VIA got the license because of legacy licenses from the IPs they held with Centaur and Cyrix. Even then Intel argued it wasn't valid anymore and Intel only lost the case because back then VIA got lucky as Intel was being sued for anti-trust for unlawful behavior against AMD after a pretty big scandal. "the FTC extended the Via license agreement with Intel for a total of 15 years, it now ends on April 7, 2018. Via now has clear sailing until the patents in question are expired, a huge win for the boys in Austin."
So it wasn't even until last year. The license ended almost 4 years ago. Which means any new patents, any new instructions, Intel developed since then, VIA don't have a license for them.
I know that they "inherited" x86 licenses through buying Cyrix and Centaur. I was not talking about getting new instructions and other tech from Intel, I was talking about the ability and knowledge to produce x86 CPUs. Considering that KX-6xxx entered mass production in 2019-2020, those are still valid. Obviously, I would not be talking about the ability to receive updates from Intel considering that we are talking about potential chip design sanctions there. Plus x86 license was shared with THATIC when AMD established a JV with the latter (those Zen 1 copies), so it is not the only source for China.
Man. Did you even read the article you linked to?
"Attention: we are testing an engineering processor with a partially disabled cache, the production sample should have better results."
And like I said Elbrus2K is optimized for technical computing. Look at the MP MFLOPS or HPL GFLOPS numbers. It blows the other processors out of the water in that. The HPL test is basically similar to the test used to make the TOP500 supercomputer index. Of course it won't do well in branch heavy, or, multi-threaded code. The architecture isn't made for that. That is what processors like the Baikal S are for.
Those are the only publicly available results. So you are saying that with the full cache its single-thread performance would skyrocket 2-2.5x times? I find it doubtful, the disparity is too big.
What other processors does Elbrus blow away in HPL? It is only compared to i7-2600 and older Elbruses there. You are also saying "look at HPL GFLOPS" there but then say that the results are BS, lol. Even if we go by MCST's claim of 750 GFlops on double-precision load and assume that they meant on HPL, then again it is on the
, not EPYC 7643.
It is basically a slower clocked slightly hotter 48 core with like 85% the performance of a similar Kunpeng 920.
More than enough for a bank. Not too bad considering Baikal S was made on 16nm process and Kungpeng 920 was on 7nm.
This is again what is claimed by Baikal itself, not a real test. Plus it uses ARM cores, meaning that it is as good as dead under chip design sanctions. Like HiSilicon Kirin and Kunpeng are basically gone for now, and mostly live off the stock.
You know, there are other VLIW architectures than Itanium. Philips used to have the TriMedia processor for example. It was pretty popular back in the day to encode and decode video at a time other CPUs couldn't do it at any appreciable speed.
Itanium was just an example, the point is that there is no domination of VLIW-based processors in parallel computing meaning that being more efficient in theory does not necessarily translate to practice because writing such a compiler is very hard.
The MIC uses CPUs for technical compute. They do not run SQL databases and Java servers on weapons systems.
We are talking about real-time or as close to it as possible complex computations of trajectories, signal processing, neural networks, and crap like that.
I never disputed that usage of older Elbruses paired with additional co-processors is a viable thing for military applications. Moreover, Russia can be self-sufficient for MIC chips because they use older tech nodes. We are talking about the broad case of chip design denial - simply having indigenous chips for MIC is not enough in that case because unless you want your economy to go into a dumpster, you absolutely need more powerful CPUs for customer, corporate, and HPC markets. There, the application of co-processors is limited to the stuff you've mentioned - NN acceleration and the like.
t is like this. There are catalogs of chips. Yet it seems like there is nothing on commercial systems or system offers which uses them. The government funded the development. The architecture was originally designed for the MIC. So, what do you think. Do you think they just design CPUs, and compilers, and make a test batch with one wafer. And call it quits? Because you know what. The weapon systems only use the 90nm processors made in Russia. They don't use this 28nm or 16nm stuff. So where is it being used? Wafers are typically ordered in thousands minimum at a foundry. Each wafer can fit at least a dozen CPU dies in it. Just use your brain a bit.
Originally designed for MIC but later it was "reconfigured" as a broad import substitution solution. They were supposed to be used in commercial systems -
and they were already working on the Elbrus 16C-based servers. Elbrus 2Cs were used in MCST-Kraftway KM-4 monoblock PCs for office work, etc. So no, they ARE being used - you just don't hear about those systems much because all of the Elbrus parties were rather tiny.
It blows the PowerPC processor used in the US F-35 flight computer out of the water. That is enough for the MIC.
The Su-57 flight computer has boards with two Elbrus-2SM chips and like 4 DSPs chips in it.
And it is quite useful for FP compute. The internet cloud sector and web services were never the target of this.
MIC self-sufficiency is a necessary but not sufficient condition. Both China and Russia are more or less self-sufficient in MIC chips, hence the main talk was about the other markets. The modern economy needs computing power and a lot of it. There China simply has a lot more options - Zen 1 licensed copies, Zhaoxins, Loongson, manycore SWs, etc. - for a variety of markets (MIC, HPC, consumer), instead of a single Elbrus. Hence it is fair to say that it is better positioned to withstand pressure.
So you are telling me the Russian MIC should be buying Zhaoxins fabbed at TSMC? Yeah that would work. Not.
Like I said. Russia has Baikal S for crap like that. And there are people working on RISC-V in Russia too.
You seem to be obsessed with x86 compatibility for whatever reason. I mean even Apple doesn't care about it.
The MIC write their own software for the weapon systems. They don't use other people's.
I was not even talking about MIC, you switched to it yourself. MIC stuff is primarily the older Elbruses, they don't use even 8Cs there. As I've said, MIC is a necessary but not sufficient condition when it comes down to indigenous chips.
Is there any doubt the real reason Huawei got banned was competition to Apple and Qualcomm? It was only when Huawei got competitive in the smartphone market that they came up with the BS they had to kill it because of security issues in every single system they made.
Of course not, it was obvious from the start. It is just funny how Americans start whining and using administrative apparatus the moment they get outcompeted - the same shit happened with Japan in the 80s when the latter got cucked by Plaza Accord because Muricans were lagging behind.
TL;DR: If you were talking strictly about MIC chips, then sorry - I misunderstood you and you are correct about Russia's self-sufficiency in that. However, in other sectors, China has a lot more to work with. I think we are derailing the thread now, hope we don't get banned, lol.
