This is huge. It looks like the only architectures widely-deployed in ten years will be x86, ARM, Power, and RISC-V (maybe also SPARC64 in Japan, although that's rare in the US).
+ many proprietary GPU instruction sets
MIPS Becomes RISC-V
11–20 of 225 posts
Re: MIPS Becomes RISC-V
#12"Development of the MIPS processor architecture has now stopped"
Is anyone still developing SPARC?
Re: MIPS Becomes RISC-V
#13This is huge. It looks like the only architectures widely-deployed in ten years will be x86, ARM, Power, and RISC-V (maybe also SPARC64 in Japan, although that's rare in the US).
The big innovation in architectures is in the SIMD world. AVX512 (x86 512-bit), SVE (ARM 512-bit), NVidia PTX / SASS (32x32-bit), AMD RDNA (32x32-bit), AMD CDNA (64x32-bit). 64-bit cores (aka: classic CPUs) are looking like a solved problem, and are becoming a commodity. SIMD compute however, remains an open question. NVidia probably leads today, but there seems to be plenty of room for smaller players. Heck, one maj…
Fujitsu seems like a leader for SIMD via their A64FX. I wonder if they will ever venture outside of the supercomputing niche.
Re: MIPS Becomes RISC-V
#14This is huge. It looks like the only architectures widely-deployed in ten years will be x86, ARM, Power, and RISC-V (maybe also SPARC64 in Japan, although that's rare in the US).
The big innovation in architectures is in the SIMD world. AVX512 (x86 512-bit), SVE (ARM 512-bit), NVidia PTX / SASS (32x32-bit), AMD RDNA (32x32-bit), AMD CDNA (64x32-bit). 64-bit cores (aka: classic CPUs) are looking like a solved problem, and are becoming a commodity. SIMD compute however, remains an open question. NVidia probably leads today, but there seems to be plenty of room for smaller players. Heck, one maj…
SIMD today is only really helpful with a few usecases. If you want to encode some video, decode some jpegs, or do a physics simulation quicker, it's really going to help. It won't boot Linux any quicker tho.
I suspect for consumer uses, SIMD is already used for nearly all the use cases it can be.
Re: MIPS Becomes RISC-V
#15I just read the official statement[1] that's linked to in the article.
Just so I get this straight: Wave Computing, the company that bought the remains of MIPS, is now (after bancruptcy) spinning it off as a separate company, that is going to work under the name MIPS, holds the rights to the MIPS architecture, but is doing RISC-V?
[1] https://www.prnewswire.com/news-releases/wave-computing-and-...
Re: MIPS Becomes RISC-V
#16"Development of the MIPS processor architecture has now stopped" Is anyone still developing SPARC?
No one worth mentioning AFAIK.
Re: MIPS Becomes RISC-V
#17"Development of the MIPS processor architecture has now stopped" Is anyone still developing SPARC?
Fujitsu for general purpose servers, and Atmel (maybe others too?) for rad-hardened Sparc.
Re: MIPS Becomes RISC-V
#18This is huge. It looks like the only architectures widely-deployed in ten years will be x86, ARM, Power, and RISC-V (maybe also SPARC64 in Japan, although that's rare in the US).
Why do you assume Power will last another 10 years?
Re: MIPS Becomes RISC-V
#19Earlier quoted context omitted.
The big innovation in architectures is in the SIMD world. AVX512 (x86 512-bit), SVE (ARM 512-bit), NVidia PTX / SASS (32x32-bit), AMD RDNA (32x32-bit), AMD CDNA (64x32-bit). 64-bit cores (aka: classic CPUs) are looking like a solved problem, and are becoming a commodity. SIMD compute however, remains an open question. NVidia probably leads today, but there seems to be plenty of room for smaller players. Heck, one maj…
SIMD today is only really helpful with a few usecases. If you want to encode some video, decode some jpegs, or do a physics simulation quicker, it's really going to help. It won't boot Linux any quicker tho. I suspect for consumer uses, SIMD is already used for nearly all the use cases it can be.
Are you sure about that?
The original SIMD-papers in the 1980s show how to compile a Regex into a highly-parallel state machine and then "reduced" (aka: Scan / Prefix-operation: https://en.wikipedia.org/wiki/Prefix_sum).
A huge amount of operations, such as XML-whitespace removal (aka: SIMD Steam Compacting), Regular Expressions, and more, have been proven ~30 to 40 years ago to benefit from SIMD compute. Yet such libraries don't exist today yet.
SIMD compute is highly niche, and clearly today's population is overly focused on deep-learning... without even seeing the easy opportunities of XML parsing or simple regex yet. Further: additional opportunities are being discovered in O(n^2) operations: such as inner-join operations on your typical database.
Citations.
* For Regular Expressions: Read the 1986 paper "DATA PARALLEL ALGORITHMS". Its an easy read. Hillis / Steele are great writers. They even have the "impossible Linked List" parallelism figured out in there (granted: the nodes are located in such a way that the SIMD-computer can work with the nodes. But... if you had a memory-allocator that worked with their linked-list format, you could very well implement their pointer-jumping approach to SIMD-linked list traversal)
* For whitespace folding / removal, see http://www.cse.chalmers.se/~uffe/streamcompaction.pdf. They don't cite it as XML-whitespace removal, but it seems pretty obvious to me that it could be used for parallel whitespace removal in O(lg(n)) steps.
* Database SIMD: http://www.cs.columbia.edu/~kar/pubsk/simd.pdf . Various operations have been proven to be better on SIMD, including "mass binary search" (one binary search cannot be parallelized. But if you have 5000-binary searches operating in parallel, its HIGHLY efficient to execute all 5000 in a weird parallel manner, far faster than you might originally imagine).
----------
SIMD-cuckoo hashing, SIMD-skip lists, etc. etc. There's so many data-structures that haven't really been fleshed out on SIMD yet outside of research settings. They have been proven easy to implement and simple / clean to understand. They're just not widely known yet.
Re: MIPS Becomes RISC-V
#20The progression of headlines is funny:
1) MIPS Strikes Back: 64-bit Warrior I6400 Arrives https://news.ycombinator.com/item?id=8258092
We are still in the game
2) Linux-running MIPS CPU available for free to universities – full Verilog code https://news.ycombinator.com/item?id=9444567
Okay, we are not doing so great, maybe we can get young kids hooked?
3) MIPS Goes Open Source https://news.ycombinator.com/item?id=18701145
Open Source is so hip and pop these days, lets do that!
4) Can MIPS Leapfrog RISC-V? https://news.ycombinator.com/item?id=19460470
Yeah, sure, that'll happen
5) Is MIPS Dead? Lawsuit, Bankruptcy, Maintainers Leaving and More https://news.ycombinator.com/item?id=22950848
Whoops
6) Loose Lips Sink MIPS https://news.ycombinator.com/item?id=24402107
And there is the answer to the question from previous headline
And now we are here.