A third world engineer responds to “RISC-V: They should have known better”

I don't really understand the author's conclusions about cost and shipping, and how RISC-V is cheaper and more accessible to people outside the US and Europe. He first talks about how getting $1 worth of chips can cost $60-$200 in shipping for him due to his location... but then by the end claims that RISC-V gives him "an architecture that arrives in my country at ten cents a part".

I don't get how both things can be true. The cost to ship something to Trinidad and Tobago has nothing to do with whether it's ARM or RISC-V; shipping the same weight of either kind of chips should cost exactly the same amount. Yes, maybe the chip cost itself of a particular ARM-based model is $0.15 while the equivalent RISC-V chip costs $0.10, but if the issue (for him and other people who live outside US/Europe) is that shipping costs are orders of magnitude more expensive than the thing being shipped, that chip-cost difference becomes irrelevant.

I think he does make a great point about how fragmentation does give you better optionality: ARM is sort of like cable TV where you have a small number of product categories that each bundle a particular set of features, whereas with RISC-V you design something bespoke that does exactly what you need and nothing more. But again, this isn't going to affect shipping costs.

You have exceeded the allowed number of requests.

Your Requests: 500 Request Limit: 500 per minute

You have exceeded the allowed number of requests.

Your Requests: 500 Request Limit: 500 per minute

with no vpn

Here are a few random things I know of:

* Tenstorrent Ascalon has a neat optimization for certain LMUL>1 SIMD operations. LMUL=2 effectively unrolls the SIMD operation making it read two SIMD registers from every source and write two SIMD registers to the destination. There are however some instructions where LMUL=2 only needs to write to one registers, those are narrowing instructions (e.g. 64-bit to 32-bit truncation) and comparisons (which write to a LMUL=1 register with packed bits). When those SIMD instructions have to .vx form, which means one argument comes from a GPR, they now only need to write one SIMD register and need to read two SIMD registers. This matches what regular SIMD instructions need and because the silicon for the execution is much cheaper than register file ports, Ascalon can exexute these instructions in a single operation. So you can compare twice as many SIMD elements against a scalar, then you can against another SIMD register.

* Ventana (now under Qualcomm) talked a tiny bit about their fetch-block-optimizer and something that sounded like a L1i-trace cache. The fetch-block-optimizer would go to certain hot L1i entries and "optimize" them, with agressive instruction fusion including fusion of non-adjacent instructions.

* NextSilicon: Idk any details yet, but they said they handled RVC without increasing latency and that they've found a good solution for implement RVV and especially LMUL, which is a challange in out-of-order designs.

* OpenXiangShan: The fastes open-source CPU, is working on doing 2-ahead instruction fetch (the thing Zen5 added).

Now that being said, Ventana was bought by Qualcomm, we know the RISC-V team is still alive, but who knows if we'll ever see anything from that outside of Qualcomm?

The Tenstorrent Ascalon devboard is way behind schedule and on 12nm TSMC instead of a 4nm node the processor was designed for and is now supposed to clock at 1.38GHz. Though I think the delay has more to do with TT management problems then with the actual design.

While the scalar part of OpenXiangShan looks really good, the RVV imolementation is currently basically unusable. They want to have fix for the problems until the end of the year, but we'll have to see.

I assume the 3am-9am figure was in US Pacific time, in order to emphasize the difference from the working hours of people on the US west coast.