| ▲ | pizza234 4 hours ago |
| I'm quite skeptical; their baseline is a in-order CPU, instead of a modern out-of-order design, and they're testing a small amount of programs. Also, I'm not an expert in CPUs, but the underlying idea doesn't seem entirely novel - WaveScalar explored dataflow-based architecture 20+ years ago. |
|
| ▲ | kragen 3 hours ago | parent | next [-] |
| They're claiming better performance per watt, not better absolute performance, and out-of-order does worse on that metric than the superscalar in-order design they say they beat by 4.59×. That's why, for example, PULPino is in-order. A third interesting metric is performance per transistor or per area, and generally the best designs there are the long-forgotten MISC designs like the MuP21 and the GA144. |
| |
| ▲ | adrian_b an hour ago | parent | next [-] | | The performance per watt of very slow CPUs is always very good, so that alone is not impressive unless much more data is provided to prove that is really better than what is typical. On the other hand, the performance per area of the "traditional" CPU that is used for comparison, which I assume that it is some kind of RISC-V, is very poor, so that the performance per area of the new CPU is likely to be also poor, even if much better than that of the strawman. Their "traditional" CPU has an area that is slightly bigger than that of Cortex-A720 made with a CMOS process almost twice as dense. So if they were made with the same process, the "traditional" CPU and the new CPU would have an area somewhat bigger than a half of a Cortex-A720. But Cortex-A720 is an out-of-order 4-wide superscalar core with advanced SIMD and scalable vector extensions, with a good branch predictor and with many other enhancements, for which the area ratio in the same process is expected to be much greater than 2 in comparison with the CPUs evaluated in TFA, while its performance must be several times greater than that of a 2-wide in-order CPU core. So both the "traditional" CPU and the Frostbite CPU appear to have a much bigger area than expected for their performance. Moreover, the performance of in-order CPUs is very sensitive to the benchmark selection and to the quality of the compiler optimizations. So the performance obtained for the benchmarks selected by them, like FFT or LU decomposition, is expected to be much better than for benchmarks that would execute an application not dominated by such regular loops, but having lots of hard to predict branches, e.g. a compiler benchmark. They say that the ISA will be published. At that time it will be easier to appreciate whether this CPU architecture is really innovative. For now, they have done a lot of work, by fully implementing the CPU core, but the data presented is not sufficiently convincing to demonstrate that it is indeed an architecture that will be able to compete with SOTA CPU cores, e.g. with those from the Arm company. | | |
| ▲ | kragen an hour ago | parent | next [-] | | > The performance per watt of very slow CPUs is always very good Generally this has not been my experience. Typically both very slow CPUs like the ATMega328 and high-performance CPUs like an Intel i7 are in the neighborhood of 1 nanojoule per instruction. (For operations more than 8 bits wide, the AVR is significantly worse.) Low-power CPUs like the STM32L line are commonly around 0.2nJ/insn. Ambiq's subthreshold logic gets them closer to 0.05nJ/insn. I agree with the rest of your comment. | |
| ▲ | jkhdigital 27 minutes ago | parent | prev [-] | | I am of the opinion that even if this novel architecture doesn’t yield dramatic improvements in peak performance across broad classes of workloads, it may still be a welcome improvement in the legibility of software performance down the stack. By this I mean: currently, transforming a program written in a typical high-level language to optimized ISA code involves what can only be described as black magic inside the compiler. If you don’t know the implementation details of modern superscalar OOO branch-predicting microarchitectures then you may have no clue why two very similar-looking functions could have dramatically different running times. Cue Matt Godbolt tech talk… What I see as the more promising side of this type of development is a tighter coherence, and thus more transparency in expected runtime performance, between the software above the ISA and the hardware below it. Essentially we now have a linear machine with a linear source language, and with the popularity of Rust nowadays we actually have a critical mass of developers who are already used to applying ownership discipline in program design. |
| |
| ▲ | csande17 3 hours ago | parent | prev [-] | | > They're claiming better performance per watt, not better absolute performance I shouldn't have to read this far into the tea leaves to figure out what performance claims a company is making, but these facts combined sound like they're claiming much better absolute performance, too? > Each Frostbite quad (a two-by-two grid of tiles) paired with four L3 nodes is 0.792 mm². > This TCPU [the one they benchmarked against] is 0.895 mm² at the same process. > Across all the workloads, we average an 8.77x performance-per-area It got 8.77x performance per area, and was 88% of the area, so that's 7.76x actual performance, right? Not that "a parallel processor outperformed a single-core processor on parallelizable tasks" is the most earth-shattering relevation in the world, especially when they won't say which single-core processor "for legal reasons". | | |
| ▲ | kragen an hour ago | parent [-] | | The second paragraph of the linked page contains their first performance claim. It reads: > On a set of 25 initially chosen workloads, including 15 held out until the final evaluation, we achieved a 4.59x performance-per-watt multiple when compared to a modern, silicon-proven superscalar (dual-issue, in-order) CPU core. They do start claiming performance per area in the following paragraph, but that's not an interesting claim for a couple of different reasons: - As I said above, MISC designs get dramatically better performance per area than any mainstream CPU. - We're already in the dark-silicon era, where even the relatively low performance-per-area designs used in laptops and cellphones must thermally throttle on a regular basis to avoid burning themselves up, and contain many specialized accelerators used only a small fraction of the time (AES-NI, AVX-512, saturating 8-bit arithmetic, popcount, etc.) instead of additional small general-purpose cores that could run all the time. Their claims would be a lot more interesting if they provided some kind of absolute numbers to compare against. |
|
|
|
| ▲ | rhdunn 3 hours ago | parent | prev | next [-] |
| If you are interested in understanding how CPUs work there are a number of "build your own CPU" style video series on YouTube: - Ben Eater's series "Building an 8-bit breadboard computer!": https://www.youtube.com/playlist?list=PLowKtXNTBypGqImE405J2... - Fabian Schuiki's series "Build a Superscalar CPU" (breadboard and custom boards): https://www.youtube.com/playlist?list=PLyR4neQXqQo5nPdEiMbaE... - Polymath Unlimited's series "The CPU you can build (Tutorial Series)" (logisim): https://www.youtube.com/playlist?list=PLe8IZRJAn3oD3E2tDhNpI... - mattbatwing's series "Let's Make a Redstone Computer!" (Minecraft): https://www.youtube.com/playlist?list=PL5LiOvrbVo8nPTtdXAdSm... |
| |
| ▲ | caned an hour ago | parent [-] | | Odd to see this downvoted. I'd like to see more of these projects. |
|
|
| ▲ | shioyama an hour ago | parent | prev | next [-] |
| They are using _interaction nets_. Interaction nets date back to the 1990s, they are a fundamentally different programming paradigm which is parallelisable by their nature (graph rewriting that can be done locally in any order). https://en.wikipedia.org/wiki/Interaction_nets |
|
| ▲ | petra 4 hours ago | parent | prev | next [-] |
| Yes, their comparisons are weird. They also compare their architecture simulation language to RTL, instead of comparing it to existing arcitecture sim languages. |
| |
| ▲ | chrisjj 4 hours ago | parent [-] | | > Yes, their comparisons are weird. Generated by ChatGPT, probably. | | |
| ▲ | smolder 3 hours ago | parent [-] | | It's slop. Token spend that doesn't accomplish anything because the prompter doesn't know what they're doing. | | |
| ▲ | chrisjj 2 hours ago | parent [-] | | It accomplishes convincing gullible readers. "AI"s primary use case. |
|
|
|
|
| ▲ | txdv 2 hours ago | parent | prev | next [-] |
| I think that 20 years ago creating your own processor as a startup wasn't really feasible, maybe with AI it has become? Some companies will try it out and history will be the judge of that |
| |
|
| ▲ | amelius 3 hours ago | parent | prev | next [-] |
| Isn't this more about work-per-joule than about throughput? At least that benchmark graph is in performance-per-watt. |
| |
|
| ▲ | checker659 4 hours ago | parent | prev | next [-] |
| Maybe it's Itanium reincarnated |
|
| ▲ | motbus3 4 hours ago | parent | prev [-] |
| Yeah. Something smells fishy |