Probably because the article is looking at silicon that's actually available today, and big-iron RISC-V silicon akin to Xeon/EPYC/Graviton/Ampere/etc is more of a hypothetical at this point.

Because it essentially doesn't? At least when I tried looking a few months ago, it was really hard to find any commercially available RISC-V SoCs, let alone high performance ones. Sure there's little hobbyist boards going for $200>, but that's about it.

Cloud world is really slow. Imagine writing about Zen3 Milan in Aug 2023 when AMD has already announced Zen4c Bergamo. Actually we were "ahead of the curve" since we got access to T2D before it was made publicly available in Tokyo region, and even then it took several months to get enough capacity in Tokyo to fully migrate our production Kubernetes cluster.

I really wish we could test out RISC-V SoCs from the likes of tenstorrent, but it's a long journey

> Cloud world is really slow. Imagine writing about Zen3 Milan in Aug 2023 when AMD has already announced Zen4c Bergamo.

AWS has Zen 4 in preview. Azure and Oracle have Zen 4 available. It's only Google cloud that's been behind this release. The cloud world isn't slow.

It takes very long time for going from preview to actual production usage for anyone. We had T2D preview access more than a year ago, it took several months to get enough stock in Tokyo region (US always gets preference in such cases whenever a new machine type comes out). GCP already has Zen4 in preview for some US customers. Also, us being one of the largest GCP customers in Japan made things even slower

RISC-V is important, and I use it every day for work, but there are no mainstream mature server platforms using it and that's what this article is about.

The context is AWS and there's no AWS RISC-V instances yet.

But worry not, RISC-V is inevitable.

> Why do these kinds of articles essentially refuse to admit RISC-V exists?

Because as much as I like RISC-V myself, it hasn't built up the scale needed to supplant x86/64 or ARM. It's still a long way to go before the following are achieved:

- Similar/better performance to x86/64 or ARM, with at least 80% of their performance

- Similar/lower prices compared against x86/64 or ARM

- A win in either price-to-performance or power-to-performance against x86/64 or ARM at some point.

>a long way to go

Veyron releases later this year. Ascalon next year.

It's going to get exciting.

Because in the great scheme of the universe, it doesn't matter.

It is an architecure cheered by FOSS folks, that ignore cloud offerings will be just as closed, and no one is selling RISC-V computers at the shopping mall down the street.