Forgot your password?
typodupeerror

Comment Re:Oh Look! (Score 1) 26

With an EBITDA margin of 67% (76% gross margin), a saturated market, and outsourced manufacturing I don't think they need many shenanigans to achieve their 13.48 P/E.

If it was possible for them to expand their business by borrowing, they could probably pump their P/E up to 200 in the current irrational market. They're at about 50% of their peak price, equal with 2.5 years ago.

From a financial health perspective the lower the P/E the better. From an investor perspective, low P/E implies lack of growth potential. Also the risk of the AI bubble tanking them is priced in. These aren't the same type of investors as the ones who invest in AI directly.

Comment Re:If true... (Score 1) 35

There are no practical consequences because it was already understood to be true in examined cases. It's not a discovery of anything; it's a mathematical "proof" of the general case. Anything practical had already been tested and found to be true over those ranges, which rules out practical considerations.

The purpose of the Polya conjecture was merely to go the other way and derive the Riemann Hypothesis, but I think it was already disproved.

Mathematical proofs haven't been discoveries for over a hundred years. They're just a type of achievement award, puzzle prizes.

Comment Re:If true... (Score 1) 35

NS models a fluid as if it is continuous. But we know that it is made of molecules.

It would be brain-dead to think it didn't have a limit. And if you're working with effects close to that limit, you were already not using it.

So "proving" the limit is/was not actually useful to science or engineering. Having a proof for the existence of a limit doesn't help you model anything near the limit. It doesn't improve your analysis of boundary layer separation.

The problem solved wasn't about how to model past the limit. It was merely "proving" that the formula could give nonsensical results there, which was already understood to be true conceptually. It's a numbers puzzle, a type of game. Finding the proof is not "science" or "engineering." No experimental result was predicted. No understanding was advanced. No formula to construct something was defined.

Comment Re:Google AI (Score 1) 36

I don't know what you mean by "Google AI" since that is a division, not a product.

If you mean Gemini, do you mean the absolute shit version they use with search results, or the one you have to pay to access? Because the one that does the search results is absolute. fucking. trash. and anyone impressed by it is an absolute. fucking. idiot.

Or do you mean the text prediction in the Android keyboard, which Google also has claimed is "AI"? Because it's shit too. It keeps thinking I want to write "Scott" when I am writing "about" and has been fucking this up for years. It allegedly uses old-school ML but I've seen no evidence that it works.

Comment Re:Excellent (Score 1) 61

(Quite why Europe hasn't retaliated yet I'm still unsure of, I would be subjecting Microsoft to fines in the millions per day if I were European Dictator.

So, you're running Microsoft and suddenly countries who are completely dependent on your software start fining you into the ground, and the regime has been pressuring you to fuck them over, and threatens to start deporting your visa hires...

Slashdot Top Deals

The difference between a career and a job is about 20 hours a week.

Working...