Don’t DM me without permission please
First it was for performance, OpenSuse back in like 2004, since then its just what I’m familiar with. I don’t feel like I need to fight the OS like I do on windows machines.
Sometimes I have to use windows in a VM or I need some Software that only works on 2000/XP (psxn00b sdk) and for that I have an old machine that I use.
Is that the House from American Football’s American Football?
Grep can accept input from stdin as with a piped cat, but I it can also just call the file directly.
In 99.999% obviously made up stat is obvious of situations its fine.
The real issue is a piped cat into grep will fork the process. Why open two process threads when one would do the job?
Edit: it was mentioned by but to expand a bit: piping cat into grep can also mask quite a few errors. It masks them because of how the shell handles error reporting on piped processes. IIRC, if the file is missing for example, you won’t necessarily know that because while cat will throw a not-found error, that gets piped into grep who gladly accepts the error (which was piped to stdin) as its input and greps through the error, reporting back that your content wasn’t found in the search material, not that the file was missing.
Node and react. Giant frameworks that seem to be the standard nowadays. They’re huge, bloated, and largely overkill for most things. I personally suspect they will be losing popularity soon due to the memory shortages.
Formerly of ohio, now of the southwest:
Ohio’s Mexican food is fine. Better than it has any right to be, not great though. I’ve found better tex-mex in Toledo, OH than in Houston, Tx (Houston’s tex-mex is kinda shit though).
I have never seen real Mexican food in the US worth having outside of my half-cousin’s Abuelito’s kitchen.
Depending on where in the world you are:
There is a resale store called Buffalo Exchange with a bunch of locations who always has the best messenger bags IMO. I’ve purchased several vintage messengers from them for carrying art supplies.
My favorite is a rigid leather satchel the size of a briefcase. Its new enough to have a laptop pocket, but old enough where the laptop would be expected to be gigantic. I use that slot for canvasses. Then it leaves the main pocket open for my mini-easel and brushes. External pockets act to replace my purse on days I carry it. All this covered with a giant leather flap which locks closed. V protec.
Edit: fixing autocorrect mistakes
Could a caterpillar play ddr? They’re kinda… Squishy
As long as it wears deodorant!
I’d hate to have the knowledge that a centipede big enough to play on a standard DDR machine exists.
They need a new Guitar before the blues catch up and kill them, gotta get that shit out and there’s no tool better than a fucked up 6 atring
Explore weird OSs! I got an old Celeron D workstation just for playing around with weird old operating systems.
Its got a 32 bit bios but 64 bit celeron, so the grub stuff has been fund 😅
You know we don’t like corn syrup right?
The only thing I know for sure in regards to why nvidia kinda blows on linux is that system RAM sharing doesn’t really work the way it does on windows or with AMD cards on linux. This means that particularly VRAM intense workloads will suffer. Unreal 5 games in particular suffer because of this.
That’s a brilliant idea, I might actually propose it to my local library as a prospective volunteer for the service.
You don’t need to be rude.
My original comment was in reply to someone looking for this type of information, the conversation then continued.
Disengage: I don’t want to deal with it today frankly, I don’t have time for rude people.
Yes, precisely.
If you’re trying to use large models, you need more RAM than consumer grade nvidia products can supply. Without system ram sharing, the models error out and start repeating themselves or just crash and need to be restarted.
This can be fixed with CPU inferencing but would be much slower.
An 8b model will run fine on an RTX30 series, a 70b model will absolutely not. BUT you can do cpu inferencing with the 70b model if you don’t mind the wait.
Today I made vegetarian salisbury steaks using impossible patties, store bought broth, and fresh veggies and herbs (and some stuff I had laying around). I spent less than $15 total (costco, price per unit) on the ingredients. It took 2 hours of cooking.
Assuming a wage of $25/hr, lower than adequate but relatively high in service fields in the US (those who work enough that delivery is super tempting), my meal cost me $65 including my labor. That’s less than it’d cost for delivery of a similar meal, is higher quality than I could get for delivery, and I’ve got leftovers for tomorrow, which I wouldn’t get with delivery.
Delivery is a scam. Gig economy Based delivery doubly so.
8b parameter models are relatively fast on 3rd gen RTX hardware with at least 8gigs of vram, CPU inferencing is slower and requires boatloads of ram but is doable on older hardware. These really aren’t designed to run on consumer hardware, but the 8b model should do fine on relatively powerful consumer hardware.
If you have something that would’ve been a high end gaming rig 4 years ago, you’re good.
If you wanna be more specific, check huggingface, they have charts. If you’re using linux with nvidia hardware you’ll be better off doing CPU inferencing.
Edit: Omg y’all I didn’t think I needed to include my sources but this is quite literally a huge issue on nvidia. Nvidia works fine on linux but you’re limited to whatever VRAM is on your video card, no RAM sharing. Y’all can disagree all you want but those are the facts. Thays why AMD and CPU inferencing are more reliable, and allow for higher context limits. They are not faster though.
Sources for nvidia stuff https://github.com/NVIDIA/open-gpu-kernel-modules/discussions/618
https://forums.developer.nvidia.com/t/shared-vram-on-linux-super-huge-problem/336867/
https://github.com/NVIDIA/open-gpu-kernel-modules/issues/758
Are animals dicots?