twtw

I always used to use various dusty old netbooks for this kind of stuff, and would run many applications on the same one.

A dedicated raspberry pi in a nice little case for each is certainly modern and cool, but it doesn't give me quite the same nostalgia as a ancient laptop shoved in a corner...

Or just use a programming language instead of a calculator with various features bolted on.

[Speaking of cvxopt, originally written in matlab] "Fortunately, it has since been rewritten in a real programming language." - Stephen Boyd

No offense intended, but ... that isn't the most glowing recommendation.

I read it as "This book is great, really helped me understand the concepts. In fact, it helped so much I failed the class."

I'm not sure I follow the reasoning in the article. It says if ray tracing flops, AMD will take the lead, but then goes on to say that AMD believes ray tracing will be important and is working on it.

But the question is could it be made to be good for that domain?

See https://arxiv.org/abs/1702.04024 [cited by this paper] for promising results doing linear algebra using aws lambda.

Is it accurate to say that rank-n tensor is to n-D array of numbers as linear transformation is to matrix?

I would have appreciated some historical context for the z80 in this post. 8,500 transistors (equiv to ~2800 nand gates in depletion-load nmos logic * ) and not really designed to be a compiler target. I'd say the z80 is pretty impressive - but certainly with quirks.

I would have enjoyed that kind of framing much more than "lol the z80 sux!"

* Not saying the z80 was implemented with all nand, just providing the figure as a reference.

Hasn't this worked via HSAIL for years?

Gets energy from the power plant to your light bulb without heating up stuff along the way quite as much.

> 20 year old (!)

This paper is close to 20 years old too. The (2018) in the title is wrong.

Tangential, but the formatting on this is awful. It's like a high school lab report - every equation is a different size and aligned differently.

This appears to be a GCN backend intended for use with OpenMP and friends, so not really comparable to cuda (more comparable maybe to the llvm nvptx backend).

It's shown in the 8th image in the rotating set of images at the bottom. Some text in one of the images references Tetsuo Kogawa, so I guess the circuit is from here: https://anarchy.translocal.jp/radio/micro/howtosimplestTX.ht...

Sure, that's how it is now, but it definitely hasn't always been that way. The original public companies definitely were "early stage, pre-revenue," though you can argue that those were long enough ago to be irrelevant.

Intel's IPO was 2 years after the company was formed.

> you can move it off your property

And where should I put it? I don't want it on the sidewalk, and I don't want it in mystreet. I can't put it in the trash because it's someone else's property. That's essentially what I meant by "anticipated solution."

Note I never suggested it would become mine, so a good portion of your comment is responding to someone else. And I'm not sure who you are referring to by "you guys," but I can honestly say I have never before been in a situation prior to these scooters that someone just left their valuable stuff on my private property and I had to deal with it. Are there equivalents to towing companies I can call for random valuable crap that belongs to someone else?

Is that moving goalposts, or just recognizing another problem?

It seems like a problem, so is it a bad thing to solve it once we've (according to you) solved hunger and malnutrition?

> You can't just steal bikes because they've been left on public land

What is the anticipated solution when you leave your bike on my private land? I've seen loads of bird scooters left on private property, and I don't see what the property owners are supposed to do. Call the police every time?

> I think the VC model itself is a great development from the last century. The fact that someone can raise millions (more than most people earn in a lifetime!) with an idea enables a lot of innovation that would be hard to nurture otherwise

Isn't this what the public stock markets used to be used for, before an IPO became a way to cash out?

> comparing GPU to naive implementations (and sometimes, implementations written in dynamic languages) instead of to highly tuned BLAS or MKL implementations

In your experience, are these comparisons using naive GPU implementations as well? If so, that still seems like a valuable comparison.

It's looking like that "something cross-platform" will be cuda, given that AMD's latest approach is to implement a compiler to convert cuda to "hip."

So that's why Xeon Phi has been so successful!

/s

Real talk, though. Why do you think larabee is the "right" hardware?

Addresses.

It's interesting to see how people react so strongly to this phrase.

By default, I assume "boring" is a verb. Maybe it says something about the reader whether they see a verb or an adjective.

What would "from whole cloth" even mean for this?

I wonder what world you live in (career, culture, etc) that makes you feel that a post with title "coolest things I learned in 2018" should come with "academic substantiation."

Also, the whole point of the words for emotions thing is sorta that those are new (i.e. not pre-existing, which maybe means not "real" to you). The point is to think about what feelings we don't have words for, and what impact it might have to make words for them.

Re 2.

I like this with the second interpretation mentioned on the page. Avoid boring (v.) other people, rather than avoid people who you seem to be boring (adj).

Those two comments come up frequently, but it's worth noting that in the whole preceding thread 'cperciva was being a pretty massively arrogant jerk.

That thread is a classic example of why the most capable people aren't necessarily the ones you should want to work with.

The alternatives you mention are things that could happen in the future. The alternative 'p1necone brought up is the status quo.

This slippery slope argument is valueless. It's perfectly possible to have socialized healthcare without communism.

You seem to be responding to the title, not the essay.

Dyson describes the theory as "simple and intelligible" once you accept the concept of fields, which was very much foreign in Maxwell's time.