[albanread]

People should zoom right out and think about the whole RISC-V project. When our phones have billions of transistors, are we seriously supposed to believe that RISC philosophy still matters. Personally I greatly prefer the user programmable 68000 family of processors. The marketing of RISC-V is perhaps the most impressive thing about it. Each to their own, I can see why giant SSD manufacturers want to use a license free design and share the cost of compiler development. Is there really anything else?

[socialdemocrat]

I loved the 68k. It was what got me started with Assembly. But one of the key reasons I have an interest in RISC-V today is for its educational potential. I know how x86 killed my interest in assembly coding. ARM honestly isn't all that much better.

RISC-V gives people a way to learn and understand what a modern CPU is like. Remember Donald Knuth's books. He teaches algorithms on an imaginary CPU. As CISC architecture got superseded by RISC, he started using an imaginary RISC CPU in teaching.

His point is that people implementing stuff need to have some sense of how the hardware works to understand tradeoffs. RISC-V is in my view a great CPU arch to give that kind of understanding for somebody who is not necessarily interested in writing assemblers, compilers or what not.

Beyond that RISC-V really fits well with the heterogenous computing trend we are moving towards where specialized hardware is increasingly doing more and more of our tasks. I would say it is and advantage that these different specialized chips have some commonality between them. RISC-V is giving people a way of creating a whole ecosystem of chips for a variety of purpose which share a lot of instructions, debuggers, profilers, compilers and other tools.

There is no way x86 could be part of that revolution. x86 is stuck as a general purpose CPU. RISC-V on the other hand will power desktop computers, smart phones, micro-controllers, AI accelerator cards, super-computers and just about anything.

[monocasa]

> When our phones have billions of transistors, are we seriously supposed to believe that RISC philosophy still matters.

The point isn't just saving gates because it's cheaper. Less gates means less critical path length, meaning less power consumption, and/or higher overall performance when compared apples to apples.

[kortex]

Yeah, absolutely. Personally, when I zoom out, and look at the trends of engineering in general: simpler modular systems that compose well together vs bespoke solutions, RISC-V precisely follows the trend. Reduce global state. Make it easier (for humans and algos) to reason about control flow. Have a simple core with optional extensions. This all makes building multi-core solutions way simpler. We are fast running out of transistor density improvements. But we are getting way better at building coprocessors. There's clear value in "doing more simple things in parallel".

[panick21_]

> RISC philosophy still matters

What matters is not 'RISC philosophy' but that it is an Open Standard that allows for Open implementation.