Rendered at 14:35:56 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
titzer 3 hours ago [-]
Ohhh look, value prediction. Exactly the kind of thing that led to Spectre. There will be a cottage industry of information leaks and mitigations for a decade.
twoodfin 16 hours ago [-]
I don’t think picking a handful of SPEC benchmarks that approximate today’s most common agentic workloads (compiling code, interpreting Python) and then calling them “agentic benchmarks” is misleading at all.
That you need a whole lot of “ordinary” compute to benefit from the scaling properties of agents is the reason Nvidia is making this chip in the first place.
jcranmer 15 hours ago [-]
The four benchmarks selected are cppcheck, llvm, cpython, and gcc [1]. These are all essentially compiler benchmarks... and all of the compiler benchmarks in SPEC cpu2026! This makes the benchmark selection somewhat suspicious to me, since it's not particularly representative of a diverse set of workloads.
I also don't buy that it's a particularly representative set of tasks you might do with agents. Also included in the SPEC benchmarks are multimedia codecs, lossless data compression codecs, sqlite (i.e., database), all of which are going to be things you should easily throw into the sets of tasks an agentic workload might do. Cherry-picking just the compiler benchmarks instead of all of SPECint... again, it just raises a couple of eyebrows.
[1] To be honest, I'm kinda surprised that both gcc and llvm are in SPEC cpu2026.
fibonacci112358 14 hours ago [-]
The code in compilers is the closest to your typical app you can get in a benchmark like SPEC, eveerything else is actually far more specialized. Compiler code is full of small basic blocks, lots of branches, indirect memory access; it's actually harder to get good performance for such code, both for CPUs and compilers (that was part of the death of Itanium too).
jcranmer 13 hours ago [-]
That's true of most of the applications in SPECint (SPECfp is a different matter); there's nothing special about compilers there.
Where compiler code is going to get really unusual, I suspect, is that compilers tend to be a little mono-focused on relatively few data structures. I know I was able to get measurable (single-digit percent!) performance differences in LLVM making very small tweaks to layout in llvm::Value. By contrast, when I was working on Thunderbird, the only similarly small change I could think to make that kind of difference would be to "oops, all string functions are now a cross-DLL call" (and even then, only because string handling is so dominant in that kind of application). Another kind of difference is that the compiler-based benchmarks are going to be quite light in virtual or indirect function calls (there's more of an emphasis on switch-based dispatching than vtable-based dispatching in most compiler implementations), which is going to make it a poorer proxy for some kinds of applications.
14113 6 hours ago [-]
Seconding this - having worked on LLVM and Firefox, the performance tuning of each application was very different. Even measuring the performance of an application like Firefox (in a meaningful way) is non-trivial, wheras compilers are much more approachable with traditional profilers (either tracing or sampling).
14113 6 hours ago [-]
I mildly disagree - depending on your definition of "typical app". Most applications have much greater use of multi-processing and concurrent cross-thread (or cross-process) communication. Compilers, aside from high-level parallelism across modules, tend to be quite single-threaded applications.
If you're solely interested in single-core performance, then I would agree that they are a good stress test, but I think for a processor that is being sold on it's parallelism, they are not a great benchmark.
lstodd 13 hours ago [-]
Death of Itanium was that it was a) VLIW and b) Intel was too arrogant. So it went to the same destination as later Larrabee and ATI/AMD attempts at VLIW GPUs.
That is, nowhere.
Also you are wrong and anyone sizing up an arch to put their loads onto must first try that load on it and not rely on "bah, compilers compile on it".
speed_spread 12 hours ago [-]
VLIW works for DSP applications, it's not an instant dead end. It's a good fit in cases where code path and memory accesses are predictable, like shader code.
Symmetry 2 hours ago [-]
Essentially every cell phone out there has a VLIW DSP like Qualcomm's Hexagon cores (though AFAIK Qualcomm is the only one who lets you run your on programs on their DSP).
pjmlp 9 hours ago [-]
Itanium only died because AMD exists, and due to various licensing reasons they were allowed to come up with AMD64.
wtallis 7 hours ago [-]
In a hypothetical world where AMD wasn't allowed to do AMD64, and Intel stayed committed to Itanium: Itanium would still have sucked, and both PowerPC and SPARC would have out-sold Itanium by an even wider margin than they did in this reality. Itanium could only have succeeded if AMD64 wasn't possible and literally all of the competing 64-bit architectures were killed off by their owners so they could jump on the Itanium bandwagon. Itanium managed to kill off PA-RISC and Alpha and (high-end) MIPS roadmaps, but it still had competitors that were not just viable but actually more successful.
hajile 20 minutes ago [-]
> Itanium would still have sucked
I'm not completely convinced of this.
If you ignore VLIW, you just have a very unexciting RISC ISA, but because of the VLIW, you get extra scheduling info that most RISC designs don't provide which might be advantageous. The real question is actually about the code density of 41-bit instructions and if it can be offset by the 128-bit package (and perhaps something like allowing new 24-bit compressed instructions).
Poulson already somewhat proved part of this as it added back a traditional frontend and even added some OoO capabilities and 4-way SMT. It wasn't earth-shattering, but it wasn't absolute garbage either.
pjmlp 5 hours ago [-]
Except you would never had Windows running on either PowerPC and SPARC.
Remember, the very first Windows XP 64 bit release was on Itanium.
my123 1 hours ago [-]
Windows NT on PowerPC was an actual product, although running in 32-bit mode.
torginus 7 hours ago [-]
I think the choices of these workloads are deliberate, considering this is a large core count CPU linked to a FP-monster GPU with a high speed, low latency datalink.
The former implies per core memory bandwidth is probably not great, meaning SQLite wont perform as well, the latter meaning FP workloads are better done on the GPU, so video encoding wont be a high point. The idea is to run branchy integer workloads that fit into RAM imo, which is what these benchmarks measure.
amluto 13 hours ago [-]
In today’s “agentic” world, everyone seems to have forgotten approximately everything we used to know about security. And this new CPU is going all in on value speculation. Delightful.
Maybe if “cyber” models get good enough at exploiting speculation attacks, people will start demanding equipment that is less prone to these attacks.
dzaima 4 hours ago [-]
aarch64 has a CPU mode, DIT (Data Independent Timing), specifically for allowing software to request all fancy value prediction stuff to be disabled for the duration of processing of sensitive data.
(doesn't help when the attack target is general-purpose/user-controlled code leaking things, but if you're relying on a process not leaking memory plainly available to it without full careful control of what the process runs, you've already been fully-SOL on that for decades and nothing has nor will nor can change about that)
amluto 1 minutes ago [-]
No way, ARM screwed this up less than Intel and at least allows user code to access the control bit. Intel’s equivalent, DOITM, is not accessible at CPL3.
scotty79 1 hours ago [-]
I think security is going to get devalued in the near future. The safest strategy is going to be to need as little as possible of the stuff that you need to keep secret. And you won't keep that stuff on a device that is shared in any manner, or maybe even connected anywhere.
archipelago123 13 hours ago [-]
Attacker: I can run any code on this machine? Time for speculation attacks!
Attacker: Oh wait, I can run any code? I already own the machine...
samrus 6 hours ago [-]
Did you forget about spectre and meltdown?
debugnik 8 hours ago [-]
Not really, Spectre showcased an attack from within a JavaScript sandbox, which isn't considered to own the machine.
Any side effects from speculation machinery can easily become a side-channel to infer values across security boundaries.
transcriptase 14 hours ago [-]
Let’s not forget the fact that despite their new market cap and being the beneficiary of having a near monopoly on making incredibly complex pickaxes during a gold rush… Nvidia is still the company with a long and consistent history of misleading their customers via marketing. Their greatest hits include:
- The digital equivalent of the VW emissions scandal where drivers detected when they were being benchmarked and altering rendering for better results
- Selling GPUs as having 4gb vram when it was only 3.5gb usable, the remaining 0.5 being absurdly slower and causing performance loss when used
- Using intentionally misleading naming schemes to obfuscate things like memory bus width being drastically different between what superficially appeared to be similarly spec’d cards
A dozen other less egregious but similarly disingenuous decisions
But to be clear, I’m a huge fan and continue to run Nvidia because their products are generally incredible regardless
podocarp 3 hours ago [-]
That 4gb vram deal sounds ahead of its time. CPU caches are tiered, why not ram? Looking forward to future systems with 8gb ddr6 and 8gb ddr5. "Swapping to ram" would become a thing.
lostmsu 3 hours ago [-]
You forgot their tensor core performance numbers "with sparsity".
KeplerBoy 7 hours ago [-]
AMDs marketing is just as sketchy. Their new Zen6 server page claims 3.3x performance per watt over vera on "agentic workloads" for a 100 kW rack. Maybe they are comparing a CPU heavy rack to an nvidia vera rack with 50 kW of GPUs sitting idle, who knows?
That doesn't clarify it and AMD says as much "Because these estimates rely on published results, internal measurements and projection-based
scaling factors, they are intended to provide directional comparison rather than direct measured rack benchmarks."
It's marketing after all and nobody should make buying decisions based on that.
davoneus 16 hours ago [-]
I'm glad to see the competition; nothing provokes meaningful change without it. Not surprised at Nvidia's blatant fabrications though; more of the same we've seen time and again (Superchip anyone, with 2+ year old designs).
foota 17 hours ago [-]
This is neat. I'm only halfway through but some really cool discussion about the internals of the chips and what they offer.
pjmlp 9 hours ago [-]
So typical big corp marketing material disguised as whitepaper.
brcmthrowaway 15 hours ago [-]
So, is this rip AMD
wmf 15 hours ago [-]
AMD announced a faster CPU two days later.
allenrb 14 hours ago [-]
It’s a time-honored tradition to compare your upcoming product to the competition’s old news.
That said, looks like an impressive core and complete CPU built with it. Would love to see a smaller, affordable version.
jnaina 11 hours ago [-]
Olympus’s raison d’être is to orchestrate GPUs—handling the control-heavy, latency-sensitive workloads that keep Rubin fed and the AI factory running without bottlenecks. Including tool calling and data marshaling.
They are not competing in the CPU space. Different markets.
That you need a whole lot of “ordinary” compute to benefit from the scaling properties of agents is the reason Nvidia is making this chip in the first place.
I also don't buy that it's a particularly representative set of tasks you might do with agents. Also included in the SPEC benchmarks are multimedia codecs, lossless data compression codecs, sqlite (i.e., database), all of which are going to be things you should easily throw into the sets of tasks an agentic workload might do. Cherry-picking just the compiler benchmarks instead of all of SPECint... again, it just raises a couple of eyebrows.
[1] To be honest, I'm kinda surprised that both gcc and llvm are in SPEC cpu2026.
Where compiler code is going to get really unusual, I suspect, is that compilers tend to be a little mono-focused on relatively few data structures. I know I was able to get measurable (single-digit percent!) performance differences in LLVM making very small tweaks to layout in llvm::Value. By contrast, when I was working on Thunderbird, the only similarly small change I could think to make that kind of difference would be to "oops, all string functions are now a cross-DLL call" (and even then, only because string handling is so dominant in that kind of application). Another kind of difference is that the compiler-based benchmarks are going to be quite light in virtual or indirect function calls (there's more of an emphasis on switch-based dispatching than vtable-based dispatching in most compiler implementations), which is going to make it a poorer proxy for some kinds of applications.
If you're solely interested in single-core performance, then I would agree that they are a good stress test, but I think for a processor that is being sold on it's parallelism, they are not a great benchmark.
That is, nowhere.
Also you are wrong and anyone sizing up an arch to put their loads onto must first try that load on it and not rely on "bah, compilers compile on it".
I'm not completely convinced of this.
If you ignore VLIW, you just have a very unexciting RISC ISA, but because of the VLIW, you get extra scheduling info that most RISC designs don't provide which might be advantageous. The real question is actually about the code density of 41-bit instructions and if it can be offset by the 128-bit package (and perhaps something like allowing new 24-bit compressed instructions).
Poulson already somewhat proved part of this as it added back a traditional frontend and even added some OoO capabilities and 4-way SMT. It wasn't earth-shattering, but it wasn't absolute garbage either.
Remember, the very first Windows XP 64 bit release was on Itanium.
The former implies per core memory bandwidth is probably not great, meaning SQLite wont perform as well, the latter meaning FP workloads are better done on the GPU, so video encoding wont be a high point. The idea is to run branchy integer workloads that fit into RAM imo, which is what these benchmarks measure.
Maybe if “cyber” models get good enough at exploiting speculation attacks, people will start demanding equipment that is less prone to these attacks.
(doesn't help when the attack target is general-purpose/user-controlled code leaking things, but if you're relying on a process not leaking memory plainly available to it without full careful control of what the process runs, you've already been fully-SOL on that for decades and nothing has nor will nor can change about that)
Attacker: Oh wait, I can run any code? I already own the machine...
Any side effects from speculation machinery can easily become a side-channel to infer values across security boundaries.
- The digital equivalent of the VW emissions scandal where drivers detected when they were being benchmarked and altering rendering for better results
- Selling GPUs as having 4gb vram when it was only 3.5gb usable, the remaining 0.5 being absurdly slower and causing performance loss when used
- Using intentionally misleading naming schemes to obfuscate things like memory bus width being drastically different between what superficially appeared to be similarly spec’d cards
A dozen other less egregious but similarly disingenuous decisions
But to be clear, I’m a huge fan and continue to run Nvidia because their products are generally incredible regardless
https://www.amd.com/en/products/processors/server/epyc/9006-...
You could try reading the footnotes, which include a link to https://www.amd.com/content/dam/amd/en/documents/solutions/a...
It's marketing after all and nobody should make buying decisions based on that.
That said, looks like an impressive core and complete CPU built with it. Would love to see a smaller, affordable version.
They are not competing in the CPU space. Different markets.