addaon

28 ms·

481.
by addaon 1y ago
Not affiliated with EPC, although I've been super impressed by their GaNs since first using them in lidars nearly a decade ago now. I've been keeping an eye on their integrated circuits (GaN + drivers), and just noticed today that
482.
EPC23102: V, 35 A EPower Stage IC(epc-co.com)
2 points by addaon 1y ago|1 comments
483.
by addaon 1y ago
> That path fits that apparent definition. Then saying it "isn't actually chaotic" is mis-aligned with the layman's understanding, and the disagreement is not explained (either by explaining a more technical definitio
484.
by addaon 1y ago
> …which we know isn't actually chaotic. It's just a path defined by mathematical functions. I don't know what function is being presented, so I can't speak to whether it demonstrates chaotic behavior -- but the whole
485.
by addaon 1y ago
> but it's still there Unless you evaluate branches in the second stage of the pipeline and forward them. Or add a delay slot and forward them from the third stage. In the typical case you’re of course correct, but there are many ap
486.
by addaon 1y ago
> I know branch prediction is essential if you have instruction pipelining in actual CPU hardware. With sufficiently slow memory, relative to the pipeline speed. A microcontroller executing out of TCM doesn’t gain anything from predictio
487.
by addaon 1y ago
Yep. The Xilinx tools are very, very good; but they're definitely proprietary.
488.
by addaon 1y ago
> If you want the bare chips and not full assembled labels the usual packaging is uncut wafer and cutting out and handling the individual dies is your problem. This does not match my experience, although I imagine it's true in some
489.
by addaon 1y ago
At a higher price point but with more capability, Digilent has a one-week 20% sale on their FPGA boards this week. Some good options (Artix 7 and Spartan 7) within spitting distance of $100.
490.
by addaon 1y ago
> The connotation of a non-profit is that it's being audited. This is very geography-specific. In the US, 501(c)(3)s (what most people think of when they say "non-profit" where I am) have no general requirement for audits.
491.
by addaon 1y ago
> are more difficult to develop for (8-bit AVR ouch) I think this just comes down to philosophy. ARM gives you a way richer tool-chain… and you need it. The complexity to do anything skyrockets. You have a compiler in the loop so achievi
492.
by addaon 1y ago
They're cheap, robust, readily available, super easy to develop for, have a well-chosen set of peripherals that can reduce BOM even further (e.g. the CCL)… what's not to like?
493.
by addaon 1y ago
Negative, just myself. I suspect I've mentioned my physical location on HN previously -- southern Utah.
494.
by addaon 1y ago
> I rarely have questions of others but I always question myself. There's only so many questions I have the ability to answer myself. Of those, there's only so many that I have the lifespan to answer myself. We stand on the sho
495.
by addaon 1y ago
I have the answers available (I have a phone and good connection), I just am tactical about when to pursue the answer in realtime and when not. If it feels like it's going to open up a wider field of questioning -- or if it feels like
496.
by addaon 1y ago
> But I, as a human, rarely have questions to ask. Wow. This just does not match my personal experience. I do an hour or so walk around the reservoir near my house 4-5 times a week, letting my mind wander freely -- and I find that I stop
497.
by addaon 1y ago
Such cynicism! I’m proudly celibate, like my father before me, and my grandfather before him!
498.
by addaon 1y ago
C++ pre- and post- C++11 are really different languages. It’s by far the biggest break (so far in my career) in how the language is used. I’ll gladly debate the detailed trade-offs of many of the newer features, but going back to a world be
499.
by addaon 1y ago
Cells. Interlinked.
500.
by addaon 1y ago
> Is this a common experience for others? I should think about whether my experience generalizes. The user seems to have had a different experience. Stopped reasoning.
501.
by addaon 1y ago
> For the same reasons they didn’t let Boeing die out. Right — so that extends the time window for pillaging, and means that you can raid future taxpayer money, not just current assets. But there’s still no “turn around,” it’s just diffe
502.
by addaon 1y ago
> But this seems like the easiest one to implement. Even easier: Just fail. In my experience the ChatGPT web page fails to display (request? generate?) a response between 5% and 10% of the time, depending on time of day. Too busy? Just i
503.
by addaon 1y ago
> For example, NASA spacecraft run critical calculations in triplicate - multiple processors run the same operation, and if one disagrees due to a stray bit flip, the other two out-vote the process. Triple modular redundancy is one way t
504.
by addaon 1y ago
So sort of the projective real line?
505.
by addaon 1y ago
> You don't have to drive, though it is kinda far from anything else. It’s right at Dulles, easy to cap onto a trip in or out of the DC area, or even a long-ish layover.
506.
by addaon 1y ago
> How effective would Gelsinger be Effective at what? Intel’s done, it’s on the Boeing trajectory, it’s just about grabbing what you can as the ship goes down.
507.
by addaon 1y ago
> He was literally closer to God If you subscribe to a religion that not only assigns a physical known location to God, but puts that location at a significant distance away from humanity either in a specific direction, or in a general “
508.
by addaon 1y ago
> Depends on what you're doing. Trying to get an accurate answer (best correlated with objective truth) on a topic I don't already know the answer to (or why would I ask?). This is, to me, the challenge with the "it depend
509.
by addaon 1y ago
> a smart and fast model that answers most questions, a deeper reasoning model for harder problems, and a real-time router that quickly decides which model to use based on conversation type, complexity, tool needs, and explicit intent (f
510.
by addaon 1y ago
> Output: $10 / 1M tokens It's interesting that they're using flat token pricing for a "model" that is explicitly made of (at least) two underlying models, one with much lower compute costs than the other; and wi

More ›