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 2021-06-19 00:00:00Z and 2021-06-20 00:00:00Z

command evaluation is weird

00:40

not always the order you expect

i think they just all execute at the exact same time, and sometimes 1 command might just take longer than the others

02:05

not sure ab that tho

right yea

04:51

sometimes that behavior is good

04:52

cus like kill takes more time apparently so i have to do toggle/kill backwards but it works

Modern browsers are so nice

Image attachment

I wouldn't blame the browser. Websites are just huge. At least my Firefox is usually around 1-2 GB which is ok-ish

what were u doing

09:55

i never ran out of memory

Well 72g is just insane. The browser should never let this much leak no matter what

Oh, I assumed you just had some thousands of tabs open

Ryozuki

what were u doing

Nothing interesting, it was just a firefox instance running for a long time. Just like 2 dozen tabs

about:memory helps I guess

Well the computer was pretty dead at that point so I didnt get to check which website caused it

Learath2

Nothing interesting, it was just a firefox instance running for a long time. Just like 2 dozen tabs

16tabs 3 youtube the rest just text

wut? how much memory do you have?

12 tabs, 1 is twitch which is kinda heave, 500mb only

i'd say memory leak

Chairn

wut? how much memory do you have?

16g, anything more than that it starts swapping

So you got like 60G of swap :D?

Deleted User

i'd say memory leak

I had like 2 userscripts running, I'm guessing one was leaking

for sure

no wonder the thing is slow

i have 13 tabs open but task manager says opera has 32 sub tasks

iirc macOS doesn't limit swap in the traditional way

why do we have chuvash and greek as language files but not in index.txt

12:17

hmm, looks like teeworlds does that too

Try adding it to index.txt, see if it works, make a PR

12:18

I don't know a reason not to include them

12:18

the font might miss some chars?

well teeworlds has chinese included in index for ages before they had a CJK font

12:19

i'll ask teeworlds people too, just to be sure.

12:22

nvm, teeworlds do have them, github search was lying to me

12:22

i'll add them to index.txt

[ddnet/ddnet] Pull request opened: #3920 Add chuvash and greek into language index

Both look ok ingame.

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 with [ASan+UBSan or valgrind's memcheck](https://github.com/ddnet/ddnet/#using...

[ddnet-web:master] 1 new commit

fa45893 Update server stats - def-

13:12

[ddnet-web:master] 1 new commit

52a3019 Remove USA2 - def-

13:14

[ddnet-web:master] 1 new commit

fc645d9 Update year in funding - def-

[ddnet:master] 2 new commits

b1b212c Add chuvash and greek into language index - TsFreddie 95f3969 Merge #3920 - bors[bot]

13:30

[ddnet/ddnet] Pull request closed: #3920 Add chuvash and greek into language index

[ddnet-web:master] 1 new commit

bdabe8a Add 49 € donation by heinrich5991 - def-

heinrich rich

richheinrich

hein rich

@heinrich5991 still can't get them to troubleshoot SSL cert. but i did gave them a ddnet-serverlist-urls.cfg with an http mirror. it worked.

14:50

should we change this to macOS too

SPOILER

Image attachment

14:50

..oops accidentally clicked spolier

Yes, will do

@Learath2 any idea why? tried both steam version and from ddnet.tw/downloads

Image attachment

[ddnet-web:master] 1 new commit

1f5541a Mac OS X -> macOS - def-

TsFreddie

@Learath2 any idea why? tried both steam version and from ddnet.tw/downloads

seems like 15.5 on mac is broken? i don't have my mac right now. can't test it myself. 15.4 worked for him

That would be bad. I don't think we changed anything

15:19

segfault sounds like someone with a mac should build their own version and try with a debugger

[ddnet/ddnet] Issue closed: #3919 DDNet 15.5.2 not compatible with Fcitx 4

TsFreddie

@Learath2 any idea why? tried both steam version and from ddnet.tw/downloads

hm, my mac isn't 15.5 either, I'm not really certain how we debug that

huh?

15:38

you can't debug ddnet 15.5?

ah I'm silly our versioning looks so similar to macOS versioning I thought 15.5 was the newly released macOS version

15:40

I'm up to 15.7 on catalina

Which servers are the ones hosted on Davides vps?

GER3

thanks, I just saw the server names :D

[ddnet-web:master] 1 new commit

a283f75 Livestream by Ryozuki - def-

15:54

[ddnet-web:master] 1 new commit

dadb1cf Livestream by LCSG47 - def-

looks like 90% of generated server stuff is useless

does client use server_data.* ?

@TsFreddie I knew I forgot something. I forgot to disallow HTTP

ddnet

suggesting ddnet

ddnet

back the average time the reason: "median time" always give us time shorter than what we will do really. if he give us 30 mins median time we quickly understand finish this map need more than 30 mins and this useless either " average time" was almost give us time longer but this better than shorter if we know we will finish in under a hour it's so helpful

1

4

troll

1

20:13

banhammer

hussainx3

ddnet

suggesting ddnet

ddnet

back the average time the reason: "median time" always give us time shorter than what we will do really. if he give us 30 mins median time we quickly understand finish this map need more than 30 mins and this useless either " average time" was almost give us time longer but this better than shorter if we know we will finish in under a hour it's so helpful

always give us time shorter than what we will do really

is that just your own impression or have you actively asked other players? for me that time seemed like a better estimate

i see it with my eyes

20:41

last one i remeber was solo map was median time 4 mins

20:41

and was 10 mins not even enough

20:43

better estimate in first finish? or

i saw somewhere median time was more accurate for the average player

1

21:06

i think a good way to do it could also be to only count the first finishes of all players, then take away the extremes but that might be expensive to calculate

1

Its not possible that you can satisy everyone with any algorithm

@jao on $ready, with which function does the ddnet bot find the most recent map file in the channel to use?

Exported 80 message(s)