DDraceNetwork - developer

DDraceNetwork

Development / developer

Development discussion. Logged to https://ddnet.org/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 2023-11-24 00:00:00Z and 2023-11-25 00:00:00Z

github-merge-queue[bot]

[ddnet:master] 2 new commits

00f2c8a Show previous commands in the console's scrollback buffer - jtbx cb22fbf Merge pull request #7536 from jtbx/scrollback-history - heinrich5991

00:08

[ddnet/ddnet] Pull request closed: #7536 Show previous commands in the console's scrollback buffer

chillerdragon

@furo: fellow 0.7 porter! Keep up the epic prs

at this point its going to be harder to add new features without making spaghetti code

Voxel

at this point its going to be harder to add new features without making spaghetti code

not sure what you're talking about. we added features that are useful independent of 0.7

[ddnet/ddnet] Pull request closed: #7535 Quote names with spaces when auto completing names in commands.

00:22

github-merge-queue[bot]

[ddnet:master] 2 new commits

7ca10e2 Quote names with spaces when auto completing. - furo321 3fff2f9 Merge pull request #7535 from furo321/quote-names - heinrich5991

heinrich5991

not sure what you're talking about. we added features that are useful independent of 0.7

I mean with so many 0.7 reliant updates I feel like it's going to be a harder focus to balance that instead of cleaning up what we already have.

can you point me to such a "0.7 reliant update"?

00:30

I don't think we have any

[ddnet-maps:master] 1 new commit

5eb40d1 A Paralyzed - ddnet-maps

morning from gym

08:11

Fridays rust

gm

@Jupstar ✪ lets nullify latency! gogo implement this https://en.wikipedia.org/wiki/Quantum_teleportation

Quantum teleportation

Quantum teleportation is a technique for transferring quantum information from a sender at one location to a receiver some distance away. While teleportation is commonly portrayed in science fiction as a means to transfer physical objects from one location to the next, quantum teleportation only transfers quantum information. The sender does not...

Teero

@Jupstar ✪ lets nullify latency! gogo implement this https://en.wikipedia.org/wiki/Quantum_teleportation

it's not clear yet if we can use that for communication but yes, let's go

08:46

gimme 1 billion $ please, then i start

2

we got anice team

09:05

jupstar, heinrich, chairn, learath maybe robyt and me

sos @heinrich5991 my new vps gets master1 errors

go make a quantum computer

09:05

anti ddos solution

09:05

for ddnet

09:06

poggers2

09:06

Chiller u got banned

classic

NekoEvil

https://paste.zillyhuhn.com/uu

09:08

curl works just fine

09:09

curl https://master1.ddnet.org/ddnet/15/register

09:09

{"status":"error","message":"HTTP method not allowed"}

09:09

probably some bitsh as firewall i got

what do u got

09:13

maybe ur server is wrong coded

works on other vps

http method not allowed

09:14

either u not sending post or cors is bad

ah no

09:14

that was just a test to see if the firewall lets me talk to it

09:14

that test passed

a

heinrich i emailed u full traffic dump

09:15

Embedded image

2023-11-24 04:05:09 I http: https://master1.ddnet.org/ddnet/15/register failed. libcurl error (7): Failed to connect to master1.ddnet.org port 443 after 199 ms: Couldn't connect to server

09:15

this looks like a problem on ur vps

yes

09:15

but weirdly normal curl works

xd

09:16

does ddnet tell libcurl version

09:16

is this debian

ChillerDragon

heinrich i emailed u full traffic dump

*Meisterservierer (edited)

deb12

nice mail

09:17

ur so corporate

<ChillerDragon> :D

u needed to say goodbye

09:17

and xoxo

<ChillerDragon> lmao teero yes and also tru dat ryo

this goodbye: Cordially, Your resident bot maker.

<ChillerDragon> 2023-11-24 04:04:52 I http: libcurl version 7.88.1 (compiled = 7.88.1)

09:18

<ChillerDragon> ._.

xd

09:18

today black friday

09:19

thats racist

<ChillerDragon> time to buy shit we dont need

09:19

<ChillerDragon> like a new server that cant do master server connections axaxax

i bought a 2tb m2 pcie4 ssd

<ChillerDragon> pog

09:19

<ChillerDragon> how much

09:20

<ChillerDragon> oh and the trollest thing is that the server shows up in my clients browser just fine

09:20

<ChillerDragon> but it spams my logs

100

<ChillerDragon> 185.223.31.160 this is me

should this work?

09:33

Embedded image

<ChillerDragon> maybe its because master1 failed it fell back to this

09:36

<ChillerDragon> running a client on that vps shows an empty server browser ._.

09:41

<ChillerDragon> same here this works fine curl https://master1.ddnet.org/ddnet/15/servers.json

no

installed the new m2

10:03

replaced the one i had windows on

10:03

i decided to ditch that for now

nowadays i only use windows to test bugs of other windows users

i got 4.5tb of fast ssd space

gg

6600mb/s theorically

10:05

xd

for like 100ms xdd

well better than a non m2 ssd

10:05

and a hdd

yeah

its pcie4 too

u got 2 m2 slots?

ok 2tb of those is a normal ssd

10:05

i got 2 m2 slots

10:05

and 1 normal ssd

then copying between these two might be slow

i gotta reformat that microsoft data partition

Image attachment

bcs they share same pcie lanes

do they

it's very often

they both pcie4

dunno how expensive your mb is xd

i think it costed me 200

Ryozuki

they both pcie4

they both are connected to a pcie4 lane yes

10:06

but to the same xD

❯ sudo dmidecode -t2
# dmidecode 3.5
Getting SMBIOS data from sysfs.
SMBIOS 2.8 present.

Handle 0x0002, DMI type 2, 15 bytes
Base Board Information
    Manufacturer: Micro-Star International Co., Ltd.
    Product Name: B550-A PRO (MS-7C56)
    Version: 2.0
    Serial Number: 07C5622_L41E756275
    Asset Tag: To be filled by O.E.M.
    Features:
        Board is a hosting board
        Board is replaceable
    Location In Chassis: To be filled by O.E.M.
    Chassis Handle: 0x0003
    Type: Motherboard
    Contained Object Handles: 0

10:06

https://www.msi.com/Motherboard/B550-A-PRO

Best AMD AM4 B550 ATX motherboard, lightning PCIe 4.0, lightning M.2, Extended heatsink, USB 3.2 Gen 2, M.2 heatsink, powerful overclocking, MSI PRO

Image attachment

noooo

10:08

xd

10:08

i should switch places of the m2 then

10:08

i have the 500gb on the pcie4

10:08

and the 2tb on the pcie3 then

10:08

i think

10:08

how can i know? xd

10:09

the bother is that the pcie4 m2 slot is under the gpu

10:09

zzz

Ryozuki

how can i know? xd

i guess u could benchmark xdd but i'd suggest if you do that only benchmark reading

meh its ok

10:10

they are fast enough

i mean, normally u wont notice anyway xDD

yeah xd

"∙ When installing PCIe SSD in M.2_2, PCI_E3 slot will be unavailable"

10:12

this is important for you tho xD

i see xd

10:13

i also had a pcie adapter

10:13

for another m2

10:13

i guess that wont happen

just so u know which is pciee3

i know thats sata

10:14

xd

oh yeah wrong pic

10:14

Image attachment

pcie x4 is like a small slot like gpu

no dual GPU then

10:14

its not worth

10:14

dual gpu

yeah

even less on a x4 lane

10:14

xd

10:15

@Jupstar ✪ is there pcie5 m2 ssds?

i bet so, but probs cost way too much

10:15

tbh pcie5 came so early

10:15

it was lke pcie4 was skipped xD

10:15

pcie3 was here for like 10 years

10:16

ddr5

yeah

it's funny that *4 is always skipped xD

also pcie6 is out there

wtf

https://pcisig.com/pci-express-6.0-specification

is that now their new strategy

PCIe 6.0 Specification Features 64 GT/s raw data rate and up to 256 GB/s via x16 configuration Pulse Amplitude Modulation with 4 levels (PAM4) signaling, levraging existing PAM4 already available in the industry Lightweight Forward Error Correct (FEC) and Cyclic Redundancy Check (CRC) mitigate the bit error rate increase associated with PAM4 signaling Flit (flow control unit) based encoding supports PAM4 modulation and works in conjunction with the FEC and CRC to enable double the bandwidth gain Updated Packet layout used in Flit Mode to provide additional functionality and simplify processing Maintains backwards compatibility with all previous generations of PCIe technology

10:17

Image attachment

https://www.gigabyte.com/de/SSD/AORUS-Gen5-10000-SSD-1TB#kf

AORUS Gen5 10000 SSD 1TB Besonderheiten | SSD - GIGABYTE Germany

GIGABYTE, Chassis, Thermal Solution, Power Supply, Optical Device, TV Tuner Card, gamer, PSU

10:18

this is a pcie5 ssd

10:18

xDD

10:18

active cooler

10:18

or fat passive cooler

10:18

dunno

damn

10:18

the cooler

10:19

not much more write speed

10:19

or read tbh

10:19

i would expect x2

10:19

but its not even x2

i could imagine with such a cooler, it can at least hold it a bit longer

10:20

i mean it's not like i would not test it out it's simply, that it is wasted money xD

10:20

their awesome, unlabeled graph (edited)

Image attachment

xd

10:25

@Jupstar ✪ what filesystem should i put

10:25

ext4?

10:26

While EXT4 generally has slightly faster read and write speeds and lower latency than BTRFS, BTRFS is a better choice for systems with large amounts of data and high availability requirements

10:26

ok ext4

yeah no hipster fs for me xd

10:32

filesystems are too annoying to deal with xD

10:32

ext4 is the mainstream for linux

for_window [class="^steam$"] floating enable
for_window [class="^steam$" title="^Steam$"] floating disable

11:10

makes steam somewhat usable on a wm xd

Wot language ryo

i3config

11:34

Image attachment

11:34

justatest

11:35

@Jupstar ✪ when process vulkan shaders

Ryozuki

@Jupstar ✪ when process vulkan shaders

steam does that for ddnet too

nice

kek

Image attachment

ld: unrecognised emulation mode: elf_x86_64 Supported emulations: elf_x86_64 elf32_x86_64 elf_i386 elf_iamcu

12:51

??????????????????????????

<ChillerDragon> @Mr.Gh0s7 works on my AI

13:08

<ChillerDragon> @Ryozuki xd

Mr.Gh0s7

kek

not reproducable on my AI

im surprised no one tried to send malware to gpt to ruin its database

Voxel

im surprised no one tried to send malware to gpt to ruin its database

that's because you haven't used it yet

and why should i

Voxel

and why should i

to send it viruses to destroy it ofc

but thats still touching it

yeah but see the real haters dont use it

14:38

that's why nobody did it yet

14:38

also you would think the ai has like, good security (edited)

14:40

idk ill google later

if there is a will, there is a way

14:40

14:41

Voxel, leader of the human revolution

something is off

Image attachment

the render order is wrong

Patiga

something is off

i assume you do this on purpose right? is this a planned feature of your demo renderer?

nah I just forgot to clear a buffer

15:02

yeah I'm not sure what happened in the drop part

15:02

ah you mean render order between tees an map?

yeah

15:02

xd

was too lazy for that so far :d

but between tees is weird too

but Pog is

15:04

that apart from tees initially being at 0, 0 all the time, the entire render pipeline worked first try

(after the shader syntax fixes)

ez

honestly I would've bet against that xd

Patiga

honestly I would've bet against that xd

when u have skins working, u gotta release a version

sure :)

i'm interested into your solution now with skin texture arrays etc

eeeh

15:08

I'm going with the ddnet approach now

ah lol

15:08

ez

https://gitlab.com/Patiga/twgpu/-/tree/tee-rendering?ref_type=heads

15:09

this is the last commit with the weird gpu-based approach

15:09

you can try it out in the demo-viewer thingy

15:10

but yea it isn't all that far, I think I got until hook and then stopped at the ninja hadouken effect

15:10

https://gitlab.com/Patiga/twgpu/-/tree/new-approach?ref_type=heads

15:11

this is the current wip

ah nice

15:11

for effects i'd really like to introduce multi texturing support

twgpu/src/sprites is the new component

there are often only few particles, then others (with other textures) and since render order matters, this causes lot of draw calls

15:12

same for quad layers actually

15:12

players are not really a problem, since u usually dont have more than like 64p or let's say 256p in a worst case

15:13

but particles sum up very fast

whats wrong with the quad layer?

yeah quad layers are maybe not that bad. i guess i just want to test out, if batching makes it a bit better

15:14

as said in our call, streaming quads is mostly as fast as ddnet's approach

15:14

so this kinda sucks

hm :/

for particles i'd say it's worse tho. ninja particles, gun trails etc. they are often mixed, since they are on a per player basis, batching all together would be nice

hm, batching most of those doesn't sound dramatic in terms of visual changes

Patiga

hm, batching most of those doesn't sound dramatic in terms of visual changes

but it is a visual change

yea ^^

i guess most of the time u'd not notice indeed

@Jupstar ✪ friday rust question when

Ryozuki

@Jupstar ✪ friday rust question when

i dunno. when will i not waste my time thinking about design choices in rust xdd

15:20

i still often struggle when i want idiomatic vs perf etc.

xd

15:21

well thats on u

15:21

both are safe

15:21

if u dont use unsafe

"Now i am become Rust, Destroyer of Unsafe code"

15:22

brb making rusty coffee

15:22

lately i learn a lot more about llvm apis

15:23

specially on how to make a shared library programatically

15:23

from mlir

15:23

without clang

15:23

the only thing i dont know how to do programatically is to call a linker api

15:23

there is no docs

15:23

for now i call the linker as a process

yeah yeah, thanks for reminder that you have a cool job

like a command

15:23

xd

15:23

ok i wont tell more

15:23

justatest

15:24

my point is, iwanted to make a blog post

xD

Ryozuki

ok i wont tell more

i said this but tbh i cant shutup xd

15:24

oh

yeah

15:24

it's ok xD

im in the llvm discord

15:24

ill ask there

if you use blockworlds as entities the unused tiles button becomes black, but if you click no and yes again it becomes red

Image attachment

Image attachment

15:35

Bug??? maybe?????? (edited)

15:36

also kind of surprised fddrace isnt an option in the entities

cyberFighter

Bug??? maybe?????? (edited)

Yeah, it's a bug. "Allow Unused" is only possible for DDNet entities. I'll fix it.

[ddnet/ddnet] Pull request opened: #7537 Hide "Allow unused" button when not using "DDNet" entities.

Previously it would just grey out the "Yes" button, but if you were to click "No". It was possible to enable it again. Which shouldn't be possible as "Allow unused" is only supported for "DDNet" entities Before !image After !image

Checklist

[x] Tested the change ingame
[x] Provided screenshots ...

Jupstar ✪

not reproducable on my AI

you know what first came to mind my when I saw yours and Chillers msg xD

<ChillerDragon> wor

16:11

<ChillerDragon> wot

16:11

<ChillerDragon> and who is "yours" xd

it's a reply to this

[3:30 PM]Jupstar ✪: not reproducable on my AI

16:13

I was taking a nap so my mind was clouded but after seeing those msgs I slept a lil better :)

https://github.com/emilk/egui/blob/master/CHANGELOG.md#0240---2023-11-23---multi-viewport

egui/CHANGELOG.md at master · emilk/egui

egui: an easy-to-use immediate mode GUI in Rust that runs on both web and native - emilk/egui

16:24

https://seanmonstar.com/blog/hyper-v1/

I’m excited to announce v1.0 of hyper, a protective and efficient HTTP library written in the Rust programming language. hyper provides asynchronous HTTP/1 a...

16:24

hyper v1

16:24

one of the most important libs

16:26

https://github.com/gluesql/gluesql

GitHub - gluesql/gluesql: GlueSQL is quite sticky. It attaches to a...

GlueSQL is quite sticky. It attaches to anywhere. Contribute to gluesql/gluesql development by creating an account on GitHub.

16:26

@Jupstar ✪ rust friday done

16:26

xd

github-merge-queue[bot]

[ddnet:master] 2 new commits

daa0609 Hide "Allow unused" button when not using "DDNet" entities. - furo321 edee3d5 Merge pull request #7537 from furo321/allow-unused - def-

16:31

[ddnet/ddnet] Pull request closed: #7537 Hide "Allow unused" button when not using "DDNet" entities.

ChillerDragon: try the masterping tool

Jupstar ✪

but between tees is weird too

ah I just interpolated from newer to older snaps accidentally, wrong direction

i decided

17:09

expect a blog post about llvm and mlir soon

17:09

poggers2

17:09

title = "Intro to LLVM and MLIR with Rust and Melior" description = "Learning MLIR with too many dialects."

17:09

rust pun intended

@heinrich5991 you forgot the readme again

17:27

cargo run ipv4 https://master1.ddnet.org/ddnet/15/servers.json this?

17:28

Embedded image

17:28

yea i got no ipv6

17:29

wait i do

17:29

hm

17:29

maybe thats the issue?

17:29

ill try turning off ipv6

17:30

hm didnt help

I planned my entire rework of the curl multi pr around curl_multi_get_handles, turns out it was added in 8.4.0 AAAAAAAAAAAAAAA

18:29

I will become unreasonable if I have to rewrite this one more time

why can’t you use 8.4.0

Way too new

18:35

I'll just go back to the ugly unordered_map whatever

uh, 8.4.0 is actually only 13 days old (released 2023-11-11)

18:37

how did you even know about this API?

heinrich5991

how did you even know about this API?

I wanted to get rid of the map of running requests, took a look at the api docs, there it was

18:37

Should have trusted that past me wasn't stupid

the more i use all these awesome well tested rust crates, that are way too ez to use, the more i hate how much ddnet reinvents the wheel xd

ok

yeah

18:45

@Ryozuki was right

18:46

all the time

18:46

since eternity

18:46

for eternity

owo

patrick, does god love us

It's a shame that std doesn't have a way to go from a pointer to an item in a std::list to an iterator at that item. It should be possible in O(1) in all implementations of it I can imagine

Ryozuki

anti ddos solution

quantum doesn't help in this case afaik

Patiga

something is off

solitary online on teeworlds

19:40

has any of you ever written a soft float library?

19:41

my attempt is working but for some case, im always off by 1 in the mantissa...

no, I have not

19:48

sounds fun though

19:49

can you give us the relevant snippets?

heinrich any idea wat else i could try? :c

can you check whether you have IPv6 connectivity anywhere?

i turned it off

19:51

should work if its off right?

19:53

ah okay nice sv_register ipv6 fixed the server log spam

19:53

ipv4*

19:54

seems like my ipv6 is bugged but i dont need players to connect via ipv6 right now anyways i think

19:54

okay only question left is how do i fix the client browser

19:54

that explains why the server was logging errors and still showing up. ipv4 worked and ipv6 didnt.

what's the issue with the client browser?

i dont see any servers :(

can you access https://master1.ddnet.org/ddnet/15/servers.json in the browser?

i can access it via curl from the vps just fine

19:56

thats the weird thing

19:56

it also says task done

19:56

i feel like i have a filter on or something

can you send a screenshot?

<ChillerDragon> https://zillyhuhn.com/cs/.1700855896.png

Embedded image

19:58

<ChillerDragon> its a fresh vps install so ddnet should have any filters active other than searching for "Noivce" right?

that seems like it's not an issue with the official ddnet client

19:58

please specify that

<ChillerDragon> nah its more a issue with my vps

19:59

<ChillerDragon> i specified that

19:59

<ChillerDragon> somehow the master connection is bugged client and server side on this vps

20:00

<ChillerDragon> does it make any sense to you that even if ipv6 is majorly bugged the browser wont work?

if something bad happens on the network layer, it should say "error", not "done", I think

<ChillerDragon> yea right?

20:03

<ChillerDragon> it also hits my api just fine

20:03

<ChillerDragon> and my api server could also speak ipv6

20:03

<ChillerDragon> does the master server promote ipv6 somehow?

no

<ChillerDragon> can i force the client to use ipv4?

you could disable IPv6 on your machin

20:04

e

<ChillerDragon> thats what i tried earlier

20:04

<ChillerDragon> let me try again

20:05

<ChillerDragon> echo 1 > /proc/sys/net/ipv6/conf/all/disable_ipv6

20:05

<ChillerDragon> this usually does the trick for me when i have ipv6 issues on some machine

20:05

<ChillerDragon> not this time :(

20:05

<ChillerDragon> the ip is gone in ip a

20:06

<ChillerDragon> but the log still shows stun/6 i assume thats ipv6

I don't think you're facing an issue that is related to IPv6 at all

<ChillerDragon> i mean on the server side i got rid of error messages with sv_register ipv4

20:07

<ChillerDragon> https://zillyhuhn.com/cs/.1700856475.png

Embedded image

20:07

<ChillerDragon> stun4 OK stun6 err

20:07

<ChillerDragon> ok fair i guess its off right now xd

20:08

<ChillerDragon> ok turned it back on same result

20:12

<ChillerDragon> ok nvm all those logs are useless

20:12

<ChillerDragon> i get the exact same logs on my machine if i turn off ipv6 and the browser still works

20:12

<ChillerDragon> yea dont know im out of ideas

20:12

<ChillerDragon> time to add random db_msgs

@heinrich5991 Is there anything better the register thread could be doing instead of running the http requests blocking?

[ddnet/ddnet] Issue opened: #7538 Image Render Tool

Make a tab (or smth like that) for rendering tee skin with custom colors, hammer, gun, sg, and else like in brawlhalla https://pbs.twimg.com/media/ES7DBadWAAINvg8.jpg Vid where you can see how it looks like - https://www.youtube.com/watch?v=0Amr188Uy7M

thonk

1

heinrich5991

can you give us the relevant snippets?

erf, it's on my work computer

<ChillerDragon> oke @heinrich5991 i managed to print a bunch of server names in the log so the data seems to arrive. Seems to be an issue in my client or config. Weirdly can not reproduce it on any other server so i still would like to blame some bugged ipv6 or firewall xd

20:51

<ChillerDragon> but seems like this one is on me now

Learath2

@heinrich5991 Is there anything better the register thread could be doing instead of running the http requests blocking?

not really. I guess it could not run 4 different threads

I need to somehow add a way to block on requests just for that usage

ah

The previous iterations used an extra mutex and a condition variable but idk how I liked that

21:18

I'd rather have a semaphore but we don't have anonymous semaphores on macOS, nor do we have C++20 for the standard ones

21:19

Or ideally I'd have a futex there but that's also either lots of work for something that'll be used once (or behind C++20 where you can efficiently wait on atomics)

21:20

I don't think spinning on it is a good idea as http requests are rather slow, I could sleep on it, idk

I mean the "blocking run" is basically only used as a way to run some code on completion @Learath2

Hm, if what is run on completion isn't slow maybe I should just run it in the curl thread and call it a day

21:25

eh nvm, it needs to acquire a lock, that's a bad idea

<ChillerDragon> aaaaaaaaaaaa

21:26

<ChillerDragon> i can reproduce

21:26

<ChillerDragon> with ddnet

21:26

<ChillerDragon> on my local machine

21:26

<ChillerDragon> fuck yeah!

21:27

<ChillerDragon> its not my client nor the vps omg xd

give us the details!

<ChillerDragon> im opening the issue dw

21:28

<ChillerDragon> what a detour omg

21:28

<ChillerDragon> ipv6 xd

Idk how this keeps happening, everytime I finish this up I hate what I've created

[ddnet/ddnet] Issue opened: #7539 Fresh ddnet install has an empty server browser

To reproduce:

mv ~/.teeworlds ~/.teeworlds_backup
mv ~/.ddnet ~/.ddnet_backup

then run latest ddnet and spam enter until you arrive in the browser !image

which commit are you on?

<ChillerDragon> wait you cant reproduce?

21:31

<ChillerDragon> omg ur right xd

21:31

<ChillerDragon> wait xd

21:31

<ChillerDragon> dude im tured

21:31

<ChillerDragon> tired

I haven't tried to reproduce

<ChillerDragon> dont bother

21:32

<ChillerDragon> gimme a sec

this is just standard "a user has opened an issue" questions ^^

21:32

especially regarding stuff that had fixes recently

<ChillerDragon> i see

[ddnet/ddnet] Issue closed: #7539 Fresh ddnet install has an empty server browser

<ChillerDragon> i merge quite frequently so i just assumed im recent enough to open an issue. Seems like the bug was fixed somewhen in the last days nice

Tbf, imo you should never open issues on our issue tracker when you are using your own fork

<ChillerDragon> agree

21:44

<ChillerDragon> i just tend to discover the bugs in the software i actually run and then i check wethere i have to fix it or not by running ddnet once

nothing as amusing as chillerdragon

<ChillerDragon> im close to mental breakdown xd

pls no

<ChillerDragon> this debugging session was no fun

do a reset

21:46

this debug session never happened

certified chillerdragon moment

this debug session never happened

21:46

this debug session never happened

21:46

this debug session never happened

21:46

did it help?

<ChillerDragon> yes

21:46

<ChillerDragon> thank

[ddnet/ddnet] Pull request opened: #7540 Added possibility to edit chat size and width

Closes #7531. To make this work, I had to refactor the code that does chat preview. I also made sure to always have some minimal ratio between chat font size and chat width so that we don't have a 5 character wide chat possible. This should also work when cl_chat_old is set to 1. https://github.com/ddnet/ddnet/assets/13364635/310d75ba-c38d-4773-96eb-c02debb00616

Checklist

[x] Tested the change ingame
[x] Provided screenshots if it is a visual change
[x] Tested in...

https://reshade.me/

ReShade - Advanced post-processing everywhere.

23:24

@Jupstar ✪ do u know about this?

23:24

so in ffxiv its common to use reshade

23:24

and im running ffxiv on linux

23:24

i found this https://github.com/DadSchoorse/vkBasalt

GitHub - DadSchoorse/vkBasalt: a vulkan post processing layer for l...

a vulkan post processing layer for linux. Contribute to DadSchoorse/vkBasalt development by creating an account on GitHub.

23:24

reshade is made to apply post shaders to make a game look nicer for gameplay or pics

cool

23:53

is ffxiv on linux natively

Exported 459 message(s)