🤖│community_dev

[bright valley]

Ive already seen snapkey being sold as hacks ;-;

[quick bough]

If it's not analogue, I don't care 😛

[strong siren]

bare win32 cpp :p

[bright valley]

I wonder if it's bannable though with vanguard

[quick bough]

also already did a thingy like this and immediately let it become unmaintained: https://github.com/Sainan/analogue-socd-resolution

[bright valley]

Couldn't you just put a qmk script into keychron to do it that way?

Idk if it allows you to modify it or not

[quick bough]

the funny thing about that rappy snappy is basically the same as what most games do when they receive analogue input (subtract)

[quick bough]

Well, technically this could now also support Drunkdeer keyboards, except... unmaintained. 😄

The one downside of not using the Wooting Analog SDK... gotta recompile for new keyboards

[bright valley]

rip

[quick bough]

Just compile from source I guess 😛

[bright valley]

lol ig

[quick bough]

yeah the thing is I just don't care about SOCD like at all

don't play FPS games, so...

pressing A and D for me means: vehicle goes straight

and that makes sense

(both for digital and analogue input)

[quick bough]

Okay, I decided to do the not lazy thing and recompiled it and pushed a new release

tho I didn't do the ultra-not-lazy thing of actually testing if it works

it's probably fine tho

[bright valley]

famous last words

of a cpp developer

[bright valley]

ah lmao

[quick bough]

Why C++ specifically?

[bright valley]

idk i feel like writing something in python is harder to mess up

than in cpp

[quick bough]

If there's a bug, I can quickly test and fix it. I'm guessing no one cares tho and hence won't run into issues

[bright valley]

lol maybeee

[quick bough]

That depends. In Python, you will only find out at runtime if you messed up

[bright valley]

you just gotta get better and use cython

[quick bough]

Eh, I don't use Python myself at all, just infering from how scripted languages be

not that C++'s compilers will prevent you from writing horrible code, but at least I can be confident I don't have horrible mistakes

[bright valley]

I'm not there with writing c++ code etc.

still pretty early in the writing outrageously bad code lmao

[quick bough]

What's your native language?

[bright valley]

None really tbh, I know java node etc. but I don't really care to use them as much, so I'm in a all out learning random languages and not getting good enough in any of them

[quick bough]

HUMAN language

[bright valley]

ew whats that

[bright valley]

english

why lmao?

[quick bough]

Okay, I think you have autism

[bright valley]

It doesn't exist if you don't acknowledge it

frfr 100% guranteed

[quick bough]

You misunderstood me twice now on very common phrasings

[bright valley]

that's very subjective ¯_(ツ)_/¯

[quick bough]

Well, 3 times if we count this

[bright valley]

You mistook me on very common answers

[bright valley]

human language is very subjective anything is a language if you set your mind to it

[flat wigeon]

Sure, now it requires behavioral analysis to be detected

[quick bough]

I mean, at first I thought you could just use the analogue interface of the keyboard, but there's actually an even easier solution: Just check how many times the person is perfectly A/D swapping.

Like, no human could ever get a frame-perfect A/D swap like 3 times in a row without some sort of assistance

[bright valley]

or can they

insert vsauce music here

[vital raptor]

Hiya!
I've been working for the past few weeks on a bit of software for Wooting keyboards. It's a spectrum analyser (something that takes audio and converts it into a graph of frequencies), making the LEDs on the keyboard react to the music you're listening to, which creates a nice effect.

It works really well for my setup and I'm really happy with how it looks, but I can only test what I have, so it would be nice if some people could also try it out and see how it goes. Hopefully it works for most Windows setups, so I can start the attempt to bring it to Linux.

And all feedback is very welcome, I'm hoping to get this into a very usable state ^^

Anyway here's a small showcase video and the GitHub repo, for those who want to check it out:
https://github.com/PrimmR/wooting-spectro

[lusty dagger]

looks great 🙂

[vital raptor]

thank you!!

[paper crow]

That's super cool 🤩

[serene lynx]

Im definitely gonna try this on my twohe later

[vital raptor]

fantastic! fingers crossed that it works on that model ^^

[wide skiff]

That is super cool, I wanted to do that on a board since forever!

[serene lynx]

updating you here, works flawlessly on the twohe!

[vital raptor]

brilliant!!

[serene lynx]

Its absolutely stunning stuff, awesome work!

[quick bough]

[karmic thorn]

I like this 🙂

[unreal pasture]

WHOEVER made that usb beta fireware its a good step in the right diection

[molten breach]

What is that?

[bright valley]

It’s beta firmware for usb

[vital raptor]

Nicee!!
Do all the themes seem okay? I had to modify a few to work with 6 rows instead of 5, but hopefully I didn't miss something ^^

[quick bough]

Everything works flawlessly unless a bug report proves otherwise 😛

[vital raptor]

So true!!

[serene lynx]

Themes looked just fine on my end

[vital raptor]

:DD

[quick bough]

Apparently CS2 devs added a detection for snap tap to the game

[balmy iron]

Unsurprising, valve does like tracking all the stats

[quick bough]

At this point I don't see why they don't just add a "SOCD resolution" option to the game, although that would be like shipping cheats with the game

Otherwise I guess they could add a warning message like "Hey, it looks like you might be using an input method that is resolving SOCD before reporting it to the game, which we consider to be cheating. Stop this or you will be banned."

Either way, they gotta level the playing field somehow, lol

[midnight frigate]

That pink is gorgeous!

[vital raptor]

Hehe thanks!

[mellow copper]

Hello all ! This is just a quick question, I know that it is somewhat alluded to in the FAQ section, but I wanted to ask, I have been looking at the 2HE model to get, but I am “concerned” that I will get it and then a new revamped model will be released or put into production, leaving the keyboard I just got somewhat behind, whether in physical or software capability. Are able to have any sort of insight for that ? Thank you in advance !

[unique goblet]

this is a shot in the dark but does anyone know if decoupling capacitors are actually needed for the linear hall sensors on the wooting? i noticed the back of the wooting 60he doesn't have any but most of the datasheets ive seen have them included in the recommended implementation

[strong siren]

I don't know when it comes to hardware, but the software side of things should be fine. We're still supporting the original wooting one from 7 years ago.

[floral crane]

Is wooting firmware encrypted and/or signed?

[quick bough]

Wooting's legal team preparing to sue you

[sudden jungle]

I imagine the only hardware upgrade would be 8kHz polling but I’m not sure it would have much of a noticeable impact

[quiet root]

so software wise we havent left any of our devices behind yet

even our wooting one gets updates with new features as long as the limitations of its hardware allow it

also this doesnt belong in here but more #💬│general or #1141096226884956181

[quick bough]

When are we getting a TKL with HE keys

also the linear keys were kinda based, the sound profile was nice

[granite ore]

@vital raptor

[granite ore]

This happened when I ran cargo run --release

And I cannot find the executable

[livid trellis]

Hopefully they start banning it.

[deep light]

is there a way to have multiple actions while holding down a singular key depending on the amount of time its held down?

i use shift for up arrow mod tap

i want to be able to hold shift for it to function as shift, hold for longer to function as up arrow pressed down

200ms for shift, 500ms for up arrow

[vital raptor]

compilation probably failed for some reason, was there an error in the terminal

[granite ore]

Other than what I showed above no.

If needed I can screenshare with you and show you what I do and what it throws at me.

[balmy iron]

You need to install libclang and friends, and make sure cargo can find it. As to how you do such things on Windows? No idea!

Apparently it's in winget so should be quite easy (famous last words) https://rust-lang.github.io/rust-bindgen/requirements.html#installing-clang

[vital raptor]

aah thanks for finding that!
i already had Mingw64 before starting so I didn't notice it was needed

[granite ore]

Will you update the build then or?

[vital raptor]

Nono, that's something you need to install

[granite ore]

How can I install it

[vital raptor]

Because the actual wooting library is in C, it needs certain C libraries to install, even though my program is in Rust

[vital raptor]

My recommendation would be to install winget and then use the command given here

[balmy iron]

Winget should be preinstalled these days, unless maybe you've run a debloater script that removes it

[vital raptor]

oh fantastic!

[cyan saddle]

Is there a conical way to get the keys for a keyboard on the RGB and analog APIs?

[quick bough]

conical?

[cyan saddle]

Like a correct and supported way of doing it. Getting the count, locations and labels of the keys.

[quick bough]

Well, not really, no. The firmware only provides the keycodes as mapped in the Wootility (for analogue and digital input), whereas the RGB interface is positional.

Of course, this isn't really an issue if we assume a not-keypad (like the UwU) and a more-or-less standard layout modulo some differences with ANSI vs ISO.

[cyan saddle]

The goofy website for managing the devices knows all of this stuff, it really should be exposed somewhere so you can have an accurate respresentation of the devices in your programs that support them.

[quick bough]

You can definitely read this stuff out yourself. I already did some light reversing on this front: https://gist.github.com/Sainan/0d285907961a22c8cafc106aa36513e5

[cyan saddle]

I could definitly dig into that stuff, maybe I should.

[quick bough]

The RGB SDK is definitely a good place to start if you wanna look at how you talk to the keyboard.

[cyan saddle]

I just want to be able to have a correct representation of the device in software. :V

This was a discussion years ago and it was never resolved and now it's more important because there are more than two keyboards.

[quick bough]

There is still the issue tho that you won't know which key is pressed if the same logical key is mapped on multiple physical keys. (Unlikely but possible.)

[cyan saddle]

Do the mappings of the keys matter if you are grabbing the raw values from the board?

[quick bough]

The analogue interface reports keycodes (logical keys).

[cyan saddle]

/** @brief Retrieve information about the connected Device

This function returns a pointer to a struct which provides various relevant
details about the currently connected device. E.g. max rgb rows, columns, etc

@ingroup API

@returns
This functions returns a pointer to a `WOOTING_USB_META` struct which contains
relevant Device Information
*/
WOOTINGRGBSDK_API const WOOTING_USB_META *wooting_rgb_device_info(void);

Noice.

That helps a lot.

[quiet root]

so count you can get with the sdks, layout as well but labels are completely irrelevant for both rgb and analog

the digital keymap is a separate thing that wootility just renders on the keys regardless

[cyan saddle]

It would be nice from a GUI standpoint to have them.

[quiet root]

you can retrieve the keymap

[quick bough]

layout as well
Well, it can tell you if it's ANSI or ISO, but not the full mapping.

At least I don't see any such thing in the RGB SDK

[quiet root]

the key layout is in one of the sdks

[cyan saddle]

Eh, I'll figure it out when I actually add the GUI to my game.

I have a couple of big blocking things in my engine for right now, so I'll just stick with my wrapper and some test cases for it.

[quick bough]

Trying to name keys is a problem you can only despair at

The best you can do in this regard is ask the OS if it can maybe be helpful with regards to the user's current layout

[flat wigeon]

Well it's a typical bug in games where the input should be bound to a position on the keyboard that the mapping depends on the current layout instead

[quick bough]

I don't think that's a bug per-se

I mean, yeah, if you set some default bind on Y and/or Z, then some European users may have a different experience trying to hit those keys, but you can account for that by changing the defaults based on the layout, for example.

Or you just always treat input as if being mapped in a QWERTY layout

[bright valley]

imo we should all just throw away qwerty to dvorak

[quick bough]

Very cool opinion that totally doesn't require billions of people to relearn how to use a keyboard

I switched from QWERTZ to QWERTY and that alone was a rough transition

[bright valley]

average people don't type without looking at their keyboards, so it'll be more like the 10% that'll have the hardest time transitioning

most of them do the whole find and click typa action

• if we continue using qwerty, itll make billions future people also adopt it

and rather than only having our current billion people, itll be even more people now using it

[quick bough]

I get the sentiment, but you'll have to sell people who never typed before on this first

[bright valley]

if more people who type faster adopt dvorak, and just make it a standard of typing fast use this layout it'll eventually get to everyone

like your average IT guy

coming from someone who can type 210wpm on qwerty

[quick bough]

I highly doubt this is true, especially your figure of 90% of people not remembering where keys are. This just comes naturally with usage, and I think most people use keyboards enough that they could at least blindly write their name.

[bright valley]

Doubt it, you underestimate phones

[quick bough]

I think WPM is more of a vanity metric than you realise. I myself am a professional software developer, very successful, but I'm only at like 100WPM, I don't even use 10 fingers to type.

[quick bough]

Well, I still do look where I type on my phone because smartphones do just suck, but I know where the key is, I don't have to search for it.

[bright valley]

I have had 160+ wpm, ever since I've had a keyboard. It's pretty natural for me to type that fast I never really practiced to become faster at typing ¯_(ツ)_/¯

Idrk how someone can't type faster, to me its just moving your fingers faster

[bright valley]

Software developers don't need 100+ wpm, while thinking you reaching over 140wpm is pretty low

[quick bough]

It's about typing at a pace you can maintain for like 16 hours, lol

[bright valley]

I

I'm a professional ctrl c + ctrl v typer

[quick bough]

I do think faster than I type sometimes, but it's more of a lockstep-process of "think -> type -> think -> ..."

[bright valley]

mhm which is why I said over 140

:D

plus intellisense and language server

and add copilot into it

[quick bough]

I use none of that crap

[bright valley]

Sucks to be you then ig

[quick bough]

Why

IntelliSense knows less C++ than I do

[bright valley]

I'm not typing system.out.println

5000x times

ill much rather type sout

[quick bough]

Copilot spits incorrect nonsense like 90% of the time

Just use macro software to setup sout as a shortcut for Sytem.out.println

[bright valley]

It's dependent per language, I mainly was saying to use it as a way to autofill what your typing

[bright valley]

intellisense

[quick bough]

I mean, any decent editor will like autocomplete stuff for you

[bright valley]

yeah

and you said you don't use any of that

;-;

[quick bough]

I didn't say I use notepad.exe

I just say I don't use any tooling that knows the code less than me

[bright valley]

Who ever the hell blocked . needs to get banned

[quick bough]

IntelliSense literally errors on perfectly valid C++

[bright valley]

Change it to fit your style of valid

[quick bough]

I mean it literally doesn't recognise syntax that's in the standard

It just errors because it doesn't know C++ lmfao

[bright valley]

what ide do you use?

[quick bough]

Visual Studio

You say IntelliSense

IntelliSense is a thing in Visual Studio where it parses the code and spits out incorrect nonsense 😄

[bright valley]

I use CLion

Soooo

¯_(ツ)_/¯

[quick bough]

CLion seemed so useless to me

[bright valley]

Visual Studio looks ugly

atp id just use nvim

[karmic forum]

Yeah all of their products seem to specific for me to have to use so many

[quick bough]

Trust me, I do not use Visual Studio for the looks, lol

or even because it's a good editor

[karmic forum]

I was interested in rustrover for a bit because i heard it was pretty good

[quick bough]

I use it because it's how you maintain medium-large C++ projects on Windows

[bright valley]

They have a new one thats ment for everything, its called Fleet

[bright valley]

I use rustrover daily as well

Fleet is a visual studio code competitor

and clion is a visual studio competitor

[karmic forum]

I might look at it at some point idk dont really have a reason to though you know

[quick bough]

Fleet is just like VS Code, so not really an IDE

[bright valley]

imo vscode is much better

[quick bough]

(also actual malware btw unless you don't value your file assocations)

[bright valley]

having vscode server on a remote server

[bright valley]

use linux

[quick bough]

I prefer Sublime Text, not M$'s knock-off

[bright valley]

ew

sublime text sucks, vscode has much better extensions

[karmic forum]

Oh does it associate like a stupid amount of file types to it by default?

[quick bough]

Sublime Text is snappy, VS Code is not

[bright valley]

its optional

[bright valley]

nvim is snappier

sublime text is also paid

[quick bough]

I don't care about extensions because ST actually works out of the box 😄

[bright valley]

? sublime text doesn't have nearly any features compared to vscode

[quick bough]

It's free to use

[bright valley]

annoying popups

[quick bough]

You can afford a licence if you're even a half-successful software dev lol

[bright valley]

Never said you couldn't

Just said it is paid

and you said its not

it is paid

[quick bough]

That's subjective

[bright valley]

not really

objectivelly vscode has more features than sublime text

[quick bough]

It's got a lot of bloat, it's very slow, and it doesn't do anything particularly well

[bright valley]

Thats subjective

vscode barebones has almost nothing, and its opensource you can remove anything you want

and only install the things you'd like to add

[quick bough]

Sublime Text is much more focused and thus does those things a lot better

[bright valley]

use Nvim

its more focused

and does things a lot more better

[quick bough]

I don't know nvim

[bright valley]

Everything has a first

[quick bough]

Doesn't run natively on Windows apparently

[bright valley]

?

https://neovim.io/

https://www.lazyvim.org/

[quick bough]

Ah, they hid it, lol

Umm, is this it?

It's literally just vim

[bright valley]

Yeah

That's it

Look at how much more focused it is

[quick bough]

Somehow even worse than vim

[bright valley]

?

[quick bough]

Can't do :q, have to do :q!

[bright valley]

you can change that

It's in the config file you have to make your own keybindings for everything and install every feature you'd like

[quick bough]

Well, I can appreciate vim, but it's not something I personally find is worth the time to properly learn

[bright valley]

¯_(ツ)_/¯

you wanted focused and snappy

you don't want to use the focused and snappy solution

[bright valley]

lazyvim is a preconfigured one if you do get into it

its not as good as making your own profile

but its smt ig

[quick bough]

I can write my code, then I press Ctrl+B and it runs

Sublime Text gets this

Visual Studio gets this

VS Code doesn't get it

[bright valley]

change keybinding

[quick bough]

that's not even the problem lol

VS Code has no concept of build systems

[quiet root]

vscode has a general task system which can definitely build

[quick bough]

Yeah, I looked into "tasks", it's a pretty half-assed system, imo

and then the hotkey is Ctrl+K+B or something like that

[quiet root]

idk how a system as flexible as this can really be half assed but ok

its ctrl+shift+b by default but can be easily overridden

{
    "key": "ctrl+b",
    "command": "commandhere",
    "args": "argshere"
}

done

[quick bough]

Well, my only experience with VS Code is making an extension to support our language, and there was no way for an extension to add a "task"

We can only add a "command" and we have to write JS for that (cringe because language support is otherwise entirely data-driven)

[quiet root]

that could be but i rarely work on my own extensions nor would i want extensions to touch tasks

[quick bough]

So, I can do Ctrl+Shift+P and select this

Then I get this awful output here

[quiet root]

and you can stuff those commands into tasks as well

[quick bough]

I mean, it certainly works, but wow

[bright valley]

I think you're a bit ignorant to other solutions, so you are stuck in the sublime is good, because that's all you're used to/know about

[quick bough]

Bro, I literally made an extension for VS Code

(and am still actively maintaining it)

[bright valley]

https://github.com/Sainan/fivem-vscode?

? that one

[minor flint]

Having a problem with the SDK. I’ve followed the instructions on the GitHub for both Rust and C++ but get “undefined reference” errors, which makes it seem like the sdk isn’t visible to the program. I have the wrappers properly configured and SDK fully installed. Probably a newbie mistake somewhere. Can anyone offer some wisdom?

#include "wooting/wrapper/includes-cpp/wooting-analog-wrapper.h"
int main() {
    int x = wooting_analog_initialise();
    return 0;
}

[quick bough]

Well, I've made more than 1, but the one I'm still actively maintaining is https://github.com/PlutoLang/Syntax-Highlighting

Point being, it's not a matter of me not having used VS Code enough

Heck, I can probably even appreciate it for some TS codebases, but I'd argue that's an issue on the maintainers for having shitty ESlint rules with so many false-positives that it's a pain to work on that codebase without essentially an IDE.

[bright valley]

Isn’t that for sublime?

[quick bough]

This is for sublime, vs code, and visual studio. Language support is mostly standardised on the TextMate formats, which can luckily be shared, but there is some stuff specific for each editor, too.

[quick bough]

VAC source code leaked

Funny thing is that even at this ~1khz poll rate I can see some false-positives

This would be a good way to check if someone is legit, because this happens constantly with legit input, less often with rappy snappy, and not at all with last input priority/snap tap.

[balmy iron]

I doubt they would just poll keyboard state like that since keyboard and mouse input is event-based

[quick bough]

I mean, games are tick based, so they will have something like this already in place, just at a slower poll rate (tick rate)

[balmy iron]

Yes but the inputs come in as events so no need to run something at a high tick rate or whatever, just check on keydown whether the other key is also down

On the other hand, with how clueless the janitor working on CS2 is, who knows

[quick bough]

Well, alternatively you could check when the input event comes in, if the other key is still down

And maybe check when the last event from that other key came in

Same idea, just different implementation for an event-based input system

[flat wigeon]

In my opinion it's cool to have such features... but not baked into the firmware. And an input API that is capable enough to flag such behavior as legit or cheat

[minor flint]

Can anyone help with this?

[quick bough]

You need to link against the library

The .lib file should be with the download

I'm not sure if that was for the DLL or static lib tho

Just checked, it's for the static lib

Here, this works:

#include <cstdio>

extern "C"
{
    #include "wooting-analog-wrapper.h"
}
#pragma comment(lib, "wooting_analog_wrapper.lib") // we are linking in a Rust static library
#pragma comment(lib, "Userenv.lib") // so there's
#pragma comment(lib, "ntdll.lib") // just a few
#pragma comment(lib, "Bcrypt.lib") // system libraries
#pragma comment(lib, "Ws2_32.lib") // that we'll need
#pragma comment(lib, "Advapi32.lib") // as well :)

int main()
{
    printf("wooting_analog_initialise returned %d\n", wooting_analog_initialise());
    return 0;
}

Ah yeah, this fun stuff

Freebie:

// Prevent Wooting Analog SDK writing to our console.
SetEnvironmentVariableA("RUST_LOG", "off");

I'm confused why cbindgen doesn't produce extern "C" by itself tho

Should be very easy with #ifdef __cplusplus or so

Even Lua does it and Lua targets C89

Apparently it needs a cpp_compat flag set to true in some TOML file that doesn't exist in the SDK project and the default is false

[flat wigeon]

Imagine changing the meanig of a key depending on the press speed

[quick bough]

No, please, no

[quiet root]

every few years

like clockwork

[quick bough]

I can imagine a difference between half-pressed and full pressed... but even that would require specialised switches (similar to the adjustable triggers on the DS5), otherwise it would be very difficult to input precisely

[quiet root]

this whole speed based events (different keys or such) and setting os volume by press depth come up way too often

and the os volume one is highly impracticle

[quick bough]

I don't think I ever saw this idea

but hey, you have analogue keys, might as well use them for slider-ish input 😛

omfg, I forgot I enabled rapid trigger on some of my keys

I thought the switches were broken lmfao

[flat wigeon]

Do they? Also I'm not interested in Volume control anymore because I have a dedicated slider for that anyway.

[quiet root]

oh yeah in the past people have asked this as well

they made their own code

and quickly abandoned it after

im not even sure if we still have the channels archived where those repos were posted

[flat wigeon]

Hahaha, the volume control part has little impact anyway

[quiet root]

i mean both the velocity actions and volume stuff

[quick bough]

I did colour input on a DS4 by using the 2 triggers for hue & value input, and it's about as usable as you'd expect

[flat wigeon]

Regarding the velocity I'm not really sure how to handle this yet

[quick bough]

on_tick:
velocity = pos - prev_pos

[flat wigeon]

Well it's not so simple

It needs to be practical

[quick bough]

or just measure the time between first non-0 value reported and "full" value reported

that should be a pretty consistent way to determine if a key is going down fast or slow

[flat wigeon]

Basically I don't want a short/long press

But press depth alone feels unreliable

[quick bough]

How so?

[flat wigeon]

How precise do you reach a certain depth and not more?

[quick bough]

With stock keys, not at all probably

[minor flint]

This worked in VS 2022, thanks a bunch. Makes me wonder why the system libraries aren’t mentioned in the readme🤔

[quick bough]

I mean, you'll notice from the linker errors

It depends somewhat on your development setup, so which system libraries are already included by default

and the Wooting guys didn't even realise that you need a .lib for both the static library and the DLL, so... don't expect an insane amount of foresight on that front 😄

[flat wigeon]

advapi32.lib is unnecessary since it's linked in per default

[quick bough]

I should mention when I collected those headers I was not using VS and instead basically invoking clang myself, and I only pass -luser32 -lgdi32 by default

Nowadays I do prefer #pragma comment(lib, ...) over the linker flag tho since it's more organised

Yeah, seems like VS includes these here by default

[regal fractal]

How can I change Keyboard profile without starting the software or website?

[balmy iron]

Press the A1, A2, A3, or mode keys. Check your keyboard mapping for what they are

[merry olive]

hey does anyone know of a programmatic way of changing key(code) mapping? some code examples or reference will be appreciated. i cannot find any "ioctl" or "write" to devices in wooting's github repos

so i want to assign a key to a special command (i.e. sending some web requests) not using macros. my plan is to assign the key to some reserved keycode and intercept on the hid level

[quick bough]

You will have to reverse-engineer the Wootility

I've already done some work on this: https://gist.github.com/Sainan/0d285907961a22c8cafc106aa36513e5

So I can already tell you you won't be able to assign custom keycodes

but you can use niche ones like F13-F15

[merry olive]

thanks. will do

[merry olive]

but how do they handle "SWITCH_TO_DIGITAL_PROFILE" things?

i don't think it's a standard keycode?

😄will know soon

[quick bough]

That's a keycode that doesn't have to be sent to the OS, it can be handled directly in the firmware

[merry olive]

yeah i mean no special driver is installed, so how?

[quick bough]

It will depend also on your firmware version I think. Be careful, you might need to factory reset the keyboard when messing with this raw stuff.

[quick bough]

The firmware is the code that runs directly on the keyboard

[merry olive]

it also sends something to the os while, say, toggle between analog and digital

[quick bough]

I don't think it does

[quick bough]

It might report that keycode to the analogue interface, but definitely not on the standard keyboard page

[merry olive]

this is something i got when pressed D1

[quick bough]

Yeah, that looks like an analogue report

[merry olive]

i see

[quick bough]

You can use the analogue SDK to read those reports, which would include these custom keys

[merry olive]

the sdk does not map all keys so i guess i'll need to figure out how to assign a key to send a special report

[balmy iron]

Wouldn't it be easier to just use bog standard keys and then exclusively open the device? Although this might be a neat trick that's only easy to do on Linux

[quick bough]

Wdym? The SDK simply gives you back whatever the device sends, so this includes these custom keys.

[merry olive]

simply want to doing it in the hard way

[quick bough]

What exactly are you trying to accomplish anyway?

Because I can already tell you that you will not get custom keycodes

Given that you mentioned the UwU, I can only recommend using F13, F14 & F15

These are very, very unlikely to interfere with anything

And you can read them to cleanly tell what you're pressing on the UwU

[merry olive]

https://github.com/WootingKb/wooting-analog-sdk/blob/8a74d30818baeb35f43499fe3786bf55162d3a32/wooting-analog-sdk/src/keycode.rs#L7

e.g. keycode 4 is mapped to scancode 0x001e which both represents A, but like 0xb6 which is "switch next song" is not mapped here

[merry olive]

so what i'm trying to do is to find out how to let my key send [4, 6]

or any uint16

[quick bough]

Guess what, you can simply not use this map?

https://github.com/WootingKb/wooting-analog-sdk/blob/develop/SDK_USAGE.md#keycodes

By default, it reports the raw HID codes that the keyboard reports

Without any remapping

[merry olive]

i know

[quick bough]

You can't let it send just any u16, it must be one of the keys coded into the firmware

[merry olive]

have you tried?

[quick bough]

I'm gonna murder you Feel free to try it, and let me know what you find. 🙂

[merry olive]

of course. thanks for the help

[quick bough]

To whom it may concern: I made a website that can show analogue input from Wooting keyboards & co using WebHID: https://analogsense.org/JavaScript-SDK/demo

[paper crow]

Nice!

[minor flint]

I noticed that the SDK cuts values off at 10%. Is there a way to disable this and get all raw values?

Also the keys seem to be in the wrong place, scancode 0x10 (q) is at m, etc

[quick bough]

Could you share the code you're using to interface with it?

[minor flint]

It's just

while (true) {
  float q = wooting_analog_read_analog(0x10);
  std::cout << "Q: " << q;
  std::this_thread::sleep_for(std::chrono::milliseconds(1));
}

Changing polling rate doesn't seem to affect the values

[quick bough]

How do you init it?

Although, just double-checkig, 0x10 is indeed the HID scancode for 'm'

So, not sure what exactly made you think that it would be 'q'

[minor flint]

https://pastebin.com/cVLxEEL8

[minor flint]

yeah... I'm not sure either, lol.

[quick bough]

Oh hey, I recognise some of that code, haha

I mean, if you want the actual source, see page 53 of https://www.usb.org/sites/default/files/documents/hut1_12v2.pdf

[minor flint]

Thanks, that's useful

[quiet root]

theres other easily digestible sources as well but they immediately show you that keyboards work in some weird ways sometimes

https://www.win.tue.nl/~aeb/linux/kbd/scancodes-1.html
https://www.win.tue.nl/~aeb/linux/kbd/scancodes-10.html

[quick bough]

This is not correct for HID

This seems to be for PS/2

Which is an entirely unrelated/different standard

Tho one that seemingly Gage here also initially attempted to use

[quiet root]

crap they had usb hid codes in the table on page 10 at some point iirc

it was a nice comparison between ps2 and usb

[quick bough]

I mean, you can just google "HID scancode" and find some decent-ish resources

but I figure just for a quick idea of e.g. "which button is Q", the actual standard works fine 😄

[quiet root]

but thats the thing they are all just ok not super good

[quick bough]

https://github.com/calamity-inc/Soup/blob/senpai/soup/HidScancode.hpp

[quiet root]

if i just want a list of keyboard scancodes in the usb spec id just use https://gist.github.com/MightyPork/6da26e382a7ad91b5496ee55fdc73db2

[quick bough]

Of course, this is ignoring the fact that Wooting adds its own scancodes (at >0xFF) 😄

[quiet root]

https://gist.github.com/BigBrainAFK/8a9e29bc5b40a76eec32a26d1f679de2

[quick bough]

[quiet root]

im not entirely sure what digital 1 is

[quiet root]

wut

did simon release them on GH?

[quick bough]

No, this is my own collection

[quiet root]

ah

well imma make a gist to keep track of it and update it

[quick bough]

banger

- `0x3B5` = Next Track
- `0x3B6` = Previous Track 
- `0x3B7` = Stop Media
- `0x3CD` = Play/Pause
- `0x403` = Profile 1
- `0x404` = Profile 2
- `0x405` = Profile 3
- `0x408` = Profile Switch
- `0x409` = Function Key (Fn)

[quiet root]

ill check next week what the new ones are and will update the gist accordingly (replaced the stuff up to)

the media keys should be standard

not from us

[quick bough]

I don't think it is lol

[quiet root]

well if they werent in the usb spec then the os wouldnt respond to them

[quick bough]

Yeah, I suppose, but I just don't see it in the spec

[quiet root]

i can inquire about that as well next week

[quick bough]

Well, there is this it seems

Tho unclear where the 3 comes from

[quiet root]

wonder if its to do with usage pages

[quick bough]

possibly

Yeah, I guess it's 0xB5 sent on usage page 0x0C

whereas usage page 0x07 would be for "normal" keys

[quiet root]

yeah i guess simon uses 0x3YY to be on 0x0C usage page

and 0x4YY is custom codes

[quick bough]

I tried studying QMK a bit but I did not get much of an understanding of the "talking to the PC" part lol

[quiet root]

yeah our hid descriptors are also a mess and cause issues with some KVMs and BIOS/UEFIs

theres a fix in testing for this but its pretty complicated to get right

[quick bough]

I thought that [the KVM issue] was just your keyboards drawing too much power

[quiet root]

no some kvms have it work but only once the OS is booted

theres some quirks in USB due to the 6KRO boot interface

this has its own set of rules so BIOS didnt have to have a full USB HID implementation

[quick bough]

lol, just mindlessly scrolling this spec

[quiet root]

yeah

the spec has some odd stuff

im glad we have someone dedicated to fixing usb descriptors for example and researching/fixing odd quircks in the firmware

[quick bough]

you what

[quiet root]

we have a person just for these kind of things

they also made the XLAT firmware compatible with keyboards not just mice

[quick bough]

ah, nice

I am somewhat hyped for the Finalmouse Centerpiece, but I guess "coming soon" is like dad returning with the milk at this point

or Wooting releasing a TKL

[quiet root]

i mean 80he is tkl

[quick bough]

[quiet root]

so is 60%

[quick bough]

Yeah, I actually have a TKL by Wooting

but it only has 3 buttons

[quiet root]

technically correct

https://tenor.com/view/correct-futurama-the-best-kind-of-correct-yes-yep-gif-5787390

[quick bough]

I mean ngl I can appreciate designs like the DrunkDeer A75

but it's also really hard to get used to typing on this if you're used to a "normal" keyboard

[quiet root]

id say the worse thing is keycap compatibility

[quick bough]

How so?

[quiet root]

gl finding those homecluster keys in the row heights they use

[quick bough]

uhh, I mean, they look like normal keys to me, idk

[quiet root]

yes but the profile is different

[quick bough]

the most "modding" I will do on my keyboard is compile QMK from source and flash it 😄

[quiet root]

most keycap sets use profiles keys so they arent all exactly the same

the fudge happened

[quick bough]

man had a key-stroke

[quiet root]

you can see some profiles are uniform but most are "sculpted" so you can just use R2 keys on R3

itll look and feel weird

[quick bough]

you say "uniform" even tho the space is clearly different in shape

[quiet root]

yeah uniform refers to anything RX not space row

[quick bough]

but yeah I see what you mean

just saying I can appreciate the design, not saying it's practical 🙂

it might be more practical if they would put Lekker Switches in it 😛

[quiet root]

well then imma dip and go do irl things

[quick bough]

have fun

[minor flint]

@quiet root , do you happen to know why the values seem to be rounded down like this while using the sdk?

[quick bough]

This is the number you're seeing

With my Two HE, the lowest it will report raw is 5

might be a calibration thing

also depends on the key

I figure they calibrate it at the factory. And they don't give you any option to do it yourself.

Which is fair enough because no one wants to do that

[quick bough]

Looking some more into it, it's only half-right. Because "Open Browser" would be 0x223 on usage page 0x0C, but it's reported as 0x323 on the analogue interface.

So, it's more like 0x300 | (scancode % 0x100) for usage page 0x0C

or & 0xff instead of % 0x100 😛

[quick bough]

[sturdy aurora]

]

[minor flint]

[

[dusk helm]

I found out that the wooting 80 HE still send US key Code I use the kanata project for keymapping and I use the Interception version which means it intercepts the keys before windows can even translate to german keystrokes just want to put the text here when other non US developer get the same problem

[quiet root]

wha

keyboards dont send different keycodes because they are iso or ansi or a specific iso language

the reason windows/linux whatever translates is cause they all send the exact same data

[dusk helm]

yes this true what I found it out now but some project tell you need a completly different configuration when you use ISO layout https://github.com/jtroo/kanata/blob/main/docs/locales.adoc https://github.com/jtroo/kanata/blob/main/docs/config.adoc#deflocalkeys and for people that develop but got into keyboard development dont know that from the beginning thats why I send this message

[sly crown]

@quick bough FYI: analogsense does not work in Cyberpunk, if you have Cyber Engine Tweaks installed (which is basically a cornerstone of cp modding). Dunno if you want to do something about it, just thought I'll let you know

[tame spire]

I have one usecase for wooting that's giga niche but I have no clue if anyone made something like this or not

so basically, yk how pens (on graphic tablets) can have pressure that affects the brush? (Higher pressure = Bigger dot just like with a real brush)

I really want to convert HE switches' depth-meter into pen-pressure, so the art program reads my key input as the pressure

mostly want it so I can draw with my mouse with my right hand; whilst using my left hand to control the pressure

[fading python]

Hey everyone. Anyone happen to know what kind of firmware the Wooting keyboards use?

[quick bough]

Do you have any more details about this? I assume it's a pattern scan failure somewhere?

[quiet root]

custom self developed one.

its not a modified qmk or such

and its closed source (just in case the next question is about source code availability)

[quick bough]

The 80HE is confusing me so much

[strong siren]

what about it

[quick bough]

The default layout for JIS has "Int 1" and "Int 3"?

and there's no analogue SDK or RGB SDK update available for it yet afaict

Also, the profile indicator is divded into 5 sections?

but it only has 4 on-board profiles?!

[quiet root]

have you tried dividing 10 into 4 and get a nice even interger result

no overlaps

[quick bough]

have you heard of blending?

you can take 2 colours

get a new colour

by interpolating

dedicated IME keys is cool af tho

[quiet root]

ok so we place the middle on the 1.25th led and then blend left and right sound good

then in 2.5 increments

[quick bough]

Sounds good

but yeah the key mapping is really confusing because I have a "Kana" keycap but no "Kana" option in Wootility, and a few others like this

[quiet root]

so #1285255683511681176 with images

[quick bough]

just make a ticket on your jira

haha, the key press indicator does use blending to display more than 10 steps

firmware bug: deleting light indicator effect on profile 1 does not black out the light

[quiet root]

same thread

this channel is for people making stuff for our keyboards not really reporting bugs of wootility or the firmware

[quick bough]

okay, I'm reporting the bugs there for someone else to read it, I guess 🙂

[quiet root]

yes the people actually fixing those 2 things

[quick bough]

well, from a "community dev" perspective, there is nothing to say about the 80 HE

[quick bough]

it seems both int 1 and int 3 are reported as 135 / 0x87, that's probably also not intended like that?

Int 3 should be 137 from what I can gather from the USB standard, guess I'm also reporting that

tho unsure if my Windows configuration is incorrect or if these keys are not reported to the OS normally at all

[quick bough]

my apologies for making horrible assumptions about your workflow... make a ticket on your notion

[serene harness]

Wootverlay is not working on 80HE

no registration

can we get an update for that?

for personal requests i would also like 1ms update and for "instant transitions" to not be off by default (every time i open OBS i have to interact with browser source, go into the menu and turn it on)

[quiet root]

the sdks arent updated as we wanna finish up firmware and wootility first

[quick bough]

If you don't mind replacing the wooting-analog-plugin: https://github.com/AnalogSense/universal-analog-plugin/releases/download/0.2.3/universal-analog-plugin.with-wooting-device-support.dll

Honestly I don't get why they didn't just add their VID and called it a day

Product name can be gathered via HID

That's what I do anyway ¯_(ツ)_/¯

[quick bough]

wait... I'm pretty sure this is already how that works

https://github.com/WootingKb/wooting-analog-sdk/pull/80

[quick bough]

Wowie, my PR being merged? On the same day, no less?!

[placid ledge]

rko out of nowhere

Need to find some time to properly go through the rest, but seen this one pop-up and it was easy enough to get it sorted

[quick bough]

Now you just need to make a new analogue SDK release and you can call it a day for the 80HE 😛

[rich tulip]

Still a lot of stuff to be added to the analog monitor! Can't call it a day yet

[quick bough]

What?

[rich tulip]

Additional features

[quick bough]

The only noteworthy thing would be if it reported positions instead of scancodes but that would be firmware-side

[quiet root]

analog sdk isnt analog monitor

[rich tulip]

<--- dumb ignore

[quiet root]

uneducated in a topic doesnt make one dumb

[rich tulip]

<--- uneducated ignore

Actually, I just didnt read his response fully and assumed lol

Mainly because I saw Simon respond and I blew up his DM and hadn't heard back so I got eager 😁

[quick bough]

tbf analogue SDK is extremely niche, no one really uses it afaict

well, I guess Woot-verlay exists

but, like, games or other interactive media... not so much

[rocky wharf]

Sadly, it is surely a great addition

[quick bough]

I mean, it's a bit of a balance to make sure the input is not too drastically different I guess

so, you probably don't want just a linear curve

I've been working on getting analogue support for keyboards from Razer (Huntsman series), DrunkDeer, and some other brands

but I think the hype for it has kinda passed and it's unsure if it's even an issue of lack of hardware access for game developers or more just them not believing it's sufficiently worth it

[rocky wharf]

Once more keyboard manufacturer adopt to HE Keyboards, it will become more important i think. I personally think that this is slowly starting as i see more and more named gaming product manufacturer producing HE keyboards.

[quick bough]

Well, from where I am standing, I'm not seeing a huge amount of interest in HE specifically, certainly not much interest in analogue input...

SOCD was the latest hype train, and no one even cared about the analogue aspect of it

Obviously, in terms of implementation, it's much more solid when you do have analogue keys, but that doesn't change how people perceive these things

[quick bough]

What I mean with this is that I see many analogue keyboards that simply do not have the firmware provide the analogue input in any way usable by software (specifically the ones from Corsair, Steelseries, and Akko)

[rocky wharf]

Acutall so much untapped potential

[quick bough]

Keychron is an interesting case because their firmware is open-source so I managed to compile it myself with a very nice analogue interface, but imagine trying to tell someone who wants to play your game to first flash custom firmware on their keyboard

I've collated more or less all of my work in this area on https://analogsense.org/ if you're interested

[rocky wharf]

Actually, i also see potential for Wootomation to use something similar to DKS.

[quick bough]

At this point I do just have like 15 analogue keyboards lying around here

[rocky wharf]

The 80HE i've ordered is the first one to be honest

[quick bough]

Damn

tbh it does look like a solid keyboard with the right key mappings

so I guess I see why they might not want to make a TKL

but I personally still need my numpad for the dedicated menuing keys 😛

[rocky wharf]

What i'm happy about is that most of there open source stuff is written in Rust as i fell in love with the language. Originally i was coding C# for a while.
And i tend to code a lot of stuff for my devices, ...
So being able to have a SDK is amazing.

[quick bough]

I code most of my stuff in C++

[rocky wharf]

I needed to use it for my Operating System Module at the university and i've used C for the FlipperZero ^^ C++ is not one of my favorites, and python is on the bottom list just for its syntax xD

[serene harness]

Most people don't know about analog keyboards still.

[quick bough]

Yeah, I think that's kinda related to what I was saying with SOCD

[serene harness]

The top post on PCMR today (probably the largest casual PC gamer subreddit) was people raging at this

because obviously this doesn't exist and the developer is an idiot

and people talking about hall effect analog boards etc were not even in the top 3 comments

and they are yet still a level closer to us than the general PC user crowd

it will take a lot to really raise awareness

i asked some friends about putting analog support in their indie game - we need to show off the cool stuff that you can do. A guy modded KSP to use analog input alongside digital commands seamlessly without much work.

[strong siren]

Honestly I personally don't think it's very easy to use. Might just be skill issue but every time I tried to get used to analog movement I found it very difficult

[rocky wharf]

Well we were used to either press it or not for a lot of years, getting used to fine grained input with keyboards will take a while i guess.

Probably muscle memory related

[quick bough]

Well, for me the 2nd comment talks about it, and it seems people are generally enthusiastic about it?

Tho the real crime is that no one mentions the analogue SDK that game developers could use to add it to their own games

but that's like asking an average person to know the Windows API

This is actually a really good feature and I think games would have to implement something like this into their games because a linear curve is not exactly desirable

(funny thing to note in this regard is that the default state of this curve is seemingly not a preset)

[rocky wharf]

I mean, most games with first/third person character movement and controller input have the ability to adjust the walkspeed based on how far you go with the joystick.
That's the simplest case where using such sdk would be great.

[serene harness]

The hardest thing for me is unlearning the compensatory mechanisms that i trained myself to do for digital controls. E.g. to apply a 60% input, approximately, i will tap the key on and off and on and off at a rapid rate, trying to average around 60%. Of course it's much easier and smoother with analog, better flight results, but i have to apply conscious effort to not do the stupid digital thing.

[serene harness]

Yes, that is me 😛

[amber juniper]

Hey i'm using a wooting 80HE and love to code. Is it possible for me to add some form of functionality regarding "SOCD"? what i want to implement is some sort of offset. meaning, if [KeyA] is pressed [KeyB] should deactivate, but after a certain amount of ms setup via the offset. further more maybe even add some variability to the offset.

I saw some possibilities of creating plugins (for 60HE), but maybe i am mistaken.

[quiet root]

we have the analog sdk and rgb sdk but what you suggested would basically be just any old macro via AHK or such

[quick bough]

Yeah, there's the analogue SDK, and with the Windows API, you can submit your own key events. It's of course a bit hacky to do it in user-mode like that, but it can work. I've already done user-mode SOCD resolution at https://github.com/Sainan/analogue-socd-resolution. This project is now unmaintained, but it may help you get started.

[amber juniper]

@quick bough @quiet root thank you! I gonna take a look at this

[tame spire]

I'll take it as a no 😔

[quick bough]

I doubt anyone did

but I guess you could with some reverse-engineering

or if there's an open-source program...

also for some reason I thought Wooting were gonna release a mouse next and was gonna say you should wait for that...

idk if HE buttons on a mouse would be that good

[frail wedge]

<@&182822816075218944> Can i Have help traking my order

or @quiet root

<@&490097707487330305>

[jagged echo]

Please, don't ping EVERYONE
Also wrong channel
Also you can track your parcel here https://hub.wooting.io/my/orders

[sly crown]

No, I haven't got time to check it any deeper, I just managed to isolate the culprit. I'll get back to you if I learn anything new.

[quick bough]

Oh, it doesn't show that in release builds, does it

Welp

Welp2

I'm sure I'll get to it at some point in the future

Just think in terms of Rust developer speed, not C++ developer speed, haha

[quick bough]

uhhh

[strong siren]

The analog thing must be awesome with that mod that adds flying cars

Double cool because that mode was made by the same guy that created qmk iirc

[quick bough]

Yeah, I'm actually unsure now. At first my ASI mod was being loaded but it crashed, and now it's just not being loaded with CET.

I guess just do what I did and use a DLL injector like Xenos to get it to swallow the damn mod. 😄

[strong siren]

The game uses razers chroma animation file format internally for the razer chroma effects

[quick bough]

Without CET it gets loaded, so idk

[strong siren]

Technically if you manually change them, convert to base 64 and embed them into the database thing with wolvenkit you can change them :D

[strong siren]

Yeah idk the modding scene is a bit of a mess imo, with the whole cet or redkit or whatever other modding tools

[quick bough]

Oh wait, there's a version.dll here

No shit they use the same ASI loader

That might explain it

They use the same ASI loader but with a different config

Yeah, okay, so, @sly crown the fix is to simply put the AnalogSense.asi into the "plugins" folder created by CET and discarding the dinput8.dll (CET uses version.dll)

I guess I could use the same config as them, but that's not typically how ASI mods are used in GTA V

[unborn summit]

hi (not a software engineer)
is there a programmatic way to trigger change of keyboard profile? ie without having to click things in wootility; something CLI triggered so can be fired via {some automation, out of scope}

https://github.com/ShayBox/Wooting-Profile-Switcher is very close, if i was a rust dev i'm sure i could use that code as inspo to make something much dumber (take CLI input, orchestrate wooting as output).. alas i am not so gifted

[quick bough]

idk if C++ is more your flavour, but this would be the only thing needed for that:

static void sendProfileSwitchToHardware(uint8_t profile)
{
    for (auto& _kbd : kbRgb::getAll(false))
    {
        if (_kbd->isWooting())
        {
            auto kbd = _kbd->asWooting();
            if (!kbd->isUwu())
            {
                std::cout << "Sending report to " << kbd->name << "...";
                {
                    Buffer buf(8);
                    buf.push_back(/* 0 */ 0); // HID report index
                    buf.push_back(/* 1 */ 0xD0); // Magic word
                    buf.push_back(/* 2 */ 0xDA); // Magic word
                    buf.push_back(/* 3 */ (uint8_t)WootingCommand::ActivateProfile);
                    buf.push_back(/* 4 */ profile);
                    buf.push_back(/* 5 */ 0);
                    buf.push_back(/* 6 */ 0);
                    buf.push_back(/* 7 */ 0);
                    kbd->hid.sendFeatureReport(std::move(buf));
                }
                SOUP_UNUSED(kbd->hid.receiveReport());
                std::cout << " Success.\n";
            }
        }
    }
}

Oops, just realised one thing missing here: ActivateProfile = 23

Also feel free to ignore the conditions, you basically only need to find a HID with vendor ID 0x31E3 and usage page 0x1337

[restive yew]

An unexpected error occured, please let us know about this through discord or email

TypeError: Cannot read properties of undefined (reading '5')
    at https://v5.wootility.io/assets/index-f86de06b.js:563:14114
    at Array.map (<anonymous>)
    at https://v5.wootility.io/assets/index-f86de06b.js:563:14104
    at useSelectorWithStoreAndSubscription (https://v5.wootility.io/assets/index-f86de06b.js:80:13138)
    at https://v5.wootility.io/assets/index-f86de06b.js:84:688
    at useWootSelector (https://v5.wootility.io/assets/index-f86de06b.js:84:1391)
    at useRemapActiveKey (https://v5.wootility.io/assets/index-f86de06b.js:729:18999)
    at RemapKeyboardKey (https://v5.wootility.io/assets/index-f86de06b.js:740:30000)
    at Nh (https://v5.wootility.io/assets/index-f86de06b.js:38:18930)
    at Vk (https://v5.wootility.io/assets/index-f86de06b.js:40:48722)
    at Uk (https://v5.wootility.io/assets/index-f86de06b.js:40:43974)
    at Tk (https://v5.wootility.io/assets/index-f86de06b.js:40:43897)
    at Ik (https://v5.wootility.io/assets/index-f86de06b.js:40:43738)
    at Nk (https://v5.wootility.io/assets/index-f86de06b.js:40:39691)
    at Ek (https://v5.wootility.io/assets/index-f86de06b.js:40:40591)
    at jg (https://v5.wootility.io/assets/index-f86de06b.js:38:3534)
    at https://v5.wootility.io/assets/index-f86de06b.js:40:37832
Reload the Wootility

[serene harness]

The KSP mod does

it combines analog and digital KB inputs depending on context

[flat wigeon]

Well makes sense that way. Otherwise I would need to handle the pressure point manually

[quiet root]

you wanna be in #1285255683511681176

[quick bough]

that's cool, I made my own mods, but it's not natively in games

[flat wigeon]

I really believe for analog keyboard input to lift off, it needs an input model, where this comes more or less for free. Like: You want overhead car control? Here take this example, it will map nicely to keyboard, analog keyboards, gamepads, steering wheels and can be mapped to other devices you don't even know

[quick bough]

Maybe I should have to buy myself a steering wheel and pedals, parse the HID input descriptors to generalise, and then make an input library that nobody will use. 😄

[balmy iron]

Yeah, well game engines already do tend to treat all inputs as analog so we'd just need a standard way of having the OS provide analog key inputs.. or a library everyone uses

[quick bough]

Wait, this product category confuses me. There's some around 50~100 euros, and then there's logitech's for 200+ euros.

[flat wigeon]

Force Feedback and stuff

[quick bough]

Like the weird triggers on the DS5?

[flat wigeon]

Well... you need the Steering Wheel itself, you need force feedback and this can go quite hard, then you have pedals, the shifter

Also probably small production batches

[quick bough]

yeah but I mean you can get all that for 50 euros

meanwhile logitech sells it for 200-300 euros

[flat wigeon]

Not sure about that. You probably pay extra for the PS Logo and for it being from Logitech ^^

Maybe the force feedback motor is expensive

twin engine force feedback

I considered buying one too, but I'm undecisive considering the price point and the fact that there are many possible configurations

[rocky wharf]

Steering wheels are effectively in the Joystick category, so the descriptors should be the same https://usb.org/sites/default/files/hut1_5.pdf#page61

[flat wigeon]

Every time I read this, I'm amazed what in theory is included in the standard

Imagine the chassis of your PC as an USB device

[serene harness]

do we have any approx timeline for 80HE support in the SDK?

[quick bough]

I already PR'd it for the analogue SDK, just a matter of when it gets released I guess

but yeah I guess now they're busy rewriting they Wootility 🙂

I hate this company so much sometimes, it just makes me scream internally how they can't stop shooting themselves in the foot because they keep making the wrong decisions and focus on the wrong things

[paper crow]

https://tenor.com/view/the-dude-yeah-well-you-know-thats-just-like-your-opinion-man-gif-10470601

[quick bough]

Sure, never said it was anything but. 😛

[strong siren]

classic sainan moment

[quick bough]

What are you doing now at Wooting btw?

If I had to assign you, I'd say you were best fit to work on the upcoming desktop app. 😛

[strong siren]

i wrote the two new rgb effects we added with wootility v5 for example :p

[quick bough]

they let you near the secret sauce firmware, damn

Although yeah the noise effect definitely reminds of artemis

[strong siren]

yeah that's basically where it's from

a faster (and sadly crappier) version that runs okay ish on the mcu

[quick bough]

How is the noise generated? LCG?

[strong siren]

it's simplex noise if you're curious

[quick bough]

I'm guessing that's the algorithm you use to map a "1-dimensional" rand function to the 2-dimensional plane of the keyboard, but not actually what I asked

[strong siren]

i generate 3d noise and map it to 2d for the keyboard actually

[quick bough]

...wtf

I thought it has to run well on the small CPU

[strong siren]

i'm sure you're cooking up a much beter implementation using the rgb sdk already

[quick bough]

haha, no, my idea of a much better implementation is just a static colour, no distraction in that 😛

Plus, idk if they ever fixed that bug where the firmware would just ignore RGB commands until the keyboard is replugged

kinda stopped bothering with it then, not very cash money when I have to replug my keyboard while playing Warframe or something

[flat wigeon]

Sainan: "[...] focus on the wrong things"
diogotr7: "[...] two new rgb effects [...]"
lol

[quick bough]

Well, I'd be happy if my subordinates just delivered anything of value to the product, but yeah, certainly not what I would have them focus on.

[earnest tulip]

Anyone know where I can get some cool keycaps

[quick bough]

Well, you have come to the right place, this is the #cool_keycaps_marketplace channel

Oh wait, it's not

Have you tried wooting.io? I hear they sell keyboards, even some accessories for those keyboards 😛

https://wooting.io/our-products?category=Keycaps

Honestly I kinda like the wave keycaps, but once again they're discriminating against us 100%ers

[gritty lake]

yup here

https://www.alexotos.com/keyboard-vendor-list/

do note that keycaps from these vendors tends to be pricy

[quick bough]

hmm, the keyboard keyboard vendor list, so you can find a keyboard for your keyboard?

[gritty lake]

otherwise just find a decent doubleshot pbt one from aliexpress

[gritty lake]

it list what each vendor sells

like so

[quick bough]

are all keycaps compatible with all switches?

[gritty lake]

nah but 99% are

just check for mx compatible keycaps

and mx stems switches

which is the plus shape thing

[clear flint]

hey gamers, is there a way to write modified rgb to a profile with the rgb sdk? i managed to write one profile's rgb onto another, but haven't found how to save rgb that has been set via the sdk
like, this code here will set my D key to green, but once the program closes it goes back to the color it was before

also, what's the difference between commands and reports? (from wootility verbose logs)
i'd think reports are data from kb -> pc but that's not what it seems like in the javascript

[strong siren]

What is your goal with this?

[clear jay]

买了wooting还在制作中，如何申请退款？

[paper crow]

请通过 https://wooting.io/support 联系我们的客服。

[quick bough]

If you want to save RGB set via the SDK, you will likely have to replace the SDK DLL to buffer the data currently being shown as otherwise there's no way to query the firmware for it.

[quick bough]

Wooting report -> HID input/output report. Wooting command -> HID feature report.

[clear flint]

making a tool to set the base rgb to a gradient. i'd rather not keep the program open 24/7.
if there's a way to set and save rgb with javascript on the wootility page, that'd be fine too.

[winged bolt]

will there be a 8k hz refresh of the 60he? If there are plans on it, I think it would be really cool to update the pcb so it can support gummy oring mount

[flat wigeon]

There is nothing wrong with keeping a process running in background... as long as you don't use a resource hog like Electron or .Net MAUI

[flat wigeon]

This minified stuff reads like crap

Why is there an extra command to get the serial number?

[quick bough]

urg, my table is like a mm too thiccc for the G29

[quick bough]

It seems it's not properly supported in Motorfest nor FH5, so I tried installing G Hub and it seemingly doesn't stop loading

[strong siren]

sounds like GHUB alright

[quick bough]

it finally loaded after 10 years

it was very important to ask me for my discord, my email, and my firstborn

seems that opening G Hub made it magically work in Motorfest

I absolutely hate this control scheme

[clear flint]

got it working c:

huge shoutout to friend from another server. they did 90% of the work and had a really smart solution to hijack the javascript functions.

will give the script a nicer ux and post soon

[flat wigeon]

I thought you already had it working by issuing the save command

[clear flint]

only thing i could get save command through sdk to do was overwrite one profile's rgb with the active profile's unmodified rgb

[quick bough]

Oh hey, I didn't know they had a talk about the Wooting Analog SDK

[full current]

How hard would it be to integrate the analog inputs to something like FL studio

Like the further I press down a key, the higher a modulation slider for a synth goes

(Is this the right channel for this question?)

[quick bough]

You might be looking for wooting-analog-midi

[full current]

I see

[clear flint]

https://github.com/Provoxin/WootilityCustomRGB
probably doesn't work on wootility v4, haven't tested

couple of examples

[quiet root]

you could also write a tool that fetches a profile from the share code, modifies the rgb and then reuploads it.

[quick bough]

I'm a bit disappointed that there's seemingly no extension to provide WebHID on Firefox

I'm just switching away from Chrome now because it seems they're serious about killing a vast majority of their extensions ecosystem out of sheer stupidity

[quiet root]

can i ask you not to use the r word

[quick bough]

Suggest a replacement

I guess 'stupidity' works?

but I think I missed a memo, when did the r word become bad?

[quiet root]

thats not how this works

also its been used as a derogatory slang for ages at this point

[quick bough]

Well, yeah, I'm clearly not praising Google in my message

[quiet root]

which is fine but no need to use that specific word

[quick bough]

Is the f word still okay? What about the s word?

[quiet root]

if you mean a word commonly replaced with fudge and the one starting with shi then yes

while i personally dislike them in my own vocabulary they arent anything bad

[quick bough]

Okay, so it's not a matter of this server trying to be PG

[flat wigeon]

R*** is now a bad word?

[quick bough]

I guess social sentiment around the r word has just shifted to make it closer to the n word?

[quick bough]

R word as in the slang for handicap

^ put this message in google to find the wikipedia article dedicated to the word, lol

[flat wigeon]

Ah, the longer r word

[quick bough]

Now I'm confused as to what r word you meant

[flat wigeon]

The 4 letter r word

[quick bough]

Is it a crime?

[flat wigeon]

You know... what you are regularly complaining about

[quick bough]

No, I don't follow

[quiet root]

i assume rust

[karmic forum]

Nooo

[quick bough]

[karmic forum]

Rust is the future

I love rust

[quick bough]

I agree, we should ban people like Mr.Ender for using that R word in that way

[karmic forum]

[flat wigeon]

I'm not native english speaking, so that was the first word beginning with r that came to my mind 😄

[quick bough]

Nah it's okay, one of these days you will try C++ and never look back

[quick bough]

Well, I was just thinking of "bad" words since you censored it, so I only thought of ||rape and retard|| (please no ban)

[karmic forum]

Im still waiting for C+++ before trying anything

[flat wigeon]

The latter has 5 letters ^^

[karmic forum]

||this is a joke||

[quick bough]

I'm waiting for C++ ||code to compile||

[karmic forum]

You think thats bad try compiling rust

[quick bough]

[flat wigeon]

I doubt. C++ really has a legacy issue that you need to sort out first before you can have fun

[karmic forum]

I love rust but like it takes forever

To compile

[quick bough]

Eh, I think the legacy is not that bad. Just don't rawdog new or malloc (use RAII, kids!) and you're good

[flat wigeon]

Well... seriously. How that was the joke about C++. And now... Angular. Takes much longer then C++ ever did.

[quick bough]

Imagine compiling your websites

(I still write HTML, CSS, and JavaScript.)

[flat wigeon]

In the meantime I press F7 for my C++ projects and wait a few seconds

[quick bough]

I press Ctrl+B

[flat wigeon]

Also funny is .Net

It compiles so super fast... yay

Try AOT compilation...

[quick bough]

but then you get bytecode, no?

[flat wigeon]

Sure

Of course it compiles fast if you skip half of the process

[quick bough]

A lot easier to emit bytecode than (optimised) x86 assembly lol

When I made a toy compiler, I basically just turned x86 into a stack machine, which is pretty meh for performance, but still better than a bytecode-parsing VM, I guess

[flat wigeon]

But start your service... send a request thumbweed thumbweed thumbweed response

And AOT compilation in .Net is not only slow, it is incredibly slow

[quick bough]

Also results in me being able to optimise the code by just doing a string replace on the bytecode

[flat wigeon]

Never waited that long for C++... except maybe for Firefox to compile

Really I should compare everything to a firefox build

How can a simple application take as long to compile as a full fledged browser

[quick bough]

Which .NET version are you using btw?

I'm still firmly on 4.5 because everything after that just seems like shit. It's not even pre-installed on Windows anymore.

[flat wigeon]

Well AOT was... I guess .Net 5/6

Currently on .Net 8 but AOT compilation doesn't work anymore

[quick bough]

I guess you like your executables to be big 😄

[flat wigeon]

With AOT compilation that would actually not be too bad, since it can trim dead code. At least in theory

[quick bough]

Well, I have no idea about .NET AOT compilation tbh

[flat wigeon]

Ahead Of Time compilation like C++

[quick bough]

I just know that after .NET 4.5 it's turned to shit and the executable sizes have balooned

ik

[flat wigeon]

You get a native binary

[quick bough]

Well, that's some good progress, I suppose

[quick bough]

Whoo, a basic thing is working

[quick bough]

Well, the Wootility is not happy no matter what I do

[quick bough]

I guess what should happen when you filter by usage or usagePage is that you get a HIDDevice with only the matching collection? At least that's what I gather from reading this horrible obfuscated JS.

[clear flint]

wooting give sourcemap pls

[quick bough]

I actually hate this API. It's so simple yet designed in such a counter-intuitive way. It just does not map cleanly to the underlying OS primitives.

Or maybe I'm overcomplicating it.

Well, that's some progress

[clear flint]

are you using js overrides to get logs

[quick bough]

This is Firefox

[clear flint]

ye firefox has overrides

[quick bough]

Do you know what Firefox doesn't have?

[clear flint]

other than webhid, no

[quick bough]

Exactly 🙂

[clear flint]

how are you getting console logs into the wootility script tho

[quick bough]

It just logs stuff by itself?

[clear flint]

o, thought those looked different than the ones i've seen. ig not

[quick bough]

some logs are my own

from my WebHID implementation

[clear flint]

ic

[quick bough]

Okay, it works a bit more now. Can't change RGB, tho 😦

Uhh idk if this is a Firefox bug or a Wootility bug lol

[serene lynx]

A!

[quick bough]

Well, it works decently-ish, but it seems RGB changes are still not commited correctly

Gonna try the firmware update feature now so I can RMA this board

[quick bough]

It looks like the data being sent to the keyboard here is correct?

oh, wait, I'm on the fancy new firmware with bigger report sizes, but wootility doesn't seem to realise this

Okay, it RGBs now!

Still kinda afraid to do a firmware update tho haha

[quick bough]

Wait, but how is it that Wootility V5 is able to access the analogue data without ever asking for it?

Oh, I guess it's that while requestDevice respects the filter strictly, when you then do getDevices it shows everything for the same physical device without respect to usage and usagePage.

These are some weird semantics ngl

[quick bough]

Insane security mechanism

but I'm also gonna block all FIDO keys and shit and that should be fine

[quick bough]

Actuation point thingy is working

[quick bough]

Okay, I've published my initial version at https://github.com/Sainan/WebHID-for-Firefox

Still some TODOs there, but it works alright

[sweet peak]

is it a plugin with a server component?

[strong siren]

that's what it looks like, pretty cool

[quick bough]

Yeah, there's a server that you need to run seperately

Not gonna get much better in that regard unless Mozilla add native support for WebHID in some capacity

[quick bough]

Whoo, extension published on AMO

[quick bough]

I'm still a bit unsure if this is a good idea

Worst case scenario, I can use the "troubleshooting" thing in Chrome, right?

I'm just unsure if a borked firmware update might turn it into a brickboard

[quick bough]

Turns out, I can!

Scary, tho

Trying to set my profiles up again, it seems it sometimes takes too long to respond on Firefox?

[paper crow]

You cannot use Wootility web on Firefox, as it's not possible to communicate to USB devices

[strong siren]

it is known, Sainan hacked it together with a helper tool that runs alongside firefox. very cool!

[quick bough]

I'm switching to Firefox so I had to add WebHID support myself lol

[mental bluff]

you're beautiful i love you

[halcyon zephyr]

Hello I need help trying to find my order I made yesterday

[mental bluff]

erm this is not the right channel

[halcyon zephyr]

What channel is it

I been trying to look for it

[halcyon zephyr]

?

[paper crow]

https://wooting.io/support

[halcyon zephyr]

How long does it take for them to reply back

[mental bluff]

#🤖│community_dev message

[mental bluff]

sent you some ideas via dm. I’m not official support though so take with a grain of salt. just suggesting checking your payment method’s histories if available

[quick bough]

Damn, this goes hard!!

Welp, guess I'm trying Woot-verlay

don't know how to get that thingy to render

Welp, guess I should've figured that Wine wasn't gonna discover the .so

Dire situation for analogue input on Linux

I feel like something's missing here

Welp, I guess Microsoft Java is less cross-platform than I had foolishly believed

[quiet root]

no its because its winforms

the sdk itself is pretty crossplatform but winforms isnt

[quick bough]

I mean from what I understand it should just be a WS server

but idk maybe bad code

[strong siren]

a ws server wrapped in forms is still forms

yeah https://github.com/DjCrqss/Woot-verlay/blob/b4228c228d3d3cd9d6b43173965317a6f9cb229f/Woot-verlay.csproj#L9

[quick bough]

So, given that wooting-analog-midi doesn't render or play sound on Linux, and Woot-verlay doesn't run, how am I to test if an SDK plugin works on Linux?

[strong siren]

artemis maybe? i havent tested the analog sdk on linux so it's possible that it's broken

the rgb sdk should work

i feel like artemis is kind of overkill for this, is there nothing else that targets the analog sdk?

[quick bough]

Also I only just realised this guy implemented WebSocket from scratch, sheesh

[strong siren]

sheesh

[balmy iron]

Nice

[quick bough]

[strong siren]

that one should come preinstalled

[balmy iron]

Dunno, I just downloaded the analog midi appimage and it just works on my system x d

[quick bough]

The only Wooting thing I see is this here in the install wizard

and Artemis just crashed when I toggled "enable feature"

[strong siren]

[quick bough]

Appimage would be MacOS, right?

[balmy iron]

No

[quick bough]

what

can I dpkg -i an appimage

[balmy iron]

No you just run it like any other binary

[quick bough]

do I need to uninstall the .deb first?

[balmy iron]

No

[quick bough]

[strong siren]

have you never used appimages before?

they're pretty cool

[quick bough]

Clearly not

[quiet root]

basically exe files but linux

[strong siren]

and one of the reasons i really dislike linux for desktop (well not the appimages themselves, just the fact that they need to exist)

yeah

[quick bough]

Can't you just statically link an executable

[strong siren]

tell that to distro maintainers

[balmy iron]

You can, indeed this is what all Rust things do and is why they are so chonky

[quick bough]

Well, okay, it seems that works

Curious as to why the .deb doesn't work then

[quiet root]

have you tried googling the error message

[quick bough]

What error message

[quiet root]

the one labelled error

[quick bough]

That was because I disconnected the device

I had to give it back to Windows to write the message

Also "hidapi error: Success" 👌

[strong siren]

lmao

[balmy iron]

A classic case of either forgetting to stash errno, or maybe just assuming a 0 read is a failure when it might not be considered so by the API

[quick bough]

I think for HID a 0 read is a failure/disconnection

[balmy iron]

Dunno, would have to look at the code to see why it be like that

[quick bough]

When we getting ABI v2 so I don't have to link against a Rust blob

Seems like my plugin works 😄

tho I still don't hear any sounds from it

might just be that Linux doesn't have a synthesiser built-in like Windows

[balmy iron]

Linux systems generally don't have a default midi synth just waiting, so you'll have to run one yourself

[quick bough]

This is why I paid 200 EUR for my Windows 10 Pro licence

Welp, at least the linking is rather simple on Linux

Might still have to look into HID report ids stuff for Razer keyboards

Oh right, I just remembered that Synapse doesn't exist on Linux

[jovial bane]

Openrazer is the closest thing on linux https://openrazer.github.io/

[quick bough]

Man, just got another analogue keyboard. The firmware has an analogue interface, but it only says how far a key is being pressed, not which key is being pressed

Guess I'll look into Razer on Linux stuff, tho already not a good start when they don't support the Huntsman V3 series

[balmy iron]

Clearly you just have to guess the key based on normal key events! But wait, what if more than one key is down at the same time..

[quick bough]

Yeah... unfortunately this firmware is unsupportable for me

What's most interesting to me is that the reports are 1:1 with those of another vendor, tho the firmware behaves a bit differently

but that's what I get with China crap

Ah man, Linux is funny. Can issue shutdown 0 as a normal user, but gotta be SU for reboot 0