DDraceNetwork - developer

DDraceNetwork

Development / developer

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

Between 2022-08-01 00:00:00Z and 2022-08-02 00:00:00Z

[ddnet/ddnet] Pull request opened: #5688 Added menu entry for cl_showrecord

Before: !image After: !image

Checklist

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

[ddnet/ddnet] Issue opened: #5689 Game freezes with Intel® Xe Graphics on Linux

I'm currently trying to play DDNet on an Intel XPS 13, but it randomly gives mini freezes that end up with gameplay. I believe it is a problem with the DDNet in particular with the intel integrated card as the hardware is good and new. I play Bloons TD 6, League of Legends, Osu! and CS without those freezes.

CPU: 11th Gen Intel® Core™ i7-1185G7
Graphics Card: Mesa Intel® Xe Graphics (TGL GT2)
Version: 16.2.2 (Flatpak version)
System: Fedora 36

**I'm getting those freezes u...

[ddnet-maps:master] 1 new commit

57671b2 M Cuboid, M Geneque, M Houll, A Pyramid-freeze-race - ddnet-maps

[ddnet/ddnet] Pull request opened: #5690 Replace mem family function with typesafe functions (fixes #5228)

This adds more type safety into the code. Generated code should be similar with no performance impacts. Fixes https://github.com/ddnet/ddnet/issues/5228. I kept mem_zero and mem_copy on default language type.

Checklist

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

i love reading code that i know nothing about! (not even joking, im starting to dive deeper into these prs and stuff)

brownbear

2

[ddnet/ddnet] Issue opened: #5691 Integration tests doesn't work on windows

Probably because fifo is not implemented on windows.

too bad i don't do anything "useful" from the user point of view

07:33

i'd like to implement lazy loading when i have enough free time

i mean i get that, but thats the charm of it, you get to study what actually goes on under the hood and what precautions are needed

07:45

even if im unaware of any of it it's really interesting now

[ddnet/ddnet] Issue closed: #5232 Run Integration Tests on Windows

yikes i burned another SSD :( does anyone of you ever made linux chill the writes to the ssd? So it lifes longer?

pepedead

10:22

typical chillerdragon dosing his ssd (edited)

10:22

xd

Huh, all my SSDs are still alive. I have the problem of having some 128 GB SSDs in each system now and they are taking more physical space than the disk space is worth

11:10

ChillerDragon: What is writing to them? I once did it on OpenBSD, because on CF card writing is really bad, no wear-levelling. So I mounted the file system read-only and put writable stuff into ramdisk

11:10

worked well, just remounted with write for software upgrades

11:14

Maybe monitor what is writing to disk so much using iotop

11:16

My Samsung SSD SM951-NVMe still runs fine, has 75 TBW expected life, 92 TBW actual, no problems yet

11:17

(after 6 years)

11:20

The one where I build DDNet nightly (Crucial BX100) runs fine after 7 years too, 23/72 TBW currently

OpenBSD is damn fucking awesome

@deen its just the ppl in the hardware store tell me every year i show up with a new dead SSD that linux is to blame since it writes too much to the disk and then they die at some point cuz ssd short lifespan

12:05

i couldnt find much online saying run this command to not break your ssd :D

12:05

only this trimming thing

I’m not aware of linux doing more writes than windows or macOS. I’d say the hardware store people just don’t know what they are talking about

How about u just don't tell them what u using xd

What ssd u get

12:52

Maybe get a higher qual

i can imagine swap, /tmp and logs being a lot, but monitoring that for a few days and also how many tbw you get in that time should help

Yea I really should check my writes from time to time. The disk that died now is a SanDisk. Not sure how dead it is. Dead enough to freeze my os and now I can’t even mount it anymore

13:38

BbbbutI USE ARCH BTW (@Jupstar ✪)

How about u just don't tell them what u using xd

13:39

That guy knew dd, testdisk and ext4 I kinda trust him :D but yea maybe he just talked crap

https://www.cvedetails.com/cve-details.php?cve_id=CVE-2021-43518 The above vulnerability applies to Teeworlds. Does this apply to DDNet somehow?

CVE-2021-43518 : Teeworlds up to and including 0.7.5 is vulnerable ...

CVE-2021-43518 : Teeworlds up to and including 0.7.5 is vulnerable to Buffer Overflow. A map parser does not validate m_Channels value coming from a map file, leading to a buffer overflow. A malicious server may offer a specially crafted map that will overwrite client's stack causing denial of service or code execution.

13:47

I see this patch for FreeBSD Teeworlds package (https://lists.freebsd.org/archives/dev-commits-ports-main/2022-March/016758.html) but I haven't found similar variables like m_Channels

13:49

https://github.com/teeworlds/teeworlds/issues/2981

Stack buffer overflow (write) while loading map in CMapLayers::Load...

The client crashes when an invalid map (14.map.zip) is loaded. Such a map can be delivered to the client by a malicious server. Tested on Ubuntu 20.04 x86_64, Teeworlds version: f35da54b6f2c5f9fd4b...

hi, anyone knows matrices and trigonometry here? I need some help around the render.cpp, I have a few questions about DrawCircle.

wew when reinstalling my os somehow i ended up with some cheap vscode ripoff :D

14:43

the colors are so low in contrast omagawd its the same theme as my laptop but its barley readable

14:44

any good vscode theme recommendations? :)

[quakenet] ChillerDragon

any good vscode theme recommendations? :)

was thinking of using vscode is it any good for c++?

Has any of you ever used Coq to formally verify code you wrote?

14:47

https://coq.inria.fr/

14:47

https://dylanj.xyz/posts/rust-coq-opaque-types/

Formally Verifying Rust's Opaque Types

Introduction The other day I was reading this blog post covering existential types in Rust (also known as impl Trait or opaque types). In that blog post, the author makes the following claim. We’re going to have to take a slight diversion into type theory here because it motivates a result that is perhaps intuitive. The following proposition hol...

Rafael Fontenelle

I see this patch for FreeBSD Teeworlds package (https://lists.freebsd.org/archives/dev-commits-ports-main/2022-March/016758.html) but I haven't found similar variables like m_Channels

No, looks like our env points are a bit different

1

Ryozuki

Has any of you ever used Coq to formally verify code you wrote?

a while ago

@bubliman ye its good for c++

14:48

imo easier to properly configure for c++ than vim

14:49

install c++ extension and you done

14:49

i might just need a new screen

ok will give it a shot

what are you using right now?

qtCreator and Kdevelop

never tried kdevelop

14:50

looks like a full blown IDE

14:50

i prefer editors that spin up without any waiting time

I started with c++ yesterday, it was pretty hard to get a project started and compiled ngl (edited)

ye

14:53

thats c++ tho

Hmm

14:53

gcc file.cpp

14:53

Ez

duh

14:54

thats c :p

./a.out

and not a project

concerning my previous question, should i go ask somewhere else?

wait till maffs man heinrich is here

Use vscode

14:54

Ngl

14:54

Better

or jupstar the render engine

Im rusty but just ask

14:55

Dont ask to ask

i use vscode to start a terminal in it and then use emacs within that terminal to spawn a vim instance

https://dontasktoask.com/

Don't ask to ask, just ask

Ryozuki

Dont ask to ask

lmao ok

Image attachment

14:55

kek

14:55

this website is so true

well might be me one sec i need to put my thoughts together

wot

15:01

first i thought my old screen just loses colors

15:01

but screenshots should not be affected by matt screens right?

15:01

Embedded image

15:02

Embedded image

15:02

there is clearly a difference but its the same theme. How can that be :c

Different scaling on the screens based on different dpis

yea scaling

And you can also have different subpixel orientations on screens

but one is visibly brihter

rgb vs bgr for example

15:04

brighter because the text is thicker probably

its thinner

15:04

look at the screens

15:05

im getting bamboozeld here

15:05

maybe its the oss code vs the m$ code :D

those are totally different fonts even

maybe its a font issue

15:05

but i feel like it could be solved only with colors

15:05

it looks a bit greyed out

Your background is grey, so antialiasing will mix gray with the foreground color

So concerning this DrawCircle, my guess at what its doing is that it draws a number (Segments/2) of triangles and rotates them around the centre to draw a circle, the for loop is i += 2, because because each triangle puts two points into the circumference the rotation of the triangle is achieved by cos and sin using a rotation matrix is this correct? i dont really understand what the a1, a2, a3 resemble, what formula is used to calculate them and why would FSegments be doubled? thanks for help here is the code for DrawCircle:

void CRenderTools::DrawCircle(float x, float y, float r, int Segments)
{
    IGraphics::CFreeformItem Array[32];
    int NumItems = 0;
    float FSegments = (float)Segments;
    for(int i = 0; i < Segments; i += 2)
    {
        float a1 = i / FSegments * 2 * pi;
        float a2 = (i + 1) / FSegments * 2 * pi;
        float a3 = (i + 2) / FSegments * 2 * pi;
        float Ca1 = cosf(a1);
        float Ca2 = cosf(a2);
        float Ca3 = cosf(a3);
        float Sa1 = sinf(a1);
        float Sa2 = sinf(a2);
        float Sa3 = sinf(a3);

        Array[NumItems++] = IGraphics::CFreeformItem(
            x, y,
            x + Ca1 * r, y + Sa1 * r,
            x + Ca3 * r, y + Sa3 * r,
            x + Ca2 * r, y + Sa2 * r);
        if(NumItems == 32)
        {
            Graphics()->QuadsDrawFreeform(Array, 32);
            NumItems = 0;
        }
    }
    if(NumItems)
        Graphics()->QuadsDrawFreeform(Array, NumItems);
}

ChillerDragon: You can tweak some font settings including aliasing and dpi scaling in gnome-tweaks-tools, although I don't think that applies to electron apps

@bubliman To me it looks like the circle is being split into 32 segments. a1, a2, a3 are the points on the edge of the circle, while (x, y) is always the center. 2 * pi is commonly used in circle calculations, like area = 2 * pi * r

a1, a2, a3 are the angles of the circle segments in radians, hence they are in the range [0;2*pi[

it seems like it actually draws Segments/2 primitives

Robyt3

a1, a2, a3 are the angles of the circle segments in radians, hence they are in the range [0;2*pi[

so a segment is a triangle?

no, they seem to be quads

I think I found a youtube tutorial to explain the algo

https://www.youtube.com/watch?v=YDCSKlFqpaU Source code is similar to TW emthod too: https://github.com/SonarSystems/OpenGL-Tutorials/blob/master/Drawing%20A%20Circle%20Using%20A%20Triangle%20Fan/main.cpp

OpenGL Tutorial 12 - Drawing A Circle Using A Triangle Fan

OpenGL-Tutorials/main.cpp at master · SonarSystems/OpenGL-Tutorials

Contribute to SonarSystems/OpenGL-Tutorials development by creating an account on GitHub.

with one point in the center and the other three on the circles radius

(with triangle instead of quads)

deen

I think I found a youtube tutorial to explain the algo

https://www.youtube.com/watch?v=YDCSKlFqpaU Source code is similar to TW emthod too: https://github.com/SonarSystems/OpenGL-Tutorials/blob/master/Drawing%20A%20Circle%20Using%20A%20Triangle%20Fan/main.cpp

Well I watched the video before and didn't notice any similarities. I'll have a look at the source. So this circle isn't using any form of a matrix? I thought the array is a matrix... Is it just a one dimensional array ?

the array is just each quad

the array is just for storing the quads because it more efficient than calling the graphics API for each individual quad

1

they are also more like freeform quads, not what u excpect, they just share 1 point in the mid, so basically 2 triangles teeworlds(vanilla) had no triangles support, so i guess someone just did it like this

normally u'd probs use a triangle fan or smth like that

the function and the round rect drawing should be moved to IGraphics since it's just generic

i have a version for the fng scoreboard that also can map texture coordinates in the circle

is it possible to do antialiasing just for circles and round rects?

15:32

MSAA I mean

15:33

on vanilla anything more than 8 MSAA samples make the text blurry

depends

15:35

well i think i explained it somewhere before cant find rn

15:35

but we could do smth different

15:35

blur edges.. if there is a transition to transparent is almost like anti aliasing

15:36

ok not blur, more like a transition xd

15:36

that's defs cheaper than having multiple renderpasses

like a shader for corners?

yeah i guess u could say that

Jupstar ✪

they are also more like freeform quads, not what u excpect, they just share 1 point in the mid, so basically 2 triangles teeworlds(vanilla) had no triangles support, so i guess someone just did it like this

normally u'd probs use a triangle fan or smth like that

well that explains a lot

15:40

anyways, thank you guys @Jupstar ✪ @deen @Robyt3 ! I was clearly lost...

‍

I need some time to process all of this info

LMAO

15:58

these automated mails xd

15:58

Embedded image

15:59

im sure this guy is really curious about code reuse in that teeworlds .demo file :D

What is the difference between void CGraphics_Threaded::LinesEnd() , void CGraphics_Threaded::LinesBegin() (edited)

16:01

and void CGraphics_Threaded::LinesDraw(const CLineItem *pArray, int Num)? (edited)

its basically batching

16:03

begin = start collecting lines end = stop and draw

16:03

and linesdraw adds them to the batch

is it for better performace to batch multiple lines together?

theoretically xd

16:04

i mean we never really draw lots of lines

16:04

i think only debug hud, maybe hook coll line, if it doesnt use quads nowadays

well im about to

with vulkan it probs makes almost no difference

16:06

GPUs are really good in accepting multiple draw calls if no state is changing, e.g. pipelines etc. and vulkan puts all streamed data into one single block of memory

16:06

little less CPU overhead tho with batching

16:06

but best is you dont use the legacy API

16:07

try to build all lines once, never change them if possible

16:07

if u need rotation, add it to a shader instead

Jupstar ✪

if u need rotation, add it to a shader instead

is that what hook col does?

nope it uses legacy API, i dont think he have smth else for lines

if(NumItems == 32)
        {
            Graphics()->QuadsDrawFreeform(Array, 32);
            NumItems = 0;
        }
    }
    if(NumItems)
        Graphics()->QuadsDrawFreeform(Array, NumItems);

how is there an if statement with no condition?

implicit bool conversion

16:15

but u can always write the statement out, its generally easier to read imo

16:16

but if u lazy af i accept that xd

16:18

also theoretically, even if not happening here, NumItems could be negative, which might not be intended, but at this point u probs have some unwanted state anyway ^^

if(NumItems != NULL) is that the readable version then?

null is usually used for ptrs

16:21

also NULL is a c macro and considered legacy

16:21

nullptr is c++11 version

16:22

just != 0

16:22

or > 0 makes more sense (edited)

16:22

readable and correct ^^

ok now i get it thx

16:22

and there are no { } why is that?

if u dont do it u dont open a new scope, but that only works until the next ; which ends the current statement

ok makes sence

bencie

is it shift-select?

+1 shift select in editor crashes game

try nightly

lol since [] is an operator didn't know that was possible

#include <iostream>

int main() {
        int a[] = {10, 20, 30};
        std::cout << 2[a] << std::endl;
}

this is cursed af

ReiTW

lol since [] is an operator didn't know that was possible

#include <iostream>

int main() {
        int a[] = {10, 20, 30};
        std::cout << 2[a] << std::endl;
}

This is because a[b] is just syntax sugar for *(a + b)

yes as it is an operator

17:36

but even 3 + a is kinda sus

how can i disable strong/weak hook on a server like KoG

RoyalMello

how can i disable strong/weak hook on a server like KoG

soon there will be a feature for it, probably next release

Jupstar ✪

soon there will be a feature for it, probably next release

great

[ddnet/ddnet] Pull request opened: #5692 Refactor `GetClientVersion`, remove many `version.h` includes, fix potential use of undefined value

Some refactoring to reduce duplicate code and to remove includes.

Checklist

[X] Tested the change ingame
[ ] Provided screenshots if it is a visual change
[ ] Tested in combination with possibly related configuration options
[ ] Written a unit test (especially base/) or added coverage to integration test
[ ] Considered possible null pointers and out of bounds array indexing
[ ] Changed no physics that affect existing maps
[ ] Tested the change with [ASan+UBSan or ...

@ChillerDragon somewhat crisp circles ingame

Image attachment

Voxel

i imagine if this was in the actual client it'd look something like these

So what is the verdict on this, could something like this be in the official ddnet client?

18:51

Should i ping ppl over this?

[ddnet/ddnet] Issue opened: #5693 Extend test with editor code

Not sure how we could do that. We could try to open a map, apply some action (resize, select & rotate) and save it and check the output checksum.

[ddnet/ddnet] Issue opened: #5694 Document special parameters passed to `IServer::GetClientVersion`

https://github.com/ddnet/ddnet/pull/5692#discussion_r934930241

Please document that not only 0 to 63 are valid parameters to this function; it'd be best to also explain the special case of SERVER_DEMO_CLIENT.

bubliman

So what is the verdict on this, could something like this be in the official ddnet client?

Is it useful? Seems like we could add it to aim line

@bubliman when can you hit someone? when they touch the circle?

21:36

I guess it can't be made skin-independent?

deen

Is it useful? Seems like we could add it to aim line

yes improves my aled chance by some amount

heinrich5991

@bubliman when can you hit someone? when they touch the circle?

yes thats how it is the hammer range under this method is 42.0f

when the center of their tee touches?

but there are smaller and bigger skins

21:40

how could it work skin-independent?

i personally dont support these types of things (including hook collision)

skin size doesn't change hammer range

makes it feel less like a game and more like a planetary orbit determination program

but when the skin size changes whether the other tee touches your circle or not @Chairn

heinrich5991

how could it work skin-independent?

yeah not sure about that

hence my question ^^

could have a smaller circle rendered on every tee

hmm, doesn't the circle change color just like hookline?

when the circles colide then u are in hammer range

Chairn

hmm, doesn't the circle change color just like hookline?

easily could be done

heinrich5991

@bubliman when can you hit someone? when they touch the circle?

@Chairn that was in response to this question

bubliman

yes thats how it is the hammer range under this method is 42.0f

plus this answer

that's obviously cheating kek

kek

one sec lemme get screenshot

[ddnet:master] 4 new commits

9a01231 Move and extend GetClientVersion to remove many version.h includes - Robyt3 24b3043 Handle CServer::GetClientInfo return to fix use of undefined value - Robyt3 ee339eb Minor refactoring of GetClientInfo and SetClientDDNetVersion - Robyt3 b4914d5 Merge #5692 - bors[bot]

21:43

[ddnet/ddnet] Pull request closed: #5692 Refactor `GetClientVersion`, remove many `version.h` includes, fix potential use of undefined value

Image attachment

Skeith

that's obviously cheating kek

kek

i mean not exactly

21:44

just pointless imo

Nah, it'll save your life on occasion, it's definitely an advantage. The curve shit he posted earlier

oh yeah thats sus

its rendered using a object i made DrawCircleOutline,

not the hammer thing tho

You can jump to abort whatever you're doing

Skeith

You can jump to abort whatever you're doing

we arent talking about that...

Yep, just giving my input, don't post that anymore feelsamazingman

feelsamazingman

well it looks cool

it looks cool but do we want to move to a fully "automated" (not the right term) client?

No lol

21:47

Enough cheats baked into the client as is

looks like extra HUD just like in the military

i definitely dont want it, it just looks neat lol

Chairn

it looks cool but do we want to move to a fully "automated" (not the right term) client?

how is this fully automated? i dont get it

An advantage is an advantage, I don't care how small

bubliman

how is this fully automated? i dont get it

they were talking about the arc prediction thing

please can u just stop with that, i think its cheating too...

bubliman

how is this fully automated? i dont get it

not the right term, but it just does the computation for you rather you think about it

21:52

hookline was primarily made, in my memory, to save tee with hook through corner from far away as it was a real pain in the ass to do without it

sure, isn't aled a pain in the ass too?

I like the idea of showing the hammer range because as a new player is very hard to figure out how far it goes, the hammer is so big but the range is kinda small. But also I think it makes the game easier so idk

if you give people an advantage when starting the game, they'll want to keep it

should i make it playable for u guys, I don't think its a big advantage, if done right it could make the game more accesible imo

21:57

@heinrich5991 not sure how to deal with skin sizes, maybe change color when player enters it?

change color so that it's like hookline

red=nohit, yellow=teeinrange?

I suppose you could actually get hammer range indicator without client modification if you change gun texture to display hammer range then use shoot hammer swap bind

bubliman

red=nohit, yellow=teeinrange?

nah, red is too intense color imo

xd but thats how the hookline is

but hookline is not on permanently

21:59

and red hooklines pisses me off

22:00

especially when other people uses it

idk i would imagine this having a toggle bind

i think i have it disabled, not sure

The design of it is ugly, if a person felt forced to use it, I don't think it'd be a good experience

@heinrich5991 maybe ur interested https://www.jackos.io/rust-kernel/rust-for-linux.html

Rust linux kernel development | JackOS

Programming and OS Tutorials

I liked Voxels design

Skeith

The design of it is ugly, if a person felt forced to use it, I don't think it'd be a good experience

Prototype, could even be animated, just wanted to know if this has potential

Voxel

i imagine if this was in the actual client it'd look something like these

This

Tater

I suppose you could actually get hammer range indicator without client modification if you change gun texture to display hammer range then use shoot hammer swap bind

could just use an entities background as well

Tater

I liked Voxels design

yeah i liked his desing aswell, except it doesnt actually represent the true hammer hitbox

Feels like something from a cheat client, no matter which way you make it, don't really think it belongs

well, I think it only feels like that because it's not part of the ddnet client already

louis

could just use an entities background as well

Yeah but it renders behind everything idk

sometimes new cool features feel like that

22:05

e.g. remember autohammer after freeze? ^^

Ya, but idk. Visual guides are usually what you see in cheat clients lol

Skeith

Ya, but idk. Visual guides are usually what you see in cheat clients lol

Autoedge is what i would think of in a cheat client

@deen are you around and have some minutes to try to diagnose the rust PR?

if it also were to show the direction of aiming, this is what that shape would look like, sharpest point furhest away, then it becomes larger closer to a tee

Image attachment

22:11

such a triangle could be made pretty interesting graphically imo

I think if you wanted to invest time into making the game easier for new players, you should probably make some new novice maps.

22:11

Far more valuable

Skeith

I think if you wanted to invest time into making the game easier for new players, you should probably make some new novice maps.

sadly my maps are deemed unoriginal, not much i can do about that

22:12

and more novice maps doesnt make the game more accesible...

Novice maps should have different rules @ tester

22:13

It makes it more playable

22:13

We only have like 4? super basic novice maps

22:13

We should have 100

heinrich5991

@deen are you around and have some minutes to try to diagnose the rust PR?

sure

deen

sure

okay, for the linux x86-64 build, can you add a VERBOSE=1 to the make command line?

22:16

like make VERBOSE=1

22:16

so we can see what command it tries to execute

hm, the nightly is still building for half an hour

alright, nightly is indeed nightly ^^

22:17

do you still have the CMakeCache.txt for the failed build?

if i were to fully develop this feature into something looking nice and working well, could we do a community vote on whether to add it?

We should at least decide if we'd consider it a cheat before you implement it

22:19

for me it seems fine, maybe others disagree

deen

We should at least decide if we'd consider it a cheat before you implement it

vote on that then?

It's a very minor cheat

it looks similar to hook line to me

And hook line is a cheat, if you're not using ddnet client, you're disadvantaged

Skeith

And hook line is a cheat, if you're not using ddnet client, you're disadvantaged

idk its open source anyone can run it

Skeith

And hook line is a cheat, if you're not using ddnet client, you're disadvantaged

you're not using the commonly accepted definition of cheat

22:21

if you use that, then freeze prediction is cheat

22:21

^^

Anything that ddnet uses is fine, we're talking about if it's a cheat, it is

bubliman

idk its open source anyone can run it

that's not a good argument, if someone made an open source cheat client we still wouldn't accept it

Skeith

Anything that ddnet uses is fine, we're talking about if it's a cheat, it is

okay, so what do you consider a cheat?

Anything that gives even a slight advantage

over what?

deen

that's not a good argument, if someone made an open source cheat client we still wouldn't accept it

true but his argument sounded like its p2w or something

Skeith

Anything that gives even a slight advantage

entities

Vanilla client I guess? But basically no one uses that, you're backing me into a wall for no reason, the question is, is it a cheat?

22:23

Yes.

22:23

If you don't use it, you're not getting information that someone else has

Skeith

Vanilla client I guess? But basically no one uses that, you're backing me into a wall for no reason, the question is, is it a cheat?

I feel like this definition of cheat isn't useful

my argument is that playing ddnet should feel teeish and without too many unnecessary lines or distractions, i dont think this is a cheat but i think its unnecessary

yeah, I'm also wondering if it actually helps you while playing

new players should see the game for itself, not a bunch of lines over the place

it's basically saying "every gameplay feature that the vanilla client doesn't have is what I call cheat"

I never wondered if I can hit someone at this distance

i.e. you're saying "bubliman's feature isn't present in the vanilla client"

22:24

this is true, but doesn't give a lot of information

It doesn't really help people that are experienced, and I think anything around the tee would probably do more harm than good for newer players anyways

22:25

If it was somehow added to the crosshair, maybe it'd be usable

a player using various methods to create an advantage beyond normal gameplay, depends what u think is normal gameplay i guess

louis

new players should see the game for itself, not a bunch of lines over the place

yes thats why it should be optional just like hookline

even hookline is kind of overused I think

I think saying it's a lot like hook collision is basically agreeing that it's a cheat, I think everyone would agree with that

I wanted it just for those annoying spots where you have to hook through corner to get someone out

22:27

because continuously shooting is annoying

Even if it's an almost useless advantage

Skeith

I think saying it's a lot like hook collision is basically agreeing that it's a cheat, I think everyone would agree with that

yes, I think it's like hook collision; it's not present in the vanilla client either, and it helps your gameplay

22:28

maybe we can get away from the word 'cheat' if we disagree on its definition

22:28

we can simply substitute your definition instead

"unfair advantage" ?

Skeith

I think saying it's a lot like hook collision is basically agreeing that it's a cheat, I think everyone would agree with that

well if you want a hardcore category to exist, that could happen i guess raw teeworlds without any help

Chairn

"unfair advantage" ?

"unfair" is not really a neutral word; then we'd have to figure out what is unfair

Chairn

"unfair advantage" ?

a player using various methods to create an advantage beyond normal gameplay

"disagreed advantage" ?

22:31

i like bubliman's definition

22:31

now we have to agreed on "normal gameplay"

well, apparently that's the teeworlds client for Skeith

Chairn

"disagreed advantage" ?

yes we are a community i guess so we agree on whats fair

It's a fair base

so yes, hook collision is like the hammer hit box, it's an advantage over vanilla gameplay

normal gameplay at this time is probably the newest release of DDnet (edited)

22:35

normal gameplay is what the comminity decides it to be?

22:35

idk how u guys decide stuff here, if its only the devs or not, im new to #developer

It doesn't automate anything, it just gives you a visual cue, the word advantage is what should be asked. I think deen more or less meant that when asking if it was a cheat

ah, it's an advantage over not having it

22:36

yea, I guess

I think most people would agree that yes, it does give some minor advantage.

to be honest, hammer range is not the biggest advantage you could get

Skeith

It doesn't automate anything, it just gives you a visual cue, the word advantage is what should be asked. I think deen more or less meant that when asking if it was a cheat

@deen did u mean that?

probably helps for aled though

no, I don't think deen meant that

22:37

I think deen meant whether we consider this as something that should be allowed or not

however, shotgun/laser line with reflection and grenade curve is, in my opinion, too much of an advantage

22:38

the question is: if we allow hammer range, where do we stop?

Grey area, honestly

22:38

Should it be added to ddnet? I really don't think so

Imo freeze bar is a bigger advantage for aled because the difference in opacity let's you know when a tee is in freeze

what?

@bubliman I guess you could try opening an issue at this point

22:39

then the discussion would be a bit more structured

Maybe it could be added while in ctrl shift d as an information type deal, but I don't think it should be an option to help with gameplay

Freeze bar changes opacity when tee is in/out of freeze. Imo it makes aled much easier

Skeith

Maybe it could be added while in ctrl shift d as an information type deal, but I don't think it should be an option to help with gameplay

so as a debugging tool? if its that big of an advatage, ppl could play just with the debugger, ultimately helping with gameplay

22:41

@heinrich5991 will do! I just need to figure out how, shouldn't be too hard

https://github.com/ddnet/ddnet/issues go here

@bubliman https://github.com/ddnet/ddnet/issues/new

click new issue

you'll need a github account though

Chairn

@bubliman https://github.com/ddnet/ddnet/issues/new

this link displays a login form for non-logged in users, I usually link /issues instead

yeah im logged in so its fine

i dunno, im always logged in ^^

Image attachment

22:43

(that's all you see on /new)

Should I write my opinion in the first comment?

bubliman

so as a debugging tool? if its that big of an advatage, ppl could play just with the debugger, ultimately helping with gameplay

Idk, it seems annoying to feel obligated to use it, I guess it's a personal thing for me. But I think most people wouldn't use it for an advantage if it was only displayed in debugger

22:44

so the advantage is less harmful

22:44

it's just information

bubliman

Should I write my opinion in the first comment?

do it like you want to

Kinda silly argument I know

I can see how you don't want to be 'forced' to use it because it's allowed and because you find it ugly

debugger is already useful to display your actual position anyways btw

22:45

So it'd be a similar bit of information alongside it

22:46

I'd argue it makes the most sense to be there

Skeith

debugger is already useful to display your actual position anyways btw

you can now have this information in the HUD without debug mode

I meant your shadow tee

22:47

In real time

I fear even that is possible to show without debug HUD

22:47

but I'd have to look it up

Image attachment

yea that one

Ya, I don't know the command, if there is one

Personally I think the worst thing allowed is the aimhack bind which allows people to autoaim to cardinal directions

funman

Personally I think the worst thing allowed is the aimhack bind which allows people to autoaim to cardinal directions

thats based, though removing it is probably not possible, unless you kept all ranks achieved with it as still valid (edited)

its literally cheating idc if people got ranks by cheating

22:56

but more so for me people have started to use it in ictf

22:56

which is sooooooo dumb

i also think this is cheating, but it's not an intended cheat, more like a feature abuse

how does that even help in ictf

Skeith

I meant your shadow tee

You can get that without debug

23:23

I forgot the command but there is one

ah? neat

[ddnet/ddnet] Issue opened: #5695 Hammer Range Indication and Cheating

Hammer Range Indication and Cheating

1. Should we allow a on screen HUD that displays the hammer range of a player? 2. What constitutes as cheating?

1. HUD hammer range

Pros

Increased accessibility to players
Clear visual indication of what is going to happen
Could make the game less dominated by pro players
Easier to learn how to hammer (throwing, aled, etc.)

Cons

bloat
Unfair advantage, towards players who choose not to use it and Vanilla Clie...

GitHub

Click to see attachment 🖼️

Let's move our discussion here. I'm going to sleep, Good Night! (edited)

good night

louis

how does that even help in ictf

it is instant autoaim to horizantal

23:26

so imagine you are hooking ceiling

23:26

so crosshair placement is off

23:26

instant flick with no possibility of overshoot

Exported 399 message(s)