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 2021-01-17 00:00:00Z and 2021-01-18 00:00:00Z

Yeah there where some nice pages

Old screenshot from one of those

Image attachment

We maxed out our bandwidth (5mbps) on chn3 yesterday

03:22

So maybe add more on the XiangYang one which ever that one is

@deen please make a vote about giving everyone strong on everyone without changing in-game physics at all. I know only stupid people will be against it, but I'm done getting frustrated of that weak hook and I would fix it if u guarantee it implemented

without changing in-game physics at all

the tee that hooks first gets the strong

04:58

so nothing changes but the order of calculation

04:58

which doesn't matter

pepesweat

maybe do two physics routes with a config toggle and update some map that specifically needs weak hook if there is any.

/weak is much easier

if there is none i can't see the downside

05:01

is it possible to give only one person weak hook> (edited)

there are some shitty crappy maps

ye add /weak

like 500 years old with all that "jump over freeze 300 times" that sometimes require specifically weak

is it possible to give only one person weak hook?

what do you mean by that

if it is a physics processing order thing, i would be quite weird to untie the "fixed" physics batch and process only one character out of order.

I should think about the system around it. In my mind there is a system which I already described. The one that hooks player first gets the strong on him

it would be*

in this case nothing changes in game phyiscs

05:03

every other one gets weak on him until he rehooks

05:04

so it works the opposite. If someone asked system to have weak then everyone gets strong on him

for the maps that require weak just wait until someone reports it then u can just fix the map

genius

theres like maybe 3 of those maps in total

genouis

louis

for the maps that require weak just wait until someone reports it then u can just fix the map

I though about just deleting to the fuck that parts

05:05

adding time penalty tile

05:05

then we can just have everybody strong

the only downside it that the soothing voice of louis explaining weakhook would become irrelevant

justatest

3

fuck that, weak hook sucks a lot

05:06

no good reason for it to stay

don't fuck louis plz

louis cool

05:08

probably should test the case when both tees have strong on each other and see how it feels and differs

reverse speedfly good

louis should submit ddnet to GDQ sometime after the covid thing.

05:09

pepesweat

much more parts are weak impossible

TsFreddie

louis should submit ddnet to GDQ sometime after the covid thing.

fastest 100 points on new name random seed glitchless

i thought almost all ddnet tech is a "glitch"

no they're features justatest

justatest

kek

"glitchless" is kinda impossible

[ddnet-maps:master] 1 new commit

f6eae53 A acq - ddnet-maps

just joined random map and there is already a tee trying to make weak inpossible part

Image attachment

05:41

not sure if it's really impossible or just insane hard

05:47

possible

weak is good

07:07

dont remove

07:07

and accounts bad

Does anyone know why I have "no matching overloaded function found" error while calling from the same header file and it appears to be ok before build attempt?

07:31

Image attachment

07:31

Image attachment

07:32

everything is called in the gamecore.h

07:33

Oh I guess I see that

07:38

No I don't

@heinrich5991 ye thanks for the support i fixed my system :) i downgraded to linux-lts paired with nvida-lts and now all works again seems like there was a driver update that broke my system and that seemed to happen while i reinstalled my system so i blamed my self instead of the driver

08:58

i wonder how such a thing is not on the archlinux news tho i am still a bit confused

09:00

oh wait it has but that was some time ago maybe i followed the arch news back then and downgraded to lts already

ChillerDragon Do you have time :X?

@TsFreddie I have stopped some CHN3 servers and run CHN4 again

great. thx

yea @Trafalgar Law

10:21

you can also pm me on irc btw via the web thingy https://webchat.quakenet.org/?channels=#ddnet

10:21

hm that link is broke idk how to create a proper one xd

Дядя Женя

@deen please make a vote about giving everyone strong on everyone without changing in-game physics at all. I know only stupid people will be against it, but I'm done getting frustrated of that weak hook and I would fix it if u guarantee it implemented

lol only stupid ppl against it?

10:27

Weak is part of the game...

weak is bug

10:27

pretty annoying bug

yes but maps adopted it

10:27

some parts are almost not doable with strong

10:30

But that's a whole other discussion. I just find it pretty rude that you say all the people against it are stupid pepeH

pepeH

alias /spec to /weak (don't actually tho

Don't let the Russian trashtalker change our hook (edited)

4

10:48

ChillerDragon lets go on a tw srv and write there

Is it somehow possible to fix laser door skipping with too much speed e.g. using jetpack?

11:13

Need it for my mod

you'd have to change the code for the check

11:19

currently it just checks your current position I guess

11:19

you'd have to check the distance the tee moves and whether that intersects with a laser door

[ddnet/ddnet] Pull request opened: #3517 Fix weak hooks

Again disappointed in the community, so doesn't care. But here is the working version. The way it works: -If you hook someone you gain strong -If you hook someone while he's already hooking you - you gain weak. And stay weak for this entire hook until you rehook while not being hooked by that person. Complicated, but the idea is that you're always strong except for the times when two tees hooking each other at the same time. Then someone is weak to simulate current physics behavior so i...

@deen ok thx

11:42

btw, this weak hook fix is not good

11:42

Does not make any sense

11:42

if the tee is frozen you cant get weak

yeah

11:42

appart from community not wanting the change, the change itself is flawed

Yea

bad change

im amazed he thought this was a good fix

11:43

-If you hook someone while he's already hooking you - you gain weak. And stay weak for this entire hook until you rehook while not being hooked by that person.

11:43

this is the stupidest thing ever

Can we run a couple servers with the actual proper fix?

no

11:49

it has been voted no

I want to see the effects of it

11:49

Not this fix. This one just sounds goofy

it's good but you'd better run a test server with it first

I mean a proper saturated add after equilibrium

because I'm not sure if it can't crush anything

11:50

@Ryozuki stop crying)

lmao

11:50

u should just leave forever

like your voice matter something

like urs matters

I don't think anyone really knows how the actual fix would feel like

everybody is annoyed by u

I fear it would feel awkward, but maybe it doesn't?

Ryozuki

like urs matters

of course it is. I've achieved something in that life

lmao

Learath2

I fear it would feel awkward, but maybe it doesn't?

it's just like you have strong and that's all

ur a cringefest

As said I don't mean the weird strongswap

@Ryozuki how old are you? Feels like acting of 14 yo child who thinks he means something in this world

ur reflecting mym an

11:53

my man*

11:53

or projecting

Learath2

As said I don't mean the weird strongswap

but what do you mean then?

I mean a proper fix like teeworlds#2478

hilarious

teeworlds/teeworlds#2478 that is

https://github.com/teeworlds/teeworlds/pull/2478

fix #541 : remove weak hook using by nheir · Pull Request #2478 · t...

Use SaturateAdd after summing all the dragging velocity

i would say u fit more of a 14y old pushing this after people voted no

11:53

and saying ppl who vote no are retarded

11:53

just an educated guess

This should give everyone the exact same hook strength

11:54

I feel it might break speedfly, but who knows, maybe it doesn't. No way to really know without testing

Learath2

This should give everyone the exact same hook strength

you can't because it will feel a very different then it was

11:54

I fixed it the way it doesn't change anything at all. That's why solution is so dirty

How do you know? I don't because I didn't test it

11:55

I assume you either tested it or you are god

Firstly I've tried to make it normal and created an Update method supposed to be called once per tick to update hook physics, but that turned out to be wrong

11:55

also it's obvious

11:56

because your velocity is calculated everywhere before and after core update when hook influence being calculated

11:56

so it would be different no matter what you do

It's not obvious at all to me what the impacts of it would be. It's only obvious that it will be different

I've tried hard all day on imaging the solution that would keep existing approach but be clear and normal, but there is no way. The way physics implemented now is so wrong that no matter what you do you will fix something accidentally

Now that's bad for people with allergies to difference like Konsti but differences on the order of 1 or 2 ticks don't concern me, they are imperceptible to all but the pickiest of players

11:58

and maybe not even them, which we could find out if we run a test server

@Trafalgar Law you can mostly find me on fokkos server i even have tw pings on currently but might be afk

@Learath2 i have applied that tw pr in my mod and it works nice, also added a config var to simply calculate the old way again

12:03

so i can toggle it

@fokkonaut have you tested whether it breaks speedfly? I've been investigating why speedfly happens at all and it seems to have some relation to the strength of the hook

speedfly only works with strong

nope, you can reverse speedfly too

Yea

12:04

but that wont work ofc anymore

12:04

reverse speedfly is with weak

What I'm wondering is what happens if everyone has the same hook strength

That doesnt matter, its about the strength of the top player

and as I read the code everyone should end up with a hook little weaker than the strong hook of past, sooo can people still speedfly as if the top player has strong with the patch?

i think they have exactly the same strength

12:06

It looks like it at least

12:06

when hooking a tee over the ground without moving

hm, I thought it'd be slightly weaker, but if it's the same I guess it should work still

12:07

@fokkonaut do you have a server running with it right now? I'd like to see if it feels different

Just leave everything as it is, then there won't be any unforeseen new bugs

I did a test like here: https://wiki.ddnet.tw/game-mechanics/hook/

Hook :: DDNet Wiki

12:09

@Learath2 and i got the same results for both tees, but i didnt check position in the code

12:09

just checked it with my eye xd

Learath2

@fokkonaut do you have a server running with it right now? I'd like to see if it feels different

yeah my server, it has weak hook 1 tho, let me quickly set it to 0, just connect

that's good enough, if I was thinking about it correctly it would have differed by a tile or so, you'd easily see that by eye

good

12:10

sv_weak_hook disabled now

12:10

you can try it

you seem to have atleast sort of fixed fast download for ddnet clients

12:10

20k is not bad

I did nothing

12:10

:D

I guess I'm just dropping less packets today

I tried to fix fast download twice now but i gave up

12:11

it always worked on LAN

12:11

but it got stuck on production

Cøke

Just leave everything as it is, then there won't be any unforeseen new bugs

it is a bug that makes it very hard to reason about things though, it'd be nice if things were symmetrical and consistent

12:12

I've been looking at 4 demos of people speedflying, with top hammering, reverse speedfly, normal speedfly etc. and with the asymmetry of the physics it's really hard to think about what actually happens

It's like u want to fix the hf bug

12:12

Xd

hf is not a bug, hf is a direct result of how the physics were imagined

12:15

it has very clear dynamics

No more like the bug that you can move around in a 64x64 box

12:15

Better let us fix all of these bugs and broke the game troll

troll

(edited)

@Learath2 97 players xd

12:16

@Cøke a tee is 28 units long and a block is 32 (edited)

you only need hook to start speedfly, later the tees don't need to even touch

My ping isn't too bad but your server feels rather slow. I think 97 players is taxing the server loop quite a bit

@Learath2 yeah maybe xd

@Ravie are you sure? I haven't finished up the tools to check this yet but I had a feeling it was the result of the tees oh so slightly clipping into eachother and the physics pushing them out

no

12:18

lol

12:19

Speedfly is that you dont really touch at all, thats why the bottom tee hammers so fast

12:19

So you cant even touch each other

12:19

and your hook with strong keeps both tees stable

not "lol" it's not an obvious interaction at all, unless you analyzed it I don't see how you could confidently make that comment

:D

your strong hook while it can keep you stable doesn't generate any velocity

Well, I saw and worked with the code for pushing tees out

12:20

Thats why i instantly thought its not this

@Learath2 well at least when I do it using deepfly bind, after the first hammer there's a visible gap between tees sometimes

Good to see there's still DDRace development.

ddnet

2

its fng

Ravie

you only need hook to start speedfly, later the tees don't need to even touch

you don't have to hook at all

you can speedfly without hooking?

yes

How would that work? Tee needs to get lifted

12:22

bottom one

Oh I remember that odd video where the guy did it

Show me

12:23

Ah I think with 2 hammerers

12:23

Hammer is creating up velocity

@murpi true but in reality u never do that

Learath2

Oh I remember that odd video where the guy did it

I sadly don't have the tools to extract my memories into webm format yet

xd

I think the top guy hammering is slightly pulling the tee below in that case

with some rng bounce timing and deepfly bind you can do it from tower

Lol how you know such things @Ravie

im abit confused, why would players be opposed to having strong given the vote in #announcements. Something im missing?

Ravie

with some rng bounce timing and deepfly bind you can do it from tower

yes

Luca

Lol how you know such things @Ravie

I've spend hundreds of hours messing around and trying to make weird new parts :p

SPYRES

im abit confused, why would players be opposed to having strong given the vote in #announcements. Something im missing?

People don't like changes in general, which is common in old niche games like this one

1

And also weak hook is a major part of the game in my opinion

yes

12:26

i wouldnt want to fix it too

I don't understand how people can just say "it's good how it is so don't touch it" and not strive for improvement

2

Well I mean in gores for example it makes sense. But in ddrace it's a "feature"

Not directly a feature but its a bug thats so well integrated that it even has official ddnet support xd

It's not about denial of improvements. That's why it is discussed. It's simply seen as a game mechanism

Luca

And also weak hook is a major part of the game in my opinion

is it really? we have a handful of parts that use it ever, it mostly only causes confusion in new players, old players either use /spec or /save to just trade it anyway, it doesn't even make anything impossible it only makes it harder

fokkonaut

Not directly a feature but its a bug thats so well integrated that it even has official ddnet support xd

Yeah whatever you like to call it these days

xddd

I will look for a part where you can see the use of weak @Learath2 (edited)

There are a couple, some drag parts are easier with weak e.g.

yes

1

To "look" for a part shows how its not important xD

I "think"(tm) it would be impossible to do that part where you hang off of someone with strong on Stronghold 3, but maybe not even that

well you know I find it makes things interesting if you have to deal with such things like weak

fokkonaut

Not directly a feature but its a bug thats so well integrated that it even has official ddnet support xd

It has "official" "support" because it's a completely invisable game mechanic. We added an indicator for it recently and we have prediction support for it because the prediction would be wrong without it

But also I am not playing that much anymore so maybe ideas have changed in the tw world

Anyway, we did a vote and people don't want it changed, I'm fine with that, but doesn't mean I won't argue my point. I honestly can't wholeheartedly say I either support or disapprove the change, because I haven't tested it

the vote is basically "do you guys want a change?" community: change bad

from what I've seen on fokkonaut's server, it didn't feel drastically different to fly, throw or hh people

probably super irrelevant here but i deleted weak hook on fng a couple years ago and people have either actively liked the change or not noticed

12:34

noone has ever complained about it ever

yeah I think you either hate weak hook or not care about it, so it would still make more people happy to fix it

In fng

12:35

XD

In fluid pvp gameplay you would never notice it

read the first three words of my msg x_x coke (edited)

I'd only be concerned about tricks and timings

nah i notice the weak hook when i play on default fng servers now

in fng you basically drag a dummy to get him into spikes

I would probably need to readjust for a day or two to be able to do freezehammers again e.g.

Ravie

in fng you basically drag a dummy to get him into spikes

yeah that or hooking close enemies when u shoot them

12:36

and when ur dragging the frozen tees weak is just annoying

Ah I know one place where weak and strong would be a decent mechanic. If you can somehow force the weak person to do a long deep drag

especialy if ur moving fast and tryin to throw them as u go by

also when you hook a frozen tee while you're both on the ground or from below, slide physics really suck (edited)

with weak?

It's significantly easier to drag someone when you are strong

++

Though as I said, since there is almost no part that is impossible to do with weak or strong, it's very hard to force such a situation, so I'd be extremely surprised if any maps have such a mechanic

12:39

I guess a deep drag part, followed by a swap, followed by another deep drag part can force both people to do a deep drag and thus atleast make one tee suffer dragging with weak

12:39

but again, of the collection of hundreds of maps we have, how many have two deep drag parts following eachother with a forced swap in between

many hard drag parts are really crap or nearly impossible to do with weak

also true.

yeah, which would be a concern of mine if the mappers actually intended to make people suffer through them as weak

12:41

usually you either get /spec or you get to choose who goes into deep, because as you said it's crap to drag with weak :P:

in drag-heavy maps /spec is really a must imo

12:42

but /spec is not always a good thing, it lets you do tee collision fuckery and sg bug

Anyway, my point is I don't think it's a major feature, I think it's just a quirk of physics that we learned to live with

12:43

We could always look into making spec not allow you to unspec if you are colliding with someone as it's unlikely people will change their minds about strong/weak changes

12:43

though that would allow a new form of blocking

maybe just add a command for giving strong hook that doesn't make your tee disappear

I guess we could just reinsert the tee to the top of the entity list

@fokkonaut you seem to have an issue with ingame serverinfo packets for ddnet clients btw

12:53

seems you only report 63/64 there

not possible

12:59

Has to be clientside

12:59

Because I 100% dont do

Sorry, I must have had a partial complex seizure and saw 63/64 instead of 97/128, because that's more likely than a mistake in one of the most ridiculously complex parts of the code riddled with dozens of branches handling different types of serverinfo lol

Its not that deep for fddrace

13:01

https://github.com/fokkonaut/F-DDrace/blob/b24dabe21b2f683c814b17b99fd978056c14d996/src/engine/server/server.cpp#L1723

fokkonaut/F-DDrace

F-DDrace is a server mod for Teeworlds 0.7 developed by fokkonaut. - fokkonaut/F-DDrace

13:01

Check SendServerInfoSevendown

Hey, I can only tell you what I saw, it can be a client bug too.

Thats what I said

What is it with people being extremely defensive and confident this month? How are you all so sure you make no mistakes?

I am just confident about this situation, lol. I know my code well

13:06

And @Learath2 I havent seen that issue before

Bravo, 10 years working with ddnet code I would never be so confident as to concretely claim much of anything about it as it's riddled with spaghetti and UB

13:07

I aspire to be as confident as you and jupeyy one day

lol

13:07

as someone with a mod based on ddnet i could never be that confident

13:07

in my own code or otherwise x_x

13:08

ie i would assume its a server bug at first

Aaanyway, I have much real analysis to learn. I bid thee farewell. I'll take a look at the serverinfo parsing later anyway, will keep an eye out for a bug on the client

I dont know how you guys can shit talk like this now. I changed this part of the code to specifically NOT report 63/64, but x/128, then before I said its not a bug in my mod I even checked the code before and sent the link

who is shit talking

13:10

lol

noby

as someone with a mod based on ddnet i could never be that confident

mimimi

thonkery

We aren't shit talking, atleast I'm not. I'm just very impressed

Dont make fun of the situation now

13:11

I doubt u r impressed by anything

Seriously, I honestly don't feel that confident about anything I write in the context of ddnet

13:11

I've coded many parts of the code to specifically do one thing and it ended up doing something else through the most convoluted means imaginable

i spent 30min fixing ddnet compiling on macbook just to test a 1 line pr i submitted why do. u think im kidding

i had times like this too

13:12

but not too often

i mean i was fairly sure it would work

13:13

but still xd

odd, but still admirable

Showing 241 changed files with 68,858 additions and 6,231 deletions.

@Learath2 @fokkonaut since the player count in the server info tab comes from snapshots, I think the maximum amount of players there would be 63 or 64

Just a small mod

Add another number like ID that shows how strong every tee is

13:19

1 is the weakest, 64 the strongest

@heinrich5991 that explains it

@fokkonaut I don't know how you can tell @Learath2 that he's looking the wrong way though

@ᶰ°Konͧsti we added something similar for the next version I think

HMM

even if you coded it "the right way", there can still be bugs

oh wow you guys are pissed because i said its not possible and has to be clientside xD

13:20

Like

13:20

????

We have a lot of x + duration * tickspeed >= tick in the code, I wonder if there is a way to prettify that

SecondsPassed(int secs)

13:21

bool

fokkonaut

oh wow you guys are pissed because i said its not possible and has to be clientside xD

We are not pissed at all. I don't see where you see me being pissed

I see heinrich complaining for nothing

I just told you that I am oddly surprised at how confident you are in your code

Lol

13:21

I am not even super confident

13:21

I just checked the code

13:21

sent a link

13:22

and said it has to be clientside

13:22

relax a bit

if someone told me my ingame serverinfo was wrong, I'd first look at where the serverinfo tab gets the player counts, then I'd look at where I send said information, then I'd look at how I collect the information

13:23

I'd also be concerned about the fact that there is something called SERVERINFO_INGAME sounds like it could have something to do with it, maybe legacy baggage?

The fact that I know what I send and how I send it led me to that its a client issue

13:24

and i was right

13:24

where is the problem

I don't see where you see heinrich complaining either, he gave a reasonable explanation to the issue

13:24

See what I mean by everyone being so defensive this month?

xD

You are defensive aswell :)

13:24

We both are

How am I defensive? I just told you it could be a bug in the client for all I know. I haven't tested it

am i late to the party

Learath2

How am I defensive? I just told you it could be a bug in the client for all I know. I haven't tested it

Isnt that defense too?

13:25

Just like I do

13:25

Clarifying the own pov

TsFreddie

am i late to the party

Just take popcorn and read it :X

@Learath2 anyways I dont wanna contribute to this discussion anymore, its useless and definitely not leading anywhere

How is it defense? I gave you a bug report, something I encountered on a mod you wrote. I didn't say it was your bug, I didn't even say which client I used to join. You just instantly said made a deduction that I thought was overly certain. I pointed out that it was rather odd that you were so confident

where are the rendered demos stored?

u said u used ddnez

Yeah, as always. Ditch me in the middle of an argument to make me look like the goof

LOL

I'll go review the backport PR

@Im 'corneum %APPDATA%/Teeworlds/videos

thanks

@Learath2 I really dont want to make it look like this, lets continue then

https://secret.club/2021/01/15/bitlocker-bypass.html

BitLocker Lockscreen bypass

BitLocker is a modern data protection feature that is deeply integrated in the Windows kernel. It is used by many corporations as a means of protecting company secrets in case of theft. Microsoft recommends that you have a Trusted Platform Module which can do some of the heavy cryptographic lifting for you.

POG

1

13:28

amazing site btw, check out the other articles as well

@Learath2 Now you said you didnt even say what client you used, but before you said "for ddnet clients"

No please, go right ahead. I'm quite used to it with heinrich and Ryozuki pulling the same thing almost weekly

❤️

1

1

heartw

1

Also, as I said I am only mirroring my pov right now, because I dont get how you can be so offended or whatever. I just said its a client bug and I even sent a link

13:29

Idk whats wrong about it

https://secret.club/2020/06/19/battleye-packet-encryption.html o_o xd

If you work almost daily on your own mod then you just can be this sure, at least I am

Learath2

No please, go right ahead. I'm quite used to it with heinrich and Ryozuki pulling the same thing almost weekly

when am I gonna be on the list

fokkonaut

Also, as I said I am only mirroring my pov right now, because I dont get how you can be so offended or whatever. I just said its a client bug and I even sent a link

Again, I don't quite see where I'm offended. I just pointed out a handful of oddities

@noby check out the other one where battleye has anticheat modules specifically for reverse engineers

nobot

1

nobot

1

nobot

1

windows sucks

fokkonaut

If you work almost daily on your own mod then you just can be this sure, at least I am

At least in this situation, because as I said I recently worked on that to NOT report /64

heinrich is going to make the ultimate anti cheat bypass

13:31

feelssurprized

I'm not the author of that blog ^^

13:31

although I'm aspiring to be as good as these guys. I probably never will be ^^

uh oh

13:31

lol

@TsFreddie all you need to join the club is join in on a discussion I'm having and disappear after giving an argument that's impossible to counter

@Learath2 and since thats the only place where I send the serverinfo or generate it, I KNEW it was not my server

An appeal to authority is the most common choice

@Learath2 would it not be clear for you that its a client thing here?

Learath2

@TsFreddie all you need to join the club is join in on a discussion I'm having and disappear after giving an argument that's impossible to counter

i bet u dont even know how every line in snapshot.cpp works from memory ok bye

lol

@fokkonaut no, as I don't really know where the serverinfo for the serverinfo tab ingame comes from. I guess you didn't either given it took @heinrich5991 to point out where it could be coming from

who cares where it comes from @Learath2, thats no information we need here

13:33

Because

13:33

again

13:33

I just said it has to be clientside

13:33

NOTHING MORE

13:34

And i am mr super duper confident now for i xd

13:34

u*

@heinrich5991 btw did you see I got a commit in ghidra 9.2? I've officially helped the USA in their never ending quest in world domination

1

Nvm then i will smoke a cig xD

mine probably still didn't get in

13:35

ddnet helping with drug addiction every day

@fokkonaut it's relevant information, how else would you debug said issue?

heinrich5991

ddnet helping with drug addiction every day

If you don't know the source of the faulty information, you can't know why it happens nor where it happens

https://github.com/NationalSecurityAgency/ghidra/pull/1889

Autodetect the ___chkstk_ms symbol by heinrich5991 · Pull Request...

Fixes #1888.

13:35

still open, not looked at

@Learath2 I see why heinrich and othes would leave your discussions

NSA

monkaS

13:35

monkaS

13:36

they collected heinrich data

13:36

monkaS

@Learath2 have you checked the secret.club link I posted above?

You are kinda like chillerdragon, you cant win against you

13:36

not possible

yeah, very hard to argue when I'm doing nothing but stating facts

me too

Learath2

yeah, very hard to argue when I'm doing nothing but stating facts

i remember u complaining about jupstar doing this

13:36

monkalaugh

I give a completely reasonable argument, I even gave reasoning as to why I think this is fact unlike Jupstar

@Learath2 no matter how much i look up to you this discussion IS POINTLESS. I gave my facts and arguments too

If you don't agree, by any means go ahead and tell me where I'm going wrong in my chain of reasoning

I did so many times now

I agree it is pointless, you are working on the assumption that I'm attacking you somehow or I'm offended

false

Which I've told you like 5 times throughout that I don't care

pepesweat

@heinrich5991 https://github.com/NationalSecurityAgency/ghidra/pulls/Learath2 I still have 3 left, will probably never get them in

NationalSecurityAgency/ghidra

Ghidra is a software reverse engineering (SRE) framework - NationalSecurityAgency/ghidra

i would pr with a alt acc just in case pepeH

pepeH

I want the world to know that I've contributed to imperialism

13:40

Maybe they'll hire me

they won't. you need to be US citizen

13:40

probably since birth or so

@Learath2 Let me explain it a last time: You told me about a bug where ingame the server info reports 63/64. And I said its not possible and has to be a clientside bug, then gave a link where you could see what the server sends. Because I recently changed it so that ddnet clients receive x/128 I knew that it was not a servers fault. And that was the only thing I said, knowing that it has to be clientside. Now where is your problem? I dont get it. Its not at all about fixing this bug right now or debugging it, it was just about that situation where you said I was oddly confident.

yeah

13:41

i doubt they hire anyone on the nsa who isnt natural born in usa

13:41

https://en.wikipedia.org/wiki/Natural-born-citizen_clause

Natural-born-citizen clause

Status as a natural-born citizen of the United States is one of the eligibility requirements established in the United States Constitution for holding the office of president or vice president. This requirement was intended to protect the nation from foreign influence.The U.S. Constitution uses but does not define the phrase "natural born Citize...

13:41

this

damn

idk why when i see a eagle i can only think of fascism

13:42

maybe because the eagle in spain flag is fascist

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

Coat of arms of Germany

The coat of arms of Germany displays a black eagle with a red beak, a red tongue and red feet on a golden field, which is blazoned: Or, an eagle displayed sable beaked langued and membered gules. This is the Bundesadler (German for "Federal Eagle"), formerly known as Reichsadler (German for "Imperial Eagle"). It is one of the oldest coats of arm...

13:42

don't think I'd call today's germany fascist

this is spanish fascist flag

Image attachment

13:43

look the resemblance

13:43

its incredible

fokkonaut

@Learath2 Let me explain it a last time: You told me about a bug where ingame the server info reports 63/64. And I said its not possible and has to be a clientside bug, then gave a link where you could see what the server sends. Because I recently changed it so that ddnet clients receive x/128 I knew that it was not a servers fault. And that was the only thing I said, knowing that it has to be clientside. Now where is your problem? I dont get it. Its not at all about fixing this bug right now or debugging it, it was just about that situation where you said I was oddly confident.

I don't have a problem with the bug. I don't have a problem with you having confidence. I literally just said I personally don't feel confident about the behaviour of most any piece of code. Which is apparently not such an insane position given noby seems to have a similar experience coding within ddnet aswell

Okay, i understand that, but where the hell is our problem then xD

fokkonaut

I dont know how you guys can shit talk like this now. I changed this part of the code to specifically NOT report 63/64, but x/128, then before I said its not a bug in my mod I even checked the code before and sent the link

Well idk, you told us we were shit talking

13:45

I was only being slightly sarcastic and it was to poke a needle at jupstar as he like you is always oddly confident of his code

Ah right, because I felt offended by ur ironic seizure message

That was more supposed to be a joke, I guess it didn't go over that well through text

sarcastic, not ironic yes

See, resolved. It's not that hard to finish these

Good

13:47

I felt like you say I havent checked anything and i just said it to protect my mod or something

Here is to hoping I get a systemd discussion to the end one day

But I have no Problems in having bugs or something, I just knew this was not my bug :D Thats why it "escalated"

13:48

anyways

13:49

@Learath2 Have a good day, I'll be mobile from now on

@heinrich5991 any idea what to do about check-style? I'm not really sure how best to handle this

bors r+

13:59

it checks the style of the resulting merge

hm, now that I think about it it's kinda weird that that's not what github runs the checks on

because you have to run it at the point of merge. it's basically the whole point of bors

14:02

https://graydon2.dreamwidth.org/1597.html

14:02

The Not Rocket Science Rule Of Software Engineering:

>

automatically maintain a repository of code that always passes all the tests

I wonder why something like bors isn't part of github yet. There isn't much point to a PR being green if the resulting merge wouldn't be green

I see

Actually, why are these failing anyway?

https://github.com/ddnet/ddnet/pull/3510

Work around Pylint import problem (fixes #3507) by def- · Pull Requ...

Checklist Tested the change ingame Provided screenshots if it is a visual change Tested in combination with possibly related configuration options Written a unit test if it works standalone, s...

14:05

(false positive, not sure how you can manage to have false positives there)

https://github.com/ddnet/ddnet/pull/3514/checks?check_run_id=1711438835 e.g. this one

Also display DDNet version for 0.7 clients by heinrich5991 · Pull R...

Checklist Tested the change ingame Provided screenshots if it is a visual change Tested in combination with possibly related configuration options Written a unit test if it works standalone, s...

space at EOL

yeah but you didn't even touch that file

yes, someone merged without bors

14:07

because bors was broken with the python checkstyle

oh, that would explain it

14:09

What is the best way to fix that? Just pushing a style commit directly before trying bors r+ again?

just do bors r+ again

14:11

the style is fixed already

14:12

https://github.com/ddnet/ddnet/pull/3515

Fix clang-style by heinrich5991 · Pull Request #3515 · ddnet/ddnet

Checklist Tested the change ingame Provided screenshots if it is a visual change Tested in combination with possibly related configuration options Written a unit test if it works standalone, s...

14:12

so yea, that's the best way to fix it

14:12

by doing a PR with the style fix

[ddnet/ddnet] Pull request closed: #3514 Also display DDNet version for 0.7 clients

14:23

[ddnet:master] 2 new commits

b66e090 Also display DDNet version for 0.7 clients - heinrich5991 2363efb Merge #3514 - bors[bot]

ᶰ°Konͧsti

1 is the weakest, 64 the strongest

We have exactly that in ctrl-shift-d, try out nightly if you want it

:O nice But would ne nice to be able to show it up Always. So u know for example how much u have to hook a tee to make him go edge before fail to freeze

yeah, i did that but the current consensus is that we don't want to add more commands

15:03

and a number is not such nice ui, so instead we'd prefer to show weak hook slightly differently

15:03

maybe a thinner hook: https://github.com/ddnet/ddnet/issues/3363

Visually separate strong/weak hook · Issue #3363 · ddnet/ddnet

There are many different ways on how this could work + look. imo, a nice solution would be to have a separate ddnet_textures.png image to game.png with just a different texture for weak hook. That ...

[ddnet/ddnet] Issue opened: #3518 Votes have negative count with f-client on ddnet server

Reported by arisoda ingame. @Learath2 @fokkonaut any idea?

@fokkonaut I added a new net object for the ddnet projectile. you should be able convert to the old projectile in your client easily. I can make a PR if you want

16:11

https://github.com/ddnet/ddnet/pull/3492

Support F-Client and replace non-extended extension messages by hei...

This continues what #3439 started, and also replaces non-extended messages that are DDRace extensions with extended messages. The DDNet server still sends the old message to old clients and the DDN...

16:12

NetObjectEx("DDNetProjectile", "projectile@netobj.ddnet.tw", [
    NetIntAny("m_X"),
    NetIntAny("m_Y"),
    NetIntAny("m_Angle"),
    NetIntAny("m_Data"),
    NetIntRange("m_Type", 0, 'NUM_WEAPONS-1'),
    NetTick("m_StartTick"),
]),

Thoughts on mongodb? I'm thinking of using a nosql for my own project.

i prefer sql

what are your requirements? what do you want to do?

more data integrity

Is there anything that ill-advised for nosql

16:22

That is*

for only data integrity I'd suggest using sqlite. it's easier for users and developers @Ryozuki

Basically just player profile and player scores, thing like that

mongodb uses a not OSI approved license btw

16:23

its not open soource software

16:23

or free software

@TsFreddie sounds like a perfect fit for sqlite

sqlite is good

16:23

idk why ppl dont use it more

16:23

if they dont have to do distrbuted stuff

we also just started using it for ddnet server

16:23

but we do use it now

justatest

what if I just don't want to write queries

use a ORM

I mean SQL queries

ORM

16:24

what language?

what language?

16:24

ye

Both typescript and c++ actually

if u rly wanna use mongodb, there is mongoose for typescript

16:24

for c++ i gotta search

16:25

https://mongoosejs.com/

What is ORM actually

object-relational manager

16:25

it allows you to write queries without writing queries

basically a fancy interface

I see

check out

16:26

django orm

16:26

for a full fledged orm

16:26

https://docs.djangoproject.com/en/3.1/topics/db/queries/

Making queries | Django documentation | Django

I've used mongodb on TS before. So I'm just not sure about c++

latest_question_list = Question.objects.order_by('-pub_date')[:5]

16:27

this is a SQL query with the django ORM

Image attachment

16:27

this is what orms strive for

16:27

xd

heinrich5991

latest_question_list = Question.objects.order_by('-pub_date')[:5]

Why does that just looks like mongo query tho.

if ur making a full web app

16:28

and just want to use typescript

16:28

this one is good https://nestjs.com/

NestJS - A progressive Node.js framework

NestJS is a framework for building efficient, scalable Node.js web applications. It uses modern JavaScript, is built with TypeScript and combines elements of OOP (Object Oriented Progamming), FP (Functional Programming), and FRP (Functional Reactive Programming).

It is not really an web app.

16:29

Although it is http

what is it then

16:29

or is it secret

https://www.codesynthesis.com/products/odb/

16:29

C++ ORM that supports sqlite

Hard to say. Some piracy stuff

16:29

justatest

monkalaugh

c++
  odb::sqlite::database db ("people.db");
  person john ("john@doe.org", "John Doe", 31);
  person jane ("jane@doe.org", "Jane Doe", 29);
  odb::transaction t (db.begin ());
  db.persist (john);
  db.persist (jane);
  t.commit();

heinrich5991

https://www.codesynthesis.com/products/odb/

Noted

16:33

I do love the fact that this is becoming my go-to place to ask about these things for some reason.

@Learath2 what do you think about the answer in https://github.com/ddnet/ddnet/pull/3502 ?

Some (mostly backported) refactor by Kaffeine · Pull Request #3502 ...

Port some cleanup and refactoring from the upstream and add some custom refactoring. Changes overview: Backported angle() and direction() functions (upstream commit 7735557) Backported vector math...

16:39

I'm fine with it as is

[ddnet/ddnet] Pull request opened: #3519 Send logout packet after using timeout protection

Only send the logout packet if you're actually logged in. Fixes #3490. Supersedes #3495.

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 if it works standalone, system.c especially
[ ] Considered possible null pointers and out of bounds array indexing
[ ] Changed no physics that affect existing maps
[ ] Tested the change wit...

[2021-01-17 17:45:49][censorlist]: failed to open 'censorlist.txt' where do i get this txt?

it's only used by the chinese ddnet hosters

exposed

the rest of the ddnet servers use an empty list

[ddnet/ddnet] Issue closed: #3399 Make averege time more correct

16:49

[ddnet:master] 2 new commits

39d06c5 Display median time instead of average (fixes #3399) - def- 174eeab Merge #3504 - bors[bot]

16:49

[ddnet/ddnet] Pull request closed: #3504 Display median time instead of average (fixes #3399)

so i dont need it right?

no, you don't need it

kk thx

why is "chinese coronavirus" on the censorlist?

I actually don't know.

16:51

I don't know who complied the list

16:54

Maybe the person made the list thinks implying coronavirus has a chinese heritage is bad.

it did come from china into the rest of the world. it's not the term I'd use, but banning it if it's not required by the government seems to be too excessive

16:56

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

Chinese coronavirus

Chinese coronavirus may refer to: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), first reported in Wuhan, Hubei, China; in December 2019 COVID-19 pandemic in mainland China Severe acute respiratory syndrome coronavirus (SARS-CoV), first reported in Foshan, Guangdong, China; in November 2002

Well if it is enough to create a outrage on social media it is bad for us.

16:56

On Chinese social media ofc

@heinrich5991 Sounds fine i guess

16:56

but why?

16:57

the old projectile works fine, no?

it's easier to support all clients if all clients speak the same protocol

16:58

@fokkonaut do you want me to make the PR against f-client?

true, but whats the purpose of having the extra info projectile?

16:59

But sure, if thats the only error with the fclient, then pr it

TsFreddie

Well if it is enough to create a outrage on social media it is bad for us.

One thing to consider is that, a lot of time it is not the actual government doing the censorship.

@fokkonaut it's for prediction and knowing whether to make it transparent if not from your team

Departments that has any power over the people or the content would just do anything to stop people from talking about a topic too much to create out of proportion rage or misinformation to keep their lazy ass on their politically supervisor seat.

17:02

Supervisor = superior*

17:03

So that's why most of the time the "censorship" is over things that are insanely stupid.

17:03

Like animal crossing

17:03

pepesweat

17:04

So I feel like being as safe as possible doesn't hurt.

17:09

I mean it is only for CHN servers, I would just let Tom feel as safe as he wants. We definitely try not to influence anything outside of china.

Do I have to add Novice to the srv name that the new players see my srv

Yes

17:22

Can't say for other regions but CHN player usually need a week to realize the filter

is that why kog has "novice insane gores"?

justatest

did it really

Its because they have a vote

something like that yeah

if you vote easy then its novice easy

17:22

but if you vote insane

17:22

its novice insane

But why don't they change server name after vote

17:23

Oh that doesn't change the novice part

17:23

But if you don't want noobs, not using Novice is actually a good idea

I love noobs

Then yeah. Add novice

What was the command in f1 to bind 2 values like cl_overlay_entities 0 100

toggle?

Hell yea

17:39

thanks

bind x toggle command 0 100

17:40

+toggle if you want to activate only while holding the key

yea works

17:40

I just forget the command

17:40

Thanks

[ddnet/ddnet] Pull request closed: #3495 Send logout packet after using timeout protection (fixes #3490)

[ddnet/ddnet] Issue closed: #3490 Can't log into rcon after timeout protection used, still shown as logged in

18:08

[ddnet:master] 2 new commits

91b1f62 Send logout packet after using timeout protection - heinrich5991 cdb5a5c Merge #3519 - bors[bot]

18:08

[ddnet/ddnet] Pull request closed: #3519 Send logout packet after using timeout protection

oh no downgradiing to lts fixed it this morning but now after reboot my gpu ports broke again :/ anyone on archlinux with a nvidia gpu who can reproduce this?

[2021-01-17 20:10:28][net]: failed to bind socket with domain 2 and type 1 (98 'Address already in use') [2021-01-17 20:10:28][net]: failed to bind socket with domain 10 and type 1 (98 'Address already in use') What happend here O.o

classic its all over my logs :) everything still works tho

My Srv without connection to the database cant read the _record.dtb files

20:09

why?

because ddnet switched from dtb files to sqlite3

20:28

I wonder if there are any conversion scripts you can probably reuse my dtb to mysql script for that

20:29

https://github.com/lib-crash/lib-teeworlds/blob/master/dtb_to_mysql.sh maybe i should update this to support dtb to sqlite3

lib-crash/lib-teeworlds

Some helpers for the game teeworlds written in bash - lib-crash/lib-teeworlds

oh i think wayland is just sok and has some race condition buggos or whatever xorg masterrace fixed the day

@Trafalgar Law https://github.com/ddnet/ddnet/blob/cdb5a5c28034bc7bba87108ee536f26afe755c5f/scripts/import_file_score.py

DDraceNetwork, a cooperative racing mod of Teeworlds - ddnet/ddnet

21:46

this is the conversion script

21:46

ChillerDragon: it's in the ddnet repository

[ddnet-web:master] 1 new commit

56e7c89 Add Tournament #54 - def-

23:38

[ddnet-web:master] 1 new commit

e4b891f Fix stars - def-

23:43

[ddnet-web:master] 1 new commit

401ca45 this week - def-

23:46

[ddnet-scripts:master] 1 new commit

13b317a Add steam for tournament preparation - def-

[ddnet-web:master] 1 new commit

6e69bbe 3 stars - def-

Exported 639 message(s)