Intel's "clearwater Forest" Xeon 7 E-Core CPU Will Be a Beast
Key topics
The tech world is abuzz with Intel's upcoming "Clearwater Forest" Xeon 7 E-Core CPU, sparking debate about its architecture and potential performance. Some commenters, like jeffbee, are calling it a "gigantic Atom," while others, such as ac29, argue that the E-cores have evolved significantly beyond the old Atom line. As the discussion unfolds, key concerns center around core scheduling, with some, like devmor, questioning how the "e-core" scheduling will handle multicore workloads, while others, like api, suggest strategic pinning of VM cores to CPU cores could be a solution. Amidst the chatter, a consensus emerges that this CPU is poised to be a beast, particularly for cloud workloads.
Snapshot generated from the HN discussion
Discussion Activity
Very active discussionFirst comment
2d
Peak period
80
48-60h
Avg / period
13.6
Based on 95 loaded comments
Key moments
- 01Story posted
Aug 27, 2025 at 6:11 AM EDT
4 months ago
Step 01 - 02First comment
Aug 29, 2025 at 10:48 AM EDT
2d after posting
Step 02 - 03Peak activity
80 comments in 48-60h
Hottest window of the conversation
Step 03 - 04Latest activity
Sep 2, 2025 at 8:03 PM EDT
4 months ago
Step 04
Generating AI Summary...
Analyzing up to 500 comments to identify key contributors and discussion patterns
Want the full context?
Jump to the original sources
Read the primary article or dive into the live Hacker News thread when you're ready.
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.
I did see mention of Ponte Vecchio.
It seems like Intel is getting closer to its Larrabee vision of massive MP low power chips (which were 486 cores in the shelved 2010 project).
While Intel's E cores share some history with the Atom line, they have almost no resemblance to the Atom line you might remember from the netbook days. In fact, Intel hasnt made a consumer Atom in a decade (there are still Atom processors for the embedded market).
The first E cores came out a few years ago with performance similar to Skylake and performance has only increased from there. Intel's first Xeon E core processor from last year even outperformed the some of the big core Xeons from the year before: https://www.phoronix.com/review/intel-xeon-6780e-6766e/10
The evolution of the Atom cores has been continuous in the series ... Silvermont, Airmont, Goldmont, Golmont Plus, Tremont (also included as E-core inside Intel Lakefield), Gracemont (from Alder Lake/Raptor Lake/Alder Lake N/Amston Lake), Crestmont (from Meteor Lake/Sierra Forest), Skymont (Lunar Lake/Arrow Lake), Darkmont (Clearwater Forest).
All the cores of this series have been designed in succession by the same team and the improvements between them have been incremental. These cores have always been intended as competitors for various Arm Cortex-A cores, e.g. Goldmont Plus for Cortex-A75, Tremont for Cortex-A76, Gracemont for Cortex-A78, and now Darkmont/Skymont for Neoverse V3/Cortex-X4.
Before Tremont, these Atom cores were used only in homogeneous CPUs. In recent years only those sold for industrial applications have retained the brand "Atom". The most recent CPUs branded as "Atom" have been launched last year.
For now, it does not matter whether one refers to any of these cores as "Atom" cores or as "E-cores", because either name refers unambiguously to the same cores. "Atom" seems preferable, because it is more clearly Intel-specific. One could refer as "efficient cores" to cores in CPUs made by others.
The first E-cores were the Tremont cores of Intel Lakefield, which was launched in Q2 2020, which were paired with Ice Lake cores, and which had a performance much less than Skylake cores. Only the next generation of Atom cores a.k.a. E-cores, Gracemont, has reached performances comparable with Skylake. While Gracemont was first launched in Alder Lake, later it has become widely available in Atom-branded CPUs.
Anyway as you note these are excellent cores, and having 576 slightly wimpy cores in a box is an idea worth considering.
I wonder how they'd work for AI compared to a GPU? That's all about memory bandwidth so I'm guessing the GPU would still win unless it was an AI type that is more compute-bound.
I am looking forward to Zen6C vs Xeon 7-Ecore. Wondering about the cost / pref.
If that is the case, why do we need more DataCentre? Wouldn't it be easier to retrofit existing DC. Sometimes these sort of hyper scale is just so big they are beyond my imagination.
Makes sense.
Even more interestingly, the Knights Landing series had a PCIe coprocessor version, which ran a stripped-down Linux kernel, and you could SSH onto it. One of my friends got one for free at a conference, and I really wish I'd picked one up!
Intel's "Clearwater Forest" 288-core Xeon 7 CPU Will Be a Beast
Intel also, as far as I'm concerned, owes me $100.
Not to be confused with Intel IME.
https://en.wikipedia.org/wiki/Intel_Management_Engine
https://wiki.archlinux.org/title/Microcode
Update the BIOS/UEFI which includes an updated microcode.
Via software at runtime after each reboot (Windows/*nix).
https://www.youtube.com/watch?v=b6vQlvefGxk
- I've delid the processor and now it doesn't work. Can I have it replaced under warranty?
- dead tone
It's two B60 GPUs on a single x16 PCIe card. Nothing unified.
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.)
If it's a clone if anything, it's of Arm's “little” strategy.
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.
turns out the answer is "lots"
Instead of that, there will be a cut down variant of Panther Lake, named Wildcat Lake, with 2 P-cores and 4 E-cores.
That will clearly have a much better performance than N150, including a much better Xe3 GPU, but it seems very unlikely that Intel will sell it at the low prices that made the N-series so attractive.
I assume that Wildcat Lake will be priced similarly to N350, so it will be available in computers closer in price to $300 than to the $100 that can get you an N100 or N150 computer.
N100/N150 reinvigorated the x86 ecosystem with price/perf of Arm plus compatibility with existing software.
> Wildcat Lake will be .. closer in price to $300
That price and P-core power budget would compete with compatible x86 AMD Ryzen instead of incompatible Arm/RISC-V boards.
Why would Intel surrender a long-sought advantage over Arm, after they finally succeeded with capable E-cores?
> It's genuinely crazy how much better value an N100 is and how much better it works out of the box than a Pi for anything.. that doesn't need to talk to electronics/GPIO.. Low cost x86_64 solutions beat the pants off ARM in the PPPITA (performance per pain in the arse) department. The Raspberry Pi software ecosystem advantage nopes the moment x86 shows up to the party.
https://news.ycombinator.com/item?id=44465319
> Intel N150 is the first consumer Atom CPU (in 15 years!) to include TXT/DRTM for measured system launch with owner-managed keys. At every system boot, this can confirm that immutable components (anything from BIOS+config to the kernel to immutable partitions) have the expected binary hash/tree.
My point is Intel's accountants have a history of seeing 'fantastic value' as 'margins too low', which reduces the margin of error in engineering to ~0.
If they must have a $300 price point, offer one with many E-cores and zero P-cores. I wonder how long Twin Lake N150 will be sold, if there is no successor, https://www.intel.com/content/www/us/en/products/sku/241636/...
[1]: https://chipsandcheese.com/p/skymont-intels-e-cores-reach-fo...
I'm team AMD all the way.
I have no faith or expectation anything Intel does will matter.
Maybe 10 years of them being irrelevant will convince them of their ways?
I'd really like to see more chips with 100+gb of eDRAM.
Most things I've run that aren't memory bandwidth limited are already so fast...
erp. 8x. math hard.
I thought I'd find a chart, best I can do is here:
https://www.techpowerup.com/322317/amd-hits-highest-ever-x86...
Intel's revenue (edit) is falling year-on-year, and AMD gaining. Does anyone have a better chart?
That each have 26 execution ports, can retire up to 16 ops per cycle.
The last time I was excited about anything Intel did was the Xeon Phi. And Intel failed spectacularly to follow up on a very good idea.
Intel simply can't innovate.
1 more comments available on Hacker News