atq2119

26 ms·

451.
by atq2119 3y ago
Societies are usual torn apart from the top, because that's where there are people with power to make impactful decisions.
452.
by atq2119 3y ago
Nice writeup! Do you know around what time repeaters were started to be used for clock trees? I also wonder about the two phases. Is that still commonly used?
453.
by atq2119 3y ago
> Who’s going to fund this study? Foam mattress manufacturers?
454.
by atq2119 3y ago
This is pretty reasonable, I think, but can it be applied to GPU code? The fundamental challenge of CSPRNGs is that they tend to have high latency and each thread produces a lot of randomness at once, while typically you only need a little
455.
by atq2119 3y ago
Idea is a pretty general term. I have a bunch of patents and I would describe them all as patenting an idea (for how to achieve some goal). The implementation or execution of the idea usually takes the form of some Verilog or some C++. That
456.
by atq2119 3y ago
You do have a point, but it's not a slam dunk. Intuition isn't some fixed thing but arises from personal experience. A lot of that is common to a culture, but there are different cultures and in any case, some truly personal aspec
457.
by atq2119 3y ago
Clothing is important in that sense, but fashion as a changing thing and especially fast fashion isn't. I suppose it can be a nice hobby for some, but for society as a whole it's at best a wasteful zero-sum pursuit.
458.
by atq2119 3y ago
There could be a contributor licensing agreement in place that effectively transfers copyright. But yeah, this is a potentially sticky point. It's inherent to the license that one party is special
459.
by atq2119 3y ago
If you follow the link to the actual license, it does list done explicitly permitted uses. So yes, it does seem that you can fork it and do something useful with the fork. That something useful is restricted, which is the point of the licen
460.
by atq2119 3y ago
> who cares if a webpage is rendered in 100ms or 2s? Do you really stand by the statement of this rhetorical question? Because if yes: this attitude is a big reason for why web apps are so unpleasant to work with compared to locally runn
461.
by atq2119 3y ago
The part that I don't understand is why AMD/Intel/somebody else don't just implement at least the base CUDA for their products. HIP is basically that, but they still make you jump through hoops to rename everything etc.
462.
by atq2119 3y ago
Then the CEO should have explained that, end of story. (It's possible that they did, and the retelling is so biased as to exclude that fact. But it seems unlikely.)
463.
by atq2119 3y ago
This makes a lot of sense - but how? By definition, management does a different job from individual contributors. How do you make a manager feel the pain of a messy code base, for example?
464.
by atq2119 3y ago
Indeed, I see a big difference between developers in my team based on their general reasoning and discrete math skills. How capable are they of abstractly interpreting code in their head? Are they used to thinking about how to prove somethi
465.
by atq2119 3y ago
This is yet another article that doesn't really seem to get it. For example: it argues that many small check-ins are better for bisecting -- but enabling a development process that encourages many small commits while still allowing bis
466.
by atq2119 3y ago
I would certainly like to see this mythical tool that can support both: * Separating out different conceptual changes that are/were reviewed together and * Transparently cleanup a messy history that starts with a few base commits (corr
467.
by atq2119 3y ago
Thank you for putting my own (~10 years ago) experience with Mercurial into words.
468.
by atq2119 3y ago
As long as we're overanalyzing a joke, it's Switzerland. There isn't really a lot of wilderness there ...
469.
by atq2119 3y ago
Bergamo (server chips) are 4c-only machines, so you could look at benchmarks for those. They're pretty impressive.
470.
by atq2119 3y ago
HR is often awful, sure. But it helps get the basics right. My brother works at a small (<100 people) startup in the material sciences, and the amount of times that his and his co-workers pay statements got messed up in creative ways is
471.
by atq2119 3y ago
Yes. Without batching, every weight is only used once per token. On discrete GPUs, you generally need around ~20 flops/byte of memory bandwidth to be compute bound. I don't know where APUs are in that regard, but probably not enti
472.
by atq2119 3y ago
> Can you think of one movie or TV show which models a realistically happy working adult marriage which isn't either sentimental and idealised, or doesn't end in drama, tragedy, and betrayal? The first thing that comes to mind
473.
by atq2119 3y ago
It doesn't really matter in this case, but hummingbirds would like to have a word with you.
474.
by atq2119 3y ago
Proposal to introduce the term "villain image" for this sort of thing.
475.
by atq2119 3y ago
I think you're confusing simplicity with computational cost. Bubble sort is simpler than quicksort. It is also more computationally expensive. Universal wave function theory is simple in this sense. (IMHO the term many-worlds is doing
476.
by atq2119 3y ago
Don't hate the players, hate the game. This applies to all the actors here as long as they abstain from getting involved in politics in a way that perpetuates the rules of the game.
477.
by atq2119 3y ago
The key irony is that the empowerment of ambitious individuals that he praises is only possible in an institutional framework that makes it possible. That just allows you to sell your homebrew computer, for example. And there are plenty of
478.
by atq2119 3y ago
The worst example of this is VISA instead of visa when talking about international travel as opposed to credit/debit cards. (Hats off though to whoever decided on that marketing name...)
479.
by atq2119 3y ago
Not necessarily. GPUs simply aren't optimized around branch-heavy or pointer-chasey code. If that describes the inner loop of your workload, it just doesn't matter how well you can parallelize it at a higher level, CPU cores are g
480.
by atq2119 3y ago
> The weird part is the question is expanding into what or where. Why would there have to be an "into"? It seems like the most parsimonious expansion is that there isn't one. My limited understanding of the math is also th

More ›