Wednesday, January 30th 2019
Intel Xeon W-3175X 28-core Processor Now Available at $2,999
The Intel Xeon W-3175X processor is available today. This unlocked 28-core workstation powerhouse is built for select, highly-threaded and computing-intensive applications such as architectural and industrial design and professional content creation. Built for handling heavily threaded applications and tasks, the Intel Xeon W-3175X processor delivers uncompromising single- and all-core world-class performance for the most advanced professional creators and their demanding workloads.
With the most cores and threads, CPU PCIe lanes, and memory capacity of any Intel desktop processor, the Intel Xeon W-3175X processor has the features that matter for massive mega-tasking projects such as film editing and 3D rendering. Other key features and capabilities:
Intel Xeon W-3175X:
- Intel Mesh Architecture, which delivers low latency and high data bandwidth between CPU cores, cache, memory and I/O while increasing the number of cores per processor - a critical need for the demanding, highly-threaded workloads of creators and experts.
- Intel Extreme Tuning Utility, a precision toolset that helps experienced overclockers optimize their experience with unlocked processors.
- Intel Extreme Memory Profile, which simplifies the overclockingexperience by removing the guesswork of memory overclocking.
- Intel Advanced Vector Extensions 512 (Intel AVX-512) ratio offset and memory controller trim voltage control that allow for optimization of overclocking frequencies regardless of SSE or AVX workloads, and allow maximization of memory overclocking.
- Intel Turbo Boost Technology 2.0 that delivers frequencies up to 4.3 GHz.
- Up to 68 platform PCIe lanes, 38.5 MB Intel Smart Cache, 6-channel DDR4 memory support with up to 512 GB at 2666 MHz, and ECC and standard RAS support power peripherals and high-speed tools.
- Intel C621 chipset based systems designed to support the Intel Xeon W-3175X processor allow professional content creators to achieve a new level of performance.
- Asetek 690LX-PN all-in-one liquid cooler, a custom created solution sold separately by Asetek, helps ensure the processor runs smoothly at both stock settings and while overclocking.
- With the Intel Xeon W-3175 processor, build the "Infiltrator Demo" in Unreal Engine up to 1.52x faster compared with the Intel Core i9-9980XE processor.
Intel Xeon W-3175X:
- Base Clock Speed (GHz): 3.10
- Intel Turbo Boost Technology 2.0 Maximum Single Core Turbo Frequency (GHz): 4.3
- Cores/Threads: 28/56
- TDP: 255W
- Intel Smart Cache: 38.5 MB
- Unlocked: Yes
- Platform PCIE Lanes: Up to 68
- Memory Support: Six Channels, DDR4-2666
- Standard RAS Support: Yes
- ECC Support: Yes
- RCP Pricing (USD 1K): $2,999 (per unit in 1000-unit tray quantities)
58 Comments on Intel Xeon W-3175X 28-core Processor Now Available at $2,999
Any news on consumer motherboard actually & pricing?
Saw a couple of those cases from CES, Full Tower+ sizes here we go.
Gigabyte is MIA at the moment. Apparently they still don't know if they'll release it or not.
I still wonder why Intel bothered so late, it's only a few months until they launch Cooper Lake-SP, which will be on the same platform as Ice Lake-SP(2020).W-3175X beats 2990WX in most tests with a good margin. This doesn't mean it's worth the money though, that really comes down to the workload.
Unlike typical mainstream PC usage, workstation performance is very workload specific. Heavy loads on this many cores also helps emphasize the architectural differences between the products. You will find a number of benchmarks where i9-9980XE (18-core) beats 2990WX (32-core), so it's not as simple as product A is better than B.
Also TR4 isnt really server dedicated platform, its just platform that also can be server (which also makes it ridiculously useful for multi-purpose PC :D). Something that cant be said about 2066.
The different architectures do have different strengths and weaknesses though. AMD does have generally more cores per price, and for certain workloads that can scale better. In other cases, especially workloads which are challenging for the front-end of the CPU core, Intel scales generally much better.
It is worth mentioning though, that as new CPUs progress with higher core counts, IPC will actually become even more important. Scaling with multithreading is very workload specific. If the workload can be divided into x independent chunks and synchronization is not important, then scaling with more cores can be nearly linear. But with mixed loads, more typical for a utility like a CAD or photo editor, we usually see diminishing returns due to increasing thread overhead. Higher IPC will help offset some of this overhead.
Core to core, zen is arguably superior to skylake (now waiting for intel defence force). It's physically smaller on a less dense process, the same or more efficient on a much less efficient process and has almost the same IPC, with superior SMT due to a wider core design and FPU. Intel's branch predictor and front end is a bit better and their architecture has quicker Inter-core communication, lower latency and same for imc.
But zen scales far better, is cheaper and more modular, and zen2 addresses the front end bottleneck.
Honestly this 28c room heater is DOA sorry
www.techpowerup.com/forums/threads/building-a-2990wx-system.246902/
As for the price: this CPU is made for really expensive workstations. We're talking 10+ thousand USD. So the premium for this over TR isn't that huge in the total cost.
Moreover, it should be cheaper for OEMs to make workstations with Intel CPUs than with AMD alternatives, so some of the extra cost will be mitigated anyway.
And even if final workstations end up more expensive, they should still be acceptable for clients. In some scenarios this CPU beats 2990WX by a really wide margin.
Take a look here....
Some quotes from Anandtech page I found interesting -
At 4.3 GHz, we were hitting over 500W peak load (confirmed by wall meter), which is the limit of the cooling setup provided. Compared to the 4.0 GHz result, we calculated that the CPU actually used 17% more power overall to get a 7% increase in performance.
Here the power is overall a bit lower, but we can see that the score isn’t rising much at 4.0 GHz, again due to our CPU temperature sensor showing 110ºC very easily. In this instance, the power consumption between 3.9 GHz and 3.6 GHz increased by 14%, while the score rose 10%.
Intel sent an EKWB Phoenix cooler which is rated for much higher power consumption, but arrived too late for our testing. We’re planning on doing an overclocking review, so this should help. But what our results show is that when Intel showed that 5.0 GHz demonstration using a water chiller they really did need it. Users might look into investing in one themselves if they want this chip.
And then the topping on the cake...
But What About That 5.0 GHz? How Much Power?
We took some of our benchmark values for power and frequency, extrapolated them with a power curve, and we estimate that at 5.0 GHz, this chip is likely to be drawing in excess of 900W, perhaps as high as 1200W. Yes, Intel really did need that 1700W water chiller.
But in all seriousness, if you have a 28 core CPU, don't you expect to have a high power draw?? Maybe it's just me but common sense to me is - more cores, more speed, more heat....
As Anandtech points out, it's more of a case of a non held back performance CPU... Question is, who would buy one? :)
Yea there is optimized for Intel or to be specific "crippled for AMD". There is so many cases where AMD and Intel perform about same but in some other cases AMD is suddenly horrible and Intel isnt. And its not core count or core speed either. Not talking about AMD losing some 20%, where its expected, since one AMD core at 4GHz isnt equiv of one Intel core at 4GHz, but cases where AMD suddenly has 50% less performance for no reason.
How about letting that go, and taking a fresh approach to this, considering we are all consumers first and foremost. Has it occurred to you that people switch 'sides' all the time, too? Look around a bit and you will see. What's more interesting here is the reasoning behind those switches. As is the reasoning behind favoring one brand or the other. Who knows, maybe at some point we'll learn something.
I've an Intel 3570K with R9 390 for only gaming and a newer workstation 2950X with Vega64. I can game on the newer 2950X system, but choose not to waste resources installing games to it and limit it to software dev, 3D modeling, video encoding. To do my required non gaming workloads on a comparable Intel system, I'd have to spend at least double on an equivalent Intel system.
Is Intel evil because of this and I'm an AMD fanboi ? No! - Intel just needs to compete better to win my dollars to address my multi-core workloads I do !
Perhaps for other people, Intel HEDT solutions make more sense for the niche workloads they need to run - however I suspect that performance/dollar (non-gaming) for Threadripper represent better value.
Other family members have various Intel quadcore with nVidia graphics cards that I maintain for them.
REX, 2000$.And that's the only one available.
Especially given the fact per the Anandtech article I just finished reading Asus only seemed to have gotten an order for 500 of these.
You know... thinking about it, it still seems oddly cheap, hard to recoup any costs spread over just 500 units!?