DDraceNetwork - developer

Could there exist a pair of numbers (a, b) such that a + b causes twos complement overflow but a subsequent - a doesn’t cause overflow?

i don’t think that’s possible

you are pretty much asking if there is a number a where (a + b) - a != b

01:38

the two’s complement part seems irrelevant unless it behaves in a way other than what i know

ddnet-maps

[ddnet-maps:master] 1 new commit

62f3103 A Noobfilter - ddnet-maps

Learath2

Could there exist a pair of numbers (a, b) such that a + b causes twos complement overflow but a subsequent - a doesn’t cause overflow?

what @Ewan is saying. two's complement signed integers behave the same as unsigned integers modulo 2^bits, so you're essentially asking if there are a, b such that (a+b)-a ≠ b (mod 2^bits), which is impossible

Yeah, that's what I thought too, but someone told me there could be one

08:48

Just wasn't sure enough

https://github.com/rust-lang/rfcs/pull/3355

RFC: Start working on a Rust specification by m-ou-se · Pull Reques...

Rendered

Voxel

this was taken from a marvel movie forgot which one did tony stark code in fucking javascript LOL

time travel found in javascript justatest

09:50

Also, would be cool if he did a unique version of javascript in lore (edited)

https://link.springer.com/article/10.1007/s00371-021-02210-6

LE–MSFE–DDNet: a defect detection network based on low-light enhanc...

The Visual Computer - Surface defect detection of industrial products has become a promising area of research. Among the existing defect detection algorithms, most of the CNN-based methods can...

09:52

~~ddnet being academic~~

lol

ConfigEntry {
        description: "At which FPS the videorecorder should record demos.",
        entry_type: Int {
            max: 1000,
            min: 1,
            default: 60,
            value: None,
        },
        flags: CFGFlags(
            SAVE | CLIENT,
        ),
        name: "cl_video_recorder_fps",
        symbol: "ClVideoRecorderFPS",
    },

10:10

pog pog

Can you actually also set a custom resolution ? (edited)

Ryozuki

ConfigEntry {
        description: "At which FPS the videorecorder should record demos.",
        entry_type: Int {
            max: 1000,
            min: 1,
            default: 60,
            value: None,
        },
        flags: CFGFlags(
            SAVE | CLIENT,
        ),
        name: "cl_video_recorder_fps",
        symbol: "ClVideoRecorderFPS",
    },

awesome

Teero

Can you actually also set a custom resolution ? (edited)

in fullscreen i think you can only pick what the backend returns as supported

10:19

which should pretty much be every common resolution at every common fps at every common bit depth

Sad. Would be cool to be able to record at higher res like 4k even if you don't have a monitor like that

10:24

On Linux I can just make myself a 4k desktop at will but on windows this would be a useful feature

https://github.com/edg-l/teeconfig (edited)

GitHub - edg-l/teeconfig

Contribute to edg-l/teeconfig development by creating an account on GitHub.

10:27

wip

10:27

gotta make it load the actual ddnet settings now xd

10:27

oh its private

10:27

sec

ez

10:28

ChillerDragon: ^

10:29

its rly alpha tho

10:29

gotta make the actual binary too xd

10:29

grammar to parse the cpp file xd https://github.com/edg-l/teeconfig/blob/master/src/grammar_cpp.lalrpop

Teero

Sad. Would be cool to be able to record at higher res like 4k even if you don't have a monitor like that

does it not support your resolution

Ewan

does it not support your resolution

it does but I would like a native feature for rendering at higher resolutions than your screen supports (edited)

I don’t think your graphics card would agree to that

10:42

without some refactoring beyond just the resolution settings

10:43

why would you want that, anyway? maybe the demo rendering stuff would be helpful for you

Ewan

why would you want that, anyway? maybe the demo rendering stuff would be helpful for you

yes i ment only for demo rendering. not for usual gameplay

@Ryozuki safe code

that’s how i write bitfield enums idk about you lol

1 << n

Ravie

@Ryozuki safe code

its safe

both is fine. in the code snippet its a bit awkward because the numbers aren't aligned and the last entry suddenly messes up the amount of digits

Ryozuki

its safe

prone to human error tho especially if u don't align it

its still safe