This is more of a successor to the Xeon 6E (Xeon 6700E / Sierra Forest-SP), which had 144 cores. There was supposed to be a 288-core variant (presumably Xeon 6900E / Sierra Forest-AP), but they never released it to the public. I was looking forward to it since Sierra Forest-AP was supposed to support a 2-socket configuration (from motherboard spec). That's 576 physical cores in a single server!

That did take me a beat. 7 cores? Must be pretty fast...

jealous of the person at Intel who gets to run `make -j 288`

Is there _any_ confirmed information about Diamond Rapids, beyond one leaked (and seemingly slightly dubious) slide?

maybe update the description

Intel's "Clearwater Forest" 288-core Xeon 7 CPU Will Be a Beast

Incredibly good. A lot of main GPU board partners are making LLM-focused boards. There was a few previews on Gamers Nexus from... Computex I think? I wanna say lots of memory, dual GPU boards that looked very well built for local LLM usage.

I've also heard Intel GPUs make good cheap hardware transcoders for home media servers.

Very curious about this as well.

Though I would say the fact that they were not involved with OpenAI on gpt-oss is not a good sign.

(OAI partnered with Databricks, NVIDIA, Dell, bunch of MSPs. etc.)

IIUC these cores are a lot better than n100. They're 2 generations newer and have some fairly significant improvements.

The Darkmont cores of Clearwater Forest are almost identical with the Skymont cores of Lunar Lake, Arrow Lake S and Arrow Lake H. The only significant difference is that they are made with the Intel 18A CMOS process instead of a TSMC "3 nm" process.

These Darkmont cores have a performance very similar with the Arm Neoverse V3/Cortex-X4 cores and lower than the performance of AMD Zen 5 compact cores.

However, for applications that do not use vector operations, the Darkmont cores have a better performance per die area than Zen 5 compact.

For many applications, a 288-core Clearwater Forest will have a lower performance than an 192-core Turin dense, but for some applications it will be faster and the low area per core will enable Intel to sell it for a lower price than Turin dense.

It remains to be seen when will be the launch of Zen 6 in 2026, but it is likely that it will be a few months after Clearwater Forest.

It is difficult to call Intel's E-core server chips a “clone” of Zen 5c. Sierra Forest was announced before Zen 4c, and they're actually designed-for-area cores instead of just resynthesizing the same chip with a different library (which is the thing you do if you suddenly need a small core, but don't have time to make a proper one) and less cache, like Zen 5c is.

If it's a clone if anything, it's of Arm's “little” strategy.

The few people doing video streaming can get Intel GPUs on PCIe cards and you can have as many or as few cards as you want.

This is a claim you can easily confirm or deny, considering that 144-core, 8-channel implementations of the Xeon 6E exist in the marketplace today.

`make -j288` baby

it also has .5gbGB of cache, and that's 12 lanes of ddr5-8000

Not really, these E-Core are 1vCPU, while other P-Core or Zen 6C core are 2vCPU.

This isn't a hybrid design. It's a gigantic Atom.

There are interesting core scheduling questions with a CPU this large, considering its 4-way-shared L2 design and profusion of thermal domains, however these issues are so complex that no system I've ever heard of — certainly not Linux or Windows — attempts to optimize it.

It's was fine like a month after the first e-core chips launched, certainly handled better than AMDs split ccd vcache which still needs to be pinned.

Once AMD takes the spot, the lack of competition will encourage AMD to act in all the same ways you associate with Intel.

I'm with you brother. Intel needs to fix their ways and I don't think a couple generations of playing second fiddle will be enough. Intel burns a lot of money on marketing hoping to trick people about how great their new shit will be (anyone remember when 14th gen was released?) and it ends up being hot air. I have no faith in this one either.

I think Xeon 7 is the first generation after they woke up. We'll see if it's enough.

I find this comment surprising. They have always had better single-core performance than AMD at the same price tag.

Doesn't microcode need to be reloaded each boot? Can't you just not load the update?

Other comments miss the point entirely. Sure it's possible to circumvent this performance impact, but that's not without its risks and you are still right about being skeptical over performance claims.

Disable mitigations (at your own risk)