<pie_>
something something if its exponential you graph it on a log log plot and it will look linear
<pie_>
but ts probably easy to get this stuff wrong if you dont know what youre doing
<samueldr>
look, it's me, don't know :)
* pie_
has analysis paralysis about stats, perhaps not unjustifiably
<samueldr>
hmmm
<samueldr>
I could also look at the source and see if there's logical relationship
<pie_>
R makes some least squares fitting methods availible if you have some parametrized gueses, but im not sure how that orks
<pie_>
probably doesnt help you though
<pie_>
it could be anything :D but you need to find what apears to be a reasonable approximation
<pie_>
well, if you can look at the source ,and its not just experimental data...thats probably a lot better, unless theres some dastardly effects you dont know about##
<samueldr>
it could be a side effect of how it's computer, like they're not trying to make the curve, it just happens due to constants
<pie_>
(though i guess thats what randomization is about in experiments)
<samueldr>
and here, luckily, it's stable and reproducible
<infinisil>
samueldr: So what is this exactly?
<samueldr>
[19:56:46] <samueldr> that's the relationship between the size of a filesystem and the available space within
<samueldr>
for a specific FS
<infinisil>
What's the percentage?
<infinisil>
Ah
<samueldr>
the space available within, one of the line is as "data clusters", the other as bytes available in df
<infinisil>
I think this is
<infinisil>
It needs 2% of the space to bookkeep the space
<samueldr>
all FS are different
<infinisil>
For keeping page tables or so
<samueldr>
here it's FAT32, and the space is all accounted for :)
<samueldr>
I think the discrepancy is because of the actual root directory taking space
<infinisil>
Yeah and I was gonna add that the curve comes from some fixed cost part
<samueldr>
I'm not concerned about the small discrepancy, considering the first one is for a FS with only 48KB available
<infinisil>
So you're concerned about why it doesn't give you 100%?
<samueldr>
I have other known fixed cost parts, 32 reserved sectors, and the FAT itself, which the FAT size changes in another relationship I'm trying to figure out
<samueldr>
not at all, I have the 100% figured out
<samueldr>
(reserved space) + (number of FATs)×(size of a FAT) + (data clusters)
<samueldr>
that graph only shows data clusters
<pie_>
"Yeah and I was gonna add that the curve comes from some fixed cost part" i guess we were all wondering if this is it
<infinisil>
Sooo what are you trying to figure out?
<samueldr>
the amount of sectors per FAT, given a size
<samueldr>
(which I didn't know that was my target when I began speaking about it)
<infinisil>
I see
<infinisil>
I once learned about how FAT works, but I forgot, as one does
<samueldr>
great, I think I have the right location in the source, and it's not something that looks hard to follow
<pie_>
joepie91 no i dont think i know of anything openal#
<samueldr>
(now thinking I need to port that to bash :/)
<ashkitten>
evanjs: and setting it to the one i got from ssh lets me connect to openrazer-daemon
<ashkitten>
no idea why it's being set to that
<pie_>
joepie91 , i really dubnno , but are you suure its not a problem with the way wine is built
<evanjs>
ashkitten: yep!
<evanjs>
Setting services.dbus.socketActivated to true should fix it!
<pie_>
i mean is it one of those build time compoents that can be anabled
<pie_>
well, whats the problem
<ashkitten>
evanjs: do i need to re-login for it to take effect?
<ashkitten>
evanjs: wait, i didn't even set services.dbus.socketActivated and logging back in worked
<evanjs>
ashkitten: I believe so, as the culprit was dbus-launch, which is what starts i3 in your case
<evanjs>
I’m doing xsession stuff via home-manager which is why I didn’t encounter this issue.
<ashkitten>
it might be related to the systemd user session issues i was having before
<evanjs>
Like what?
<ashkitten>
like if i logged in after a reboot it would fail to set session variables so systemctl --user didn't work
tokudan[m] has quit [Remote host closed the connection]
sphalerit has quit [Remote host closed the connection]
worldofpeace has quit [Write error: Connection reset by peer]
<ashkitten>
i think someone replied on the nixpkgs issue saying it was a pam thing
Ralith has quit [Read error: Connection reset by peer]
thefloweringash has quit [Write error: Connection reset by peer]
colemickens has quit [Remote host closed the connection]
<evanjs>
ashkitten: well this could be related to that, yeah. I’d set that variable to true if you’re using i3, anyway
<ashkitten>
main point being, the first login after a reboot wouldn't work properly
<ashkitten>
okay
<pie_>
pet peeve: people not putting ome sort of ffuser over bright led lights
<pie_>
cant look at the ceiling because muh retinas
<pie_>
y u do dis
<evanjs>
pie_: why are you looking at the ceiling so much, anyway? 😝
<evanjs>
ashkitten: and maybe reboot once to ensure everything still works. Can’t wait until FRidh approves the PR so we can get it merged
<ashkitten>
evanjs: sure enough, if i set that it works after a reboot
<evanjs>
Oh well speak of the devil lol
<ashkitten>
evanjs: are you planning to package any of the gui tools?
<ashkitten>
wait
<ashkitten>
it's not working
<evanjs>
Yah polychromatic. Almost done but having an issue with the imports 🙄
<evanjs>
What isn’t
<evanjs>
Or like what’s it saying
<ashkitten>
oh
<ashkitten>
i accidentally git reset over adding openrazer
<evanjs>
Oh phew. Don’t scare me like that 😝
<ashkitten>
scared me!
colemickens has joined #nixos-chat
<evanjs>
Yeah I’m planning on getting polychromatic up anyway. Luckily, the 1.0.0 beta had profile support (and uses meson and is wayyyy easier to package)
<evanjs>
Anyway, sorta tempted to add the dbus bit to the NixOS wiki
<evanjs>
Omfg. I enabled fx-sync with openrazer and now my brightness keys affect the mouse, too 😯 holy crap I haven’t used this stuff for a while
<ashkitten>
it's unfortunate that the main thing i wanted out of openrazer was button remapping and that's the main feature it lacks
drakonis has joined #nixos-chat
<evanjs>
Does it? Hrm
<ashkitten>
apparently so
<ashkitten>
there's some shenanigans you can do with evdev though
<evanjs>
"Support custom mappings for all types of devices (mice, firefly, etc)"
<ashkitten>
nice
<ashkitten>
is it done, though?
<ashkitten>
no checkmark
<evanjs>
though, including the firefly makes me wonder if it's the kind of mappings we're thinking of
<evanjs>
and yeah
<ashkitten>
anyways i'd be glad to have that
<ashkitten>
i'm no stranger to writing my own stuff for peripherals (g933-utils, roccat-tools riir) but it's always nice to have a thing done for me
<ashkitten>
i need to work on g933-utils, tbh.....
vika_nezrimaya has joined #nixos-chat
<ashkitten>
need to make it a better api and some more functionality
<ashkitten>
maybe i'll add lua scripting like i did with the roccat-tools rewrite
<ashkitten>
sorry, thinking out loud
<evanjs>
ashkitten: Especially if it's a full-blown application/service. Nah I do that all the time
<ashkitten>
evanjs: the original roccat-tools has a gui i still haven't bothered to add to my rewrite, lol
Ralith has joined #nixos-chat
thefloweringash has joined #nixos-chat
sphalerit has joined #nixos-chat
worldofpeace has joined #nixos-chat
tokudan[m] has joined #nixos-chat
<ashkitten>
guis are hard
<evanjs>
Yeah especially depending on the language *cries in Rust*
<ashkitten>
but seriously, the codebase sucked really bad
<ashkitten>
yeah i rewrote it in rust :p
<ashkitten>
at least it's slightly more readable even though it doesn't support nearly as many devices and doesn't have as many features
<ashkitten>
scripting support is way better tho
<ashkitten>
*way* more readable tho
<evanjs>
ashkitten: haha very fu- *checks GH* oh wait yessss. And yeah. rust-qt-binding-generator is probably to most promising thing I've seen so far for GUI support. Now they just need to include a Cargo.lock upstream so I don't have to use my fork...
<evanjs>
which is really the same thing with a lockfile so Nix can package it...
<pie_>
i looked into thatbindinggenerator once and i thinki totally didnt get it
<pie_>
or im mixing it up with something else, or theres been a lot of progress
<ashkitten>
i decided to rewrite it when i went bug hunting one day because something wasn't working right, and then i saw it was full of repetitive code, if/elses for returning the value of booleans, code was spread unpredictably across multiple separate libraries...
<pie_>
yeah i think that was it but it looks like more now, ill have to look at it again
<evanjs>
It's great, but there's a lot in qt I'm still unfamiliar with. It's also a shame QT Design Studio is commercial only AFAIK
drakonis_ has joined #nixos-chat
ekleog has quit [Quit: WeeChat 2.4]
ekleog has joined #nixos-chat
drakonis has quit [Ping timeout: 276 seconds]
drakonis has joined #nixos-chat
<samueldr>
uh, I'd prefer to have a better proof, but I got my constant, and it's found in the code 512/520*16
<gchristensen>
:o
<samueldr>
well, it's a bit more complex... now I need to try and see if it actually helps
alex_giusi_tiri has quit [Quit: Leaving.]
freeman42[NixOS] has joined #nixos-chat
vika_nezrimaya has quit [Remote host closed the connection]
freeman42[NixOS] has quit [Ping timeout: 246 seconds]
drakonis has joined #nixos-chat
drakonis_ has quit [Read error: Connection reset by peer]
drakonis1 has joined #nixos-chat
<ashkitten>
evanjs: close!
drakonis_ has joined #nixos-chat
drakonis1 has quit [Read error: Connection reset by peer]
drakonis1 has joined #nixos-chat
drakonis has quit [Ping timeout: 250 seconds]
drakonis_ has quit [Ping timeout: 276 seconds]
pie_ has quit [Remote host closed the connection]
vika_nezrimaya has quit [Ping timeout: 258 seconds]
endformationage has quit [Quit: WeeChat 2.5]
<ashkitten>
general question: do most "dumb" phones that include texting allow for more than one line per message?
freeman[NixOS] has quit [Remote host closed the connection]
freeman42[NixOS] has joined #nixos-chat
freeman42[NixOS] has quit [Remote host closed the connection]
freeman42[NixOS] has joined #nixos-chat
<eyJhb>
ashkitten: as far as I remember, yes
<ashkitten>
hmm
<eyJhb>
Why? :p
<eyJhb>
But I'm actually not quite sure, because it is so long ago. I think someone here actually only uses a dumbphone, think joepie91 is one
<ashkitten>
eyJhb: i need to design a better syntax than the one i currently have for posting to mastodon via sms, and it turns out it's really hard to design a syntax that's concise, easy to write, consistent, easily explainable, robust, and extensible
<ashkitten>
if it all has to go in one line that's 10x worse
<eyJhb>
Do you have a length limit?
<eyJhb>
And.. Can you just like, encode stuff in json? :p
<ashkitten>
it should be easy to type
<ashkitten>
what i currently have is this:
<etu>
The limit for SMS is 160, always been and still is. But even before smartphones they managed to break it up in several messages so the receiving assembles several messages to a single one.
<ashkitten>
post.visibility[[content warning]] post text...
<eyJhb>
etu: yeah, but maybe the service allowed for this automatically :p
<eyJhb>
ashkitten: somewhat confusing what the use case is, should it be parseable on the other end?
<ashkitten>
yes
<ashkitten>
it will be parsed and sent to the mastodon server as an actual post
<etu>
eyJhb: From what I know, it's part of GSM and is made of dark magic and superglue and even people working with GSM at the time didn't understand how the messages got glued together in the right order.
<eyJhb>
"dark magic and superglue" is the best way to solve problems ;)
<ashkitten>
the most annoying thing is i've got multiple optional fields, one of which can contain arbitrary text. and the post text ofc can be arbitrary too
<eyJhb>
ashkitten: I see no reason why you couldn't put new lines, and if it does work with dark magic and superglue, you could exceed 160 chars
<ashkitten>
multiple lines would definitely help
<eyJhb>
And just do a simple HTTP style request, headers, empty newline, text
<ashkitten>
yeah
<eyJhb>
You could try it, and see? :p Best way to find out ;)
<eyJhb>
ashkitten: actually very sure you can, as I have a... phone I use for.. stuff, and that receives newlines from the provider
<ashkitten>
i don't care about newlines on the receiving end
<ashkitten>
i need to be able to compose messages with newlines
<eyJhb>
So, are you sending the messages manually? /constructing them?
<ashkitten>
i know the spec allows it, but some phones probably send messages with enter and don't let you insert a newline
<ashkitten>
the user types messages themselves
<eyJhb>
Could could always just substitute newlines with '\n', literal. But can see when I get home, or joepie91 might be able to check
<ashkitten>
so apparently ms is making exfat free
<ashkitten>
which is cool and all but i still won't trust them until they make core parts of windows and the nt kernel open source
<ashkitten>
until then they haven't actually given anything significant of themselves to foss communities
vika_nezrimaya has quit [Ping timeout: 245 seconds]
<joepie91>
ashkitten: my experience is that almost every dumbphone allows newlines, but it's often a bit awkward to type one (emoji, OTOH, are very poorly supported)
<joepie91>
'newline' is generally just a symbol in the symbol table
<joepie91>
so there's not even an 'enter key'
<sphalerite>
huh, I seem to remember my nokia phone back in the day having enter on 0 together with space
<__monty__>
For me it's space the down.
<joepie91>
sphalerite: newline, not enter, and as an alt for the {space,dot}
<joepie91>
ie. just the abbreviated version of the symbol table
<__monty__>
What's the difference between newline and enter?
<joepie91>
'enter' is a key (which dumbphones almost never have), 'newline' is a symbol, and the crucial distinction is that the former can issue UI commands whereas the latter is always a character inserted into the text (which is what ashkitten asked about :P)
<joepie91>
enter results in a newline in a lot of contexts, on systems that have an enter key
<joepie91>
but the reverse is not true; a newline symbol will not result in an enter keypress
freeman[NixOS] has quit [Ping timeout: 245 seconds]
freeman42[NixOS] has joined #nixos-chat
Synthetica has joined #nixos-chat
drakonis has quit [Ping timeout: 250 seconds]
drakonis has joined #nixos-chat
jtojnar has quit [Quit: jtojnar]
jtojnar has joined #nixos-chat
mudri has quit [Quit: Connection closed for inactivity]
freeman42[NixOS] has quit [Remote host closed the connection]
freeman42[NixOS] has joined #nixos-chat
freeman42[NixOS] has quit [Ping timeout: 245 seconds]
drakonis_ has joined #nixos-chat
drakonis has quit [Ping timeout: 276 seconds]
jtojnar has quit [Quit: jtojnar]
jtojnar has joined #nixos-chat
spacekookie has quit [Changing host]
spacekookie has joined #nixos-chat
spacekookie has quit [Quit: **agressive swooshing**]
spacekookie has joined #nixos-chat
mudri has joined #nixos-chat
drakonis has joined #nixos-chat
waleee-cl has joined #nixos-chat
freeman42[NixOS] has joined #nixos-chat
drakonis has quit [Quit: WeeChat 2.4]
freeman42[NixOS] has quit [Ping timeout: 245 seconds]
drakonis has joined #nixos-chat
drakonis_ has quit [Ping timeout: 258 seconds]
endformationage has joined #nixos-chat
drakonis_ has joined #nixos-chat
drakonis has quit [Ping timeout: 252 seconds]
<averell>
wow, a vim upgrade causes 500+ rebuilds.
<averell>
ah i'm dumb, it's just all those plugins
<eyJhb>
:D Still rebuilds? ;) :p
<samueldr>
yeah, not all rebuilds are equal :)
<andi->
"only rebuilds affecting me are valid rebuilds!" :D
evanjs has quit [Quit: Configuring ZNC, sorry for the joins/quits!]
<samueldr>
I was thinking more about the trivialness of the rebuilds; e.g. packing vim scripts is likely quick an trivial, we can probably chew through a couple order of magnitudes more than a ghc
evanjs has joined #nixos-chat
<andi->
vim ~= 0.0011 ghc rebuilds? :D
<samueldr>
it'd be nice to have something like LFS has, a "factor" of build time
<samueldr>
even if informal
<eyJhb>
I still cry a little inside everytime there is a update for VBox/Vbox Ext Pack
<andi->
Firefox went from compiling within 25min on my notebook to above an hour in just about 1.5y :/ I cry whenver I try to upgrade it.
<eyJhb>
But you have the option to use the cache, don't you andi- ?
<andi->
when packaging new version? I don't think so.
<eyJhb>
Ah
<eyJhb>
I thought you just meant on a update on channels for your system
<eyJhb>
So many things I should do, also regarding Nix, but so little time
Synthetica has quit [Quit: Connection closed for inactivity]
<eyJhb>
Anybody working on getting like, a extra five hours a day?
<sphalerite>
\o/ got a windows moved from bare metal to a nixos kvm hypervisor, with PCIe passthrough for high graphics performance :D :D :D
<sphalerite>
gchristensen++ for inspiration
<{^_^}>
gchristensen's karma got increased to 149
<sphalerite>
(this is a games machine, and the games run nicely in the VM)
<samueldr>
can the GPU handle the VM rebooting nicely?
<samueldr>
when it does, it's apparently a nice setup
<samueldr>
(my trashy old GPU can't :))
<aanderse>
gaming rig? <3 steam proton (aka wine)
<sphalerite>
samueldr: yes, seems to work fine
<averell>
does that need more than one GPU
<averell>
and a dedicated screen?
<sphalerite>
averell: yes
<sphalerite>
averell: unless you're happy with the linux being headless
<sphalerite>
averell: in this case, linux is using the intel integrated GPU and I have a monitor connected to that, but I haven't actually made use of it at all, doing everything via ssh
Myhlamaeus4 has joined #nixos-chat
vyorkin has joined #nixos-chat
<evanjs>
ashkitten: openrazer PR got merged. You should be able to drop the branch within the next few days :P
vyorkin has quit [Client Quit]
<samueldr>
even then, there is work to allow more complex setups, e.g. intel iGPU on the host, the dedicated GPU on the VM, and the display being shown on the host