You skipped over the second sentence of what you're responding to:
> Say you’re reading a paper or trying to implement some technology that uses a mathematical concept you aren’t familiar with
In such a case you're not interested in either manifold or sub-manifold or group in and of itself. So a lack of familiarity with theorems isn't an impediment.
i don't have an outrageous amount of experience here but i work at a place that designs and sells x86 chips (and i work on arch adjacent things). i'll just say this: do not underestimate the effect of institutional friction/momentum/complacency on "time to ship". point being: maybe RISC is more "agile" than CISC but also maybe Tenstorrent could have chosen a CISC approach and still iterated quickly simply because they're not an institution.
edit:
> They are working on 2-3 CPU designs, 2-3 AI accelerators, all the IO stuff, etc
also do we actually know how much they're designing wholecloth and not buying off the shelf?
edit2: i think Tenstorrent is great and i'll probably buy a devkit soon (as well as maybe apply lol...)
this is very common. lots of companies around (e.g. cadence) that sell SIPs as synthesizable RTL for easy integration into chip designs (and lots of hardware companies buy these SIPs).
1024 fp16 macs is pretty good but the 128b vector datapath is weak sauce. on the other hand 2MB SRAM is legit. i wonder how many tiles (i don't think it's in the post).
> Wigand turned his attention to improving tobacco additives, some of which were designed for "impact boosting", using chemicals like ammonia to enhance absorption of nicotine in the lungs and affect the brain and central nervous system faster. Wigand believed this process was a deliberate attempt to increase addiction to cigarettes.
> It's easy to assume you're great especially working under "passion coders" who keep stuff afloat.
i work on compilers+hardware so that's the backdrop here.
here's a real hypothetical for you (ie it happened but i'm not going to use specifics): our internal proprietary compiler is spitting out incorrect atomics instructions that deadlock our internal proprietary multi-core DNN accelerator. this incorrect code is downstream of a big, lucrative, customer's (you know which one) LLM model.
now the question: is it the "passion coder" that will solve this or the grinder?
> well because they have other priorities and responsibilities in life that make it necessary to maintain a strong work ethic.
It's amazing how many people in tech can't imagine that the same people that can grind med school or IB or big4 accounting or white-shoe law somehow can't grind the same way in tech. Newsflash: the majority of people in FAANG are grinders not "passion coders".
Personally I hate this job but I'm very good at it and it was either law school with my 98% LSAT or tech. I picked tech because reading and writing briefs all day seemed somehow worse.
> Side-note: Mojo has been designed by the creator of LLVM and MLIR to prioritize and optimize vector hardware use, as a language that is similar to Python (and somewhat syntax compatible).
Are people getting paid to repeat this ad nauseum?
they're not the same - versal acaps (whatever you want to call them) have AIE1 arch while phoenix has AIE2 arch. there are significant differences between the two arches (local memory, bfloat16, etc.)
you see this kind of wishful (delusional?) thinking very frequently on hn:
> Microsoft won't ever be able ... their business model is ... have had enough.
statements made with complete certitude, as if these things aren't nebulous to begin with or in the process of being decided. it reads like the opening line to a christopher nolan superhero movie (i.e., a call to arms) rather than a position or even an opinion.
i personally don't understand what people get from the act of writing/posting such things - i mean i'd get it if we were on some hill about to rush into battle but this is an anonymous/faceless tech news forum.
It's funny - former Google devs (whom I just maintain a good relationship with their former employer) are ideally positioned to profitably take advantage of the arb of TPU over GPU.
With all do respect, I think you're telling a half-truth here. Your LinkedIn profile shows you had two MS degrees before starting your PhD. Undoubtedly you got some of the requirements waived for the PhD course requirements. That means your total time post BS was 5 years. I did the same thing - 1.5 years (3 semesters) MS and then came back later (different school) to do a PhD and got half the course requirements waived. Finished the PhD in 4 years but total time post BS was a more humble 5.5.
> Vivado is intimidating (as opposed to Yosys/nextpnr which you seem to hate)
i never said i hated yosys/nextpnr? i said somewhere that yosys makes the uber strange decision to use C++ as effectively a scripting language ie gluing and scheduling "passes" together - like they seemed to make the firm decision to diverge from tcl but diverged into absurd territory. i wish yosys were great because it's open source and then i could solve my own problems as they occurred. but it's not great and i doubt it ever will be because building logic synthesis, techmapping, timing analysis, place and route, etc. is just too many extremely hard problems for OSS.
all the vendor tools suck. it's just a fact that both big fpga manufacturers have completely shit software devs working on those tools. the only tools that i've heard are decent are the very expensive suites from cadence/siemenns/synopsis but i have yet to be in a place that has licenses (neither school nor day job - at least not in my team). and mind you, you will still need to feed the RTL or netlist or whatever that those tools generate into vivado (so you're still fucked).
so i don't have advice for you on RTL - i moved one level up (ISA, compilers, etc.) primarily because i could not effectively learn by myself i.e., without going to "apprentice" under someone that just has enough experience to navigate around the potholes (because fundamentally if that's what it takes to learn then you're basically working on ineluctably entrenched tech).
i haven't a clue what that means but when julia's autodiff is the only one that is implemented using dual numbers i think the claim is pretty obviously false and seemingly everyone has an issue admitting that. without a doubt you'll just say something else sarcastic rather than admit it.
> Say you’re reading a paper or trying to implement some technology that uses a mathematical concept you aren’t familiar with
In such a case you're not interested in either manifold or sub-manifold or group in and of itself. So a lack of familiarity with theorems isn't an impediment.