The Intel Core i9-7980XE and Core i9-7960X CPU Review Part 1: Workstation
by Ian Cutress on September 25, 2017 3:01 AM ESTExplaining the Jump to Using HCC Silicon
When Intel makes its enterprise processors, it has historically produced three silicon designs:
- LCC: Low Core Count
- HCC: High Core Count (sometimes called MCC)
- XCC: Extreme Core Count (sometimes called HCC, to confuse)
The idea is that moving from LCC to XCC, the silicon will contain more cores (sometimes more features), and it becomes cost effective to have three different designs rather than one big one and disable parts to meet the range. The size of the LCC silicon is significantly smaller than the XCC silicon, allowing Intel to extract a better production cost per silicon die.
| Skylake-SP Die Sizes (from chip-architect.com) | |||
| Arrangement | Dimensions (mm) |
Die Area (mm2) |
|
| LCC | 3x4 (10-core) | 14.3 x 22.4 | 322 mm2 |
| HCC | 4x5 (18-core) | 21.6 x 22.4 | 484 mm2 |
| XCC | 5x6 (28-core) | 21.6 x 32.3 | 698 mm2 |
In the enterprise space, Intel has each of the three designs throughout its Xeon processor stack, ranging from four-core parts (usually cut down versions of the LCC silicon) all the way up to 28 core parts (using XCC) for this generation. The enterprise platform has more memory channels, support for error correcting and high-density memory, the ability to communicate to multiple processors, and several other RAS (reliability, accessibility, serviceability) features that are prominent for these markets. These are typically disabled for the prosumer platform.
In the past, Intel has only translated the LCC silicon into the prosumer platform. This was driven for a number of reasons.
- Cost: if users needed XCC, they had to pay the extra and Intel would not lose high-end sales.
- Software: Enterprise software is highly optimized for the core count, and systems are built especially for the customer. Prosumer software has to work on all platforms, and is typically not so multi-threaded.
- Performance: Large, multi-core silicon often runs at a low frequency to compensate. This can be suitable for an enterprise environment, but a prosumer environment requires responsiveness and users expect a good interactive experience.
- Platform Integration: Some large silicon might have additional design rules above and beyond the smaller silicon support, typically with power or features. In order to support this, a prosumer platform would require additional engineering/cost or lose flexibility.
So what changed at Intel in order to bring HCC silicon to the HEDT prosumer platform?
The short and shrift answer that many point to is AMD. This year AMD launched its own high-end desktop platform, based on its Ryzen Threadripper processors. With their new high performance core, putting up to 16 of them in a processor for $999 was somewhat unexpected, especially with the processor beating Intel’s top prosumer processors in some (not all) of the key industry benchmarks. The cynical might suggest that Intel had to move to the HCC strategy in order to stay at the top, even if their best processor will cost twice that of AMD.
Of course, transitioning a processor from the enterprise stack to the prosumer platform is not an overnight process, and many analysts have noted that it is likely that Intel has considered this option for several generations: testing it internally at least and looking at the market to decide when (or if) it is a good time to do so. The same analysts point to Intel’s initial lack of specifications aside from core count when these processors were first announced several months ago: specifications that would have historically been narrowed down at that point in previous designs if they were in the original plans. It is likely that the feasibly of introducing the HCC silicon was already in process, but actually moving that silicon to retail was a late addition to counter a threat to Intel’s top spot. That being said, to say Intel had never considered it would perhaps be a jump too far.
The question now becomes if the four areas listed above would all be suitable for prosumers and HEDT users:
- Cost: Moving the 18-core part into the $1999 is unprecedented for a consumer processor, so it will be interesting to see what the uptake will be. This does cut into Intel’s professional product line, where the equivalent processor is nearer $3500, but there are enough ‘cuts’ on the prosumer part for Intel to justify the difference: memory channels (4 vs 6), multi-processor support (1 vs 4), and ECC/RDIMM support (no vs yes). What the consumer platform does get in kind is overclocking support, which the enterprise platform does not.
- Software: Intel introduced its concept of ‘mega-tasking’ with the last generation HEDT platform, designed to encompass users and prosumers that use multiple software packages at once: encoding, streaming, content creation, emulation etc. Its argument now is that even if software cannot fully scale beyond a few cores, a user can either run multiple instances or several different software packages simultaneously without any slow-down. So the solution to this is rather a redefinition of the problem rather than anything else, which could have applied previously as well.
- Performance: Unlike enterprise processors, Intel is pushing the frequency on the new HCC parts for consumers. This translates into a slightly lower base frequency but a much higher turbo frequency, along with support for Turbo Max. In essence, software that requires responsiveness can still take advantage of the high frequency turbo modes, as long as the software is running solo. The disadvantage is going to be in power consumption, which is a topic later in the review.
- Platform Integration: Intel ‘solved’ this by creating one consumer platform suitable for nine processors with three different designs (Kaby Lake-X, Skylake-X LCC and Skylake-X HCC). Both the Kaby Lake-X and Skylake-X parts have different power delivery methods, support different numbers of memory channels, and different numbers of PCIe lanes / IO. When this was first announced, there was substantial commentary that this was making the platform overly complex, and would lead to confusion (it lead to at least one broken processor in our testing).
Each of these areas has either been marked as solved, or redefined out of being issue (even if a user agrees with the redefinition or not).
Facebook
Twitter
RSS
152 Comments
View All Comments
mapesdhs - Monday, September 25, 2017 - link
Ian, thanks for the great review! Very much appreciate the initial focus on productivity tasks, encoding, rendering, etc., instead of games. One thing though, something that's almost always missing from reviews like this (ditto here), how do these CPUs behave for platform stability with max RAM, especially when oc'd?When I started building oc'd X79 systems for prosumers on a budget, they often wanted the max 64GB. This turned out to be more complicated than I'd expected, as reviews and certainly most oc forum "clubs" achieved their wonderful results with only modest amounts of RAM, in the case of X79 typically 16GB. Mbd vendors told me published expectations were never with max RAM in mind, and it was "normal" for a mbd to launch without stable BIOS support for a max RAM config at all (blimey). With 64GB installed (I used two GSkill TridentX/2400 4x8GB kits), it was much harder to achieve what was normally considered a typical oc for a 3930K (mab was the ASUS P9X79 WS, basically an R4E but with PLEX chips and some pro features), especially if one wanted the RAM running at 2133 or 2400. Talking to ASUS, they were very helpful and advised on some BIOS tweaks not mentioned in their usual oc guides to specifically help in cases where all RAM slots were occupied and the density was high, especially a max RAM config. Eventually I was able to get 4.8GHz with 64GB @ 2133. However, with the help of an AE expert (this relates to the lack of ECC I reckon), I was also able to determine that although the system could pass every benchmark I could throw at it (all of toms' CPU tests for that era, all 3DMark, CB, etc.), a large AE render (gobbles 40GB RAM) would result in pixel artefacts in the final render which someone like myself (not an AE user) would never notice, but the AE guy spotted them instantly. This was very interesting to me and not something I've ever seen mentioned in any article, ie. an oc'd consumer PC can be "stable" (benchmarks, Prime95 and all the rest of it), but not correct, ie. the memory is sending back incorrect data, but not in a manner that causes a crash. Dropping the clock to 4.7 resolved the issue. Tests like P95 and 3DMark only test parts of a system; a large AE render hammered the whole lot (storage, CPU, RAM and three GTX 580s).
Thus, could you or will you be able at some point to test how these CPUs/mbds behave with the max 128GB fitted? I suspect you'd find it a very different experience compared to just having 32GB installed, especially under oc'd conditions. It stresses the IMCs so much more.
I note the Gigabyte specs page says the mbd supports up to 512GB with Registered DIMMs; any chance a memory corp could help you test that? Mind you, I suspect that without ECC, the kind of user who would want that much RAM would probably not be interested in such a system anyway (XEON or EPYC much more sensible).
Ian.
peevee - Monday, September 25, 2017 - link
"256 KB per core to 1 MB per core. To compensate for the increase in die area, Intel reduced the size of the size of the L3 from 2.5 MB per core to 1.375 MB per core, keeping the overall L2+L3 constant"You might want to check your calculator.
tygrus - Monday, September 25, 2017 - link
Maybe Intel saw the AMD TR numbers and had to add 10-15% to their expected freqs. Sure, there is some power that goes to the CPU which ends up in RAM et. al. but these are expensive room heaters. Intel marketing bunnies thought 165w looked better thn 180w to fool the customers.eddieobscurant - Monday, September 25, 2017 - link
Wow! Another intel pro review. I was expecting this but having graphs displaying intels perf/$ advantage, just wow , you've really outdone yourselves this time.Of course i wanted to see how long are you gonna keep delaying the gaming benchmarks of intel's core i9 due to mess rearrangement horrid performance. I guess you're expecting game developers to fix what can be fixed. It's been already several months, but on ryzen you were displaying a few issues since day 1.
You tested amd with 2400mhz ram , when you know that performance is affected with anything below 3200mhz.
Several different intel cpus come and go into your graphs only to show that a different intel cpu is better when core i9 lacks in performance and an amd cpu is better.
Didn't even mention the negligent performance difference bettween the 7960x and 7980xe. Just take a look at phoronix review.
Can this site even get any lower? Anands name is the only thing keeping it afloat.
mkaibear - Tuesday, September 26, 2017 - link
Erm, there are five graphs on the performance/$ page, and three of them show AMD with a clear price/$ advantage in everything except the very top end and the very bottom end (and one of the other two is pretty much a tie)....how can you possibly call that a pro-Intel review?
wolfemane - Tuesday, September 26, 2017 - link
And why the heck would you want game reviews on these CPUs anyways? By now we KNOW what the results are gonna be and they won't be astonishing. And more than likely will be under a 7700k. Game benchmarks are utterly worthless for these CPUs and any kind of s surprise by the reader in their lack of overall performance in game is the readers fault for not paying attention to previous reviews.Notmyusualid - Tuesday, September 26, 2017 - link
Sorry to distract gents (and ladies?), and even though I am not a fan of liquid nitrogen, here:http://www.pcgamer.com/overclocked-core-i9-7980xe-...
gagegfg - Tuesday, September 26, 2017 - link
EPYC 7551P vs core i9 790XEThat is the true comparison, or not?
$2000 vs $2000
gagegfg - Tuesday, September 26, 2017 - link
EPYC 7551P vs core i9 7980XEThat is the true comparison, or not?
$2000 vs $2000
IGTrading - Tuesday, September 26, 2017 - link
That's a perfectly valid comparison with the exception of the fact that Intel's X299 platform will look completely handicapped next to AMD's EPYC based solution and it will have just half of the computational power.