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 2023-11-29 00:00:00Z and 2023-11-30 00:00:00Z

wow!

Honestly, truly no reason to support 0.7, the only reason I don't just remove it is because it makes chiller happy

kek

Learath2

Honestly, truly no reason to support 0.7, the only reason I don't just remove it is because it makes chiller happy

LOL

remove

why u want chiller to be sad feelsbadman

feelsbadman

$ ccache-s
Cacheable calls:   21174 / 25718 (82.33%)
  Hits:             3567 / 21174 (16.85%)
    Direct:         3406 /  3567 (95.49%)
    Preprocessed:    161 /  3567 ( 4.51%)
  Misses:          17607 / 21174 (83.15%)
Uncacheable calls:  4544 / 25718 (17.67%)
Local storage:
  Cache size (GB):   6.3 /  50.0 (12.59%)
  Hits:             3567 / 21174 (16.85%)
  Misses:          17607 / 21174 (83.15%)

08:22

this is sad isnt it

https://github.com/ddnet/ddnet/pull/7515

Add finish info messages from 0.7 by furo321 · Pull Request #7515 ·...

This adds showing finishes alongside kill messages (now renamed to info messages), like in 0.7 and a few older race clients. Also now with the new Sv_RaceFinish packet, we could also localize fin...

09:23

just read this

09:24

the feature is nice, but i agree with heinrich, no numerical ids, and no reimplementing 0.7 protocol in ddnet please

09:24

we should in fact, remove sixup and all 0.7 hacks

im also against this https://github.com/ddnet/ddnet/pull/7500

Add 0.7 client broadcast feature by ChillerDragon · Pull Request #7...

Where is this coming from? Robsti suggested this could be a separate pr. And I welcome every chance of getting pieces of #5949 merged and discussed separately. So this used to be part of the 0.7 cl...

09:31

more bloat

09:31

now this started me a interesting thought

09:32

ddnet is driven by prs right, and someone doing a PR unnanounced drives ddnet towards a position, even if unwanted, which is what chiller kinda does with his 0.7 prs, so if i were to make a PR removing the 0.7 code, what would happen?

09:33

its a bit unfair, because my pr removing it would be seen as destructive, but we got no way to stop adding 0.7 if we dont want

09:33

it can be interesting to make that pr, maybe it will stir some discussion xd

well, we could do it like other projects. allow other versions of the game. This as some pros and cons: pro:

ppl like Tater can advertise their changes more easy.
ppl can decide what they like more

cons:

ppl like Tater would take supoprt and maintaining more serious
it might split the community at least partially

09:35

and i mean other versions hosted by ddnet.. so basically have a launcher and allow to switch to other versions

@Jupstar ✪ the truth is, you and me cant decide

09:35

it always feels like its heinrich ultimate word

09:35

i would go towards a standalone ddnet

09:36

no teeworlds support

09:36

and heinrich gives in a bit from time to time to chiller adding this stuff

yes, but maybe we can convince him to allow smth like i wrote above

09:36

he always says he wants more openness.. we should also profit from this openness

by other versions of the game u mean 0.7?

09:37

or another client

09:37

i feel ddnet already allows other versions if not bots

yes other clients + servers even

09:37

in the end what must fit is the ranks in the data base

09:37

physics must match or ranks must be split (that's a matter of taste of teh community)

my point is, 0.7 support should not be a goal of ddnet/ddnet, it should be done by someone out of tree as a fork

09:38

and we should focus on the core of ddnet

github-merge-queue[bot]

[ddnet:master] 2 new commits

5aa9003 Remove unused ver parameter of MACRO_INTERFACE - Robyt3 308a2bb Merge pull request #7558 from Robyt3/Kernel-Macro-Cleanup - edg-l

09:38

[ddnet/ddnet] Pull request closed: #7558 Remove unused ver parameter of MACRO_INTERFACE

adding easier moddability is a nice goal too

Ryozuki

my point is, 0.7 support should not be a goal of ddnet/ddnet, it should be done by someone out of tree as a fork

well i obviously agree. but that's why i'd say the only real solution to this is split the client. else we always have these discussions.

u mean move client and server to different repos?

purely objective we make a game for the majority of the community. and the majority would probably benefit from removing 0.7, and go into a more ddnet controlled version, so we can fix weaknesses of the current game

Ryozuki

u mean move client and server to different repos?

yes for example. in the end in steam i'd still use a launcher to select which version you want to test

09:41

ofc just removing 0.7 won't convince anybody... but in trade cleanup the network stack, make connecting faster. at least partially more ddos proof. whatever advatnages there could be could result in the community to like one way more

09:41

i personally simply cannot imagine there is a huge interest in 0.7

09:41

all 0.7 mods basically also exist in 0.6

@Ryozuki but some features like the 0.7 skin system could atleast be ported. I think it's very neat

Teero

@Ryozuki but some features like the 0.7 skin system could atleast be ported. I think it's very neat

that's not directly the point

09:42

we'll have to support both. 0.6 and 0.7

09:42

that will be a mess

yes.

KISS

0.7 can join 0.6 servers right? But not the other way around

https://en.wikipedia.org/wiki/KISS_principle

09:43

i begin to be a @Ryozuki

Teero

0.7 can join 0.6 servers right? But not the other way around

yeah, but it also only works like 95%

09:44

there were some bugs, dunno if they fixed

09:44

and always needs 0.7 maps hosted too

Jupstar ✪

https://en.wikipedia.org/wiki/KISS_principle

It should be common knowledge xd (edited)

the amount of backends we support is also a problem. that increased complexity too much

09:46

i was young, naive

github-merge-queue[bot]

[ddnet:master] 2 new commits

a55a696 updated russian.txt - 0xfaulty 8ae5f89 Merge pull request #7554 from 0xfaulty/russian_fix - edg-l

09:46

[ddnet/ddnet] Pull request closed: #7554 updated russian.txt

Teero

@Ryozuki but some features like the 0.7 skin system could atleast be ported. I think it's very neat

i dont disagree to porting "ideas" from 0.7 to ddnet using the proper ddnet clean protocol, not using sixup hacks or 0.7 shit

Jupstar ✪

the amount of backends we support is also a problem. that increased complexity too much

yes, but that will be harder to remove

09:52

we should support only vulkan, and maybe opengl 4

09:52

or 3.3

09:52

you know dota 2?

xdd

it removed opengl support

09:53

you can only use vulkan

ez

they are so based at valve

it's simply because it's annoying to maintain such different APIs

opengl users should use a layer or smth

i can understand it

vulkan to ogl xd

or vulkan software renderer

09:53

60fps are enough

true, more if ur gpu is trash it cant support vulkan

would make development much easier

we limited by: backcompat, 0.7, teeworlds, old hardware, chiller

XDDDDDDDD

and heinrich

09:54

it would also be easier to support only linux

but thats another

09:54

discussion

09:54

*

we should also support kernel 6+

09:55

and use modern kernel features

09:55

and default compile using x86_64_v3

that's something where i could imagine learath having fun with even xdd

Image attachment

from you i expect nothing less than a nasm version that does not use any libc calls

true

09:58

libc is bloat

09:58

i believe libc could be more efficient if not limited by backcompat too

09:59

anyway what we need is a good stable syscall interface

09:59

not defined by C

09:59

kernels define their functionality with syscalls but they dont make em stable, only linux

09:59

they provide a stable interface through unholy C

yep. IMHO we could easily write emulators for old software to be compatible. i don't really like this aggressive backward compability thinking. I can understand that a kernel should not change every other month. but after 30 years. stuff simply changes

10:00

and if the emulated software looses 15% pref, that would be negated by CPU hardware improvements

i think a kernel could update its functionality by providing a kind of translation layer for some updates

10:01

giving time to update to ppl

10:01

and then remove the layer

10:01

and so on

10:01

or deprecation notices

apparently linux did that very rarely even

@Jupstar ✪ the system V call convention is also not the most optimal nowadays iirc

10:02

System V AMD64 ABI

i dunno what that is

10:02

CPU feature?

no

10:03

its how u make function calls at assembly level

10:03

how it expects the arguments

10:03

and return values

10:03

and stack

ah ok, for 64-bit i assume then?

in x32 it used the stack cuz not many registers

10:03

in x64 it passes some parameters with registers

10:03

iirc

yeah, i only know 32-bit intel syntax

10:04

never worked with 64-bit asm

rust itself

10:04

doesnt follow system v for calls between rust methos

10:04

this is why u need extern "C"

10:04

iirc

10:04

https://github.com/rust-lang/rust/issues/85641

Rust does not follow System V Amd64 ABI calling convention for smal...

Consider the following fn return_result_ok() -> Result<u32, u32> { Ok(7) } fn return_result_err() -> Result<u32, u32> { Err(7) } fn main() { let r = return_result_ok(); println!(&...

10:04

when interfacing with C

ez

10:05

i take that 1% more perf

https://en.wikipedia.org/wiki/Calling_convention

Calling convention

In computer science, a calling convention is an implementation-level (low-level) scheme for how subroutines or functions receive parameters from their caller and how they return a result. When some code calls a function, design choices have been taken for where and how parameters are passed to that function, and where and how results are return...

10:05

more here

10:06

One calling sequence, defined by Microsoft, is used on Windows; the other calling sequence, specified in the AMD64 System V ABI, is used by Unix-like systems and, with some changes, by OpenVMS. As x86-64 has more general-purpose registers than does 16-bit x86, both conventions pass some arguments in registers.

@Jupstar ✪ https://llvmweekly.org/issue/517

10:31

my blog post is on the llvm weekly

10:31

gigachad

Ryozuki

@Jupstar ✪ https://llvmweekly.org/issue/517

Nice even with your real name as notice xd

xd

10:35

well my web has the real name

mornin fellas

Ewan

mornin fellas

what time is it

10:54

did u sleep

No

10:54

i have to do the flip flop

10:54

woke up at 5 pm

xd

how are u doing

Ewan

how are u doing

nice, got into a llvm newsletter

cool

11:18

i bought the cosmic crisp apples

11:18

these ones even look pretty shitty and a little brown like they've been sitting for a while

11:18

but they still taste better than all other kinds of apple

11:18

but they are expensive

im not fan of apples

like $5.50 for 3 lbs

Ryozuki

im not fan of apples

they are a good snack

11:19

if the alternative is something shitty that will leave u feeling shitty

i guess i just dont snack

weird

11:19

i dont eat meals like i should

but i eat quite conscious

that would require a routine past get up -> go to computer -> get brothers from school -> go to computer -> take dog out -> go to sleep

sometimes i eat nuts with whipped cheese, the one thats 0%

11:20

xd

11:20

wait

whipped cheese

idk if its nuts

like cream cheese ? or wat

whipped cheese = queso batido?

11:20

idk the word

what's batido

it just got proteins

11:20

and no fat

11:21

batido is like steered?

11:21

Image attachment

11:21

literally this

11:21

+ nuts

does it taste like cheese

nuts have the healthy type of fat

Ewan

does it taste like cheese

ye, but its almost neutral flavour

it sounds like a regional delicacy

as in it doesnt taste much

i am imagining cheese yogurt

but the nuts give it enough taste for me to be edible

11:22

xd

11:22

its like yogurt yes

11:22

but not quite

lol i see

11:23

i like nuts

11:23

but they are very expensive

are american nuts as healthy tho

11:23

in spain its rly common

are they not the same nuts...

Image attachment

walnuts

idk

11:23

3.7€

11:23

according to google

you put the least flavorful nut in your whipped cheese

30g

11:24

350g

11:24

the most healthy

11:24

xd

probably not by a wide margin

11:24

though i like walnuts

11:24

i like a mixed nuts more tho

11:24

almonds cashews walnuts

11:24

maybe peanuts but not usually. i dont like peanuts themselves

11:25

they taste chalky

i like walnuts at this point

11:25

also the cheese fills you easy

it's a high protien food

gotta get the muscle

the most sustaining macro nutrient

11:25

each protien is like 9 calories or something

11:26

based on my recollection of freshman year biology

xd

it's not necessarily about muscle

idk yeah

11:26

its for sure more healthy than having pizza

11:26

for breakfast

11:26

i read this today on a whatsapp group

11:26

xdd

yes you need lots of protien to build muscle but you need to eat lots of protein even in a deficit or when trying to stay the same bcs otherwise you just lose muscle

11:27

also why it's counterproductive to eat empty carbs/sugars and fats, and expect to keep muscle while gaining weight

11:27

bulking as u might call it

Ewan

yes you need lots of protien to build muscle but you need to eat lots of protein even in a deficit or when trying to stay the same bcs otherwise you just lose muscle

yeah

gotta balance your diet .....

thats why u want high protein low calorie

yea

i think

11:28

im no expert

depends on what your goal is

11:28

based on my understanding high protien low calorie = lose fat, sustain and build muscle high protien high calorie = gain weight, sustain and build muscle

i just think it would be hard to do on merica

11:28

or more expensive

yeah

they give u giant sized portions

11:29

they use bad oil

11:29

iirc

corn and seed oils are by far the most popular

or butter

don't knock butter it's awesome in moderation

11:29

unlike shitty seed oils which serve no purpose other than to be cheaper than a good oil like avocado or olive or whatever

i only buy good oil

11:30

luckily i live in the country that exports the best oil

same but there is no guarantee for other things we buy at the store

i use extra virgin olive oil

they are usually made with Shittiest oil

Ryozuki

i use extra virgin olive oil

us too

and well i try to not use too much

11:30

non stick pans best

no way

11:31

as long as you're gentle to them they are fine

11:31

a well seasoned cast iron or stainless steal skillet will let u use the same amount of oil + have the same nonstick properties as a teflon pan

11:31

they are just harder to maintain

xd

11:31

u sound like a expert

they have superior heat retention

11:31

so ur food heats more evenly

i just have my pan ok

11:32

i got it for 30€ i think

also much safer to ingest a flake of polymerized food oil than a flake of teflon once your coating goes bad

justatest

i own many pans

11:32

prob 45lbs of cast iron in my cabinets

damn

11:32

give some

xd

they so expensive

11:32

i only got 2

not hard to find at second hand stores but expect to do some restoration

11:33

we got some caraway pans a few months ago which are nonstick but they are also not teflon

11:33

better for the environment and safer to ingest

11:33

good in a pinch

started recently living alone tho i guess thats why

yea

11:33

i will have to get rid of a lot of my possessions when i move out

trying to make a windirstat ish app

Image attachment

11:47

ms fluent guidelines rly do make it hard to make space-efficient UIs

11:47

the original app has tables with a bunch of columns that are all totally visible at a reasonable size

Image attachment

11:47

but if i busted out a table that complied with the fluent guidelines i would just get 2 chars per column and then ... xd

xd

11:49

imagine using windows

11:50

make it with gtk

11:51

@Ewan how can u live with just 700gb

this just the windows frontend

Ryozuki

@Ewan how can u live with just 700gb

i have 8tb across all disks

11:51

windirstat reading is weird

but they are ssd?

11:51

or m2

3 m.2 which total 1.5TB

11:52

then i have 2.5 tb of sata ssd

11:52

4tb hdd

i got 2.5tb m2

11:53

and 2tb ssd

nice

11:53

i lied

11:53

i have 2.5tb in m.2 and 1.5 in sata ssd

11:54

1 pcie 4.0 500GB boot drive and 2 pcie 3.0 1tb disks in raid 0 (edited)

11:55

the sata disks are in a migratory period so to speak

11:55

it’s my old drive so sometimes i just go there to find a file or whatever but they don’t rly get used

i will sit on this design for a little bit

Image attachment

13:17

i am out of salt for my eggs

looks modern

13:22

and windowsy

13:22

i need t o make android apps to become rich

like i said this is just the windows frontend

13:25

it will look less windowsy but largely the same on other platform

man doing assembly with nasm rly helped me

13:30

use gdb and understand stuff

[ddnet/ddnet] Issue opened: #7559 Add more console commands for demo player

Needed by #7556. While we already have demo_play, demo_speed, demo_slice_start and demo_slice_end, these commands are not enough to allow more binds for demo player, with the ultimate goal to remove hard-coded binds. I suggest to add the following console commands for demo player:

demo_speed_adjust i[direction]: adjust the speed of the demo, using an array of speeds
demo_seek_skip i[direction]: skip time using an array of time skips
demo_seek_marker i[direction]: se...

Ryozuki

man doing assembly with nasm rly helped me

i always used ollydbg like a caveman

13:41

and by the time I discovered there were better tools interest had shifter

13:41

d

[ddnet/ddnet] Issue opened: #7560 Add more console commands for menus

Needed by #7556. I believe we have some console commands related to menus, like quit or ui_page (even though I think this one does not work correctly?), but I also believe we need more. For example, we could add console commands to go to specific pages like "Play", "Settings", "Demos", to be able to rebind the keys P, S, D respectively. I am not sure if this is really needed, but I also think that it won't hurt to remove those hard-coded binds as well. @Robyt3 as you suggested t...

Image attachment

13:48

gm

Leak

13:51

The geo guessers will know xd

deen already has my address

13:53

it would be easier to ask him

13:55

are these colors better .. .. .. .

Image attachment

Ewan

trying to make a windirstat ish app

haha wtf i thought that was the screen panel where you set up your screens from windows settings at first XDD

wait you don’t have 2,220,302 monitors

Ewan

wait you don’t have 2,220,302 monitors

managing ddnet spaghetti code is hard i know but....

14:09

XD

today im playing with gdb at work

Ryozuki

today im playing with gdb at work

you mean gdb will play with you

finding a bug in a shared library made with mlir

Teero

you mean gdb will play with you

xd

14:11

gdb with gef is niceee

never heard of gef until now

14:20

looks nice

i heard about it from heinrich

14:32

i use it daily

@Ewan ur a audio pro

14:51

i got my 100€ bose speakers but

14:52

i feel i can get more quality

14:52

do u recommend some good

14:52

good bass, and maybe good with music that sounds like synthwave? xd

14:52

up to 200€

14:58

You should always try to control the volume at the last part of the chain, which in this case, would be your speakers. Set the DAC and computer volume to max. If your speakers can't handle it(some speakers can't handle high input), then turn down the DAC volume a little bit.

14:58

ok i havent been doing this

well ppl don’t often make quality oriented computer speakers that aren’t monitors

15:00

you can get some bookshelf speakers and they will sound good but you’ll probably have to wire it up

@Ewan i have a dac/amp fiio

also if your room is very small or you are close to the speakers then that will suck

i think the speakers i have are active

15:01

so they also have a internal amp

15:01

should i get passive speakers?

depends on what your amp can do

the room with pc is not small

15:01

im a bit close to the speakers i think

15:01

xd

yeah i wouldn’t go much bigger

15:02

do you want accurate audio or a ‘fun’ listening profile

idk if accurate means good

15:02

idk

15:02

xd

15:02

changing the dac to full and adjusting volume at the speaker seems better now

KEF Q150 and a subwoofer

i also put the speakers a bit further

Q150 expensive

15:04

also concerned bookshelf speakers will provide a bad experience if he is rly close to them

or Q100

firstly what outputs does ur amp have

fiio k3

15:08

xd

15:08

remember in the past

15:08

i go tthis

15:09

Image attachment

my results came

Image attachment

15:23

apparently that's 95 days

15:23

more than 1/4 of the year feelsbadman

feelsbadman

(edited)

ewan u gotta put a computer image in there so this doesnt get deleted

uhhh

15:23

flpflpflpf

isn't it better to get a soundsystem with a built in power amplifier?

no

15:24

maybe for a tv or something

15:24

ryo ur amp doesn't have any kind of raw out

there are rly good cheap ones

i suppose you could split your line out but idk most of what you'd want would take RCA, 1/4 or raw speaker wire

15:25

you could check out some monitors

15:25

with smaller drivers if you're close to it and the room is small/medium

15:26

presonus eris 3.5-5.25 (the number is the diameter of the driver in inches)

15:26

that would be about your budget

15:28

you ought to look up a review since i dont buy new audio equipment

15:29

im also not sure your amp would power any passive speakers

15:30

i think it’s a headphone amp

15:30

good news is most monitors are active

Ewan

i suppose you could split your line out but idk most of what you'd want would take RCA, 1/4 or raw speaker wire

why do u meantion split

15:32

mi fiio has 2 outputs

15:33

the one at front, for headset

15:33

the line out at back

15:33

for my speakers

15:33

or does a passive speaker needs 2 inputs?

wat

im confused

15:34

idk most of what u said xd

15:34

what is raw out

15:34

why do i need it

speaker wire output for hooking up speakers that don't have other connectors

15:34

which is most

15:34

it can look a bunch of different ways

i see

15:35

i have my speakers on the line out at back

15:35

it looks like a jack

15:35

small i think

it is 3.5mm jack

ye

just a headphone jack

its bad right

that will not power passive speakers (edited)

ah

15:36

but my bose speakers got a amp im sure

xdd

i can adjust volume there

yea

they are powered to current

15:36

too

but i thought u were trying to get new speakers

15:36

which usually aren't powered

yes, if they are better tho

they rely on the amp/receiver to power them

i see

unless they are monitors which are designed to be used at a computer

15:37

which expect to be hooked up to an audio interface but the dac you have will work just as well

i like following unix philosphy with this

just need an adapter

if its standalone its usually better

15:37

1 thing but good

15:37

monitor with speakers doesnt sound like that

15:37

xd

depends on use case

15:37

sry by monitor i mean

15:37

studio monitors

15:38

they are speakers that aim to produce accurate sound for producers/engineers and purists

15:38

not like ur screen xd

xd

15:38

im a consumer

basically look up some reviews

15:39

a reddit thread or something for ppl in ur situation

actually turning the dac to max and controllign volume on speaker kinda sounds better

good

i had the dac near min xd

15:39

if u put my speakers at max i will blow off the neighbour ear

15:39

lmao

yea u should take all the steps u can to make sure ur own sound is still good

15:39

what sample rate does ur pipewire run at

it was at 48khz but i changed it now to default.clock.rate = 192000 default.clock.allowed-rates = [ 192000 96000 48000 44100 ]

15:40

but

15:40

i need to restart

15:40

brb

i see

15:40

wait

15:40

192 is tryhard sample rate

what should i use

15:40

96?

sample rates higher than 48k are really only good for audio analysis that like, isn't done by your ears

before both options only had 48000

15:41

ok

15:41

so what do i put

depends

Image attachment

15:41

48?

anything other than 48k or 44.1k will require resampling which is an inherently inaccurate process and can produce artifacts in itself. you can choose your resampler to remedy this but you should probably only have a high sample rate if you listen to music which is distributed at this rate

15:42

like listening to a 44.1 cd rip or youtube video or something on a 96k system will just stretch those 44.1 thousand samples and interpolate the rest the best it can

15:42

it will not be as good as listening to a recording with 96 thousand samples per second. which is rare to find

ok

so unless u have like lossless rips or something then i would stick with 48

but allowed rates means if a source has that rate it will use it?

15:42

and not resample?

yea let it do whatever it wants there

ok

15:43

so default is for what then

all that should matter is the one that your audio system actually runs at

ah

default should be 48 imo

ok

pro audio is actually a huge scam

Image attachment

15:43

so this is good

yep

what is quantum

listening quality gains are usually not found on the software side

15:43

no idea what quantum is

do u suggest other values than default

idk i spent weeks ricing out my pulseaudio configs and then it all fell out of my head

its related to buffering and latency

oh

15:44

turn it down then

15:44

if u want lower latency

https://gitlab.freedesktop.org/pipewire/pipewire/-/wikis/Config-PipeWire?version_id=9578bfa2b3bdc7a30a957cb4e254133df7f96844

Config PipeWire · Wiki · PipeWire / pipewire · GitLab

Multimedia processing graphs

doesn't matter much 1024 seems high by default

15:45

if u have a good cpu then u can probably turn it down to 256 or lower... below that you will see minimal gains but probably popping from buffer underruns

15:45

depending on how high ur cpu utilization is

i dont do any latency sensitive task

15:46

ill leave it as is xd

smart

15:49

defaults are usually good

Ewan

defaults are usually good

not in linux kernel

15:51

linux kernel uses defaults for ancient computers

15:51

i had to raise my ulimits for sane stuff

they are most well rounded for the industry

esyncf

15:51

esyncfs

15:51

https://github.com/lutris/docs/blob/master/HowToEsync.md

docs/HowToEsync.md at master · lutris/docs

Contribute to lutris/docs development by creating an account on GitHub.

15:52

Esync removes wineserver overhead for synchronization objects. This increases performance for a lot of games, especially ones that rely heavily on multithreading.

so true

"We think there is a fundamental misconception about piracy. Piracy is almost always a service problem and not a pricing problem," he said. "If a pirate offers a product anywhere in the world, 24 x 7, purchasable from the convenience of your personal computer, and the legal provider says the product is region-locked, will come to your country 3 months after the US release, and can only be purchased at a brick and mortar store, then the pirate's service is more valuable."

16:30

why is gaben so wise

Ryozuki

"We think there is a fundamental misconception about piracy. Piracy is almost always a service problem and not a pricing problem," he said. "If a pirate offers a product anywhere in the world, 24 x 7, purchasable from the convenience of your personal computer, and the legal provider says the product is region-locked, will come to your country 3 months after the US release, and can only be purchased at a brick and mortar store, then the pirate's service is more valuable."

https://www.reddit.com/r/IndianGaming/comments/17g2m4y/steam_removing_regional_pricing_from_argentina/?rdt=63109

From the IndianGaming community on Reddit: Steam removing regional ...

Explore this post and more from the IndianGaming community

Jupstar ✪

https://www.reddit.com/r/IndianGaming/comments/17g2m4y/steam_removing_regional_pricing_from_argentina/?rdt=63109

Um... It's not a removal of regional pricing - it's a removal of the regional currency.

16:38

still, steam service as a whole is more than owning a game

16:38

friends, library, proton, etc

most people pirate for pricing reasons in my experience lmao

@louis look at it like this

16:39

netflix is a service

16:39

u can also pirate movies

16:39

there will always be ppl who pirate

16:39

but it provides a comfortable service for the generic population

16:39

most end up getting netflix

16:40

xbox game pass is so popular cuz this too

that's still a pricing issue and i don't think that relates to ur quote

its not a pricing issue

16:40

u can pirate the series, but u choose the comfort of having it at 1 click

not if i'm broke

if ur broke u wouldnt buy it anyways

16:41

ur not the target

if im broke i'd pirate it correct

there will always be pirates

if i rly wanted to watch it

the point is putting stuff anti piracy is bad

16:42

because real customers will have worse experience

16:42

u make those customers with money thinking about it convinced with a good service

Ryozuki

xbox game pass is so popular cuz this too

spotify

16:42

pirating music will never be as easy as using spotify

yep

this is why i have over 3 months of listening this year

Ryozuki

the point is putting stuff anti piracy is bad

i mean i agree with that

16:43

i think urq uote has nothing to do with that though

16:43

he makes it seem like the issue of piracy is with region locks or unavailability of products which is just not true most of the time

Ewan

pirating music will never be as easy as using spotify

true but i'd say thats different

16:45

caues spotify can be a free service too and i'm pretty sure u can block ads anyways

‍

️

‍

️

16:48

https://pyret.org/index.html

The Pyret Programming Language

Ewan

pirating music will never be as easy as using spotify

spotube

17:04

https://github.com/KRTirtho/spotube

GitHub - KRTirtho/spotube:

Open source Spotify client that doesn'...

Open source Spotify client that doesn't require Premium nor uses Electron! Available for both desktop & mobile! - GitHub - KRTirtho/spotube:

Open source Spotify client that does...

Ew

Better than shit electron

https://github.com/mrpond/BlockTheSpot

GitHub - mrpond/BlockTheSpot: Video, audio & banner adblock/skip fo...

Video, audio & banner adblock/skip for Spotify. Contribute to mrpond/BlockTheSpot development by creating an account on GitHub.

What electron

And instant download

Wdym

Ewan

What electron

https://www.electronjs.org/de/

Build cross-platform desktop apps with JavaScript, HTML, and CSS | ...

Build cross-platform desktop apps with JavaScript, HTML, and CSS

I know what electron is

17:08

Where is the elctron

Spotify

17:08

Uses

Spotify is not electron

Electron

17:08

It is

It isn’t

It is

It quite literally is not

does it use >100MB ram?

17:09

then it's electron

No

17:09

It’s CEF

17:09

and the devs have brains

Chromium

17:09

Same thing

Literally not

17:09

Separate projects developed in parallel by separate people

its fking slow

Fast af for me

but it's web based

17:10

and web based is always JS runtime

17:10

and html+css engine

17:11

so how much RAM does spotify use @Ewan

Relevance

leak

Let me see

Send ss

17:12

No photoshop pls (edited)

chrome for 3 tabs

Image attachment

17:14

xdd

17:14

but i have the RAM

17:14

it should use what it needs

17:14

i like to share

Jupstar ✪

it should use what it needs

Limit to 1mb and see what happens

firefox enjoyer here

me too

Ok it was at 900

MB??

it was open for days and i had hella music cached

17:15

i restarted and it's 300

ff is even worse @Ryozuki

Image attachment

A GIGABYTE?

mhm

i got 32gb

Ryozuki

i got 32gb

and you like to use what you bought

17:16

Taking the horse to the track xD

Image attachment

17:17

lok at RES

17:17

1gb for rust analyzer

17:17

firefox 861mb

17:17

xd

Jupstar ✪

ff is even worse @Ryozuki

loli_dance

Image attachment

how only 8gb

17:17

in 2023

Ewan

how only 8gb

Here is expensive

damn

where is here

Compressed (0mb)

my ram was $300

Ddr5?

yea

u use a new cpu bro

17:18

ofc ur ram is expensive

17:18

its also 5600mhz

so true

17:18

naw it's 6000

Ryozuki

where is here

latam

my ram is 3200mhz it costed me 110~~€

17:18

for 32gb

expo

17:19

ovahclocked

17:19

not rly overclocked if it comes like that tho is it

17:19

that's just the clock

u talking about urs or mine

mi ne

idk

The Crock

if ur cpu supports 6000mhz then be it

well yea

17:19

just saying they call it an overclock

17:19

even tho it applies the expo preset by default

In my first oc I burned my cpu

17:20

ChikaChikaChika

lmao

17:21

back when hardware had no overheat protection

Jupstar ✪

back when hardware had no overheat protection

it's not a bug, it's a feature

ye

17:23

it tells u to buy new hardware

@Yek-H can you PC run vulkan?

17:24

lol

17:24

troll

17:25

u have good pc

Jupstar ✪

@Yek-H can you PC run vulkan?

Sure

https://discord.com/channels/252358080522747904/295908390956433410/1023401770032037938

XDDDDD

17:25

troll

owo

@Yek-H so how much RAM do you actually have

17:25

xDD

Jupstar ✪

@Yek-H so how much RAM do you actually have

8

and your GPU?

17:26

how much VRAM

4

lol

17:27

i also have double the RAM than VRAM 32 vs 16

17:27

But now I will reinstall windows, I modified the system so much that it became unstable

17:30

waaaa_pb

Linux ftw

Doesn't run some programs

what do you need?

Teero

what do you need?

Raidrive

Yek-H

Raidrive

that's all?

17:32

i bet there are 1000000 alternatives to that xd

https://rclone.org/

Rclone syncs your files to cloud storage: Google Drive, S3, Swift, Dropbox, Google Cloud Storage, Azure, Box and many more.

17:33

first one i found

17:33

opensource and free

Teero

https://rclone.org/

Doesn't work well with emby

raidrive is just an FTP client no?

then don't use emby

SFTP and mount a directory as local

i hope you dont have emby on a windows server justatest

justatest

Teero

i hope you dont have emby on a windows server justatest

justatest

In windows emby doesn't use resources because it's direct transmission

https://emby.media/emby-theater-linux.html

media server for personal streaming videos tv music photos in mobile app or browser for all devices android iOS windows phone appletv androidtv smarttv and dlna

17:35

xd

Teero

https://emby.media/emby-theater-linux.html

Sure but the amd driver on linux dead

17:36

RageJK

no

17:36

nvidia is shit.

17:36

amd is always latest (edited)

17:37

all excuses kek

kek

In linux makes conversion of multimedia = cpu 100%

17:38

My GPU is not detected properly

huh?

Transcoding

what gpu do you have?

RX 5500 XT

u need vaapi drivers

Teero

https://emby.media/emby-theater-linux.html

i read that as enby and almost smiled

why would it make you smile

today was the day where using 0xbeef as a value helped me debug

nobyisthebiggesttroll

0.7 can join ddnet servers that have the bridge active. But only because ddnet added support for that

Replying to @Teero 0.7 can join 0.6 servers right? But not the other way around

18:37

It’s not about interest. It’s avoiding the hard fork and uniting the playerbase again

Replying to @Jupstar ✪ i personally simply cannot imagine there is a huge interest in 0.7

18:38

._.

Replying to @Ryozuki we limited by: backcompat, 0.7, teeworlds, old hardware, chiller

there is no playerbase to unite

18:39

0.7 is chiller itself

18:39

also we shouldnt avoid a hard fork

I play 0.6

18:39

lol

we should have clean ddnet, and if others want clean 0.7ddnetorwhathever

chillerdragon

I play 0.6

more reason u give xd

I wanna play ctf with the squad

18:40

Vanilla squad

18:40

Using ddnet client

18:40

Not mouse bugged vanilla client

chillerdragon

It’s not about interest. It’s avoiding the hard fork and uniting the playerbase again

uniting 6 people

Yes uniting 6 people!

# NEW! ChillerDragon vs. Enemy | 0.7 support

19:17

19:18

prime time 20:15

justatest

0.7 is no bloat when 0.6 is removed

19:31

Heinrich does not want to remove 0.6 to keep back compat

19:32

Ryo and jopsti want to break back compat to cleanup protocol

19:32

But let’s be real neither ryo nor jopsti are working on ddnet so who would even break the protocol by actually implementing the changes. And what changes even?

i think if we break back compat we would be able to have more flexibility with features

Like what

19:35

What’s limiting right now?

well for one the killfeed code is mega jank

19:37

we could very easily squish it if we didnt need to have 0.7 packets

Is it?

19:38

That’s not a matter of breaking 0.6 compat that’s a matter of adding 0.7

19:38

Show me mega jank code and how clean it could be?

19:39

Im sure it would also be clean if we didn’t need to have 0.6 packets and only had 0.7

afk

[ddnet/ddnet] Pull request opened: #7561 Also print command in console when identical to previous

The command currently being executed was not being printed to the console if it's identical to the previous history entry.

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 ex...

https://github.com/rust-lang/this-week-in-rust/pull/4912

Add Intro to LLVM and MLIR with Rust and Melior by edg-l · Pull Req...

20:26

@Jupstar ✪ ill be in the next TWIR

20:26

This week in Rust

poggers2

Poggers

pepesmash

Ryozuki

@Jupstar ✪ ill be in the next TWIR

no idea wtf that is but: epic

dude

20:34

u should read TWIR

20:34

every week

20:35

its the oficial newsletter

20:35

of rust

20:35

look the repo group

20:35

rust-lanh

how can there be news if all good features stay in nightly

g

xdd

20:35

HUH

20:36

@Jupstar ✪ its one of my sources

20:36

for links i share here

20:36

and for ffr

Ryozuki

and for ffr

was about to say: which is tomorrow, but then i realized it's only the day after tomorrow :/

chillerdragon

Show me mega jank code and how clean it could be?

im back

21:23

anyways what im saying is do we REALLY need 3 packets for the killfeed? like the only reason why i added one (and furo added one) is that we cant touch the og killfeed code

21:23

or else that breaks back compat

voxel4president

id be a shit president

then you are in the average

21:26

it's already a skill if you are not a very shit president

21:26

xd

Ryozuki

0.7 is chiller itself

I'm fairly sure 0.7 is chiller and his 15 bots

chillerdragon

Im sure it would also be clean if we didn’t need to have 0.6 packets and only had 0.7

I'm confused, why would we only use the packets of a dead fork that breaks half our mod by moving stuff out of the snap? Elaborate

To stay compatible

23:14

What does the new snap break?

It breaks skin changes, colour changes, name changes just to name a few

chillerdragon

To stay compatible

To stay compatible with a dead game?

Exported 780 message(s)