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 2022-07-20 00:00:00Z and 2022-07-21 00:00:00Z

New c++? https://github.com/carbon-language/carbon-lang

GitHub - carbon-language/carbon-lang: Carbon language specification...

Carbon language specification and documentation. Contribute to carbon-language/carbon-lang development by creating an account on GitHub.

00:54

seems sorta like Rust-lite designed for existing cpp code bases https://github.com/carbon-language/carbon-lang/blob/trunk/docs/project/faq.md#why-not-rust (edited)

@deen maybe its time to update pts with vk? https://www.phoronix.com/scan.php?page=news_item&px=Zink-Lazy-Descriptors-Default Then he can test zink vs native vk xdd

Zink OpenGL-On-Vulkan Ready To Switch To Lazy Descriptors - Big Per...

Mesa's Zink OpenGL-on-Vulkan implementation is finally ready to enable lazy descriptors by default, which in turn will mean better performance for this generic OpenGL implementation across many games/applications.

What's the purpose of the 'use clipping' feature on map groups?

Jupstar ✪

@deen maybe its time to update pts with vk? https://www.phoronix.com/scan.php?page=news_item&px=Zink-Lazy-Descriptors-Default Then he can test zink vs native vk xdd

true, thought about that too

have we ever done profile guided optimization?

07:55

maybe we can get some perf

Tried everything. Ofast best option for tw

08:26

Rest only few percentage

k2d222

What's the purpose of the 'use clipping' feature on map groups?

-Ofast is dangerous, it lets the compiler mess up float calculations, probably changing gameplay

08:26

wrong reply

k2d222

What's the purpose of the 'use clipping' feature on map groups?

clipping lets you show the layer only when the player is in certain positions

heinrich5991

-Ofast is dangerous, it lets the compiler mess up float calculations, probably changing gameplay

Tw uses quantinized floats tho

08:28

It's also not dangerous. It's simply not standard conform

yes, it's not standard conformant

08:28

yes, but we had compiler options change gameplay before

08:29

so having quantized floats is apparently not enough to work against it

08:29

https://github.com/ddnet/ddnet/blob/5daa0f96f593460a594ba237646b3119d5b6c4c1/CMakeLists.txt#L303-L308

Yes i bet vsc compiles different float than gcc too

probably not, on x64

08:30

I'd guess they're abiding by IEEE, hopefully with the same rounding mode

08:30

IEEE 754

I think bsd also has a forced difference

08:31

Smth about storing redults in registers

this doesn't matter anymore in x64 AFAIK because you can use the IEEE 754 compliant sse registers and not the weird stuff people use on x86

08:32

so compilers just use that AFAIK

Anyway only server accuarcy matters more . the 2 edges u have client side prediction almost dont natter

08:33

I use it sincr 3-4 years

08:33

Never seen a problem. Runs smooth af

perhaps one could find out which part of -Ofast gives the benefit

Probs sqrt and loop hard unrools

-Ofast on gcc apparently additionally enables -ffast-math, -fallow-store-data-races, -fno-protect-parens and disables -fsemantic-interposition

08:47

I don't understand what -fallow-store-data-races does precisely from the docs

looks like people even complain about it & asks about improving the documentation

08:49

Allow the compiler to perform optimizations that may introduce new data races 
on stores, without proving that the variable cannot be concurrently accessed 
by other threads. Does not affect optimization of local data. It is safe to 
use this option if it is known that global data will not be accessed by 
multiple threads. 

Examples of optimizations enabled by -fallow-store-data-races include 
hoisting or if-conversions that may cause a value that was already in memory 
to be re-written with that same value. Such re-writing is safe in a single 
threaded context but may be unsafe in a multi-threaded context. Note that on 
some processors, if-conversions may be  required in order to enable 
vectorization.

yes, that's what I found as well (it's the result of the documentation bug ^^)

yes, "Suggested improvement" (edited)

Tater

New c++? https://github.com/carbon-language/carbon-lang

how many programming languages does Google wants to make

google using cloudflare's "quiche" (quic library): https://security.googleblog.com/2022/07/dns-over-http3-in-android.html

DNS-over-HTTP/3 in Android

Posted by Matthew Maurer and Mike Yu, Android team To help keep Android users’ DNS queries private, Android supports encrypted DNS. I...

@Jupstar ✪ wrote a script too see if the 'bounding box' optimization for tilemaps would help (if you have a better name, go ahead :p) bounding box here is the smallest rectangle around all non-zero-id tiles in the layer to measure the optimization, I took the bounding box of the game layer to approximate the area in which the gameplay happens then, for each tile layer I calculate size_of_tiles_layer_bounding_box / size_of_game_layer_bounding box to get the 'coverage' of the tiles layer on the gameplay area I then average that over all the tiles layers of the map to get the 'coverage' of the map then averaged again over all maps to get the total average coverage

10:37

(I disregarded clipping, parallax and everything else for this)

10:37

tilemap_bounding_box.py

1.41 KB

10:37

94.59 KB

10:38

Total coverage: 63% (at the end of the output)

10:39

I think this indicates that this optimization could improve the performance significantly

10:41

feel free to look over the code though, I'm not 100% sure it all makes sense

But also only if the user views the empty pixels. But yeah might help in some maps.. Is that per layer? Are bigger layers weighted more?

11:04

If u have a 2x2 layer anf its filled with 1 tile the coverage is 25%

per layer, averaged for the map all layers are weighted the same

rendering videos is bugged on 16.2.2? Im having issues :/

ah damn

11:05

you are right, but maybe its still fine as an approximation, I don't think that is something regular in maps

Yrah probs

Lagar

rendering videos is bugged on 16.2.2? Im having issues :/

More precise pls

well I also don't take positioning into account so its hard to fix that easily in the calculation I think, coz for that I'd need to somehow include the 'outer-tile-repeats' logic somwhow

11:07

ah, might actually be possible

this is the video im getting

11:08

333.13 KB

11:08

but demo is working good

I check if the rightmost column is filled, if yes, set max_x to the siez of the game layer

11:09

hm that will still not take offsets into account

11:09

eh maybe I just leave it at that

Lagar

this is the video im getting

i fixed it setting windowed screen before render, with windowed bordeless im having issues although i never had an issue, idk

Maybe buggy driver

the dma was passed \o/

12:41

users will be able to install any software they like on apple's hardware

poggers

[ddnet/ddnet] Issue opened: #5645 Using `play` to play demo files doesn't work when first launching the client

When launching the client with DDNet "play demos/my.demo" command line it doesn't play the demo when the client finishes loading and instead the following error is shown in the console: [demo_player]: playing passed demo file 'play demos/my.demo' failed: error opening demo file If you type the same command play demos/my.demo into the console it plays the demo. Would be nice if the play command also works on the initial start for integration/fuzz testing.

[ddnet/ddnet] Issue opened: #5646 Various ASan and UBSan errors when playing demos with fuzzed chunk data

``` src/game/client/components/chat.cpp:795:89: runtime error: index -1 out of bounds for type 'CClientData [64]' #0 0x55c133fd047a in CChat::AddLine(int, int, char const) src/game/client/components/chat.cpp:795 #1 0x55c133fc80b5 in CChat::OnMessage(int, void) src/game/client/components/chat.cpp:570 #2 0x55c134677688 in CGameClient::OnMessage(int, CUnpacker, int, bool) src/game/client/gameclient.cpp:770 #3 0x55c133c282d2 in CClient::OnDemoPlayerMessage(void, int) src/...

@Jupstar ✪ is there a reason why I shouldn't use vertex buffers as uniform buffers? not sure what my thought process was there, but its what I did a few times

https://developer.nvidia.com/blog/vulkan-dos-donts/

21:37

hm seems like a nice resource

ah right, I mean using instance-changing vertex buffers

[ddnet:master] 3 new commits

8e203cb - dannyhpy 4812339 - dannyhpy a93dac5 Merge #5644 - bors[bot]

22:31

[ddnet/ddnet] Pull request closed: #5644 Add Esperanto language

Exported 75 message(s)