Comment by orbifold | Hacker News Reader

Comment by orbifold | Hacker News Reader

Parent Full thread

Personally I think the POWER instruction set is better in many ways. It has a proven track record of high performance and embedded implementations. A lot of ISA design is about avoiding patents and other traps. Ultimately it doesn’t really matter because most of the value is in the implementation and the software is the commoditized complement.

Truly innovative designs don’t actually use RISC encoding and hide the details of instruction encoding. See for example NVIDIAs PTX , which gets mapped to hardware specific instruction streams, which are nothing like a traditional RISC architecture.

To me RISC-V is another example of how open source often produces lowest common denominator copy cat versions of ideas that are 30+ years old. Just like Linux. The sad thing is that this often kills innovation and locks in suboptimal designs for a long time, because it is hard to compete against something that is free.

RISC-V was not designed to be the most optimal ISA however. It was designed to be a teachable ISA which could also be used to implement fast hardware.

The need to be easily reachable puts constrains on the design but is also a benefit in that many people will know this ISA and be able to make tools for it.

POWER was designed to be a compiler writer's dream and has some sharp implementation corners.

I think I would probably recycle the Alpha ISA circa 21164 (EV-5) with maybe a CAS instruction. It was pretty balanced between hardware and software and a lot of the complications in the VLSI design (dynamic logic, mostly) are moot with a modern technology if you stick with reasonable speeds.

Presumably now that the MIPS unaligned byte access patents are expired, a whole bunch of the idiocy that Alpha had to abide to avoid that patent can just be sidestepped.

> A lot of ISA design is about avoiding patents and other traps.

Aren't patents limited to 20 years? Can't you just ignore all inventions that are < 20 years old to be safe from patents? Then you'd still be 10 years ahead of POWER.

I think anybody wanting to implement POWER nowadays means the PowerPC ISA, as implemented in modern POWER CPUs, not the original POWER ISA from 1990.

Power ISA had updates in 2007, 2009, 2013, 2015, and 2020.

https://en.wikipedia.org/wiki/Power_ISA#Specifications

Can you summarize, or point to a discussion of, the ways in which PTX differs from a traditional RISC architecture?

Reply on news.ycombinator.com