I think one of the best outcomes of a competitive CPU market is that the artificial limitations Intel introduced to segment their CPU market are going to have to die.
It's pretty clear that now AMD is speed competitive, people will look at features and this artificial segmentation will push more people to AMD.
Eg, from the article:
You may notice the Silver and Bronze tiers of Xeons are gone, as are the -M SKUs. This leaves the workstation parts crippled, you can’t get more than 768GB per socket on parts meant for heavy memory intensive tasks like CAD, FEA, and modeling large data sets.... AMD’s Epyc on the other hand can put up to 2TB in a single socket, 4TB with 256GB DIMMs versus Intel’s 756GB, an artificially enforced cap so DIMM side is irrelevant.
Also, the Xeon 8180 is $10,000?! I used to spec servers, and I don't remember seeing CPUs anywhere near that price. Is that common in the workstation market?
About a decade ago, the top proc's were 4-6k. They've trended up. 8k was the top for awhile and now 10k+ is the new norm. It's definitely Intel abusing their position, in this case.
Also, the $10k one isn't even the most expensive. The Xeon Platinum 8180M is $13,011.
> I think one of the best outcomes of a competitive CPU market is that the artificial limitations Intel introduced to segment their CPU market are going to have to die.
Actually, for me following at home ... it seams like they have a whole new segmentation scheme. So not killing it, just changing it.
This is a poorly-written article with no benchmarks – so I think it might be worth reserving judgement until we have a better idea of how this all plays out.
It will definitely be great for competition if AMD can become more competitive with Intel's latest chips.
14 and 18 core Xeon-W and i9 processors have only been announced, but are not available yet.
We already know that the Intel 18-core parts will have a base clock of 2.3 GHz vs 3.3 GHz of the one tested above, both parts have 4 memory channels.
The 16-core AMD Threadripper from that benchmark runs at 3.4GHz base clock and has 4 memory channels, the 32-core AMD EPYC runs at 2.2GHz base clock and has 8 memory channels.
-- Educated guesses and extrapolation starts here--
Even ignoring double memory bandwidth AMD is likely to increase any advantage (or catch up in AVX2-heavy workloads) over Intel.
I'm only looking at base clocks, because Intel CPUs are not able to maintain boost clock for any meaningful amount of time, even with water cooling - the package doesn't transfer heat from the die fast enough. AMDs fare quite a bit better, but few people will bother with water cooling their workstations.
I recently was wondering how much the AVX512 helped. I figured it would be pretty noticeable on FP heavy workloads like say Spec.org's CPU2016. I was pretty surprised that the Xeon 5118 ([email protected] with 1 AVX512) gets a FP score of 946. The Xeon 6126 ([email protected] with 2 AVX512) gets a FP score of 1090. So 13% clock rate gets them a 15% better score. For such FP heavy codes I was pretty shocked a second FP unit only gains them a few %.
Not a particularly impressive improvement since the cost to a dual socket node adds $1,000.
You conveniently ignored that the top Xeon-W processor has 28 cores, not 18, and 2.5 GHz, not 2.3 GHz base clock. So your calculation for Intel should be more like 2.8 * 0.76 = 2.13. We are not comparing against i9-7980XE here.
Still, EPYC wins on max memory (2TB per processor) and number of PCI-e lanes (128 vs 48). And of course the price.
Sorry, I got confused by the article. Anyway, should we compare by price or compare best Intel chip to best AMD chip? Because EPYC processors are priced closer to Xeon-W line, but have the number of cores topping Xeon Platinum :)
The comparison is somewhat weird and the space is a bit weird:
- Xeon-w cpus seem to be i9 HEDT just without disabled ECC - everything else (beside a price premium) is the same
- Epyc CPUs are server CPUs and I'm not aware of any desktop motherboards for them, but considering 32 core Epyc has the same TDP as Threadripper they could release a desktop variant (different socket)
- Xeon-W and i9s with more than 10 cores have only been announced, you can't buy them yet
so it's a comparison of parts you can't get your hands on with either parts in a different space or ones that were not announced yet
The benchmarks are out there and well known by now. This article summarizes them fairly: Intel still wins, barely, on single threaded performance. AMD of course ships more cores and more IO for a lower price.
This is an opinionated editorial by someone who knows the market pretty well, and there is a place for that.
> Intel doesn’t want to be put in the position of admitting they are charging significantly more for a new product that runs about half the speed of their cheaper competition.
Unfortunately, people are locked into the Xeon line - so it doesn't matter a ton for now. In the coming years if AMD can keep up, they'll win - over and over time.
It does get them a foot in the door though since most people never considered anything but Xeon. Personally I'll get an AMD the next time I upgrade just for variety; even if Intel would take the performance lead next generation.
(Currently sitting on two E5-2660v2 for reference)
Especially at the pricing of AMD's parts, it'll cost me $400-500 for a Xeon E5-2420v2 to upgrade the E5-2403v2 in my TD340 to something more reasonable clock-speed and core/thread-wise - not to mention I HAVE to buy two of them to get full access to the PCIe lanes and memory capacity of the thing.
I think instead of upgrading this thing any more I'll just sell it and build a new EPYC system once the parts are available outside of OEM configurations. At $600 the EPYC 7281 is a steal - 16c/32t, a full 128 PCIe lanes plus four memory channels (with 2TB of memory supported) on a single socket.
I'm not running workstation hardware, been on an i7-4790K for a few years now, but sometimes constrained on ram at 32gb, and recently went to a 1gb nvme for my primary drive + gtx1080. I will say that even Threadripper is compelling on value, and Epyc equally so. I'll probably go Threadripper for my next system, depending on how the next year or so shape up. Everyone using these systems has different needs, some need raw CPU power, others more memory, some max both. Intel has carved out too many parts to maximize profits and confuse customers. AMD seems to be sized "just right" on their parts, without killing features.
I'm still pretty happy with my ~3yo pc, I did ditch windows as my desktop OS earlier this year. (aside: win10's aggressive advertising in the OS finally irritated me too much, and I'm not the only one).
Dunno, I think talking myself into a Ryzen, but I wanted ECC and not some hand waving that it might work. Intel's coffee lake i7 6 core/12 threads (i7-8700) is looking like it will beat the ryzen on single thread and multithread workloads. At least that's the leaked benchmarks so far.
I'm hoping that the Xeon E3 flavor of the i7 is out shortly afterwards. Should make for a great workstation and guarantees ECC.
I was shopping for ECC threadripper boards and they often don't mention ECC. I'm hoping AMD fixes that and brings out a threadripper down to $350 (with the Intel x-series starts).
It's nice to see that AMD is still around so this day could come. Things looked pretty grim in the Bulldozer era when they were being pounded by Intel on one side and slammed by NVidia on the other.
> There are a few things to note here. First take a look at AMD’s 1S Epyc pricing, specifically the 32C 7551P. It costs $2100 or a mere $101 more than the 28C Skylake-X for 14C more, 1.25TB more addressable memory, and no artificial fusings of bells and whistles.
The math doesn't work out here: 32 cores isn't 14 cores more than 28 cores. I think this was a typo and the author meant to compare with the 18 core Skylake-X, which exists (whereas there is no 28 core model).
Intel wants $1,440 for its 10 core and $1,113 for a 8 core Xeon-W? Is Intel in self-destruction mode? 4 memory channel, less PCI lanes, 8 less cores, but $120 more when compared to Threadripper 1950x?
You really need to hate your hard earned $ with a passion to buy such rubbish with an Intel logo on it.
The author sounds like they have an axe to grind with Intel. That's totally understandable, since Intel really has been soaking the server/workstation market, but it's not appropriate for a comparative article, especially one with no actual benchmarks. Just the facts, ma'am.
Of course Charlie Demerjian has an axe to grind with Intel. He was the reporter who broke the news that Intel threatened Taiwanese motherboard manufacturers to _not_ release motherboards for the then new AMD Athlon 64 CPUs. The manufacturers had to _leak_ this info to him in fear of Intel's retaliation ("Ops we have a shortage of north bridge chipsets, guess you guys aren't getting any chipsets allocation for the next three months.")
He was also one of the first reporters to report Intel's secret paymenst to Dell to not use AMD Athlon CPUs, that ultimately greatly limited AMD's ability to generate revenue to increase its production capacity at its fab.
Believe it or not, Intel acted like a mini-Microsoft back in AMD's Athlon heydays, and that pissed off a lot of people in the industry.
Intel lost a lawsuit to Transmeta for similar nastiness [1] (of course the latter went out of business before the total could be collected).
Charlie holds many grudges; he pissed NVIDIA off enough that they denied him access after which his been dumping dirt on them whenever he can, justified or not.
"Semi-accurate" is an appropriate name. He's sometimes right.
What is it with this author and the word "mere". At first I thought it was legit because the AMD chip was cheaper, then sarcastic because a purely inferior Intel was more expensive then it was used everywhere so now I think the author doesn't know how to use it.
I'm confused. Xeon-W more directly compares with Threadripper. Xeon-SP compares with Epyc. Why is this comparison even a thing?
Also, I need to bring this up every time we have an "AMD is destroying Intel" post: Intel's per-core performance is still higher than AMD, somewhere around 10% depending on the application.
AMD is competitive again, which is great. But I'm still buying Intel because my use-cases demand maximum per-core performance.
You don't need to bring it up, is right there in the article.
There are many use cases, and maybe to you per-core performance is paramount, but he's making the case that for the market Xeon-Ws are aimed at, cores are more important than 10% peak performance, otherwise you wouldn't be buying a 36 core monster to begin with.
Semiaccurate is at best a clickbait hardware site. So this article is to be laughed at. AMD’s new server line is not to be laughed at but intel still commands the edge in cases where AVX is used and in gaming. AMD has become so much more competitive of late but intel can throw sheer cores and with those cores their improved IPC at the issue and enterprises will pay because of name recognition.
Intel's domination in gaming and "low level" workstation is getting tenuous at best, and that's why they need to shake their entire lineup at that level too and doing several line release in a single year (which is hilarously obvious as to how unprepared they were).
Yes, in pure per core performance intel still has a nice edge of roughly 10%, and yes games mostly care about single core perf, and yes if you want to make the best gaming machine money can buy you still go for intel.
But gaming has not been CPU bound for quite some time, and amd's proposition there is not "the same thing, little bit less expensive and little bit less powerful", it's "the same price, little bit less powerful by core but still not cpu bound, and we give you twice the cores". Intel might very well fix that by the end of the year but right now you want a 600 - 1000 € gaming machine there is a very good chance you go for Ryzen.
And of course that's why AMD did that, they did not go crazy with core count to be nice, they did it because that's what they finally needed to get over that "but intel's still ahead and there is no reason to go with amd if you don't need to save fifty bucks".
I mean, they also were able to cheaply increase core count by not stuffing an iGPU into the chip. Outside of HEDT CPU's on the Intel side you have no choice on whether you get HD/Iris/Iris Pro graphics, even if you're going to put a gaming GPU in the system anyway. Meanwhile, the mainstream AM4 platform is based around the assumption you'll be sticking a GPU in already so space on the CPU die is better used for more cores.
What benchmarks show gaming not being CPU bound, everything I have seen says otherwise. 4 - 8 cores clocked as fast as possible make for the best performance.
What I like about AMD is that they pack their chips with features. Low end ones support VT-D and VT-X, AES, etc., whereas Intel likes to segment the market a ton with arbitrary features turned off to prevent higher ticket SKUs from being replaced with lower priced ones.
IPC, clock speed and core count don't tell the whole story any more.
All these latest generation chips boost clock speed ("turbo" mode) the fewer cores there are in use. AMD is far more generous with its turbo modes at multiple tiers (read: higher percentage of top turbo speed at the same active core count) although at a significant TDP cost
Per Phoronix[1] AMD's threadripper does really good against Intel's workstation and HEDT CPUs. How could two more core complexes make it look less onesided. A Ryzen is one core complex, A threadripper is two Ryzen and an Epyc is 4 Ryzen. It is going to be the fastest until we look at Intel's highest end server chips.
Epyc is a stepping above the early Threadripper's and Ryzen. Other errata have been patched, and knowing the influence virtualization has on the server space, I'd hope they'd fix this one as well.
It's pretty clear that now AMD is speed competitive, people will look at features and this artificial segmentation will push more people to AMD.
Eg, from the article:
You may notice the Silver and Bronze tiers of Xeons are gone, as are the -M SKUs. This leaves the workstation parts crippled, you can’t get more than 768GB per socket on parts meant for heavy memory intensive tasks like CAD, FEA, and modeling large data sets.... AMD’s Epyc on the other hand can put up to 2TB in a single socket, 4TB with 256GB DIMMs versus Intel’s 756GB, an artificially enforced cap so DIMM side is irrelevant.
Also, the Xeon 8180 is $10,000?! I used to spec servers, and I don't remember seeing CPUs anywhere near that price. Is that common in the workstation market?