Qualcomm Acquires Risc-V Focused Ventana Micro Systems

A dual ISA decoder with with fuse-off options will likely have unwelcome power-perf-area and yield consequences.

Implementing ARM and RISC-V decoders might depend on licensing fine print for each licensee

Buying a team that's already working on RISCV also reduces the chances of ARM lawyers getting involved.

Correct. However you need circuitry on silicon to implement said architecture which is the expensive and time consuming part.

Qualcomm also wants the RISC-V engineers and their Knowledge and the Software Ecosystem and SDKs/Tools that Ventana has developed over the years to create CPU cores that execute that RISC-V ISA. And that includes all the design Verification tools/EDA tools that Ventana developed for their specific CPU core designs and all that non hardware stuff that takes a larger investment in dollars than just the hardware's development alone costs!

So the RISC-V ISA is Royalty free but not anyone's actual RISV-V ISA executing CPU core designs that cost millions to create and are the proprietary part of the Acquisition that Qualcomm is after. I'd imagine that Qualcomm's Nuvia engineers could more rapidly swap out the ARM ISA Instruction Decoders on any Oyron cores with some RISC-V Decoders and use most of the same Oryon Micro-op engine design that's native to the current Oryan generation cores but maybe Ventana's micro-op execution engine has something that's valuable to Qualcomm as well.

And so ISAs on modern microprocessors are abstracted away at the actual hardware level by the Micro-ops execution engine desogns that are proprietary to the ones that created them. And the reason that many license from ARM holdings is not just the ISA but the software/OS/Drover ecosystem that's built up over the decades for the ARM ISA ecosystem and that costs many times more than the hardware's costs to develop and maintain over the years. And so the ARM OS/Software/API and driver ecosystem is decades more mature than the RISC-V OS/Software/API and driver ecosystem, and that took years and 100s of billions in investments to get ARM where it is today!

But since RISC-V is royalty free there are hundreds of companies using RISC-V, including Nvidia for it's FALCON(FAst Logic CONtrollers) that are used all over Nvidia's GPUs and other accelerators. And with RISC-V one is free to implement only a subset of the RISC-V ISA or create custom RISC-V ISA extensions unlike ARM holdings where licensees have to implement the entire ARM Licensed ISA regardless of if all the instructions are needed for the task and no ISA extensions allowed.

So maybe Qualcomm is interested in the micro-controller market that's lower margin and that makes RISC-V's Royally Free more attractive! Or Qualcomm, like Nvidia, wants to develop some RISC-V Micro-Controllers for it's own in-house needs and not have to pay for ARM Holdings ISA based Micro-Controller designs. Look at Nvidia's dozens of on GPU die Controllers(Encoder/Decoder Logic,Etc) and because that's Nvidia FALCON RISC-V based that's quite a bit of savings in Royalty Payments and CPU core design payments to ARM Holdings or anyone else because FALCON is Nvidia's In-House IP and that RISC-V ISA is free to use for Nvidia or others to save billions that way.

RISC-V being freely available does not mean that implementations of it will not be patented from here to the Orion nebula and back.

They'll need to license future versions of the ARM ISA and now they know the licensor is hostile.

That's not quite what Raspberry Pi did with the RP2350 (the ARM and RV cores are wholly separate) but they did include the ability to fuse off one side or the other, so I wonder if they'll release a cheaper RV-only version at some point.

Frankly, Ventana seemed like an interesting entry in the space, but I have no idea who would have actually bought their servers at the end of the day. They taped out multiple designs, but none actually seem to exist outside their labs. I don't really see any path to meaningful RISC-V server adoption for at least several more years and by that time Qualcomm could design something on their own, assuming they are serious about re-entering the market. Grabbing the talent and any useful IP/core design components makes the most sense to me, anyway.

which means the M1 was being worked on since at least 2018, I'd bet much earlier than that, for sure much earlier than that if you count silicon which never left the lab.

reminder iphones run on apple silicon since 2010, which means they had to be working on it at least since 2008. they have a lot of experience in silicon design by now.

The difference you perceive is mostly software. Windows and Linux are really just designed for desktop machines first and foremost. MacOS was too, but when they transitioned to Apple Silicon, they replaced a lot of the internals with stuff taken from iOS, and iOS is designed with batter life first and foremost.

Getting the level of battery life out of non-apple laptops is just going to be a long, hard slog of going through the operating systems and auditing *everything* and every design decision for how it affects battery life and how much resources its using.

Gotta love paying extra for dark silicon, amirite?

I'm going to go out on a limb and guess you have't been using a CUDA-supported machine as a basis of comparison...

But more likely, the early product line will meet the same fate as the dog in "Old Yeller" (1957) in a market consolidation push. =3

They're a totally different gate count niche than a Cortex-M equivalent.

but unfortunately very in-line with the thesis that qualcomm is getting squeezed by a commodifying market where value-add opportunity is shifting outside of the SoC platform.