DDraceNetwork - developer

00:01

why is my ping so high

murpi

I tried the commands but nothing yet, maybe if i can't find the solution for it i'll go for the older versions like you suggested, thank you for the answer! Just to clarify in case it looks like, this is not a problem related to my monitor since i've checked other games and i don't have this issue

He should try updating his computer in first place, even the vms work correctly

how the f did bombay manage to get smooth emotes aswell xD

pressing "run server" in the main menu doesn't open the cfg files, is this a bug?

which cfg files aren't opened? @louis

idk i think autoexec_server

ddnet-maps

[ddnet-maps:master] 1 new commit

9621d82 M Indicud, M Rogue World, M Tropical Island, M SpookyDrag, M New Generation, A Slide 2 - ddnet-maps

def-

[ddnet-web:master] 1 new commit

fb51c39 Server stats update - def-

07:56

def-

[ddnet-web:pr-funding-2021] 1 new commit

6546a1d Update funding for 2021 - def-

07:56

def-

[ddnet/ddnet-web] Pull request closed: #108 Update funding for 2021

07:57

def-

[ddnet-web:master] 2 new commits

6546a1d Update funding for 2021 - def- 86dc58a Merge pull request #108 from ddnet/pr-funding-2021 - def-

07:59

def-

[ddnet-web:master] 1 new commit

dbc4598 Revert "Fully funded! Thanks for all the donors" - def-

@Deleted User i can not reproduce any issues with gfx_resizable 1 neither on debian+gnome nor on arch+gnome both before and after your pr it was fine

10:57

seems like something fixed it already some time ago maybe some new gnome or graphicsdrivers stuff or an older ddnet commit

ok

11:01

ty for testing

For todays aoc puzzle I want a data structure that has constant index time, can grow in both directions and deals with negative indexes. So I can nest it three levels deep and get the game of life playing field. It would be nice if it could automatically grow when I try to access indexes that don't exist yet. E.g. indexes -5 to 5 are existing and I try to access index 10. Then it should initialize indexes 6 to 10 with a default value and return that default value.

11:07

It sounds like you could implement this with a ring buffer that grows resizes automatically when out of space e.g. doubles it's size in that case.

11:08

Like Rusts VecDeque https://doc.rust-lang.org/std/collections/struct.VecDeque.html

std::collections::VecDeque - Rust

API documentation for the Rust VecDeque struct in crate std.

11:08

I wonder if I need to write the rest of the functionality (e.g. negative indexes) myself or if this is a common data structure I just don't know the name of

@timakro Just use a hash set?

11:09

indexed by (Int, Int, Int)

I thought about that but isn't it crazy to nest hash sets three levels deep

11:09

Ooohhh

11:09

Okay why didn't I think about that

nono, a single hash set of course

11:10

with a tuple of 3 ints as the index

Yep that makes sense ^^

heinrich5991

[ddnet/ddnet-web] Pull request opened: #110 Revert "Mention that the last years are already covered"

This reverts commit 92bf8587f601d10c87ec0590de429ba1d931b71a.

@timakro I also made that mistake

13:10

but I did nest hashmaps three levels deep ^^

@heinrich5991 I imagine with rusts .entry(x).or_default() this isn't even too verbose. And from a theoretical standpoint operations on the nested hashmap are still O(1) ^^

I used python with defaultdicts

13:18

so it wasn't verbose for indexing at all

13:18

(it was for summing all the fields though)

In the end I used just a hash set of tuples

yea, someone else also just used a hashset

So summing is just cube.len() ^^

I didn't realize that a hashmap<bool> with default false is just a hashset ^^

There was another puzzle were you needed to iterate over an area like for x in 0..10 { for y in 0..10 { }} I couldn't find a practical rust crate yet which let's me write this as a single loop

itertools?

13:21

https://docs.rs/itertools/0.9.0/itertools/macro.iproduct.html

itertools::iproduct - Rust

API documentation for the Rust iproduct macro in crate itertools.

From the itertools crate there is for (x,y) in (0.10).cartesian_product(0..10) {}

13:22

Omg! Nice exactly what I wanted

13:22

I just looked into the itertools:Itertools trait, I thought that was everything itertools provides

https://artsandculture.google.com/experiment/AAHWrq360NcGbw

Blob Opera - Google Arts & Culture

Create your own opera inspired song with Blob Opera - no music skills required ! A machine learning experiment by David Li in collaboration with Google Art...

13:24

quite cool

Where is the machine learning?

We developed a machine learning model trained on the voices of four opera singers in order to create an engaging experiment for everyone, regardless of musical skills. Tenor, Christian Joel, bass Frederick Tong, mezzo‑soprano Joanna Gamble and soprano Olivia Doutney recorded 16 hours of singing. In the experiment you don’t hear their voices, but the machine learning model’s understanding of what opera singing sounds like, based on what it learnt from them.

13:26

on the ?

@heinrich5991 Can Rusts macros just generate arbitrary rust code?

yes

13:27

but this is a limited macro, using macro_rules!

13:27

proc macros can execute any code on the input

https://docs.rs/typed-html/0.2.2/typed_html/

typed_html - Rust

API documentation for the Rust typed_html crate.

13:27

Hmm somehow those 4 nested loops looked pretty xD

I should not have chosen Standard ML today. The standard library feels like a wasteland and I had to implement everything myself. No sets, no cartesian product, etc: https://github.com/def-/adventofcode-2020/blob/master/day17.sml (too lazy to read in the input too)

huh

13:43

looks like haskell

ML is the predecessor of Haskell

13:44

but Haskell is just so much nicer, the only advantage ML has is that it lacks lazy evaluation which can bite you in Haskell

oh. I thought these were separate branches

Sure, but Haskell is what language researchers created after what they learned from ML

13:45

and no one really uses ML anymore

maybe OCaml would work better?

13:45

I think the original rust compiler was implemented in OCaml

yes, ocaml is definitely more reasonable and actually used nowadays, so it probably has a standard library

def-

[ddnet/ddnet-web] Pull request closed: #110 Revert "Mention that the last years are already covered"

13:49

def-

[ddnet-web:master] 2 new commits

d9d4d34 Revert "Mention that the last years are already covered" - heinrich5991 59fc4cd Merge pull request #110 from heinrich5991/pr_fix_funding2 - def-

Is there anything wrong with a /31 subnet? I thought it was fine for a point2point link

the all 0 and all 1 bit addresses are reserved AFAIK

15:36

so there are 0 usable addresses in that range

RFC3021 seems to have an exception for exactly this

TIL

15:37

https://networkengineering.stackexchange.com/questions/1547/31-point-to-point-bitmasks says yes

/31 Point-to-point bitmasks

When is it appropriate to use a /31 network in production, and is using them considered good practice? On a point-to-point link, broadcasts shouldn't be required, so is there any compelling case for