mlyle

7 ms·

31.
by mlyle 2mo ago
> High end compute also existed in the 60’s. Believe me, I know-- my dad did some work on the 360/44 and other large systems. High end late 1960s compute -- of the sort used to go to the moon or to design big nuclear weapons -- was
32.
by mlyle 2mo ago
> Because you need those calculations to "pump" nuclear weapons. They are needed to calculate both the geometry to make fusion bombs possible at all and to calculate the effect of a given geometry. They are the reason US/R
33.
by mlyle 2mo ago
It's my understanding that it was traced definitively to Taylor Farms, but the only positive test may have been a false positive. You don't always get a positive test, but this doesn't necessarily prevent you from concludin
34.
by mlyle 2mo ago
I think if you expect to be under 60K/s and suddenly find yourself at 20K/s heading for 200K/s-- you have a better problem than if you built for 1M/s and actual load is 20K/s. The unexpected success of the former w
35.
by mlyle 2mo ago
Yes, because we're currently in a GPU and RAM supply crunch, things have held value well compared to historical averages. We are increasing production of GPUs and RAM a lot.
36.
by mlyle 2mo ago
Strictly true, but finance being able to predict this pretty closely is how the entire modern economy works.
37.
by mlyle 2mo ago
That still doesn't tell you what GPUs will be worth in 5 years, because it depends upon what inference demand is like and what the alternatives are. What will an hour of NVL72 be worth in 2029? (And other things, that we know partiall
38.
by mlyle 2mo ago
I guess the point he's making is that the orbital velocity is pretty slow, so it's not too different to "get there" and "orbit there". Of course, the rocket equation often makes "just add a few percent mor
39.
by mlyle 2mo ago
> and create a narrative like we know anything about the unknown thing. It helps us feel like we understand it more. In fairness, this very often helps us understand the unknown thing more.
40.
by mlyle 2mo ago
Chunks were made with different versions, but I wonder if there's any viable approach to figure out the seed & generating version of minecraft and store just deltas.
41.
by mlyle 2mo ago
> broadly applicable FEM solver gmsh/Calculix is great. Sure it doesn't reach quite as far as Ansys, but you can solve real meaningful problems with it. > we currently don't have a free CAD kernel BRLCAD is OK and OCCT
42.
by mlyle 2mo ago
There's room for lots of improvement to our world, and "lives saved" is not the only metric.
43.
by mlyle 2mo ago
> which purports to gauge a study's rigorousness as a simple percentage That's not what it does. Some of the studies showed a causal relationship clearly. Some showed merely an association with a plausible argument about why i
44.
by mlyle 2mo ago
Which, if chickenpox is really the cause of a lot of dementia (as this article implies) might have been a pretty grave mistake.
45.
by mlyle 2mo ago
I got shingles at 45 and it was -not fun-. My arm is slightly disfigured. I think the age 50 target is dated. With reduced childhood incidence of chicken pox, we're all exposed to varicella zoster less, and it seems like the ages of
46.
by mlyle 2mo ago
Nor does he say he was the first to create them. The bro has posted hundreds or more pictures functions, many of which are well known... But well-colored and cropped into interesting locales, and described well. Including a good article t
47.
by mlyle 2mo ago
Please think about how this can be set up to track upstream Kicad. Hopefully you have a small adaptation layer and some clever build, not wholesale changes to the code. I teach HS classes where a lot of students end up doing some pretty se
48.
by mlyle 3mo ago
I think several is used by most speakers for larger quantities than few. It has the connotation of being larger, and that changes usage.
49.
by mlyle 3mo ago
Opinions vary, but: A couple: usually 2, though not always A few: 3, 4, 5 Several: 4, 5, 6, or 7.
50.
by mlyle 3mo ago
I think it's also fascinating that you can be so judgy, and felt the need to reply to yourself to announce this to the room instead of to me. Really, what's the point of that? Does it help you feel superior? I've found the
51.
by mlyle 3mo ago
I don't understand how you refer to this as "Cerebras-style". Cerebras is wafer-scale and unique. Jalapeno is an inference-optimized conventional chip.
52.
by mlyle 3mo ago
Yah well, I don't read all front-matter like that. Most of the time it's noise. Count it in the stuff that becomes cognitively invisible, like banner ads.
53.
by mlyle 3mo ago
I read it and saw LGTM and URL and was like "probably satire" but could not rule out it being real until like 30% in. It's like a modern version of Poe's law.
54.
by mlyle 3mo ago
I'm not saying that food banks never pay full price for food-- I'm sure it happens some. But for the most part food banks pay way below market rate. But, for example, if you make a >$100 donation to Second Harvest Food Bank (i
55.
by mlyle 3mo ago
I don't think you read what I said. I retired from industry to teach high school. A really big part of why I did this is because I wanted to help. I make basically nothing. There are many more personally lucrative things that I coul
56.
by mlyle 3mo ago
It's not like this must be exclusively A or B. The high school kid who volunteers at a homeless shelter and hopes it will help their college app is likely doing it both out of altruism and self-interest. (Actually, the person who hel
57.
by mlyle 3mo ago
I wonder if it goes away in sunlight. It's probably a few milliseconds difference in timing between when the (short) exposure is taken, resulting in different brightness of lights powered by 60Hz AC.
58.
by mlyle 3mo ago
> That is not in any way related to Fable (visibly) being switched to a less strong model if you’re trying to discuss certain topics. We're talking about it being invisibly moved to a weaker model if it looks like you're dist
59.
by mlyle 3mo ago
NAT was not designed as a security boundary. Sure, it may block some kinds of incoming traffic accidentally and as a side-effect disrupt some attacks. But why would you rather have an always-broken network that might block attackers inst
60.
by mlyle 3mo ago
We're finally getting there in the US, though. Top ASNs are >75% IPv6 capable. It's Optimum Communications and Frontier (my provider) that are really holding the numbers down at ~15% each. The latter is improving very slowly,

More ›