talldayo

49 ms·

511.
by talldayo 2y ago
Which is why you don't send 4 TPUs to do 1 GPU's job.
512.
by talldayo 2y ago
TPUs are like the NPU of the training world. You take a bunch of extra time, money and dedicated silicon and end up with an ASIC that barely competes on equal terms with a similarly priced GPU. Unless you've got access to Nvidia's
513.
by talldayo 2y ago
> Is there a possible way to unblow the fuse? Maybe, but it's also possible that it wouldn't do anything. Other Nvidia boards (like the Tegra in the Switch) also come with arbitrarily disabled "dark silicon", but enab
514.
by talldayo 2y ago
If the US attempted to pass reciprocal laws for every Chinese policy, America would go bankrupt within the month.
515.
by talldayo 2y ago
> AI Leader Grammarly lmfao
516.
by talldayo 2y ago
Swift doesn't support most OSes as well as Rust does. It also has a smaller/worse standard library and much fewer developers. According to StackOverflow, Rust users outnumbered Swift-using respondents 2:1. Rust learners outnumbere
517.
by talldayo 2y ago
Do you have any demos proving that this concept is even possible? Power consumption is already managed meticulously at most datacenters and they'd want to see some hard proof that your product can improve their situation.
518.
by talldayo 2y ago
This isn't a response at all. Nvidia has been making Jetson boards for longer than Apple has been making their own chips.
519.
by talldayo 2y ago
Might be time for a new iPad. My old-school iPad Air has 2gb of memory and is an absolute hog when loading content-heavy websites.
520.
by talldayo 2y ago
Good LLM-generated code comes from good developers that can provide detailed requests, filter through wrong/misleading/cruft results, and test a final working copy. If your personal bar for PRs is high enough then it won't ma
521.
by talldayo 2y ago
> This is ego speaking. I suspect you haven't seen code review at a 500+ seat company.
522.
by talldayo 2y ago
Syncthing?
523.
by talldayo 2y ago
> I’ve been using MacOS for over 25 years, and never seen a “Taboola ad” in an Apple application Try booting up Apple News. It's a recent change, but a real one and not "made up" in the slightest: https://www.ax
524.
by talldayo 2y ago
> Are there any laptops that support 256 GB of RAM that cost less than a fully speced out Macbook Pro? No, but there are absolutely laptops with more than 128gb of addressable RAM. If you did your research you might have found them, at
525.
by talldayo 2y ago
> Apple is the only company that makes integrated GPUs that are actually any good I will reiterate; if you have used AMD or Intel iGPUs recently you would know this is just plain wrong. AMD's iGPUs often outperform Apple's for
526.
by talldayo 2y ago
In my opinion - I think pineapple compliments every individual component of pizza. The acidity of the fruit cuts through the grease in the cheese and compliments the savory flavor like a wine platter. It goes excellent with a good sauce, an
527.
by talldayo 2y ago
Certified Apple Shill, Donald Trump.
528.
by talldayo 2y ago
MacOS will gladly serve Taboola ads in first-party apps and beg you to use Safari in push notifications. If that's not infiltration by advertisement, I'm not sure what is.
529.
by talldayo 2y ago
Ironic how the EU comes out on top for acknowledging sideloading as a fundamental feature of smartphones. It will take the "it's a series of tubes" guys quite some time to figure that one out.
530.
by talldayo 2y ago
> I’m setting the upper bar for my performance expectation at the DASUNG 25” monitor, which I did love, but couldn’t fit in a backpack. I haven't used one, so I can't speak specifically to how good it is. I can say that most
531.
by talldayo 2y ago
There are a few rakes there, yeah. 1. The age of Thinkpad Hackintoshes is sunsetting, and the chips that support it are getting long in the tooth. You can't buy recent Intel chips and expect them to work flawlessly, and the laptops old
532.
by talldayo 2y ago
> If it isn't profitable, they shut down and difficulty adjusts. If the number of minable coins is fixed that doesn't sound like a perpetually appealing solution. People worry about it because it's still a problem, and you
533.
by talldayo 2y ago
> This is not binning, it's segmentation. Wish there was competition. How do you know? Unless you can confirm that both dies are equally functional I think that's a silly assertion on the level of "Nintendo Switches are ne
534.
by talldayo 2y ago
Yep, there are a number of different 12gb Nvidia cards; 3060ti and 2060 also had 12gb options. But I think those are gap filler cards, at least in Nvidia's lineup. They're intended to fill the "budget 4k" market that nee
535.
by talldayo 2y ago
If Russia isn't bluffing about the Su-75 (fat chance), then supposedly they've got a design that can suit the task. But nothing is in operational service yet, so it's all one big nothingburger until then.
536.
by talldayo 2y ago
It's good news I think - more cards have to set 16gb as their baseline to future-proof them for the long term. Skipping a 12gb generation is better for everyone, and I frankly don't even think 8gb cards are that bad if you aren&#x
537.
by talldayo 2y ago
> I guess many countries on a budget look at this as a "cheap" nuke nowadays, something a few extra steps of physics could make easy reachable Countries on a budget do not have the resources to create a CERN-like accelerator. I
538.
by talldayo 2y ago
> These results are disturbing, but what’s even more disturbing is the AI developers’ apparent comfort with their products' increasing disloyalty. We've been dealing with this for literally centuries at this point. You want to
539.
by talldayo 2y ago
If you work in retail, smartphones are not the bottleneck of your productivity.
540.
by talldayo 2y ago
You can tell it's American because Russia doesn't make single engine fighters.

More ›