DDraceNetwork - developer

Patiga

because we need to switch the texture for each sprite of the tee

But there's all on the same texture?

wdym

It gets separated in memory?

everything gets cropped & chunked out afaik

00:09

before it even gets to the gpu

But you don't have to do it that way

yeah i mean you can treat it as an atlas

00:09

but that's not how it works rn

<ChillerDragon> @Barbs 🐙 could you describe the bugs in words?

00:10

<ChillerDragon> I am not sure if JSaurursRex is maintaining that modification. I for sure am not. I am maintaining the server at 88.198.96.203:8309 "ChillerDragon's CTF GER1" but it would still be good to know what kind of bug.

Idk the atlas doesn't even sound that hard right? We already have them on one texture and know the coordinates.

00:12

Is it because of the tee colors?

00:12

Those get stored in the texture itself?

you'd need a shader to change the color if not

00:13

which idk if it does already

00:13

but i would assume not because it's a fairly well known algorithm that's been reproduced in a few other software representations

00:13

that web one

00:13

idk

00:14

i wonder what it actually does yeah

They all write the colors to the texture directly I'm pretty sure

too bad it's american hours rn

Assa

I think quad trees currently don't make sense in ddnet, because the overhead is too big for only 64 players and the game doesn't contain too many entities (edited)

updating quad trees isnt that intensive seeing as you only have to do it once per tick at most. each player checks against each player for collisions, and when hooking and for each projectile turning an O(n^2) to a O(n) or less it definatley worth

00:19

specially if we go upto 128 clients

does someone has resources/a link to the new years photoshoot mod? I want to do sth similar with my friends

Gumba

does someone has resources/a link to the new years photoshoot mod? I want to do sth similar with my friends

@TsFreddie

Gumba

does someone has resources/a link to the new years photoshoot mod? I want to do sth similar with my friends

if theres not many players u can just photoshop it :P

https://github.com/TeeworldsCN/MomentCap i think its this one

GitHub - TeeworldsCN/MomentCap

Contribute to TeeworldsCN/MomentCap development by creating an account on GitHub.

00:48

@Gumba

good find

ill use that freddie ping to ask him another question then

org reminds me of the BemaniCN

00:49

that's probably totally lost on u sorry

00:49

zhn

@TsFreddie

we love ya freddie, no questions and no answers ;3

Solly

specially if we go upto 128 clients

With proper collision you can effortlessly handle thousands :)

Solly

updating quad trees isnt that intensive seeing as you only have to do it once per tick at most. each player checks against each player for collisions, and when hooking and for each projectile turning an O(n^2) to a O(n) or less it definatley worth

^

Who can create a script for fifo?

02:41

I just want my message to be displayed in the broadcast every specified time. this is implemented via fifo But I do not know how

github-merge-queue[bot]

[ddnet:master] 2 new commits

7886a73 Update spanish.txt for 18.8 - n0Ketchp c25e06d Merge pull request #9382 from n0Ketchp/patch-23 - def-

03:17

def-

[ddnet/ddnet] Pull request closed: #9382 Update spanish.txt for 18.8

@Evelyn you can

03:46

do some research

03:46

i believe in you

SollyBunny

[ddnet/ddnet] Pull request opened: #9383 Don't merge diffrent colored client msgs

Checklist

[x] Tested the change ingame
[x] 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
[x] Tested the change with [ASan+UBSan or valgrind's memcheck](https://github.com/ddnet/ddnet/#using-ad...

<ChillerDragon> nice website http://www.bricktou.com/arch/x86/include/asm/syscallsyscall_get_error_en.html

syscall_get_error

Program logic extraction. Brick Co.,Ltd.

Ewan

@Evelyn you can

I can't do it

06:54

I've been trying for 2 weeks now, I've even been looking for a programmer, I'm asking the gpt chat

06:54

Nothing helped me.

06:55

When I was looking for a programmer, no one understood what I wanted and because they didn't play ddnet

06:56

Although I was willing to pay $12

When I was looking for a programmer, no one understood what I wanted

You will become a really good project manager (edited)

3

you need to be the programmer

06:58

lol

06:58

$12 means little to most ppl in here

This command was not found.

you would probably have an easier time getting ppl to help you if you showed initiative & tried to code with them instead of just asking them to do it for you

Ewan

$12 means little to most ppl in here

2 weeks is nothing

07:26

buy yourself some c++ course with that money

07:26

better invested

I do not know how python works, I will not be able to use it.

08:06

maybe $ 12 is not enough for this job, but it's more difficult for me to find , I can find a maximum of $ 30, I don't know if it's good for just creating a script that simply enters the broadcast command into a fifo file every specified time

jxsl13

buy yourself some c++ course with that money

I've already ordered a book.

1

Evelyn

maybe $ 12 is not enough for this job, but it's more difficult for me to find , I can find a maximum of $ 30, I don't know if it's good for just creating a script that simply enters the broadcast command into a fifo file every specified time

And even then I do not know how long it will take to write a script, maybe a day, maybe a week

Souly

https://github.com/TeeworldsCN/MomentCap i think its this one

thank you!

Solly

if theres not many players u can just photoshop it :P

kinda true, but its hard for me (no skill)

Solly

specially if we go upto 128 clients

I agree, that this might be viable with 128 players, since, as you already noticed, you'll have 128^2 = 16384 collision checks for players instead of 4096 curently at most

Tater

@Jupstar ✪ why is tee rendering so slow btw?

Tbf I cannot say if the bottleneck is purely tee rendering. But generally in the player rendering pipeline you have a lot of unbatched draw calls. You first render the left hand + hook, then the right hand + weapon, then the player And this is repeated for all players on a per player basis. and even if you render one tee in a whole you still render 120 tees on its own, since all tees have different skins / weapon textures etc. (potentially) (edited)

Hmm yeah it's a lot of draw calls I guess.

Is anyone familiar with some good open source profiling tools for c++? I refuse VSC and the only other solution i found was Orbit from google

Assa

Is anyone familiar with some good open source profiling tools for c++? I refuse VSC and the only other solution i found was Orbit from google

on linux u have perf (+ flamegraph), callgrind and i think gcc even has some integrated counter

Also thinking about it, the client will suffer from the n^2 physics much more than the server right? Potentially it must do 1000ms/20=50 ticks of simulation for every actual tick due to prediction.

oh that's good, going to look into that, thank you ❤️

Tater

Also thinking about it, the client will suffer from the n^2 physics much more than the server right? Potentially it must do 1000ms/20=50 ticks of simulation for every actual tick due to prediction.

yeah, defs a big problem

does the client check for collisions?

anti ping has to

1

09:54

and local char pred

It does almost everything

09:58

It would be nice if the server and client simply shared the same physics implementation. They're not really that far apart right now.

they are mostly doing this, both client and server include collision and gamecore

09:59

but I agree, there could be a better separation

Yeah but they duplicate all entities

? they have to, they run on different machines. I don't get your point there

I mean the code

10:00

There's a duplicate file for each entity

Assa

but I agree, there could be a better separation

I want less separation lol. None ideally

I mean separation of the physics to the rest of the code

Yeah

slab a version on it and poof, backwards compatible physics

The issue is cleanly integrating all the servers sideeffects in the physics in a way the client can ignore (edited)

10:03

Idk how ddpg does this tbh

10:03

Maybe you just probe the game state instead of triggering events (edited)

I can tell you how I do it, physics are only done on base-entity-objects and entity-entity-interactions have callbacks

Tater

The issue is cleanly integrating all the servers sideeffects in the physics in a way the client can ignore (edited)

you mean like random teleporters?

Jupstar ✪

you mean like random teleporters?

No like calling the function that saves your time to the database when you touch finish (edited)

ah, yeah the client simply has a dummy database that is disabled

10:07

in the end ddnet's database is also optional

Jupstar ✪

ah, yeah the client simply has a dummy database that is disabled

But that means the physics depends on the database implementation?

Tater

But that means the physics depends on the database implementation?

database interactions are async to me

Or you need a bunch of callbacks

no, only on the interface

if they alter physics that cannot be predicted ofc xd

github-merge-queue[bot]

[ddnet:master] 2 new commits

11ee49f Don't merge diffrent colored client msgs - SollyBunny 2ec51d7 Merge pull request #9383 from SollyBunny/dont_merge_diffrent_colored_client_msgs - Jupeyy

10:09

Jupeyy

[ddnet/ddnet] Pull request closed: #9383 Don't merge diffrent colored client msgs

Tater

But that means the physics depends on the database implementation?

ah there is an interface anyway

Yeah

since wasm has no access to anything

10:10

you have to import host functions into the module anyway

10:10

and only the server gives access to anything real

How many interfaces does the server need for the physics?

10:12

I have no estimate

currently it only has database and terminal (basic logging)

10:13

maybe it will need file access too, but ddnet doesnt

10:13

(i think)

terminal includes chat?

i meant like create logging

But don't you need chat

Tater

But don't you need chat

yes, it currently only gets chat commands

10:14

but chat for me is an input into the wasm module

10:15

if you mean that, then it ofc has lot more stuff

10:15

rcon commands etc.

10:15

the other way around is more interesting bcs it is more unsafe

Input into the module is different from an interface?

well i thought you meant, what the server can call outside of what wasm offers

hmm

10:16

But if the physics wants to say "player has finished" how does that happen

10:17

Idk what my question is anymore lol

well twgame for example uses a mpsc channel. The player hits the finish line -> then the player_tick or whatever checks for events in that channel and starts a database query.

10:18

and the query itself happens async

10:18

and the database query itself uses a database interface

10:18

the client uses a dummy database which implements that interface

10:19

except for the actual query all things happen inside the physics module tho

Jupstar ✪

except for the actual query all things happen inside the physics module tho

What doesn't happen in the physics module then?

10:22

I thought it was very limited

Tater

I thought it was very limited

i mean what does limited mean. wasm allows you to basically do anything that doesnt require a kernel call (even allocations)

10:22

so you couldn't just do a http request

Snapshot packets?

10:23

ok

Tater

Snapshot packets?

ah that happens indeed outside of it

10:23

well no

10:23

50:50 the snapshot binary data is inside the physics module. but the compression / binary diff etc happen outside

How can you require a database interface on the client if this module is supposed to be sent over the network. Does that mean servers need to use the standard interface?

Tater

How can you require a database interface on the client if this module is supposed to be sent over the network. Does that mean servers need to use the standard interface?

The wasm module has imports, which are functions inside a wasm module without a state or anything else. so e.g. you do call_sql("SELECT ..") inside the wasm module. and then inside the host you can do with that information whatever you want. The client simply returns an error that this is unsupported. This is very simplified, bcs i abstracted these simple functions away using derive macros, in the end it feels more like you use a instance of a database object. so you do object.query(my_statement) and the rest happens hidden away from the user

ok that makes sense

Tater

They all write the colors to the texture directly I'm pretty sure

the color is added in the shader. the client doesn't use texture atlasses, probably because the colors might leak over (?)

Patiga

the color is added in the shader. the client doesn't use texture atlasses, probably because the colors might leak over (?)

Not even for the tile atlases?

10:33

Those get separate textures?

Tater

Also thinking about it, the client will suffer from the n^2 physics much more than the server right? Potentially it must do 1000ms/20=50 ticks of simulation for every actual tick due to prediction.

btw with 128 tees that do random inputs on ctf1 i get an avg of:

28601 t/s
29154 t/s

with the non O(n²) approach Which surprises me, this is like 2x more than i had a month ago lol. Time to git bisect xd But defs worth to optimize it ig

Tater

Not even for the tile atlases?

those are texture arrays. those are great and work well with same-sizes textures

Jupstar ✪

btw with 128 tees that do random inputs on ctf1 i get an avg of:

28601 t/s
29154 t/s

with the non O(n²) approach Which surprises me, this is like 2x more than i had a month ago lol. Time to git bisect xd But defs worth to optimize it ig

This sounds high but it limits perceived smoothness to 29000/50 "fps" if you do it on the same thread

10:38

which is probably fine

Jupstar ✪

Tbf I cannot say if the bottleneck is purely tee rendering. But generally in the player rendering pipeline you have a lot of unbatched draw calls. You first render the left hand + hook, then the right hand + weapon, then the player And this is repeated for all players on a per player basis. and even if you render one tee in a whole you still render 120 tees on its own, since all tees have different skins / weapon textures etc. (potentially) (edited)

what if all parts of tee will render batched?

Tater

This sounds high but it limits perceived smoothness to 29000/50 "fps" if you do it on the same thread

yep, kinda sad, but i'd assume ddnet does much worse even

Not many people are playing with 1000ms

is it the same in terms of visuals

zhn

what if all parts of tee will render batched?

Am i a fortune teller xD

10:38

it's first of all not easy

10:39

skins can have hd variants so texture arrays are not really usable

Jupstar ✪

Am i a fortune teller xD

you are

on high end gpus u can probably bind like 200 textures at once

Jupstar ✪

yep, kinda sad, but i'd assume ddnet does much worse even

The teero tater client has identical physics at 3x+ the speed of the github one but I don't want to publish these changes because it only helps botters lol

in one shader

Tater

The teero tater client has identical physics at 3x+ the speed of the github one but I don't want to publish these changes because it only helps botters lol

yep, but i guess yours is mostly optimized for single player

can you dynamically select one?

e.g. a pickup also goes through all characters in cpp version

Patiga

can you dynamically select one?

afaik there are at least extensions for that

oh neat

But now I'm leaking

10:41

xd

Original message was deleted or could not be loaded.

i see

Tater

But now I'm leaking

we can switch to a different chat

There's nothing too ground breaking I just look at the profiler and optimize the expensive parts over and over

yeah i basically did the same

10:43

but i cannot fully trust the flamegraph i use XD it 100% wasnt so fast a month ago

10:43

and i only rewrote other stuff than physics

Quad tree or spatial hashing would eliminate n^2 but idk if you can have collision order perfect consistency with the slow version (edited)

Tater

Quad tree or spatial hashing would eliminate n^2 but idk if you can have collision order perfect consistency with the slow version (edited)

i mean collision order in ddnet is just the client id sorted or not?

No

it always does for i < MAX_CLIENTS ..

There's 2 currently

10:45

There's client ID and week strong order

10:45

Both affect physics

yeah but week strong still exists in my version too

10:46

i only optimized the fetching of which ids are interesting

idk

i mean it scales better than lineary now. 64 tees are more efficient to calculate than 1 tee

10:49

so it must work ig xd

10:49

(if i just divide the ticks/s with 64)

You did something special for tee checking?

Jupstar ✪

i mean it scales better than lineary now. 64 tees are more efficient to calculate than 1 tee

Wait better than linear???

10:50

That's impossible lol

Tater

You did something special for tee checking?

yeah, i created a world field where every tile contains the ids of the current tees inside the tile. as long as you don't stack redicolously, like 128 tees in 1 tile, this removes the O(n²) and is O(n)

Tater

Wait better than linear???

i mean there is other overhead than physics in the whole tick

10:51

so maybe it's possible

Jupstar ✪

yeah, i created a world field where every tile contains the ids of the current tees inside the tile. as long as you don't stack redicolously, like 128 tees in 1 tile, this removes the O(n²) and is O(n)

O(n) is linear

Tater

O(n) is linear

yes

10:51

i mean the actual results xD

10:52

like with 1 tee i have 350k ticks with 64 i have 40k ticks

Well that's C + O(n)

350000/64 = 5468.75

Jupstar ✪

yeah, i created a world field where every tile contains the ids of the current tees inside the tile. as long as you don't stack redicolously, like 128 tees in 1 tile, this removes the O(n²) and is O(n)

So what method do you use for moving tees

Tater

So what method do you use for moving tees

you mean move_box?

10:54

or inside the field?

No for tee<->tee

10:54

Like if a tee moves across multiple tiles in 1 tick, you have to check all of them

to sum it up shortly: i get a list of all ids of neighbouring tiles of the current tees position

This is just spatial hashing tho right?

kinda

How it is not

i dunno, i dont have an exact defintion, but yeah it looks similar

10:58

i guess since tees are always a circle you don't need them to be in mutliple tiles at once

Rly?

i'd say so xd

I guess TW has no unskippable interactions with entities right?

11:00

So you skip needing to do intersection over buckets to find fast collisions (edited)

Jupstar ✪

i guess since tees are always a circle you don't need them to be in mutliple tiles at once

I think you're just assuming a maximum object size with this

11:04

You can use a bigger radius for all collisions to avoid figuring out how many buckets they should be in

Tater

Wait better than linear???

It's asymptotically linear but for a normal number of tees you're not reaching that asymptote

Better than linear observed performance is possible with an apparently linear algorithm because of many reasons. The algorithms are usually not linear over their whole domain but only after a certain limit. The compiler can aggressively vectorize giving similar performance for a couple n in a row

11:05

Many others

risu

It's asymptotically linear but for a normal number of tees you're not reaching that asymptote

Scaling doesn't factor the constant cost in the calculation so you ignore the overhead

11:06

Wait! (edited)

11:06

@Jupstar ✪ how do you have laser walls and those things

11:06

They will break your 1 bucket per item model

Tater

@Jupstar ✪ how do you have laser walls and those things

what is that? i implemented vanilla

Tater

@Jupstar ✪ how do you have laser walls and those things

but anyway, most entities only check for itself<->character collision or not?

11:07

i dont know any entity that does entity<->entity collision

Its still n*m (edited)

11:07

For map lasers

but why

how else?

you have n lasers, but only check for the characters in range

Jupstar ✪

i dont know any entity that does entity<->entity collision

There are currently none. But I guess you do preclude them from existing in future mods too

which is O(1)

Every tee has to check every laser once per tick

Tater

Every tee has to check every laser once per tick

but doesnt ever laser check for all characters in range instead?

11:09

or wait

11:09

is that part of the movement?

11:09

of a char

It is part of movement

i see

Currently lasers aren't checked like that iirc. They insert imaginary stoppers

11:09

It's not a real entity<->entity collision

well we'll find a solution for them if the time comes xD

11:10

but yeah they sound like they always consume a tile, similar to any tiles

11:10

ok they can be diagonal right

11:10

that ofc sucks

I mean fully correct spatial hashing is a valid solution

but in worst case you could assume their length is treated as the diameter?

Its not that hard

Jupstar ✪

but in worst case you could assume their length is treated as the diameter?

Yes

Tater

I mean fully correct spatial hashing is a valid solution

sure, could try that some day

But 1 tile size is small tbh

for now i just wanted 2000 tees on my screen xd

Quad tree is like a fancier version you get to have small and big buckets, at more cost. But it's probably not worth it at all for just the lasers

11:14

They can just update a few dozens buckets per tick

I've stared at this problem for a while now. Quadtrees are not great for this usecase. Updating them is too annoying

I mean Jupstar has apparently already proved you don't need it

11:17

Spatial hashing is good enough

I mean teeworlds has also proven that you can get away with O(n^2) for a loooong time

well with n=<64 you can do almost anything

11:18

if grenades would collide with each other this wouldn't be viable lol

Tater

They can just update a few dozens buckets per tick

Shouldn't you instead only be updating the smallest node of the tree that contains the laser? I guess you'll get less fine grained collisions that way so it's a trade-off between a more complex update and more expensive collisions

risu

Shouldn't you instead only be updating the smallest node of the tree that contains the laser? I guess you'll get less fine grained collisions that way so it's a trade-off between a more complex update and more expensive collisions

There is no tree

I'm just thinking that in realistic play scenarios the more expensive collisions won't be expensive at all

Jupstar ✪

btw with 128 tees that do random inputs on ctf1 i get an avg of:

28601 t/s
29154 t/s

with the non O(n²) approach Which surprises me, this is like 2x more than i had a month ago lol. Time to git bisect xd But defs worth to optimize it ig

So to correct my intial wrong benchmark:

6442 t/s
6678 t/s

Which is still quite good if we assume the 1p vs 128p bencmark

11:19

But still much less

Quadtrees and spatial hashing are broad passes of collision. It's to cull things that you can't possibly collide with. So they wouldn't change the "resolution" of it

risu

Shouldn't you instead only be updating the smallest node of the tree that contains the laser? I guess you'll get less fine grained collisions that way so it's a trade-off between a more complex update and more expensive collisions

Replying to this "fine grained" comment

11:21

At the end of the day you just have to do the box/sphere/capsule/whatever collision check

Learath2

Replying to this "fine grained" comment

I mean, the tree consists of a bunch of different sized quads, so depending on which size you're comparing against you'll get information that's of different levels of usefulness

One thing is for certain. if all 128p in ddnet would be active at the same time and collide with each other, we probably have super bad performance xD

point being that lasers would naturally be placed in much larger quads

so PSSST

I feel like calculating the collisions of 128 players should be no problem at all, but calculating the result of those collisions might be depending on how it's done

risu

I feel like calculating the collisions of 128 players should be no problem at all, but calculating the result of those collisions might be depending on how it's done

but you calculate it a few houndred times on the client

If your tree has quads that are very small your broadpass eliminates a lot, but your updates get expensive. If they are very large it degenerates to current ddnet situation

100*128 = ?

So there is a balance there

Jupstar ✪

So to correct my intial wrong benchmark:

6442 t/s
6678 t/s

Which is still quite good if we assume the 1p vs 128p bencmark

This is still actually quite good I didn't do the math on your previous one but it would have been insane lol

Tater

This is still actually quite good I didn't do the math on your previous one but it would have been insane lol

yeah i added spectator slots recently and didnt update the benchmark lol xD

11:24

pros at work

Jupstar ✪

One thing is for certain. if all 128p in ddnet would be active at the same time and collide with each other, we probably have super bad performance xD

I think 128 still doesn't completely break. I remember someone testing it with just tees dumped into the start of a map

Jupstar ✪

yeah i added spectator slots recently and didnt update the benchmark lol xD

How fast is your single tee

Jupstar ✪

but you calculate it a few houndred times on the client

that's still on the order of tens of thousands, should be doable in less than a millisecond if you do it right, right?

256 is where it starts breaking down and 512 is barely still working iirc

risu

that's still on the order of tens of thousands, should be doable in less than a millisecond if you do it right, right?

Wym by do it right? I'm guessing Jupstar is more talking about the current situation. There are obv ways to do it properly and handle far more

I'm just not exactly seeing the reason why we can't do it properly, so I'm assuming there's something fundamentally difficult about keeping identical behaviour

Tater

How fast is your single tee

around 350k. And the flamegraph shows that most is spend in the pickups. so since it's vanilla and ctf1 you can do your own math. If i disable that i get around 500k

ok im pretty impressed by your "super linear" performance

11:29

should I be using multiple tees in the same gameworlds to get more ticks/s?

11:29

lol

lmao

11:30

i guess it's fair to say that with few tees, other factors like player ticks etc. also have a small impact

If your previous benchmark was real it would be 3.5 million tee ticks/s lol

I think we can pump those numbers up

With multi threading yes

TPS: tees per second

Multi threaded collision is a thing as well if we want 10,000s of tees

risu

I'm just not exactly seeing the reason why we can't do it properly, so I'm assuming there's something fundamentally difficult about keeping identical behaviour

The reason is that the ROI is low on doing it properly. 64p works just fine. 128p also works pretty acceptably which is our next stop. So migrating to a full spatial hashing or quadtree solution would just be a marginal performance improvement on the server side

water simulation tee edition

Jupstar ✪

water simulation tee edition

YES

Tater

Multi threaded collision is a thing as well if we want 10,000s of tees

You can't really do this well deterministically, the only option seems to be maintaining islands of tees that you simulate separately, but you need to be careful with moving tees between them as it will require locking and blocking if not done properly

Learath2

The reason is that the ROI is low on doing it properly. 64p works just fine. 128p also works pretty acceptably which is our next stop. So migrating to a full spatial hashing or quadtree solution would just be a marginal performance improvement on the server side

it might really already be needed now. if someone has prediction_margin of like 200 and zooms out, good bye xd

What would happen if you pumped high pressure tee into the spawn point of Multeasymap? Would some of it eventually finish?

Jupstar ✪

it might really already be needed now. if someone has prediction_margin of like 200 and zooms out, good bye xd

Yeah the client running the physics aswell is a bit of an issue. I can only advise them to not do that

risu

What would happen if you pumped high pressure tee into the spawn point of Multeasymap? Would some of it eventually finish?

this is a funny experiment ngl xd

Learath2

Yeah the client running the physics aswell is a bit of an issue. I can only advise them to not do that

Its a good model tho. The client should run physics as many AAA games do

If anyone wants to do it properly it'll definitely be accepted as a PR. The hardest part is reviewing it since we don't currently have the capability of replaying teehistorian files to check nothing broke

I think the client could potentially cull tees that are far away for prediction if it becomes an issue (edited)

Tater

Its a good model tho. The client should run physics as many AAA games do

I meant as in not set prediction margin so high then zoom all the way out

11:38

Running the physics on the client aswell is sane and the correct(tm) thing to do

Jupstar ✪

this is a funny experiment ngl xd

Now the real question: How high should the flow rate be for the pumped tee fluid to achieve a finish faster than a 63 player server of average Multeasy players?

1

risu

Now the real question: How high should the flow rate be for the pumped tee fluid to achieve a finish faster than a 63 player server of average Multeasy players?

Now that is one demo I would love to see

Learath2

I meant as in not set prediction margin so high then zoom all the way out

Prediction margin is capped at 300 now so the real issue is the 4000ms ping cap

Keep pumping tees at the start. See how fast you need to spawn tees to get an average finish

i found a new elevator. The tee popping one

risu

Now the real question: How high should the flow rate be for the pumped tee fluid to achieve a finish faster than a 63 player server of average Multeasy players?

I'm not sure any number is faster with current physics, the tees don't push each other than fast

ok now make each tee in the fluid hammer in a random direction once every 10 seconds

11:41

oh wait the mazes would immediately stop the fluid on multeasy

Would be better on aip gores

yeahhh

11:42

needs to be a simple map without no collision or large areas of tele backwards

11:43

though tele backwards with ever increasing tee pressure could be funny to see

morning

hai

gm

<ChillerDragon> @Avolicious will there be a irc bridge for kog?

12:17

<ChillerDragon> [client] got disconnect from server. reason: You have been banned (IP banned! Open a ticket on https://discord.kog.tw/)

12:17

<ChillerDragon> 88.198.96.203

ws-client

<ChillerDragon> @Avolicious will there be a irc bridge for kog?

Mhmmm I guess not

12:21

You cant use Hetzner ips to join our servers

<ChillerDragon> where can i apply for a whitelist?

12:25

<ChillerDragon> @Avolicious its my ip it will not change and i host my vpn there

If we whitelisted every person having a static hosting IP, our poor admins would go monkey brain creating exceptions

12:26

I assume it's something similar for kog

I've added the IP to the allowlist for now, needs now a few minutes to populate across the system

<ChillerDragon> If you dont have the staff to create exceptions you should disable ip blocking otherwise users cant use vpns

meloƞ

If we whitelisted every person having a static hosting IP, our poor admins would go monkey brain creating exceptions

Thats true tho

1

ws-client

<ChillerDragon> If you dont have the staff to create exceptions you should disable ip blocking otherwise users cant use vpns

We dont want people with vpn

12:27

thats the whole purpose

<ChillerDragon> @Avolicious so nice of you thanks

12:28

<ChillerDragon> i assume you dont want vpns so that you can identify users based on their ip you can still identify me with the ip pointing to zillyhuhn.com

I dont care about any identification from our side ^^

<ChillerDragon> what do you care about then?

Banning vpns to make it harder to avoid bans

<ChillerDragon> yea avoiding bans is also what i mean by identifying

12:30

<ChillerDragon> if you ban my ip i can not avoid the ban

12:30

<ChillerDragon> but now i can secure my confidental kog credentials in public wifis

I dont ban per ip

<ChillerDragon> based on what do you ban?

The whole hetzner AS is banned

12:31

all their ip ranges

(because there are no ddnet players playing from inside a hetzner datacenter)

12:31

<ChillerDragon> @Learath2 there should be

12:31

<ChillerDragon> selfhosting vps is a thing

In general it's 99% of the time a VPN

12:32

For the 1% of the time where the players are selfhosting a vpn, we have a whitelist

<ChillerDragon> yes and vpns are good for the security and privacy of the player and should be allowed by the server

12:33

<ChillerDragon> @Learath2 how do i get on the whitelist?

Clearly the real solution is accounts

risu

Clearly the real solution is accounts

Its not

12:34

Accounts can be created very easily like you can use a vpn

12:34

Its just an additional layer to bypass

<ChillerDragon> paid accounts it is then

clearly we just need players to supply ID :D :D :D

12:35

$1 for an account and only one account per card

This command was not found.

ws-client

<ChillerDragon> @Learath2 how do i get on the whitelist?

You open a ticket

Chiller, please never talk about privacy if you want paid accounts

<ChillerDragon> @Avolicious pay in cash is of course an option

12:35

<ChillerDragon> @Learath2 how?

But this is also a privacy issue if you pay in cash. I can identify you then

In this day and age I think it's about time we start introducing real id solutions for some things

<ChillerDragon> @Avolicious how o.O

Avolicious

But this is also a privacy issue if you pay in cash. I can identify you then

He can mail you cash with no return address

Learath2

In this day and age I think it's about time we start introducing real id solutions for some things

ok but regulating things properly makes them less fun :(

That's how mullvad took cash payments for a while

Learath2

He can mail you cash with no return address

True, but he is on the postal camera then

12:36

Identified by Deutsche Post

risu

ok but regulating things properly makes them less fun :(

We had enough fun, people calling other people racial slurs is so 2009

Avolicious

Identified by Deutsche Post

He can wear a mask and drop it off into a postal box outside

<ChillerDragon> but you cant see the footage @Avolicious and even if you can see it you cant identify me

ws-client

<ChillerDragon> but you cant see the footage @Avolicious and even if you can see it you cant identify me

I already did & talked with you about that xD

12:37

I just dont wanna expose anything

<ChillerDragon> @Learath2 imagine going to the postal office once a week with a ski mask to buy new accounts for botting on gores

If there's a realistic way to universally deanonymize communications on the internet then everyone and their neighbour's dog is going to start requiring it to use their services...

idk if germany still has them actually, but in italy we still do have some postboxes around

But we talked some days back

12:37

about how I did

<ChillerDragon> @Avolicious what? rly

ws-client

<ChillerDragon> @Avolicious what? rly

Yes

<ChillerDragon> i dont recall at all

12:38

<ChillerDragon> are you sure you talked about it with me?

ws-client

<ChillerDragon> are you sure you talked about it with me?

Yes

<ChillerDragon> you can identify people based on a picture?

I will tell you ingame in whisper

risu

If there's a realistic way to universally deanonymize communications on the internet then everyone and their neighbour's dog is going to start requiring it to use their services...

I don't want to deanonymize in all cases. Usually all you need is a token that this is a real person, sadly no government really provides this service

<ChillerDragon> @Avolicious ok join me in game

12:39

<ChillerDragon> 88.198.96.203:8303 libre server with vpns allowed!

A token that allows you to prove that you are a single human that has never registered on my service before

12:40

one that I can submit to the govt, and the govt can verify that it is a single person without knowing who that single person is

Zero Knowledge Proof

I guess it would work if the individual had to generate a new token for every service, so they can't be tracked across services

I should sit down and figure out how one might structure such a service actually. 0 knowledge proof of being a human person

risu

I guess it would work if the individual had to generate a new token for every service, so they can't be tracked across services

The only issue is how such a token could be verified to be legitimate without exposing to the govt who that person is, so they can't track it either

Learath2

I should sit down and figure out how one might structure such a service actually. 0 knowledge proof of being a human person

Before doing that, please add this finish db so we can show the flag next to the map if a player finished

Avolicious

Before doing that, please add this finish db so we can show the flag next to the map if a player finished

I quit my job, so I'll have more time very soon

You dont need to quit your job

Learath2

The only issue is how such a token could be verified to be legitimate without exposing to the govt who that person is, so they can't track it either

oh yeah I'd be devastated if my government knew I played DDNet oh wait I connect to your servers all the time so they know anyways I guess this is how that relates to the VPN issue (edited)

Just 3 http endpoints

12:43

Add, delete, update

Avolicious

Just 3 http endpoints

That part would take like 2 hours, that's not the issue. The internal part needs some thinking

12:46

The server needs to expose an internal interface that allows our python info script to query it and that python info script needs modifying too

12:46

Anyway, not too difficult. I just did not have the time

Who can create a script for fifo?

<ChillerDragon> @Avolicious could you add this to kog o.O https://github.com/ddnet/ddnet/pull/8866

Translate greensward skin in sixup server bridge by ChillerDragon ·...

greensward is not a official teeworlds skin. But it comes pre installed in 4 custom 0.7 clients so the translation will display the correct skin for users of one of those clients: gamer ddnet F-Cl...

1

chillerdragon specific feature

ws-client

<ChillerDragon> @Avolicious could you add this to kog o.O https://github.com/ddnet/ddnet/pull/8866

as a fellow greensward enjoyer i love this, greensward should just be a default skin at this point owo

1

as a passionate greensward hater, it should not

wow greensward is a great skin I didn't even know about that one

<ChillerDragon> Well I did make an attempt to make it an official skin. And that resulted in the skin swardy being added...

5

greensward is just chiller skin at this point

https://github.com/teeworlds/teeworlds/pull/2152 feelsbadman

Add greensward skin by Whis by ChillerDragon · Pull Request #2152 ·...

Using this skin since 6 years. And can highly recommend it. Created by Whis and part of the DDNet 0.6 client. Whis gave me full permission to contribute this skin to the official client.

Jupstar ✪

i found a new elevator. The tee popping one

https://discord.com/channels/252358080522747904/295908390956433410/1006185945969725601 greenthing

Swarfey

https://discord.com/channels/252358080522747904/295908390956433410/1006185945969725601 greenthing

how u doing

18:37

My chat input clears itself while writing, restarting the client fixes it but its annoying

18:40

Anyone else experience this?

zaer1n

My chat input clears itself while writing, restarting the client fixes it but its annoying

Are you writing in a language that requires IME support? Like Chinese, Japanese, Korean, or maybe something with an Arabic script?

Learath2

Are you writing in a language that requires IME support? Like Chinese, Japanese, Korean, or maybe something with an Arabic script?

Turkish characters at most but I remember it happening while only writing english too (edited)

Hm, Turkish characters shouldn't cause this anyway. What do you mean by "clears itself"?

Learath2

Hm, Turkish characters shouldn't cause this anyway. What do you mean by "clears itself"?

for example while writing "hello world", the input gets cleared after I write "hel"

18:44

sometimes allows more characters then clears

18:44

sometimes only 1

does the chat prompt stay open or does it go away?

Learath2

does the chat prompt stay open or does it go away?

it stays open (edited)

Can you type more or is it stuck?

I can keep typing but the above problem keeps repeating

One more question, is the clearing instant or does it delete characters one by one?

instant

18:46

I couldnt record it sorry

Hm, @Robyt3 I think you touched the text input last, does this sound like anything you can think of?

Doesn't sound familiar. @zaer1n Does it work with any older version from https://ddnet.org/downloads ?

Robyt3

Doesn't sound familiar. @zaer1n Does it work with any older version from https://ddnet.org/downloads ?

Didnt try older clients

18:50

happens since 18.6 or 18.7 I cant really remember

18:50

I cant reproduce it too

18:50

just happens randomly

@Robyt3 @Learath2 It happened again, can I show it in screen share? found the trigger too (edited)

zaer1n

@Robyt3 @Learath2 It happened again, can I show it in screen share? found the trigger too (edited)

Please record a video if you can

And what's the trigger? If it happens consistently you could test with older versions to find the first version where it's broken

Robyt3

And what's the trigger? If it happens consistently you could test with older versions to find the first version where it's broken

its in the video, my shitty internet doesnt upload it

Robyt3

Please record a video if you can

zaer1n

Click to see attachment 🖼️

Looks like a bug with your snipping tool

is it tater client? maybe tater might have a clue

Learath2

is it tater client? maybe tater might have a clue

its tclient in the video

19:38

Tested with ddnet 18.7 still happens

19:38

just noticed other text inputs are affected too

zaer1n

Click to see attachment 🖼️

@Tater

zaer1n

Tested with ddnet 18.7 still happens

No it happens with normal ddnet client too ^

Did it work with any previous version? Since it happens after using the clipping tool I'll blame the clipping tool for grabbing the focus incorrectly

2

20:00

Robyt3

Did it work with any previous version? Since it happens after using the clipping tool I'll blame the clipping tool for grabbing the focus incorrectly

@meloƞ how do I try older versions

zaer1n

@meloƞ how do I try older versions

https://ddnet.org/downloads/

laughs in nixos roby kek

{ config, pkgs, ... }: {
  nixpkgs.overlays = [
    (self: super: {
      ddnet = super.ddnet.overrideAttrs (oldAttrs: {
        version = "16.0";
        src = pkgs.fetchFromGitHub {
          owner = "DDNet";
          repo = "ddnet";
          rev = "16.0";
          sha256 = "";

build once, get shasum, (or prefetch it), insert it, and run

hi, how i can set direction of walk tee? i’m trying use controls->minputdata->direction, but he not walking (Controls()->m_aInputData[g-Config.m_ClDummy].m_Direction = 1;) (edited)

20:31

20:32

he seems to be trying to walk, but he doesn't

pls dont alter your input on our official servers! - this will get you banned pretty quickly.

20:33

in general questions towards automation or input modification are usually not answered here.

meloƞ

in general questions towards automation or input modification are usually not answered here.

*not allowed

this is don’t official server

20:34

Don't Client

Teero

*not allowed

my second answer was ackhtuhally targetted towards his question, not whether or not we even allow that

(edited)

meloƞ

in general questions towards automation or input modification are usually not answered here.

oh, okay

can i ask a quick question? when will there be: levels, leveled up by completeing maps. harder maps are locked for higher levels payed skins, you only own the basic skin by default, you can buy skins with real money gatcha system: the good skins are locked behind a gatcha system, where you open loot boxes and have to open 100 to get the skin you want

@Jupstar ✪ ^ NOTED

Bors Matyas

can i ask a quick question? when will there be: levels, leveled up by completeing maps. harder maps are locked for higher levels payed skins, you only own the basic skin by default, you can buy skins with real money gatcha system: the good skins are locked behind a gatcha system, where you open loot boxes and have to open 100 to get the skin you want

finally we can introduce gambling to children (edited)

level up things is currently not planned on ddnet. you could make a #town-hall thread if you are invested in it. payed skins are also not planned. and a gacha system is a bad idea in any way.

1

Teero

level up things is currently not planned on ddnet. you could make a #town-hall thread if you are invested in it. payed skins are also not planned. and a gacha system is a bad idea in any way.

no no, we add lootboxes and hats and DDNet will be the next hot thing in no time

20:52

(I still kinda think that hats could be fun)

risu

no no, we add lootboxes and hats and DDNet will be the next hot thing in no time

yea yea exactly xD then you only need to have an infrastructure that supports 200000 tees

20:52

which we don't have

ok but imagine the infinite money

risu

(I still kinda think that hats could be fun)

hats would be cool. decorations in any form would be

Just set up more servers :D :D :D :D

risu

ok but imagine the infinite money

okay i see

risu

Just set up more servers :D :D :D :D

imagine only 50 people of those 200k would be perma ddosing :D

Teero

hats would be cool. decorations in any form would be

Add hammer skins and make the existing weapon packs only apply to default? I guess that would kinda take away the point of weapon packs... (edited)

20:54

I mean, it would half do that and half improve the situation

20:54

for people who want a specific set of weapons because they find it nice to play with it'd be bad

20:55

but for people who use weapon skins because it makes the game look more interesting it'd be a big positive

20:56

I could see a small set of people getting mad over weapon skins

Voxel

Don't Client

do not the client!

@risu hehe go try implementing all of that with backwards compatability in mind. (or go argue with deen and heinrich)

20:57

backwards compatability is evil. If we would have a version system like minecraft for example so many issues would go away and many new features could be added

1

Is it hard to add new messages that get ignored by older versions?

risu

Is it hard to add new messages that get ignored by older versions?

no thats not the problem

20:59

also hats and personalized stuff isn't really possible until we have accounts

Teero

backwards compatability is evil. If we would have a version system like minecraft for example so many issues would go away and many new features could be added

its why im eating up whatever juppey's cooking

meloƞ

@Jupstar ✪ ^ NOTED

yeah was my no1 prio, all implemented, my paypal acc is set as target ofc

Jupstar ✪

yeah was my no1 prio, all implemented, my paypal acc is set as target ofc

No that's fair, sanitize your stuff tho, I now have your name address phone number email social security number the name of your first pet and how much you make annually

And the name of my wife too?

21:19

I'd like to know it

tbh payed skins feels a bit sad but if it means it goes to ddnet servers then id be supportive of it

Jupstar ✪

And the name of my wife too?

It differs depending on how much you work out in a month

Robyt3

[ddnet/ddnet] Pull request opened: #9384 Fix server handling when vote creator leaves

Reset the vote creator client ID CGameContext::m_VoteCreator to -1 when the player that started the vote leaves instead of keeping the invalid client ID around. Previously, this was causing the server to crash due an assertion in the IServer::GetAuthedState function when a ban vote was aborted due to rcon authentication change being handled in the CGameContext::OnSetAuthed function after the vote creator has already left the server. It was possible to cause this situation as the targe...

I was going to stay quiet but since it's been misspelled three times in a row by three different people I have to speak up. It's paid, not payed.

meloƞ

No that's fair, sanitize your stuff tho, I now have your name address phone number email social security number the name of your first pet and how much you make annually

i make ~7k

1

23:36

(points per year in ddnet)