks. you making your own would be appreciated

btw whats nice about it xd do you prefer to have closed networks over ours? or simply bcs of maintaince?

14:36

if ddnet is open to host any open source mod its attractive too imho like the wario mod thing

14:36

was popular for like 2 days xd

14:36

bcs deen announced it etc

Jupstar ✪

btw whats nice about it xd do you prefer to have closed networks over ours? or simply bcs of maintaince?

it's nice that we don't have a single point of failure, that people can e.g. search for the community they want. that not everything goes through the same people, e.g. me being a discord moderator here, but people can also chat elsewhere. that new features can be tested independently of ddnet, e.g. the accounts thing. better bot protection

mh very awesome, if u dont play gores u like it

I play gores

"Rank 4981. with 262 points"

14:41

well u play it indeed

14:41

but 200 points is also not a lot

14:41

then let me rephrase: you dont take gores very serious

14:42

most points were made 2 years ago

https://ddnet.org/players/heinrich5991/

14:42

most points were probably made 2 years ago

14:42

or even earlier

14:43

I find it kinda bad that you go to people's profiles to tell them "no, you don't actually like xy"

well i'd not say you are part of the kog community just bcs u played a bit of gores 2 years ago lmao

point shaming axaxax

i basically know all ppl there and played with them, were VC with them

14:44

xDDDDDDDDDD

never saw u in gores jopsti

14:45

random

i even played with u

14:45

e

back when u had ur spin bot xDD

under wat name

14:45

watf

too many to remember

i never

chillerdragon: i remember there was a time where chiller was a famous prefix among many ppl xD what happened to your fanbase

it didnt notice any reduce

oh true chiller has fans

14:54

honestly if ddos wasnt such a deterrent i might consider making a network

14:54

maybe with the servers the davide guy shared its doable

14:55

rust based infrastructure, without docker or kubernetes

14:55

psql database instead of mysql

14:55

no python 2 scripts

14:55

no hardcoded paths

14:55

AGPL license

14:55

the dream

xD

14:55

i agree with most stuff

14:55

xD

xd

e.g. no python 2 scripts => no python at all

14:56

agpl would have prevented kog closed source

Jupstar ✪

e.g. no python 2 scripts => no python at all

true

Ryozuki

agpl would have prevented kog closed source

mh i doubt u could sue them

14:56

bcs nobody cares

u can talk to the FSF

14:56

and they can bring lawyers

14:57

i think the FSF is comitted to defend free software in court

probably not necessary

14:57

if you license under agpl, people will probably not make closed-source forks in the first place

mhh sure xD

14:59

i bet u could go to any random company and find legal issue within them bcs they use software in a way that is not legal in the way the software defines its legal usage

oh, I think it prevents a closed-source serverside antibot btw ^^

i have no problem with that

15:00

but i cant relicense my fork of tw under agpl i think

@Ryozuki give up on that, be a future human

Ryozuki

but i cant relicense my fork of tw under agpl i think

why not?

u can?

I don't know why you wouldn't be able to do so

nobody would use ur fork tho ryo xd

Ryozuki

but i cant relicense my fork of tw under agpl i think

the other way around is more interesting

go mc ryo

15:04

mc is full of gpl

how much do i have to obfuscate the code to relicense under cc0

xd

15:05

1 chat gpt prompt is enough

15:05

pls reorder code

15:05

thank

ChillerDragon

nobody would use ur fork tho ryo xd

idk chiller

15:06

im providing software to peopel who value their rights

15:06

if they dont want to use it cuz x license up to them

15:06

let me bring a quote

15:07

I make my code available for use in free software, and not for use in proprietary software, in order to encourage other people who write software to make it free as well. I figure that since proprietary software developers use copyright to stop us from sharing, we cooperators can use copyright to give other cooperators an advantage of their own: they can use our code.

gpl worst

best actually

15:13

https://github.com/yuankunzhang/charming

GitHub - yuankunzhang/charming: A visualization library for Rust

A visualization library for Rust. Contribute to yuankunzhang/charming development by creating an account on GitHub.

15:14

uh no license

15:14

"MIT/Apache-2.0"

15:14

i guess they forgot in repo xd

open an issue

done

15:19

https://www.doc.ic.ac.uk/~afd/homepages/papers/pdfs/2023/ISSTA-tool.pdf

15:19

RustSmith: Random Differential Compiler Testing for Rust

@heinrich5991 how much % r u done with quic if i may ask

client and server can connect, both via 0.6 and via quic protocol and to legacy clients/servers. needs some tweaking around disconnects. lots of smaller features missing, too, e.g. timeout protection. I have only tested it locally, so windows/mac support and cross-compilation support also still missing

i see, thanks for the status

https://no-ht.ml/

15:30

https://nocss.club/

Jupstar ✪

@heinrich5991 how much % r u done with quic if i may ask

what's the reason for your interest?

heinrich5991

what's the reason for your interest?

i want to see how well it works xd

how fast?

15:49

haven't measured that tbh

heinrich5991

how fast?

i dunno if u implemented 0.5 rrt but stuff like that how fast it responds, how fast connections are etc.

15:51

benchmarks are semi interesting.. if u have 1million packets throughput or 700k is not interesting to me xD

I implemented even worse rtt handshakes

15:52

mandatory RETRY frame for new connections, to fight IP spoofing

not optional?

15:52

i mean i can understand why

I haven't really built any options there

but if there is no direct attack

yes, that'd be a perf optimization, I'm postponing/not doing that right now

i c

although I think 0.5rtt handshakes aren't really suited for TW

15:53

they're important more important for protocols with short-lived connections IMO

well the gain on huge pings is ofc insane

15:54

tested with increasing the loopback ping

one less RTT, you mean? yea

15:54

I guess we could try cleaning up the rest of our connecetion though

Ryozuki

https://www.doc.ic.ac.uk/~afd/homepages/papers/pdfs/2023/ISSTA-tool.pdf

lol, rust compiler in c++ https://github.com/thepowersgang/mrustc

GitHub - thepowersgang/mrustc: Alternative rust compiler (re-implem...

Alternative rust compiler (re-implementation). Contribute to thepowersgang/mrustc development by creating an account on GitHub.

Chairn

lol, rust compiler in c++ https://github.com/thepowersgang/mrustc

interesting that there are ppl have this kind of motivation xD

yes, it's pretty amazing

[ddnet/ddnet] Issue opened: #6878 Editor unused tiles warning popup is only shown for game and front layers

Should also be shown for the other entities layers.

would you say if u used a proper game engine and are tryhard, u could implement most stuff for teeworlds in one week? xdd

16:29

i am impressed how far away i still am from vanilla support, i feel like i'm at like 30% only yet xD

Jupstar ✪

would you say if u used a proper game engine and are tryhard, u could implement most stuff for teeworlds in one week? xdd

I don't think so. I don't even know any game engines

ok but assuming u'd know one already

also, projects like these are always harder than they appear ^^

yeah

16:31

if u play the game u think ez simple game xD

Ryozuki

no python 2 scripts

I don't think we have any left anymore coolcry

coolcry

Learath2

I don't think we have any left anymore coolcry

coolcry

$ rg python2 servers/scripts | wc -l
49

NOOOO

17:29

Maybe none that we use actively? Some are old toys for broadcasting cubes and stuff

hmm, info.py seems to be python3 at the very least

[ddnet/ddnet] Pull request opened: #6879 Use temporary file when saving editor map

Write the map to a temporary file first. When the map was saved to the temporary file successfully, first delete the existing map file having the real filename, then rename the temporary file to the real filename. If deleting or renaming fails, show an error message popup and log an error message to the console. The implementation is consistent with the way temporary files are utilized by Microsoft Word, so this should work on Windows. See: https://support.microsoft.com/en-us/topic/d...

python2 moment

there are no python2 moments

18:49

its dead

i dont even know what you talking about

18:50

today chairn said smth about old windows versions

18:50

did he mean win10?

18:50

i only know win10 and win11

18:50

OpenG what?

18:51

rust stable?

18:51

@Ryozuki in ryoOS do forced nightly updates xDD

xd

18:52

@Chairn is it true some modern cpus u can write code that modifies the cpu itself?

18:52

physically

18:52

like a fpga

18:52

https://www.tomshardware.com/news/amd-to-fuse-fpga-ai-engines-onto-epyc-processors-arrives-in-2023

AMD to Fuse FPGA AI Engines Onto EPYC Processors, Arrives in 2023

AMD to arm its chips with FPGAs

18:52

this

18:52

well idk if its this

18:53

https://www.amd.com/system/files/documents/xilinx-deep-learning-solution-on-amd-epyc-processors.pdf

i guess, Intel did the same with Knight CPU with FPGA integrated

18:53

but they stopped doing it because it wasn't worth it

Xilinx Reconfigurable Acceleration Platform

you can't accelerate programs when the bottleneck is not the computing but the rather slow memory

f

18:57

Image attachment

18:57

@Chairn

18:58

french know?

didnt know england has it this good

github-merge-queue[bot]

[ddnet:master] 3 new commits

a7f2956 Rename variable df to Writer - Robyt3 8abf9a7 Use temporary file when saving editor map - Robyt3 8a8ffc3 Merge pull request #6879 from Robyt3/Editor-Temporary-Save - heinrich5991

19:10

[ddnet/ddnet] Pull request closed: #6879 Use temporary file when saving editor map

19:10

[ddnet/ddnet] Issue closed: #4476 Editor map saves should go into a temporary file first

guys im a dev

19:11

uwu

owo

UwU

owo

it has been years since auto demo behvaior got changed i gotta say i still miss the old one

19:56

where you could stop the auto demo

19:57

or is that still working and the button shows it wrong?

19:58

nah ye you can just start recording a manual in addition to the auto not stop the auto

19:58

that used to be a thing and was what i used before there were demo markers to mark stuff

19:58

also its nice to view demos while being in game

19:58

playing back demos while they are being recorded is always a bit wonky

[ddnet/ddnet] Issue opened: #6880 Force the reserved slot into the spectator team

Problem: When moderators use the reserved slot on full servers they might forget it and start playing the map. They dont moderate then and active moderators cant join the server. Solution: Force the player of an reserved slot into the spectator team so he cant play the map but rather can only moderate in this state. Addition: Introduce a new variable to switch between this behaviour and the normal one.

@heinrich5991 be honest did you see the impact of your gameplay change coming?

20:42

https://tube.zillyhuhn.com/videos/users/chiller/heinrich_hammer.mp4

20:42

check this out that part is intended to be more complicated than a strong hammer throw its supposed to be a swing plus hammer boost

20:43

and the new mechanics allow to just hammer the floor and gain the same amount of speed :D

I don’t see the problem

20:46

It’s normal that you gain that amount of speed when you dj and then hammer 2 freezed people under you (edited)

ye sure :D

20:48

i wanna see you pull off that move without heinrichs patch

xd

20:48

he is no oldschool player

20:48

xd

oldschool?

20:48

that happend like last week

he doesnt know the non patched version

was even post steam release wasnt it?

20:48

post vanilla steam release

really? wasnt it like 4 years ago

thats what i said

u said post

20:49

u meant pre

20:49

i see

i said post

ah vanilla

20:49

lmao

I remember i gained that exact amount of speed doing that exact thing 1-2 years ago

who thinks in vanilla trollodragon

ddnet steam release so irrelvant

20:50

idk when that happend

20:50

the whole greenloight was such a mess

20:50

tw steam release i remember

@Iza the point is, there was a time when hammer didnt trigger out of freeze when u hold mouse1

it did

20:50

but was more lucky

well u had to press it

20:50

or trigger the bind with mousewheel

Well fortunately it isn’t like that anymore

20:51

I still sometimes try to initiate a freeze hammerfly in kog only to be disappointed (edited)

woah Iza you really dont know ddnet without the new heinrich patch?

did kog patch it out?

When was the patch?

#1922

https://github.com/ddnet/ddnet/pull/1922

Allow to hammer directly on unfreeze by heinrich5991 · Pull Request...

Previously, this was only doable with frame-perfect input if you were going to get frozen again the next tick. Now, you only have to hold your hammer to immediately hammer once you get unfrozen. Th...

2019

20:52

almost 4 years

Yeah i wasnt there

20:53

I started playing ddnet almost exactly on steam release

2019 is 0.7 times already

chillerdragon: when was the irrelevant vanilla steam release

newfag

20:53

idk

20:53

sooner

@Akira

"Storage: 10 MB available space" biggest lie ever

looks at 80gb demos folder

xd

20:54

"Release Date 24 August 2015 – 10:07:00 UTC (8 years ago)"

20:54

8 years lmao

20:55

ddnet: Release Date 17 August 2020 – 08:53:17 UTC (3 years ago)

20:55

24 august and 17 august

everything after 2015 is new for me

20:55

imo also noby is a new player

20:55

said the guy who didnt even play 0.5 xd

everything before 2018 does not exist for me

20:56

make ur brain only support the last 5 years

20:56

anything before never existed

wat brain

tru

ChillerDragon

https://tube.zillyhuhn.com/videos/users/chiller/heinrich_hammer.mp4

looks fun, but you probably hate fun too

omagawd

20:59

is that a reply to wat?

Your video

thouzght so

21:00

you gotta admit its gameplay affecting tho

21:00

bro wanted to fix hammerfly start and added easy super hammer throw replacement

21:01

and then i get called our for breaking the client by adding eval_if_cmd xd

Never said it's not gameplay affecting

it disrupted the industry

21:03

now you can not hammer boats if a spike is in 20 tiles distance

21:04

nah lerato it is fun and its not too bad actually given how crazy of a change it is

21:04

but im just asking heinrich if he saw that coming

Ryozuki

french know?

know what?

@deen two more days to go

21:20

Are you hyped already xd

Jupstar ✪

@deen two more days to go

How do you know?

Because you planned it

xd

Exactly on the birthday of ddnet

21:20

Nice move

imagine u just vibin and not even born yet

21:21

and some randome dude on the internet is predicting ur bday

Xd

deen you already got instragram, snapchat and tiktok acc for your kid? so you can post from minute 1?

Nope, just an email address

lmao

21:23

u got a mail?

21:23

haha

21:23

deen.jr@protonmail.com

21:23

leak

@outlook.com troll

troll

axax

21:24

probaly more like felix@felsin9.de

21:24

do we know gender yet?

21:25

we as in public

I actually signed up for https://purelymail.com and am switching all my felsin9.de to felsing.org

interesting front end

an html site and roundcube, exactly what I'd do myself

Damn, 10$ per year, that like rly rly cheap

21:26

greenthing

woah why is 3gb so expensive

21:28

i never paid for mail in my life xd

21:28

my gmail looks like it offers 15gb for free?

21:31

deen how did you find this cute service? they are clearly niche

deen

I actually signed up for https://purelymail.com and am switching all my felsin9.de to felsing.org

good to know, they do email hosting? i need email hosting

21:31

for my domains

"they"

21:31

xd

?

We're a small company. So small, the "we" is royal.

21:31

https://purelymail.com/about

That's like really small kat

kat

ChillerDragon

my gmail looks like it offers 15gb for free?

gmail is also expensive when you want your own domain

ChillerDragon

deen how did you find this cute service? they are clearly niche

The same way I find anything computer/internet related, by searching with site:news.ycombinator.com (and site:reddit.com for the rest)

ah domain i see

21:39

is hackernews even properly indexed on search engines?

21:39

do hackernews entries ever persist?

yes

Exported 658 message(s)