sunman42

202 signers as of 12:30 PM EDT.

Compared with my ISP-provided router, configuring and using my Airport Express is day to the ISP's night. Simple, straightforward, and easy. And the Express itself is extremely reliable. And, anathema as it may be to courts everywhere, the express just works with Macs and iOS/iPadOS/tvOS systems.

And oh yes, having firmware updates pushed from outside my home by outfits with opaque security protections is not my idea of A Good Thing. Particularly when those outfits appear to exist only to increase customer costs every year above and beyond inflation.

Am I the only one who feels the current interface is vastly inferior to the previous one? Far too much fiddling and too many multiple locations for the same search functions, &c. Apple appears to have veered farther and farther off its course of earlier years when keeping interfaces clear and simple was a priority. Now it just appears to be, "How many functions can we cram into how many displayed screens?"

Or maybe I'm just a cranky, old guy who's resistant to change.

sdw2001 said:

baconstang said:

Shhhh.... Don't tell the Dotard.  He thinks he just pulled off an amazing deal....

Yeah, the dotard who conducted the greatest comeback in political history---despite the full force of the US government, media and majority of leftist global powers against him.  I understand why leftists don't like Trump, disagree with him or even hate him.  But I'll never understand how they think he's stupid.  

I believe it’s Grover Cleveland you’re thinking of for “greatest comeback,” and if the election had been two years later, the Felon-in-Chief would have been reading the results in his cell at a federal correctional institution.

forgot username said:

Before this administration, I would never have seriously suggested the following, but there are no good government rules anymore, or anyone to enforce them. He could be waiting for the price to drop low enough, and then have Musk and/or the Trump kids buy it.

Even if the companies have a defense, they would still have to deal with costly litigation if prosecuted by the next administration. Additionally, considering Trump’s bullying behavior, it wouldn’t be surprising if he later changes his mind and attacks the companies if they do something that displeases him. 

Isn’t this irrelevant? Hasn’t the Chinese government made it clear that they will never turn over the secret sauce algorithm that is central to the user experience? I know there’s one or two very wealthy white males with friends in VCwho think they could buy everything else (including the user data) and build a social media that the kids will flock to in droves, but does anyone with any sense believe that?

blastdoor said:

Stabitha_Christie said:

blastdoor said:

Stabitha_Christie said:

blastdoor said:

Even though the Mac Studio with M3 Ultra seems like a great option for LLM usage and development, there is a big drawback in terms of cost.

But compared to what? How much would you have to spend to do the same job on a PC?

This was actually covered in the video. You can rent a ton of server time for the cost of a maxed out Mac Studio. So if you are just using it for development or general LLM usage then it really doesn't make sense financially. That said, the maker of the video also gave examples of why it would be worth it to pay to run an LLM locally. Specially when it has to do with privacy.

My question isn’t the cost to rent but the cost to buy.

If that was the intent of your questions then your questions were really poorly worded as buying didn't come up at all.

Anyway, to answer your question, you could build a PC that could to this cheeper than you could buy a Mac Studio. The big deal about Deepseek was that it ran on consumer hardware. 

There are many versions and the 671 billion parameter version is not going to run on anything resembling a standard PC. So I think you just don't know. 

Well, one not too shabby version ran on my *cough* poor, old (yeah, 18 month old) M2 Ultra Studio with a mere 128 Gbyte of memory. And yet, it was still able to give reasonable answers to two of the burning questions of the last ~ 50 years: (1) What is the air speed of an unencumbered swallow? and (2) Lakes or hamantashen? It appeared to be familiar with both source texts and more recent work online. Curiosity satisfied, and not quite certain what, if anything, it might have been communicating to a mother ship somewhere, I deleted DeepSeek-r1:<whatever>. Who knows, might be the last time I ever mean to ask an AI anything.