u realize a battery cant read its mAh
#🔌│tech
fair gull
the only way to get the absolute accurate mAh is by either keeping track of it while charging or discharging and knowing the discharge current and how long it took
u can somewhat estimate a percentage by knowing what voltage the battery is at when empty and full
but true capacity and thus health is a function of time and requires active monitoring
astral haven
Yea
And ampere only reports what the phone says
Usually it says "battery health good" or "battery health poor"
wheat plume
I no see that stat
wait nvm
I do
yeh, says good, just wish it had % like iPhones
maybe you need to let it run in the background constantly so it gathers info?
fair gull
there deffo is
stark bobcat
@pale sigil you have windows on your steamdeck right? how is navigation?
empty wren
there deffo is
Accubattery
reef patrol
💯
ashen spindle
gfx1102 (Hotpink Bonefish)
I do prefer those codenames over boring lakes tbh
crimson wigeon
Nice they still do the fish names
ashen spindle
As an aside, Navi33 outperforms Intel’s top end Alchemist GPU while being less than half the cost to make and pulling less power.
crimson wigeon
pepelaugh
pale sigil
<@171594634605232128> you have windows on your steamdeck right? how is navigatio...
Without steam running pretty shit
stark bobcat
I figured no dual boot support anyway so 🤷
cursive summit
> As an aside, Navi33 outperforms Intel’s top end Alchemist GPU while being less...
Why I was saying Intel is DOA
It can't compete with midrange of this generation
And they are launching when next gen launches
Unless Battlemage is 3x the perf, they won't be relevant with next launch
ashen spindle
nah arc is kinda expecyed just late
astral haven
I was not expecting them to hit 2070 perf and they are at least raw perf wise there...
This is literally identical to larabee
About even with midrange with first gen, and could have been easily higher with software optimization
Second gen would have been significantly faster but was canned for consumer smh
ashen spindle
ehh im gonna stay hopefull
if only the hope they see the bigger investment as a sunk cost fallacy that has to pay off and maybe will
misty saffron
yep just another lanapost
misty saffron
no context, wrong channel, a remark that doesn’t clarify anything
wheat plume
xd
shadow minnow
stark bobcat
no context, wrong channel, a remark that doesn’t clarify anything
duuude lmfao
shadow minnow
Was this post posted by Lana bingo
stark bobcat
uhm
misty saffron
yeah if only it was obvious
wheat plume
VTubers are tech
stark bobcat
suuuuure
misty saffron
vtubers generally are people
stark bobcat
imo they are cringe but you do you
misty saffron
or online personas, rather
wheat plume
we dont have a cringe dedicated channel xd
stark bobcat
general very much looking like that right now
ashen spindle
pretty sure meme talk was that channel when I last looked into that hell hole last year
shadow minnow
we dont have a cringe dedicated channel xd
#🐸│memes
true quest
sucks that it doesn't exist
stark bobcat
quarantine zones need to exist
shadow minnow
^
wheat plume
LOL
misty saffron
and if you’ve passed quarantine
ashen spindle
oh right the channel that exists for whatever reason
misty saffron
cursive summit
How does one into #objectively-worse-name
wheat plume
🤔
misty saffron
you literally just click on it
cursive summit
wheat plume
:')
misty saffron
just try harder
shadow minnow
https://tenor.com/view/tyler-lie-lol-not-telling-the-truth-that-was-a-lie-gif-13...
Wdym works for me
Probably just a you problem
cursive summit
Of course it does, orange role
shadow minnow
I am not aware of the role granting you access there
Actually I think most of the time not all with the role had access? Not sure
cursive summit
Afaik it's a username whitelist
misty saffron
rename yourself to hollow then to test
cursive summit
Account whitelist, fine
But there is a role probs that autoallows the access
Doesn't BTTV actually just show hidden channels and their config?
shadow minnow
Afaik it's a username whitelist
Discord isn’t role or username. It’s always combination of both. And well, everyone who has administrator perms
shadow minnow
Account whitelist, fine
Hack hollows account
shadow minnow
But there is a role probs that autoallows the access
Wootamin config and founder I think
There used to be a role
But
It got deleted
I will never forgive for that
Would say never forgive name but idk if it was Erik or Calder now
wheat plume
stark bobcat
Instead of bringing back the role that should have been never deleted
being?
cursive summit
Even I have no clue what Bex is talking about
stark bobcat
I assure you, the chronobreak is coming
cursive summit
Who
stark bobcat
no one was just funny I remembered that quote due to the conversation about hidden channels
you dont wanna know
cursive summit
I didn't see light mode
crimson wigeon
Only those who have been here for some time remember the great role purge
stark bobcat
fat s1mple
cursive summit
That lad reminds me of snax more
vocal cipher
quick question: is one cycle of memtest86 enough if there are no errors? (or there are errors) Or should I finish at least 4 cycles?
astral haven
No number of memtest 86 cycles will tell you if there are no errors
Use occt, generally very quick to find errors
thick elk
Besides the rule that you cannot prove stability, memtest86 is not very efficient
Personally use stressapptest but different tests may load different things
astral haven
Besides the rule that you cannot prove stability, memtest86 is not very efficien...
Yea
Also this
The quickest test I've used that has found all but the most insidious of gpu memory errors is occt power
thick elk
Stressapptest works best for my cpu but only runs on linux and doesn't come with any well known distro by default.
system memory that is
astral haven
Have you tried the memtest in occt?
It's nightmarishly fast
Linpack medium is sometimes faster but also stresses the CPU and heats up my room quiiiick
Linpack large for Ryzen tho
thick elk
I have but stressapptest is faster, it can detect errors within 1 minute on a system that I used to normally game on for a day
astral haven
And occt doesn't?
thick elk
8700k but it was a while a go
stark bobcat
generally I've seen many people recommend prime95 for like 24 hours
if you are doing memory
thick elk
Nothing against occt it's just that stressapptest is tailor made by google to find memory issues in datacenters, and well optimized for large amounts of memory.
vocal cipher
but OCCT seems to run on windows and isn't that limits its test?
I mean memtest boots up on usb stick so it can test everything
astral haven
generally I've seen many people recommend prime95 for like 24 hours
Prime95 is capable but doesn't report error count or support stop on error
astral haven
but OCCT seems to run on windows and isn't that limits its test?
Nope
Paging exists
Memtest86 is simply not as much of a load on the memory :/
astral haven
Yea
It can only test the amount that isn't in use
But it will force windows to move around the memory, and after a few tests you can get something like 99% coverage
astral haven
You are probably fine for all but the most ram harsh games
If it can run for the full hour you probably won't run into an error whatsoever
vocal cipher
thanks, I guess what I have is mostly driver issues
vocal cipher
how long?
cursive summit
1up you
reef patrol
smh it's like looking at slack
stark bobcat
AHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH
frozen ingot
slack actually has dark mode now
stark bobcat
you maniacs would use light mode on steam if it had one
pale sigil
Can you split hdmi from one HDMI port
crimson wigeon
Actually there was such a steam skin when skins were still commonplace
pale sigil
Not mirror but basically have two monitors from one HDMI
@ashen spindle you're my go to expert
stark bobcat
I could test that if I had some decent hdmi cables
pale sigil
Considering sucking it up and buying this, especially if I could split
stark bobcat
probably should I heard it's good but they have some trouble getting them all out
like too much demand
ashen spindle
idk splitting video cable stuff is cursed
it should work but enough bandwidth etc. is needed and gl testing that or some cable might just not work
junior ore
Not mirror but basically have two monitors from one HDMI
No this isn't really a thing. Not without an additional brick/attachment.
Are you wanting a Switch-Box or multi-monitor?
Switch-Box lets you plug multiple devices into a single monitor* if you are just watching to switch between multiple devices.
but if you are plugging multiple monitors into a single device and the device has only one video connection : you need a dock port.
which is going to be either a USB C or universal USB 3 kind of thing.
fair gull
Not mirror but basically have two monitors from one HDMI
yes
ud just have the same on both
fair gull
displayport can daisy chain aswell
junior ore
🤔 I always forget that but yeah they can.
fair gull
i think hdmi technically has 2 video streams it can send but i wouldnt know of any device thaf can split those or send 2 monitors over hdmi to begin with
junior ore
the lack of their response causes me to internally scream because I can't help without further input because my old jobs was all about doing this aaaaaaaaa
don't even know which connector their need or what device\
stark bobcat
hdmi steamdeck I think
junior ore
ah
I've been using the heck out of a Dell Thunderbolt docking station ever since I got the device
am thinking of trying out the lenovo docking station too
its an older USB-C dock though but it does the job too pretty sure.
pale sigil
No this isn't really a thing. Not without an additional brick/attachment.
Are y...
Multi
junior ore
for steam deck?
well if you can get a Displayport 1.2, you can Daisy chain
as an example
but this requires a Displayport cable rated at HBR2 or higher, which is 1.2 and newer
and is specifically called Multi-Stream Transport, or MST
also the monitors must support it the feature.
like. it's a dedicated Output port kind of thing
shrewd thicket
well if you can get a Displayport 1.2, you can Daisy chain
If your OS supports display port multi stream transport* *cough apple*
junior ore
lol
cursive summit
Uhh
I don't think Nvidia or AMD ever dropped multi-GPU
2080Ti had the NVLink bridge
ashen spindle
they did officially drop SLI
cursive summit
I don't remember that current gen has one, but that doesn't mean that multi-GPU doesn't work
ashen spindle
Intel Arc graphics does not support multi-GPU for gaming
cursive summit
Like wouldn't I be able to slot 2 RX 6700XTs and they would just work together
cursive summit
Not at double the efficiency but with a sizeable perf boost
ashen spindle
no not really unless its an older game that has support for that
cursive summit
What about productivity then
ashen spindle
depends
cursive summit
Something like Blender could support 2 or more GPUs
cursive summit
Good that you said "almost"
deft marlin
I remember a friend telling me when his cousin gamed in the mid-2010s that SLI was a huge perf boost
astral haven
The worse the GPU the more of an improvement sli can be
deft marlin
But now optimisation is king, Moores law is barely effective now
and we’ve managed to ditch the need for SLI
astral haven
Moore's law is accelerating
deft marlin
at least in less intensive tasks
astral haven
And we are outscaling sli
That's why sli isn't supported on 30 series
Explicit mgpu over nvlink is what is supported
deft marlin
I mean all I can think of which would make sense to have SLI eternally would be cryptomining
astral haven
Nope
deft marlin
not “as is implemented”
astral haven
Sli never made any sense for mining
Sli is a set of specific technologies
Sli is basically just for syncing the GPUs together
For most gpgpu applications (render, scientific, mining) the syncing is done after several seconds worth of processing or even minutes
For games it's milliseconds
deft marlin
._.
astral haven
Also you should believe me on this as I own and have tested all the pcie multigpu consumer cards besides the 295x2
I even have 2 gtx 690s
Sli works
But it is not a pleasant experience
Now it's aloooooot better of an experience than crossfire
But on a scale of Linux to windows in terms of speeding up games it is closer to Linux
Faster in 1/100 games, equal in 80/100 and slower in 19/100
ashen spindle
and more stutters in every game aside from sniper elite lmfao
deft marlin
Sli is basically just for syncing the GPUs together
For most gpgpu applications ...
So you’re saying the improvement is barely marginal for GPU intensive workloads but great or used to be great for gaming
astral haven
and more stutters in every game aside from sniper elite lmfao
Nah I'm including stutter in that quin
astral haven
So you’re saying the improvement is barely marginal for GPU intensive workloads ...
That's not what I'm saying
I'm saying it just isn't used outside of games as it's not the correct strategy for syncing compute tasks
The stutter was mostly with older gpus, and they improved the sync and scaling massively using tiled rendering rather than afr
Tiled rendering basically doesn't add any stutter, but it is very very bad at lower resolutions (as low as 0 performance improvement at 720p in games that get +60% at 4k)
mild dome
Sli suffers a problem in two parts
- The game engine needs to support it in some capacity, otherwise you might see no improvements or even worse, a performance deficit. This creates a negative feedback loop where developers aren't willing to support something no one uses. No one is willing to use something that isn't supported.
- It wasn't very cost effective. Instead of buying two GPUs you might as well just buy 1 bigger gpu. This is probably why only the flagship cards retain SLI support, because you can't get anything bigger then a 3090 ti.
- One of the primary reasons irc that sli became popular was to drive high resolution monitors. Nowadays upper middle tier cards can do that just fine.
astral haven
1 is wrong, most of the optimization is done by selecting a few hidden settings in nvidia inspector, and that was distributed by driver and can be manually selected by you
2 is not quite correct, 2 gpus that are 60% faster is cheaper for nvidia than 1 gpu that is 60% larger thanks to the crazy scaling of defects vs die area
3 is correct but you didnt realize it, it was for high resolution monitorS dual or triple monitor setups mostly
oh also about 2, 30 series DOES NOT AT ALL support sli
it supports nvlink acceleration for explicit mgpu
thats where the myth about requiring game engine support comes from, explicit mgpu is not SLI, and it does rely on the developer to implement it
mild dome
-
Long before the 3000 series came along I had heard about spotty support for sli when researching into building a pc with dual 980 TIs
-
I don't believe that tbh. And I am talking about in terms of the consumer, not nvidia
3060 is $330
3080 is $699
Looking up performance numbers the 3080 usually gets around 100% more fps then the 3060. Whereas a max performance on sli 3060 would only net you 60% extra only in games that have very good sli support
astral haven
this is the options for opengl, which is a very short list compared to the options for directx
all of these are a binary toggle, and each of these settings can be added together for performance
astral haven
1. Long before the 3000 series came along I had heard about spotty support for s...
just because nvidia is charging out the ass for the 3060 doesnt mean a dual 3060 gpu would be that expensive
junior ore
SLI/multi GPU has been around for a stupid long time. Yes it produces questionable results that, ultimately, is going to rely on the individual game on if it supports it.
astral haven
again no
junior ore
also : there is no more SLI. or Crossfire.
astral haven
questionable results? yes. relying on the game? no.
sli does still exist, just not in new hardware, it will stay supported in drivers for a while
heck 3d vision was only dropped last year lmao
junior ore
ergo : it's irrelevant. because you have multi-GPU support elsewhere. And they found it wasn't worth diving into further anyways.
astral haven
also no
astral haven
nvidia is pursuing multigpu hard right now
astral haven
they just stopped developing explicit mgpu and they pivoted to developing transparent mgpu
its not done yet ofc
junior ore
those AI cards and quadros
astral haven
but they are working on it
nvlink is not what nvidia is working on for mgpu
nvlink is like pcie
junior ore
I'll believe it when I see it but it's not worth it, for real. On the consumer end.
astral haven
its just an interconnect between cards and cpus
mild dome
just because nvidia is charging out the ass for the 3060 doesnt mean a dual 3060...
Just because a 3060 costs a lot doesn't mean two 3060s would cost a lot
astral haven
5 years from the release of the 2080ti
mark my words
astral haven
>Just because a 3060 costs a lot doesn't mean two 3060s would cost a lot
<:Barel...
dual 3060 is not the same as 2 separate cards
that is 2 very different things
1 card, 2 gpus
shared power delivery and io
junior ore
it sounds like you are drugs.
astral haven
packaging is also alot cheaper
junior ore
Dual-GPU means. Two.
astral haven
1 card, 2 gpus
mild dome
Those existed and also cost a crap tonne amount of money, generated a lot of heat, and once again had really spotty support.
junior ore
no one here is talking about thar except you carmon
mild dome
Like if you were talking about chiplet design that would be something different, but you are talking about a two seperate GPU solution
astral haven
support was fine, they generated very little heat besides the 295x2 from amd, and they were cheaper than 2 separate gpus
700/2 < 500
mild dome
Had to lose a lot of clock to get that
astral haven
this is what dual gpu looks like
mild dome
They couldn't sell it for the full price because they knew no one would accept an SLI 580 solution clocked 20% lower
astral haven
gtx 580 sli was faster than the gtx 590 yes
but it also was a full 100w more
astral haven
They couldn't sell it for the full price because they knew no one would accept a...
except thats not true and the titan z proves it
mild dome
Those two GPUs were on seperate SLI x8 lanes. They acted as 2 SLIed cards for all intents and purposes
astral haven
except thats not true and the titan z proves it
titan z was dual titan black, less than double the power, 3x the price
mild dome
https://www.reddit.com/r/buildapc/comments/p8fbb/whats_the_difference_between_two_gtx_580s_and_a/
"The pair of GTX 580s is better. The GTX 590 is a card with two GPUs that are internally SLId that are midway in performance between a GTX 570 and GTX 580. It's a similar deal with the 6990 and a pair of 6970s. The main difference is that the top card in dual card setup will run hotter."
https://gagadget.com/en/graphics-cards/geforce-gtx-580-vs-geforce-gtx-590/
So fantastic, the 590 was outperformed by the lower priced 580 in some cases
astral haven
each gpu had a full x16 connection
instead of the 1GB/s standard SLI connection, these cards had a direct connection between each other of 8GB/s
astral haven
https://www.reddit.com/r/buildapc/comments/p8fbb/whats_the_difference_between_tw...
no
each gpu on a 590 was outperformed by a 580
in sli it was consistently better
astral haven
if you dual gtx x90 gpus tho thats gonna be a baaaaaaaaaad time and it is gonna cost a fortune
but if you are comparing gtx 580 sli to 590, the 590 is consistently faster than the price difference, and sometimes can eek out a win with the extra intergpu bandwidth
deft marlin
Now what Crossfire and NVLink
astral haven
crossfire is radeon's competitor to sli
nvlink is nvidias competitor to pcie*
*except you use it in addition to pcie
mild dome
Could I have a source on those claims like:
"instead of the 1GB/s standard SLI connection, these cards had a direct connection between each other of 8GB/s" - I recall one of the issues these dual GPU cards faced is that they still needed to go through the CPU for all communications due to the limits of PCIe at the time, killing the benefit having two GPUs on one card would supposedly bring.
"each gpu had a full x16 connection" - this is actually impossible, as the card used a single x16 slot, not a mythical x32 slot, I assume you meant they could share a single x16 slot (which really means they have to take turns with the bandwidth), but from what I recall reading years ago each GPU was physically wired with x8 traces.
astral haven
aha look closer at the pcb
whats this mysterious nvidia branded chip that has 48 pcie lanes worth of traces running in and out of it 🤔
mild dome
That picture doesn't have the detail you think it does
I can't fukin read such pixelated text
My eyes aren't a cheesy cop show, I can't yell "ENHANCE" and get an image
astral haven
this is called (not actually because its nvidias version of) a plx chip and it basically allows each gpu to communicate to each other at full x16 speed, while also allowing each gpu to talk to the cpu at a full x16 worth of bandwidth
the reason that they didnt need to share bandwidth is because data going into vram (which is the high capacity stuff) could be sent to the primary gpu then copied over
and the stuff that needed to be blasted over (frame data) wasnt sent simultaneously to both gpus
so you could send your entire frame worth of data to gpu 1, and send it twice as fast as an x8 connection would allow, then you can send your different data to your second gpu
heres a screenshot of aida64 getting very very very confused about gpu bandwidth (a 9900k has a total of 12.6-12.8GB/s of pcie bandwidth and this test is showing 49GB/s, which is faster than the system ram)
the way to recreate these results is to attach a pcie switch directly to the cpu to give yourself 2 x16 pcie slots
then you take 2 dual gpus that each have a pcie switch on them and connect to that.
mild dome
I see, never heard of this chip before, but that's likely because I mostly looked into only dual slot SLI.
astral haven
yes
and multi card sli is not the same as dual gpu cards
it doesnt help that nvidias SLI marketing is ATROCIOUSLY AWFUL
shadow minnow
At least apple managed to make multi GPUs work well
astral haven
No they didnt
They got multigpu functionality for free because noone buys a Mac and benchmarks them in eSports lol
Show me an m1 whatever vs the dual whatever with the around 20% scaling that sli gets you in CSGO and I'll be very very very shocked
And that's with 2.4tb/s of GPU interconnect bandwidth vs 2-12gb/s
arctic mantle
anyone have experiences with Moondrop IEMs ?
astral haven
Chu chu train
arctic mantle
yeah ive heard the chu is amazing, since is so so cheap, i already ordered one for shits and giggles
how about their more expensive ones
i m eyeing the blessing 2 and kato
mostly for their look
final flame
The starfield if good but on the more relaxed side of tuning if you prefer that
arctic mantle
The starfield if good but on the more relaxed side of tuning if you prefer that
that would be pretty nice
how is it compared to the aria
since they are basically the same price
also the kato and blessing, i do see a lot of ppl saying kato is worth the price up from aria, but i havent heard much about the blessing
vocal cipher
for those who have msi gaming edge wifi motherboard ryzen 3700x, don't update you BIOS above 7C37v1F-AGESA - 1.2.0.3b it miserably fails...
final flame
how is it compared to the aria
I don't know personally since I only own the starfield, but apparently the main difference is in the soundstage, the snow version of the aria is a brighter tuning
arctic mantle
no
cursive summit
Are you from Texas
arctic mantle
no
cursive summit
Have you been in Texas
cursive summit
So you slept with Texas?
arctic mantle
cursive summit
final flame
So you slept with Texas?
I wish
cursive summit
There's no point going for top end Moondrops imo
Aria is probably as high as one should go
final flame
depends on your use case, if all you use iems very commonly then it may be a decent investment
cursive summit
Why not just get a pair of Sony IEMs
final flame
always could try it out and return it if it's not good enough for you
final flame
anime girl on box drip
cursive summit
Sony has sound, though
arctic mantle
honestly, i would prioritize drip for this
like the Kato, the design is so unique
instantly stands out
final flame
not to mention aren't sony iems really large
arctic mantle
fair gull
it doesnt help that nvidias SLI marketing is ***ATROCIOUSLY AWFUL***
because sli doesnt exist anymore maybe thats why
you wont find the firmware as its not opensource. afaik its written in c/asm
final flame
Sli was cool but it had several issues, Nvidia never ended up fixing those issues and decided to just kill off sli
arctic mantle
Like is the Kato worth getting when the aria/starfield are almost as good for half the price and the blessing is noticeably better for 50% more
fair gull
that looks inefficient
chrome estuary
everyone else sends random tech shit youtube channels. so here's my contribution. dawid does some fun things
ashen spindle
the fun random stuff is always the best
reef patrol
that looks inefficient
Indeed - They're a DIY crafting/building channel who also stream games on the side. 😄
chrome estuary
i kinda want a triple monitor gaming setup. I mean I have a triple monitor setup but it's two office monitors and a gaming monitor, all in different resolutions.
thick elk
I have tried it in the past, most games just don't scale well / it generally was not worth the hassle
stark bobcat
not to mention aren't sony iems really large
Blessing 2 has a large nozzle
final flame
Blessing 2 has a large nozzle
I meant more the housing not the nozzle
stark bobcat
Well tbh I'd be more bothered by a large nozzle than a large housing
Just small ear canal things
thick elk
I know the trouble, I feel both
pale sigil
Why
wheat plume
its a network security device and can be used to crack passwords and encryption by cpaturing handshakes and shit
and auto importinf to wireshark to get more info
wheat plume
its neat to me shush xd
pale sigil
Someone used one of these in uni for their dissertation
Sat in the hub
Did it for the uni wifi
Ended up getting in a lot of trouble... They were doing it for educational reasons.
Pretty moronic
wheat plume
yep
stark bobcat
6400 
still tho the intel gpu sadly isn't worth buying I wonder how it is performing in china
reef patrol
someone didn't use the bezel adjustment tools.
chrome estuary
he just pressed the button for enabling all monitors as one
ashen spindle
ok
wheat plume
67W is such a specific wattage to make your charger work at
ashen spindle
what specific wattage is not specific?
ashen spindle
67 isn't any more or less specific than 40w, its just not a round number
thick elk
Of course arbitrary would be the better word
ashen spindle
even then one isn't more or less arbitrary than the other
if your voltage/amperage makes it so you have 67w its the same as if that result is 60w
and if you just eyeball them then both are arbitrary
final flame
Custom photoshop splash screens are cool
cursive summit
Expected from a wehraboo
junior ore
Don't you mean Weaboo? Or are you trying to use the wehrmacht one?
junior ore
sus
arctic mantle
shadow minnow
Anime art
thick elk
How do they feel?
thick elk
Deadzone really did not bother me much, as when I am playing games where this difference matters I much rather use the trackpad anyway, does seem tempting though
arctic hedge
I'd say do it when yours start to drift in a year or two then and not before
Since you'd get no benefit
thick elk
The one thing I really dislike about the deck's analog sticks it won't fix anyway.
that is the surface on top
arctic hedge
The one thing I really dislike about the deck's analog sticks it won't fix anywa...
Yeah, no good fix for that unless you don't want capacitive
thick elk
To be honest I never used the capacitive feature anyway
mostly because motion controls suck in a moving train which is a common place for me to use the deck.
Still tempting to buy them, though
stark bobcat
I wish steamdeck had more battery life
thick elk
Sure but it's still managable at least for me. If I ever use it on longer trips, I will probably use a large powerbank to top it up. Using a good framerate cap with matching refresh rate also helps a lot.
stark bobcat
Framecap on the steamdeck adds unbearable input lag
Already capping to 11w
Get ~2 hours
On a plane I could at least lower brightness more
Sadly on the train I wouldn't see shi if I did that
thick elk
on my daily commute, I use it for 1.5h max and never felt too short.
Input lag sure, but not a big deal for all games.
I can even charge it at work in between
never had to
but I can definitely see it being to short for you
stark bobcat
I pretty much only play racing games while commuting
I have a commute of about 2 hours
thick elk
Currently playing nier replicant, before that paradise killer, not exactly games that cared much
I did at some point disable the cap on black mesa when it could no longer hold a stable 60
stark bobcat
Currently playing most wanted 2005 really solid 60 with 11watts and mods
Next up in forza
Battery will suffer there for sure
I will probably never enabled steamdeck framecap unless it's a turn based game
It feels soooo bad
pale sigil
@ashen spindle what's a good guess on when and will reveal the new gen
cursive summit
Of what
ashen spindle
maybe a teaser around gamescom but probably more like last gen in september is my guess
pale sigil
Mainline
cursive summit
NEW GEN OF WHAT? 
cursive summit
New gen of AMD what products?
ashen spindle
yah that was obvious
cursive summit
CPUs is kinda already known
pale sigil
Stop being a tool
cursive summit
Aug 28th announcement September 15th launch
pale sigil
It's obvious what I'm on about
pale sigil
Yes well I'm asking for a reason as I clearly didn't know
Anyway, I'm going to hold off on a 5800x
cursive summit
We already talked about Ryzens plenty, links were shared
I'm more interested in Radeon rn, not enough info
pale sigil
You think I'm paying attention to your messages? 
pale sigil
I'd think you are paying attention to your own discord server
Have bots for that
cursive summit
That seems like gross negligence
You can give me that pretty pink role, I'm more active than you are 
ashen spindle
paying attention and reading every link is a massive difference
hell you dont even read half the articles you post yourself
cursive summit
Nobody was talking to you, so shush
stark bobcat
hell you dont even read half the articles you post yourself
Same
I heard 5800x3d prices might drop
If they do I'll buy it
cursive summit
I don't think it will drop before next 3D launch
stark bobcat
🤷 also I don't really know how that would be the case
Next gen will have better performance all around anyway
cursive summit
Yeah but next gen still won't match the 3D cache levels so it's still a different kind of product
stark bobcat
But 5800x3d is the best am4 cpu so it might stay expensive for a while
cursive summit
Also I bet 5800X3D will still be competitive with Zen 4
stark bobcat
Like intel in socket upgrades
cursive summit
So I don't think we'll have a price cut for a few months for the 3D parts
Now the other SKUs of 5000 family, yeah those might get cut
stark bobcat
Yeah but next gen still won't match the 3D cache levels so it's still a differen...
Performance is all that matters if it outperforms people who wanna upgrade their motherboard+ram+cpu
will get that anyway
stark bobcat
It's not just a node jump we will see how performance pans out
cursive summit
That is true, but imho 5800X3D will hold its own compared to Zen 4
Eh not that long to wait, few weeks
umbral knot
the flamethrower was also pretty cool
but also did people not expect it to get slower
stark bobcat
Yes but also you could watch the video
umbral knot
I am rn
shadow minnow
Hi rn nice to meet you
umbral knot
👋
wheat plume
uh
stark bobcat
seems like a linux problem to me 🤷
wheat plume
I just get it from Flathub?
then I cant use Sandbox right???
flathub vs native
native is out-dated
DAMN IT
wheat plume
that was Firefox xd
but this seems to be suffering the same faith
Chromium is updated
tbf, im getting Chrome from the ArcoLinux 3rd party repo, so whoever maintains that package is not doing a great job rn
pale sigil
slow garnet
i agree fuck all changes, but if you're on shady websites, you're gonna want that update
woops, wrong channel
i'm tired af
too hot
wheat plume
I swapped to Chromium
Which is in the main Arch repo so it should be always up to date
I wonder how many people are in the archlinux dev team
nah, probably more like at least 2, Id think
why does the browser need to reload the page when switching between light and dark theme in youtube??!!
its never done that before for me until now
dem bezels 
too thick
reef patrol
Smells like chromebook.
cursive summit
Pentium N 🤢
reef patrol
More like Pentium No
cursive summit
I like how this thing can't boost even
So it's a "burst"
So it achieves that burst clock for like 5 secs and goes back to 1.1GHz
thick elk
These processors do have their place, but definitely not in the > € 300 market
cursive summit
For 500 euros you can get something perfectly normal and functioning
Like a second hand gaming laptop from 2018
wheat plume
In Ireland you still see laptops with 768p displays and 4gb ram and 64-138Gb storage selling for €400-€500 -.-
cursive summit
Then don't live in Ireland
Move to somewhere where it doesn't suck and isn't a third world country
wheat plume
Ikr
Give me money to support that move
cursive summit
How much do you need
wheat plume
A lot
cursive summit
Need specifics
cursive summit
That sounds unrealistically large
You want to move into an apartment you bought or something?
deft marlin
Probably trying to shift the entire family elsewhere
wheat plume
Fuck no
I want to avoid my family
cursive summit
Probably trying to shift the entire family elsewhere
Ngl, knowing Lana they would bury them
And take the current house for themselves
Problem fixed
deft marlin
Well that explains the price actually
thick elk
It's basically a higher clocked cpu from the € 250 tablet pc I have. Useful for watching some tv show and very light usage but no more than that.
deft marlin
Whenever your parents catch up you bribe their close friends to let you go
deft marlin
It's basically a higher clocked cpu from the € 250 tablet pc I have. Useful for ...
Wrong channel?
cursive summit
That's why you leave at 3AM through the window and trash your SIM card
deft marlin
Oh nvm
arctic mantle
8ts been a day, and i m still debating whether to get the moondrop Kato or Blessing2/dusk
Or just wait for a new blessing
final flame
the only other option in that price bracket is the timeless
deft marlin
No way you’ve just said that
arctic mantle
final flame
if I were you I'd try the dusk or blessing and if you dont hear a difference after using them return them and get your money back
final flame
Overear headphones <:mm:638142954841833512>
I prefer iems since they're easier to carry on you because of the small form factor
arctic mantle
^
final flame
overear headphones you have 2 options to carry them around the neck or in a bag where they could get damaged
arctic mantle
Headphones.com don't have the dusk
Dusk only on shenzhenaudio and their shipping is slow af
final flame
only Shenzhen audio has the dusk I'm pretty sure
final flame
main difference between the dusk and blessing is the tuning
dusk has a bit of a bump in the bass
arctic mantle
ik, but from what I see the dusk is exactly what a wants, just slightly more bass and slightly less upper mid
final flame
you also get crinacle waifu
arctic mantle
3 weeks shipping
From 4px
Instead of giga Chad FedEx space time defying shipping
final flame
meanwhile Dhl has my package sitting in their facility for 5 days in fort worth Texas
arctic mantle
A had something shipped from Texas to Montreal in 12 hours from time of FedEx picking up the package to getting it in my hand
Btw how's the ie300
stark bobcat
too many anime avatars
arctic mantle
stark bobcat
arctic mantle
Anyone have experience with shenzhenaudio shipping ?
stark bobcat
idk I selected the cheapest shipping option and it came in under 2 weeks
when I order from china I usually expect a month so I was pretty stoked
arctic mantle
How did you select cheapest option
I can't even pay for express shipping even if i wanted
stark bobcat
no clue was a while ago
arctic mantle
It just default to free shipping
I would gladly pay another 30-40 for FedEx express 2 days
stark bobcat
wheat plume
when I made the reservation, I thought I had to pay the full price there and then
so yeah.....
didnt have enough money for the 512 at that time in my account
wheat plume
ive never preordered a thing before this
so yeah...
i didnt know that you dont pay upfront
stark bobcat
true quest
twas in the faq
cursive summit
cursive summit
That is a pretty phenomenal deal
stark bobcat
it was there from day 1
stark bobcat
https://www.hardwaretimes.com/amd-radeon-rx-6900-xt-discounted-to-lowest-price-y...
great price but that power consumption tho
wheat plume
Oop...
The "After Q3" batches got split between Q3 and Q4
Idk how they decided to choose who goes where
I'd assume reservation time
stark bobcat
yes
arctic mantle
Same price get dusk and wait 3 weeks, or get regular blessing 2 and get it in 2 days
stark bobcat
Mine is q4
so you can be patient?
arctic mantle
Well i want my iems cuz i don't have anything similar
And I don't really need the steamdeck now
arctic mantle
Yes
stark bobcat
IT IS VITAL
wheat plume
Installing mods for cyberpunk on Deck will be frustrating without Vortex to handle it
;_;
stark bobcat
vortex is ass and for pussies
arctic mantle
wheat plume
Vortex is ass to setup where it stores mods
arctic mantle
IEMs IEMs IEMs
wheat plume
But it's fine other than that
stark bobcat
worst mod manager I have ever seen tbh
I wish I could just go without mod managers entirely but load order is important sometimes sadly
wheat plume
Load order doesn't seem to matter on cyberpunk
Well maybe you'd need cyber engine tweaks and red script to load first
stark bobcat
never played cyberpunk in my life cannot comment
wheat plume
I've almost maxed out the technical and hacking specs
So I can kill everyone by quick hacking them now xd
Stay hidden
Hack them
I don't take any damage and don't need to use my weapons
It's really fun
wheat plume
how do I force Chrome to open on my second monitor?
I managed to get it to open in the same sized window each time but it only launches on my main screen
stark bobcat
linux problems
wheat plume
nah
--window-position problems
it launches on second screen now but at the top, it ignore the second coordinate
stark bobcat
why do you need this
ashen spindle
my chrome just opens in the spot I close it in
stark bobcat
same
wheat plume
So does the regular Chrome window but not the sandboxes I made
theyd efault to the half width tall window that chrome is in on first launch
this should work
wait no
added an extra 2
this works :D
wooo
shrewd thicket
Yes lmao
This is the laptop setup
I disabled the title bars because I love the meta + right/left click for window drag and resize
wheat plume
I love global menus 💜
shrewd thicket
And ofc window shortcuts
wheat plume
I disabled the title bars because I love the meta + right/left click for window ...
I do that too......
It was in my old rice
dusky aurora
shrewd thicket
It's way too big for me
wheat plume
Add some padding to konsole damn man xd
shrewd thicket
I hope there's gonna be 4080's that fit the lian li q58
shrewd thicket
Add some padding to konsole damn man xd
Yeah I definitely did that
dusky aurora
Is it gonna be possible to get a build with a 4080 under 2k
wheat plume
shrewd thicket
Is that latte dock?
Yeah latte ftw
ACTUALLY
No
I forgor
I switched to plasma default dock at the end
shrewd thicket
But probably still using latte seperator package or something idk
wheat plume
Oh hmm...
shrewd thicket
Yeah at the end I switched to a single menu bar
Just plasma default one
No latte package
I did have padding and seperator widgets tho
Idk where they're from
I remember latte crashed with my super fancy desktop management widget
In Wayland
So then I switched to plasma dock so I could use Wayland
wheat plume
This was my really old rice
shrewd thicket
Gives me vibes from my old rices on debian
wheat plume
Xd
shrewd thicket
wheat plume
How TF you blur vs code?
shrewd thicket
Some girl made a electron injection thingy majingy
For transparent electron apps
But then quit doing it
But then I figured out
U can actually do it in Linux natively
wheat plume
O...
shrewd thicket
- some custom css
But I don't have that css anymore
Also because it uses injection it needs to be re-done each update
So not worth it imo
wheat plume
Hmmm.....
Only works on qt I'm guessing
Gtk still doesn't support blur 🤦♀️
shrewd thicket
Aka unreadable
wheat plume
Ye
shrewd thicket
I loved doing vaporware cyberpunk rices
I have a bunch more but never shared them on the Unix porn discord which is basically my rice archive
wheat plume
That sub got boring to me real fast
It's just the same twm setup every post basically
shrewd thicket
It's funny how early into Linux I used to put effort into customizing every piece but eventually just found settings I liked and went for clean minimal setups
From this
To this
Most customization is just settings and a menu bar
That's it
Where before, custom discord mods, vscode mods, Spotify mods, random ass themes from the internet scraped together etc
wheat plume
It takes so long to setup plasma the way I like
shrewd thicket
Same but I've configured it so many times that I know where each setting is
And I know what it should be on
wheat plume
Yeah me too pretty much
shrewd thicket
And I'll notice it very quickly when it's not as it should
wheat plume
I'd love to go back to using it
But I'm still worried about the desktop crashing during games
shrewd thicket
I'd love to go back to using it
But I'm still worried about the desktop crashing...
Dunno man, never had issues at all
wheat plume
I didn't either until last year
wheat plume
So either kde or I did something
wheat plume
I use fedora on my laptop
Kionite is freaking amazing
wheat plume
Yep
shrewd thicket
I wouldn't use another distro even if they use btrfs tho
wheat plume
And pipewire and Wayland
shrewd thicket
Cuz the integration with opensuse is just too good
wheat plume
I think fedora uses Wayland on Nvidia systems in their latest release
shrewd thicket
Opensuse also on pipe wire and u can select Wayland in the install setup too
Tumbleweed is literally just as modern as arch and don't fall behind whilst maintaining unmatched stability and in case of emergency unmatched recovery
I simp for this distro
wheat plume
I've never used opensuse
But I wish more OS had a YAST equivalent
shrewd thicket
Yeah
Unfortunate yast is made in ruby but otherwise great tool
I love the yast tui too
wheat plume
Arco Linux made the Arch Linux Tweak tool which is kind of like yast
I use that on all my sisters whenever I use arch
Sisters?
Systems*
shrewd thicket
U use a tweak tool on ur sisters?
wheat plume
Xd
shrewd thicket
Damn arch users really do be modifying everything
wheat plume
Yep yep :')
wheat plume
Bbyyeee
wheat plume
Oop
amber marsh
i got a somewhat stupid question rn, a 2nd nvme obviously uses more pcie lanes, do the pcie lanes for the nvmes split up, or does it use "gpu lanes"?
aorus x570 elite (rev 1.0)
r7 5800x
rtx 3070 (pcie 4.0 x16 according to gpuz)
980pro 1tb
970pro 1tb
fair gull
u have 24lanes
so with a gpu and an nvme ud only use 20
should leave u with 4 for a second nvme
reef patrol
And that's just the CPU lanes
reef patrol
Yeah in essence you've got the motherboard lanes on the southbridge, which uses "hidden" lanes on the CPU to communicate between the two as needed.
amber marsh
u have 24lanes
okay, i thought it was 20 for some reason ... ("20+4" ) idk why i interpret it wrong
reef patrol
So to answer your question - Unless you've got two GPUs you generally have nothing to worry about. Except if you're mixing SATA and M.2. There's usually some kind of limitation here if you read the manual page on the M.2 sockets (if x is plugged in, then y doesn't work)
amber marsh
thank you all
fair gull
okay, i thought it was 20 for some reason ... ("20+4" ) idk why i interpret it w...
technically thats true since the +4 is whats used by the chipset but in ur usecase ull most likely not suffer any downsides
mild dome
Yeah
16x goes to one of the x16 slots (sometimes there is a switch to divide into two x8 slots if need be)
4x goes directly to a single NVME
then the last 4x goes to the Southbridge chipset, and that splits the bandwidth between additional NVME drives, SATA, computer ports, Sound, Wifi, Ethernet.
It's designed like that as consumers usually have a single GPU which needs x16 (though as we moved from PCIe 3.0 to 4.0, and now 5.0 really isn't the case)
A single drive containing the OS (So having an NVME drive with it's own dedicated lanes makes sense since the OS will be using that drive the most)
And a whole bunch of other things that don't need that much bandwidth, and usually aren't utilized all at the same time.
reef patrol
Almost no GPU needs 16 lanes.
mild dome
Hence "as we moved from PCIe 3.0 to 4.0, and now 5.0 really isn't the case"
High end ones barely needed x16 3.0 lanes, with 4.0 you could easily get away with 8x lanes with no performance penalty, and now we got 5.0? Wonder if they'll ever find a way to saturate so much bandwidth?
ashen spindle
not quite accurate as a 3090 is still doing like 98% performance on pcie 3 x8
reef patrol
I don't know about you, Qiun. But I'm personally willing to chalk up a 1-5% performance loss (in a few games, not most) up to "not needed" 😄
ashen spindle
yes yes that was in support of your argument
as in even high end does not need 3 x16
shadow minnow
Would be nice if pcie spec allowed for x6 and x12
ashen spindle
I recon next gen looks different though
ashen spindle
Would be nice if pcie spec allowed for x6 and x12
I'm not sure if the spec doesnt allow that or if that is just not a thing because it would drive costs up for no real benefit
shadow minnow
Well rn limiting GPUs to x12 would probably have no bad performance impact and allow 4 lanes to be diverted somewhere else
Especially with pcie 4.0 having it x16 with 4 lanes divertable to ssd would be nice
ashen spindle
especially with pcie 4 you can just put it to x8 and have 8 divertible lanes
shadow minnow
Well yeah but on 3.0 it can limit you a bit while x12 wouldn’t
ashen spindle
again, like 2% on a 3090 which came out when 4.0 started to be a thing
mild dome
It's added cost
Were I to hazard a guess, they realized it would be cheaper to implement a switch then to design the system to divert an arbitrary amount of lanes
shadow minnow
Okay then you can have x12 on 4.0 to get the extra 2% instead of x8 4.0
ashen spindle
so youre asking for chips to split the lanes that drive board costs for a difference that maybe exists at the high end many years after the board comes out
cursive summit
Pricing in Canadian Dollars:
AMD Ryzen 9 7950X – $1158 (906 USD)
AMD Ryzen 9 7900X – $798 (624 USD)
AMD Ryzen 7 7700X – $631 (493 USD)
AMD Ryzen 5 7600X – $435 (340 USD)
yes cool the prices are placeholder, good talk
ashen spindle
Okay then you can have x12 on 4.0 to get the extra 2% instead of x8 4.0
no because x8 is enoujgh
shadow minnow
so youre asking for chips to split the lanes that drive board costs for a differ...
No I am asking for costs to go down, rn most boards have hardware to divert 8 lanes, just 4 would make it cheaper
ashen spindle
I'm going out on a limb and say they infact are not adding lanes by lanes so that halfing it costs more for the plx to do than diverting 4 lanes lmfao
the costs comes from you using a chip that is not mass produced since its not used anywhere
mild dome
I have never seen a system with x12 lanes. I think it could be inferred that PCIE doesn't support "odd" lane splitting
It's double or nothing
ashen spindle
it does support it but its not used
mild dome
Looks like PCIe 5.0 defined x12 and x32 links, but no one has used those yet. Can't find any other mentions
Ah here
"According to the book "PCI Express System Architecture", the Byte-Striping logic in the Physical Layer is capable of 1, 2, 4, 8, 12, 16 and 32 Lane links (See pg 401 and 408)."
ashen spindle
no it was defined before as well
"up to and including pcie5" means everything including the latest which is pcie 5 has it defined
shadow minnow
Weird 24 isn’t defined when 32 is
mild dome
Best that I could find were "x24" physical slots that actually delivered a x16 and x8 set of lanes to daughterboards
ashen spindle
32 is double 16 so there is an easy way to find potential use cases if you need real dual gpu support or something like that in data center, what would the need for 24 be
shrewd thicket
Pricing in Canadian Dollars:
AMD Ryzen 9 7950X – $1158 (906 USD)
AMD Ry...
Ryzen 7 for 500 usd seems like normal price, it's the competitor to i7 right?
ashen spindle
its a retailer putting prices up based on leaks and their own expectation, nobody that far down the chain has actual prices yet lmfao
ashen spindle
yes they could be close to the prices it will cost
shrewd thicket
Yes it's a lot of money but it's not expensive
ashen spindle
but its nothing to go on to actually judge the value of the parts or if one should wait or not for that gen etc.
shrewd thicket
Gonna stick to 12th gen anyways
Unless 13th gen comes out before 4080
I'm not waiting on that
Can always upgrade later cuz same socket
12th gen supports pcie gen 5?
ashen spindle
oh dont worry, latest info is that 13th gen has another delay
shrewd thicket
Just like Nvidia
ashen spindle
well nvidia might wait a bit to clear inventory, intel might wait a bit because they cant make them earlier
shrewd thicket
Gonna need me one of these when they come out
I'm just worried about getting too many m.2 ssd's and hurting my GPU performance
I kinda wanted to do a all m.2 build but now idk anymore
ashen spindle
to have your ssds hurt gpu performance you either need to have a basically broken board or do some black magic lmfao
shrewd thicket
M.2 goes over PCI lanes tho?
ashen spindle
if your gpu has X amount of pcie lanes, it has those lanes
shrewd thicket
The rest are already dedicated to the m.2 slots?
Okay cool
Then I'ma pump my system full of 980 pro's
ashen spindle
usually you just have one ssd that shares lanes with gpu directly to the cpu the rest go over the chipset
ashen spindle
the faster one usually, chipset isn't really slower unless you saturate the bandwidth
and I assume you're not going to write to like 3 m2s at the same time at max speed
shrewd thicket
Nah
I just wanted to go 3x 2tb 980 pros or something for games
But now I see 990 pro might release in augustus, just in time before I buy parts
Hopefully
chrome estuary
damn it. I could've just dropped in a 5800x3d into this fucking motherboard? man i wish i would've known about that some time ago cause i would've. Now the 7800x3d is just around the corner.
shrewd thicket
What's that
chrome estuary
Asus prime pro x470
ashen spindle
but for max m2 spaces you might want to go zen4 since they have some boards with 5 dedicated m2 spaces and such
shrewd thicket
No 7800x3d
chrome estuary
the next step up...
ashen spindle
No 7800x3d
the zen4 version with vcache
chrome estuary
if you know what the 5800x3d is, you should know what the 7800x3d is going to be
but yeah a bios update would've just straight up allowed me to throw a 5000 series in this which i should've looked up around the time the 5000 series came out
ashen spindle
Well they are probably giving it a price drop when zen4 regular is announced or launched so maybe still worth to do then
unless you’re out for more performance than just good value
chrome estuary
i'd like to slide into am5 so i can get a few different things. DDR5, I want to get an AIO water cooler and i heard something about the am5 supporting am4 coolers but i'm iffy about that, I'd like a high performance machine due to wanting to emulate PS3, Switch, and other things and the 2700x is just piss poor at that tbh.
at least all my experiences with it has felt so.
ashen spindle
Ah yeah for ps3 emulation zen4 is going to be much better
chrome estuary
i mean compared to my i7-3770 that i upgraded from, it's blown that thing out of the water but yeah... i'm gonna wait
i mean i'm really not wanting to wait until probably next year for my new cpu
but i'm gonna.
also i thought the 5800x3d was supposed to be some kind of absolutely massive jump compared to the 5800x. it's better in some cases (10-25fps etc etc. I've not seen it in terms of emulation and other such things), but everyone hyped it to be some kind of god CPU and in some cases it's doing exactly to or even a little worse (very few cases) to the 5800x
ashen spindle
well the thing about is that in some games it brought an uplift that is about the same or better than a cpu generation jump
so now on future products like the 7800x3d the idea is that it comes out earlier so you get those for games and the jumps in games become bigger
it was the first consumer product in that regard with stacked memory on it
for emulation zen4 is going to be a big jump because it will have instructions that are really helpful for ps3 emulation specifically
chrome estuary
don't get me wrong. going from a 2700x to the 7800x3d is going to be a massive leap.
but looking at the 5800x vs 5800x3d I'd probably just take the 100$ cheaper CPU in return for a couple less frames.
i'm probably gonna buy my new cases and PSU before the CPU drops, see what kind of ram it needs and snag that, then grab the CPU and mobo at the same time once the 7800x3d comes out
ashen spindle
sure, it's a enthusiast product mostly to test the waters basically and isnt meant as an all around better product
chrome estuary
yeah you're not kidding and you're also looking at nearly a fucking 20-30w jump in power usage
ashen spindle
no because thats the maximum power draw, in games its going to be the same pretty much
chrome estuary
i made that comment watching active power draw, not maximum power draw.
the 5800x was in some cases pulling 105-110w where as the 5600x was pulling about 75w-ish.
chrome estuary
in games, yes.
it might not be the best comparison or anything but it tells me enough
wheat plume
uhh
stark bobcat
context where
chrome estuary
so other than the eurorra carbon fibre laptop being the ugliest thing i've seen. but it's a pretty cool laptop
wheat plume
and its a Clevo chassis so you can get the laptop without all the eurrora stuff for a lot less, apparently
chrome estuary
i'm not surprised it's a clevo chassis
misty saffron
context where
are you new?
chrome estuary
those dudes have been doing the whole "literal desktop replacement" for quite a while.
stark bobcat
are you new?
well compared to most people here yeah