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 2024-08-04 00:00 and 2024-08-05 00:00

Ah lol, I did a github search for that string but didn't find something, probably typo

00:00

sounded like a bot command at first, I'm not used to the latest client code

‍

https://github.com/ddnet/ddnet/pull/7253 (reply to the q about cl_sub_tick_aiming) (edited)

firing/hooking takes aim when action happened not when tick by JSau...

When moving the mouse fast there can be a discrepancy between where I aimed and where my weapon or hook actually go's. This is because of the fact that the mouse can move after I've shot/ho...

that seems like it should be on by default no?

1

it's not?

<ChillerDragon> assembly is basically as simple as go lang

03:26

<ChillerDragon> https://zillyhuhn.com/cs/.1722741970.png

Embedded image

[ddnet-maps:master] 1 new commit

042fcfd A Grell - ddnet-maps

How do we get player pos in player.cpp?

05:30

is it just a snap?

chillerdragon: did you find out how to do mafs with lea?

morning

No idea how to do maffs w lea. imul works fine

06:09

It’s in the character. So something like GetCharacter()->Core()->m_Pos.x

Replying to @StormA How do we get player pos in player.cpp?

06:10

Untested im on phone

06:10

Are you server side dev now? @StormA done with all bot clients? xaxa

Nah i need a local position for own tee, so i can implement the player trail

06:14

Just working with the graphics

Client side?

06:15

I assumed server side because you didn’t specify and mentioned player.cpp that file is server side only

06:15

So you mean players.cpp ?

06:15

Yea then it’s in the snap. In the m_aClients array

chillerdragon: if your maths can be converted in this form, you can use lea to do it in 1 instruction

Image attachment

06:16

https://github.com/dotnet/runtime/issues/75119 there're examples here xd

Does it make sense to u?

it's used for optimizations

06:17

my dumb compiler just uses imul and lives a happy life

Yes but how is getting the memory address doing multiply xd

if you pass a register, it takes it's value

06:18

if you do lea rax, [r15] it will just move value from r15 to rax

06:18

but if you do lea rax, [label] it will move address of label

Yes but moving an address is not multiplying xd

it can be just a number

06:21

not an address

06:21

i mean in register

06:23

    mov r15, 10
    lea rax, [r15 * 2]

in rax will be value 20

That’s so trol

06:24

Instead of imul you do *

06:24

Xd

06:24

How is * not the same as multiplication

06:25

But ye I get it now

06:25

I also stopped using xor to zero registers

chillerdragon

How is * not the same as multiplication

you can do more operations in 1 instruction

Isn’t the * also an instruction?

idk how that works in cpu land but in assembly you can put it in 1 intruction xd

wow

06:26

Makes it harder to read

06:26

And potentially not faster because nobody knows what happens in cpu land

06:26

There is always one dude commenting it’s slower on modern cpus

if ppl use it for optimizations, i'd expect it to be faster

I assume it was faster in some cases

06:27

But blindly copying that unreadable style seems not so smart

06:28

See the last comment on the dotnet issue you linked

it always bsod my noodle when i see "load effective ADDRESS" instruction and it's used for regular numbers xd

Yes very trol

hi

06:56

oops

06:56

wrong channel

lea is probably always faster than imul

MilkeeyCat

chillerdragon: did you find out how to do mafs with lea?

Doing maths with me?

chillerdragon

I also stopped using xor to zero registers

Y?

<ChillerDragon> lea quiqq maffs

07:58

<ChillerDragon> printing signed integers is amazing -0 pog

(fwiw you can use scaled and offset addressing for operands all over the place, but making optimal choices is very very hard) (edited)

Learath2

Doing maths with me?

did you see the message that I finished structs? xd (edited)

MilkeeyCat

did you see the message that I finished structs? xd (edited)

Not yet

Celebrate

MilkeeyCat

@Learath2 they are not read only anymore!

@Learath2

[ddnet/ddnet] Issue closed: #6975 Emote on focus lost

ws-client

<ChillerDragon> lea quiqq maffs

skrrrrr?

[ddnet/ddnet] Issue closed: #8677 Binds do not work correctly when using composite keys

was something changed with mouse4 / mouse5 binds? when i use a 2 cfg deepfly bind, holding down mouse4 will repeatedly toggle it on and off, but i only want it to toggle once per press

[ddnet/ddnet] Issue opened: #8679 "Name" string can't be translated correctly

In Ukrainian and other Slavic languages, we have two words for "name", "ім'я" for personal names and "назва" for proper nouns (for example, Andrew is "ім'я", and DDNet is "назва"). Because server browser uses the same string as everything else, it is impossible to translate the string "Name" to these languages correctly.

just remembered about this one heh

louis

was something changed with mouse4 / mouse5 binds? when i use a 2 cfg deepfly bind, holding down mouse4 will repeatedly toggle it on and off, but i only want it to toggle once per press

Works as expected for me with simple hookline bind. Does it work with other binds for you?

[ddnet/ddnet] Issue closed: #8679 "Name" string can't be translated correctly

EGYT

just remembered about this one heh

See #8434

https://github.com/ddnet/ddnet/issues/8434

Two meanings of the word "Name" · Issue #8434 · ddnet/ddnet

Yet another problem I found during my remake of Ukrainian translation In Slavic languages, "name" can be translated in two ways: as a first/given name (Andrew, Alex, Anthony etc.) and as ...

whoops, forgot that I remembered (edited)

11:50

still didn't get fixed though :/

if you make 10 more issues it won't be fixed faster

It's made more complicated because it depends on #8405

https://github.com/ddnet/ddnet/issues/8405

Localizable doesn't work with context · Issue #8405 · ddnet/ddnet

Localizable just drops the context, as such it can't work at runtime because it doesn't know its context anymore. Possible solution: Return a struct containing both the string and the conte...

[ddnet/ddnet] Issue opened: #8680 vulkan_debug: Validation Warning: [ UNASSIGNED-BestPractices-PushConstants ] Object 0: handle = 0x7b5b98722550, type = VK_OBJECT_TYPE_COMMAND_BUFFE...

I'm getting a lot of these warnings when running with dbg_gfx:

2024-08-04 13:46:51 I vulkan_debug: Validation Warning: [ UNASSIGNED-BestPractices-PushConstants ] Object 0: handle = 0x7b5b98722550, type = VK_OBJECT_TYPE_COMMAND_BUFFER; | MessageID = 0xfd5d8e3f | vkCmdDrawIndexed():  Pipeline uses push constants with 80 bytes, but byte 0 was never set with vkCmdPushConstants.

gist: https://gist.github.com/def-/94a4fbd8a7ddcf8ff331a0c277fe4e95 @Jupeyy is this expected? Could it ca...

12:00

[ddnet/ddnet] Issue closed: #470 [GUI/SUPER] Key

Robyt3

Works as expected for me with simple hookline bind. Does it work with other binds for you?

hmm.. it might be my mouse just acting up

12:01

if i bind mouse5 to +fire i get really unpredictable behavior

12:02

yeah it is, false report i guess

can i issue f1 cmds to a headless client when starting?

<ChillerDragon> with fifo yes

12:47

<ChillerDragon> or you mean on launch once?

like DDNet --connect localhost (edited)

<ChillerDragon> yea like a regular client it takes args

12:48

<ChillerDragon> DDNet "connect localhost"

ah nice

12:48

thx

ws-client

<ChillerDragon> with fifo yes

is that easy to set up or custom?

<ChillerDragon> @Teero ? is it a reply to something? idk what you mean by custom

ahy ea

12:55

the fifo stuff

<ChillerDragon> DDNet "cl_input_fifo deez.nuts"`

xdd

<ChillerDragon> echo quit >> deez.nuts

yoo nice

12:56

so cool

[ddnet-scripts:master] 1 new commit

b752acc -dnsbl - def-

13:20

[ddnet/ddnet] Issue opened: #8681 Team count wrong

From Discord:

[2:41 PM]xteriche: why 6? !image [2:42 PM]xteriche: if you count there are 8 of us [3:13 PM]RedRay32: does he count on one side only?

[ddnet/ddnet] Pull request opened: #8682 Refactor CRaceHelper, mark CCollision functions as const and add tile getters

See commit messages.

Checklist

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

[ddnet/ddnet] Pull request opened: #8683 Fix invalid team count in scoreboard for split teams

Fixes #8681. !DDNet_rcQue7LR2c

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...

13:46

[ddnet/ddnet] Issue opened: #8684 Server info should be created with CJsonStringWriter

https://github.com/ddnet/ddnet/blob/00d941a3094568535bde921855501a77224162ad/src/engine/server/server.cpp#L2326-L2332 Should use CJsonStringWriter instead of str_format to create the json object. The fixed size 16384 buffer can be replaced with an std::string as well to reduce the arbitrary size limitation. CC #8542 which increases the size to 32768, which would not be necessary then.

[ddnet/ddnet] Pull request closed: #8283 Add support for colored broadcast (0.7 compatible)

rip

rip

rip

14:38

when colored names in server browser though

feelsbadman

1

rip

Nouaa

when colored names in server browser though

fr

14:43

well, i think it might be tough with the name character limit

[ddnet-web:master] 1 new commit

49c9a91 5991 - def-

troll

1

5991 players (edited)

14:57

like heinrich?

14:57

gigachad

Heinrich's name prophesied it

GitHub

Click to see attachment 🖼️

5991 goes brr

GitHub

Click to see attachment 🖼️

justatest

15:07

wait different webhooks?

[ddnet/ddnet] Pull request closed: #8683 Fix invalid team count in scoreboard for split teams

16:01

[ddnet/ddnet] Issue closed: #8681 Team count wrong

Chiller I have a fork with 7 players on it running on a different master that supports an xml-like syntax for colored broadcasts running on 0.8.0. It’s a secret release by oy. Can you please implement that established standard?

troll

3

Nouaa

when colored names in server browser though

please never

1

16:13

that just sounds distracting as fuck

troll

1

Rainbow server names sounds very early 2000s

classic advertising problem

Learath2

Chiller I have a fork with 7 players on it running on a different master that supports an xml-like syntax for colored broadcasts running on 0.8.0. It’s a secret release by oy. Can you please implement that established standard?

justatest

@Learath2

Image attachment

Ryozuki

@Learath2

Me

17:28

Anime_irl

https://www.reddit.com/r/ProgrammerAnimemes/comments/1ejscnp/hello_world/

From the ProgrammerAnimemes community on Reddit: Hello World

Explore this post and more from the ProgrammerAnimemes community

17:29

lol

Image attachment

Learath2

Chiller I have a fork with 7 players on it running on a different master that supports an xml-like syntax for colored broadcasts running on 0.8.0. It’s a secret release by oy. Can you please implement that established standard?

I love how professional ddnet maintainers are.

jxsl13

I love how professional ddnet maintainers are.

Thanks. I try my best

just butthurt

17:51

that's it

Heinrich is the professional one, I suggest you interact with him instead if you don’t like the way I talk

just thought that maintainers are supposed to be somewhat professional, might have had a false assumption.

17:55

end of discussion.

Somewhat being the operative word

jxsl13

end of discussion.

EOF

zhn

EOF

EOL

justatest

1

You just don’t like the outcome of the issue. I’ve had this tone for decades and you’ve been here long enough to have an issue with it earlier

you seem not to like someone not wanting to implement your (complex) ideas

18:01

that's why you posted that coment above

but..but.. learath is known for these jokes around here :/

either way, please elaborate why you posted that comment above

whats goin on

I couldn’t care less who implements it. Chiller is completely free to not implement something he doesn’t agree with, as we are with not merging something we don’t agree with

pepe_ew92

2

you seem to case

18:02

care

18:02

hence the comment

jxsl13

either way, please elaborate why you posted that comment above

Banter. It’s a well known meme in this channel that chiller has a weird obsession with 0.7. He even knows and has one of my previous comments about it on his github profile

18:04

Stop getting offended on other peoples behalves. If chiller has an issue with my message he knows very well that he can just tell me here. And he also knows that I can implement my own proposal

I'm not reading this often enough, mit indded be the case. for me that did just look like toxic behavior and demotivating an active participant in the development.

18:04

might*

Image attachment

rule #1

18:06

of this discord

18:06

please read it

personally I would enjoy all the servers having rainbow formatting

jxsl13

rule #1

I did not insult anyone by any definition of insult. And I wouldn’t say it’s “lazy” negativity or even negativity

jxsl13

rule #1

I will refer you to rule #6 of this discord. DM me if you have a personal issue with me. DM another admin if you have a complaint about me

I do not have any personal problems, because I don't know you enough, I'm reading that one comment out of context of th ebig picture in here. Just trying to make you reflect a little bit.

18:11

will try to keep it at this.

I can see how it might appear out of context. As I said, if Chiller has an issue with it, he can message me and I’ll apologize but knowing the guy for a decade, I’m pretty sure it’s fine

1

I would like some rainbow tees on DDNet :)

18:16

Ya know, the one KoG has.

murpi

I would like some rainbow tees on DDNet :)

If you finish an insane map you get rainbow tee for 2 days

Or that freeze function noby has. Moderation would be so much more fun :0

troll

1

murpi

Or that freeze function noby has. Moderation would be so much more fun :0

Remove the ban commands, give mods super and a ban hammer

Blockworls(?) had this really cool feature that lets you grab people with your crosshair and move them around.

Learath2

Remove the ban commands, give mods super and a ban hammer

YES

Tater has been typing for so long I’m getting worried

I disagree that the format needs to be based on nesting tags with open and closing. I know of some games that implement text formatting where you put in sequential tags to control the formatting and then there is a reset tag is you want to go back to the defaults. I find it concise and easy to use when you're actually typing. https://wiki.trackmania.io/en/content-creation/text-styling (edited)

18:26

do you really want to type <span style="color: #c00000">red text</span> or $F00red text (edited)

Tater

I disagree that the format needs to be based on nesting tags with open and closing. I know of some games that implement text formatting where you put in sequential tags to control the formatting and then there is a reset tag is you want to go back to the defaults. I find it concise and easy to use when you're actually typing. https://wiki.trackmania.io/en/content-creation/text-styling (edited)

I’m not fundamentally opposed to the control sequence approach either. I did propose my own with a \033 in it.

Tater

do you really want to type <span style="color: #c00000">red text</span> or $F00red text (edited)

That is annoying, that’s why I proposed an xml derivative instead of strict html <c=#c00>red text</c> isn’t much worse than $C00red text$z

have you tried typing both of those ten times

18:30

I think the first one is atleast 3x slower

devs did you left the bugs so ppl can make hard maps? (edited)

18:31

or this is a feature or not even a bug

nait

devs did you left the bugs so ppl can make hard maps? (edited)

the physics, including all the bugs are considered features

1

Tater

I think the first one is atleast 3x slower

I could type either with about the same cadence. So I’d estimate the speed to be more related to the character count, so about 2 times slower, but that’s only when changing colors, long stretches of single color text or single colored chunks within text would be dominated by the rest of the typing

18:34

Anyway, as I said. I’m not fundamentally opposed to control sequences either, but I think I want the sequence explicitly terminated, so we can reserve some stuff for the future. Like ^<command>; or ^<command>^ (edited)

what the user types must not necessarily be what the server receives

are there a type of man that approve maps to be official or that make developers?

nait

are there a type of man that approve maps to be official or that make developers?

#📬submit-maps

jxsl13

what the user types must not necessarily be what the server receives

Transparently translating $b to $b$ ? It saves one character but it is weird to parse, e.g. what happens to $b$ ? $b$$?

I meant more like tranlating any easier to write format to the xml format

18:38

translating*

18:39

where xml stays flexible for anything to come in the future

devs from what idea came? if i would be programmer i wouldn't have much ideas to do this game with many features

ideas from players

cool

jxsl13

where xml stays flexible for anything to come in the future

Mh, if we do want hierarchy, stuff like control sequences are not good for expressing that to begin with. If we have to have control sequences, then I’d say we just drop the xml idea at all. A terminated control sequence is easy to parse and just as extensible

18:42

Also easy to type

I just wonder will anyone want to type that as xml into the f2 console or is there some distinction required between cheap ass user input (maybe even binds) and whatever the other part is

Depends on how important it is that colored broadcasts be typed really quickly. I think most of these would be used programmatically or within binds anyway

announcement type of messages definitly binds

Or realistically, given how rare broadcasts have become, these will only ever be used within our announcement loop cron jobs

boradcasts referencing specifi users, definitly server side

18:47

broadcasts*

jxsl13

boradcasts referencing specifi users, definitly server side

But do you think it’d be used by an admin, typed out by hand?

on smaller servers it's more typed by hand for fun stuff, I'd guess

18:48

where an admin is actively participating in the community and staying on the server to have people join

18:49

dunno how broad that use case is nowadays

Tater

I disagree that the format needs to be based on nesting tags with open and closing. I know of some games that implement text formatting where you put in sequential tags to control the formatting and then there is a reset tag is you want to go back to the defaults. I find it concise and easy to use when you're actually typing. https://wiki.trackmania.io/en/content-creation/text-styling (edited)

can we just use ascii color codes

reminds me of fokko a lil bit

18:49

or chiller

zhn

can we just use ascii color codes

Console ones with \033? They are even more annoying to do by hand than xml

they are too nieche, imo

18:50

and finding those color codes

18:50

you'd need to be half of a dev, imo

\x1b[3;5;1m

18:50

why not huh

why not powershell

We don’t even have a way to insert \033 in our console easily

cuz no windows

but we support windows

do you support windows on linux

no

troll

jxsl13

and finding those color codes

The truecolor version is actually very simple btw, just decimal r g b

but windows supports linux

i don't think linux supports ps

\033 what color is that?

it's start of color code

18:52

or it's end

that's ascii 0x33?

\033 is \x1b from my comment

zhn

\x1b[3;5;1m

hm

18:54

too dumb for that

\033 is escape, [ is the control sequence init, 38 is set foreground color, 2 is true color mode

18:55

m terminates the sequence

18:55

\033[38;2;r;g;bm

18:55

Anyway, the \033 makes it too annoying for hand use imo

1

zhn

troll

wine

18:59

this discussion is kind of at a stalemate

19:00

this is shittier to write than xml

19:01

not gonna lie xD

19:02

even as broadcast for a bind, at some point someone has to write it, at least once

@heinrich5991 could you be convinced into merging a control sequence solution that is extensible? I would be okay-ish with ^…^ with the first two being ^RGB^ and ^RRGGBB^. As I said very early on, xml is more suited for representing hierarchy. I did want to keep that possibility for the future but it is also more annoying to use by hand for a feature we might end up using. Also the xml derivative thing pretty much negates all benefits of being close to a standard

hm

19:08

if full fledged xml is not user facing by default, it would be ok

Without the terminator I’m not really a huge fan. Different versions of the parser can lead to different results. There is a reason almost every in band signaling solution you’ll find online uses a start and stop code

might have broadcast_color

19:09

broadcast

19:09

and broadcast_xml

19:10

or just broadcast with a subset of all available xml features (background, italic, etc)

19:10

and for anyone wanting more, they'd have to use xml

19:10

or whatever the more complex format is

19:11

but that's more code to maintain, which is a con imo

19:13

(because kind of code duplication for that translation layer)

19:15

a terminator is a reasonable extention. (edited)

jxsl13

this is shittier to write than xml

both suck imo

19:21

xml is even worse though

zhn

xml is even worse though

I personally would prefer ^rrggbb^ over xml and if xml is required, I'd prefer a translation layer on client side.

1

19:27

with actual xml instead of the derivative

If we have a translation layer, 100% just actual proper xml over the wire. The derivative was just a compromise so it's more typeable by hand

I tried not proposing a built-in translation layer ^^

19:59

I thought we wanted to move away from version number compatibility code.

@fokkonaut I don't know where this thought comes from

19:59

care to elaborate?

20:00

the version compatibility thing was something that works quite well IMO

Initially because you created something such as the server cap flags

20:06

There was some thinking about adding client caps aswell iirc

20:06

And for example f-client sends a ddnet client Version string, but i cant update it because it doesnt support all the following features

20:06

So i cant get to support specific things without implementing everything inbetween

fokkonaut

So i cant get to support specific things without implementing everything inbetween

yes, that's kinda the plan. to lessen the burden of backcompat

20:09

we only have to support ddnet clients by version, not for each feature individually either working or not

[ddnet/ddnet] Pull request opened: #8685 Support composite binds with + commands, fix handling of composite binds with F1-F24 keys, fix incorrect key names with multiple modifiers, refactoring

See commit messages. Closes #8314.

Checklist

[X] Tested the change ingame
[ ] 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 integration test
[X] Considered possible null pointers and out of bounds array indexing
[ ] Changed no physics that affect existing maps
[ ] Tested the change with [ASan+UBSan or valgrind's memcheck](https:/...

github-merge-queue[bot]

[ddnet:master] 8 new commits

b28bac8 Move CBinds::GetKeyId function to IInput::FindKeyByName - Robyt3 d2222f5 Add CBindSlot as return value for GetBindSlot function - Robyt3 8fd40ef Fix key names for composite binds with multiple modifiers - Robyt3 9f6cb92 Add assertions to CBinds::Bind and Get functions - Robyt3 12e5f7f Various minor refactoring of CBinds - Robyt3 dc1a483 Make handling of F1-F24 keys binds consistent with regular binds - Robyt3 3eb8122 Support composite binds with + commands - Robyt3 8d609b9 Merge pull request #8685 from Robyt3/Client-Binds-Fixes - def-

21:02

[ddnet/ddnet] Pull request closed: #8685 Support composite binds with + commands, fix handling of composite binds with F1-F24 keys, fix incorrect key names with multiple modifiers, refactoring

21:02

[ddnet/ddnet] Issue closed: #8314 Binding ctrl+key to multiple actions like +right;+jump binds to key without ctrl modifier instead

Error:

D:\Coding\DDnet_coding2\ddnet\src\game\client\gameclient.cpp(1130,15): error C2660: "CEffects::DamageIndicator": Function doesnt accept 4 arguments [D:\Coding\DDnet_coding2\ddnet\build\game-client.vcxproj]

gameclient.cpp: m_Effects.DamageIndicator(pos, dir, Alpha, Life); effects.cpp: void CEffects::DamageIndicator(vec2 Pos, vec2 Dir, float Alpha, float Life) effects.h: void DamageIndicator(vec2 Pos, vec2 Dir, float Alpha = 1.0f, float Life = 0.75f);

21:09

I dont get it

hmmm

21:11

try a clean build

i already deleted /build and remade it with cmake .. (edited)

hmmmmm

not even chatGPT could help me

do you have it on github?

no

21:12

I can publish it

21:13

but dont shame my code

can you pastebin the effects.h code?

/* (c) Magnus Auvinen. See licence.txt in the root of the distribution for more information. */
/* If you are missing that file, acquire a complete release at teeworlds.com.                */
#ifndef GAME_CLIENT_COMPONENTS_EFFECTS_H
#define GAME_CLIENT_COMPONENTS_EFFECTS_H

#include <base/vmath.h>

#include <game/client/component.h>

class CEffects : public CComponent
{
    bool m_Add5hz;
    bool m_Add50hz;
    bool m_Add100hz;
    bool m_Add200hz;

public:
    CEffects();
    virtual int Sizeof() const override { return sizeof(*this); }

    virtual void OnRender() override;

    void BulletTrail(vec2 Pos, float Alpha = 1.f, float TimePassed = 0.f);
    void SmokeTrail(vec2 Pos, vec2 Vel, float Alpha = 1.f, float TimePassed = 0.f);
    void SkidTrail(vec2 Pos, vec2 Vel, float Alpha = 1.0f);
    void Explosion(vec2 Pos, float Alpha = 1.0f);
    void HammerHit(vec2 Pos, float Alpha = 1.0f);
    void AirJump(vec2 Pos, float Alpha = 1.0f);
    void DamageIndicator(vec2 Pos, vec2 Dir, float Alpha = 1.0f, float Life = 0.75f);
    void ResetDamageIndicator();
    void PlayerSpawn(vec2 Pos, float Alpha = 1.0f);
    void PlayerDeath(vec2 Pos, int ClientId, float Alpha = 1.0f);
    void PowerupShine(vec2 Pos, vec2 Size, float Alpha = 1.0f);
    void FreezingFlakes(vec2 Pos, vec2 Size, float Alpha = 1.0f);
    void FinishConfetti(vec2 Pos, float Alpha = 1.0f);

    void Update();
};
#endif

hmmm. just to make sure. you don't have two checkouts of the ddnet repository and are editing the wrong one?

No just this branch

don't reward clickbait

True

Does my codebase include the keybinds? Before I publish it

new day, new skill issues in assembly. who could've though there's no 2-operand form for imul intruction with 1 byte destination

(edited)

why do you need 1 byte destination?

because my code was generating it xddd

MilkeeyCat

do you have it on github?

https://github.com/Apfff/ddnet

MilkeeyCat

new day, new skill issues in assembly. who could've though there's no 2-operand form for imul intruction with 1 byte destination

(edited)

TIL. I wonder why

21:31

It’s not hard to work around, but still curious

It would be too easy to make everything work first try pepeW

pepeW

21:32

Aaand I think while I was making structs I broke functions xd

I opened the intel assembly manual on my ipad. It triggered the agressive killing of all background apps

21:43

5079 pages combined manual

justatest

2 days in the back is a nasty map

22:46

everything up until the gores part is easy

22:46

and then the unhookable(!) gores part is taking us forever

every time I've done it I've just hammer hit through that part

but it's like

23:03

70% death rate

we're like 20 ppl

heinrich5991

and then the unhookable(!) gores part is taking us forever

team 0 players cant hammer hit for their life

https://discord.com/channels/252358080522747904/295908390956433410/1269791783660027935

find someone who is capable of hammerhitting and do it with them

23:11

and drag the noobs through

"unhookable gores part" lol

<ChillerDragon> @Tater xd

23:53

<ChillerDragon> wowo totar stop XSSing me !

23:53

<ChillerDragon> https://zillyhuhn.com/cs/.1722815580.png

Embedded image

23:56

<ChillerDragon> @Learath2 yes implementing 0.8.0 sounds good. And if you say its even released by the holy oy that must be good. But im not smart enough to write a xml parser.

23:57

<ChillerDragon> @jxsl13 dont worry lerato is bff

Exported 327 message(s)

Timezone: UTC+0