Conclusion: Variables Maketh the CPU

Suffice to say, Intel’s new 12th Generation Core family has added more variables to the mix than a traditional processor launch. The combination of different types of core, coming at a time where a new operating system has just been launched, but also at the inflection point of a new type of memory. Let’s break down some of the differences we’ve observed in our testing.

When we compare Windows 10 to Windows 11, purely on throughput benchmarks, we don’t find much difference. There are a few spots where Windows 11 has a slight advantage in multi-threaded workloads, but this comes down to how new threads are spawned between the performance cores and the efficiency cores. Intel stated that the performance between the two, at least for CPU workloads, should be within a few percentage points, stating that Windows 11 should have lower run-to-run variance. We can corroborate this in our testing. Windows 10 also had some scheduling issues with low priority threads, which we expect to be ironed out with updates.

Comparing the new DDR5 to the old DDR4 is a different story, as the new memory standard offers a substantial uplift when it comes to memory bandwidth. As we saw recently with the M1 Max review, sometimes memory bandwidth can be a limiting factor. In our testing, DDR5 had a minor lead in single threaded tests but in a lot of multithreaded tests, the lead was significant. For real world, we had examples of +14% in Agisoft, +10% in NAMD, +8% in V-Ray, +10% in Handbrake, and +20% in WinRAR. In SPEC2017, we saw a couple of single threaded workloads get +15% improvements over DDR5, but in multi-threaded this was up to +40% on average, or more than 40% in specific tests. This also comes down to the doubled memory channels (4x32-bit vs 2x64-bit) which can be better utilized on top of the bandwidth increases.

Now comparing the P-core to the E-core, and it’s a story of how the E-core individually can perform on par with a Skylake core. Having eight extra Skylake-class cores is nothing to be sniffed at. In a lot of tests the E-core is half the performance of the P-core, but the P-core is itself is now the market leader in performance. The Golden Cove core inside Alder Lake has reclaimed the single-threaded performance crown with an uplift in SPEC of 18-20%, which is in line with Intel’s 19% claim. This puts it ahead of Apple’s M1 Max or 6% (int) and 16% (fp) ahead of AMD’s Zen 3 core.

The Core i9-12900K

Combining fast P-cores, Skylake-class E-cores, and DDR5 into one package means that Intel has certainly jumped from behind the competition to in front of it, or at least in the mix. When you have your operating system set up just right, and no issues with schedulers, it outperforms AMD’s offering when single core performance matters, and in multi-threaded workloads, it does tend to sit somewhere between a 5900X and a 5950X.

(5-1c) Handbrake 1.3.2, 1080p30 H264 to 4K60 HEVC

It’s important to note that in some tests, the Core i9-12900K does win outright. It’s at this point we should consider how much is core related vs standards related: DDR5 has produced somewhat of an uplift, and the competition is expected to claw some of that back when they introduce it, but those products are expected more towards the latter half of 2022. For users with those specific workloads today, and willing to pay the DDR5 early adopter tax, Alder Lake can provide performance uplifts right now.

(0-0) Peak Power

Power is an interesting topic, and although our peak power numbers when all cores were loaded were above the 241W Turbo power on the box, in real world workloads it didn’t tend to go that high. The P-cores alone on the chip matched the power consumption of Intel’s 11th Generation in AVX2 workloads, but adding in the E-cores does put it over the previous generation. I’m not entirely sure what that says about Intel’s 7 manufacturing process compared to the 10SF used before. A lot of the performance gains here appear to come from IPC and DDR5, and that doesn’t seem to have come with performance per watt gains on the P-cores. It means that Intel is still losing on power efficiency at load compared to the competition.

(2-2) 3D Particle Movement v2.1 (Peak AVX)

I have to say a side word about AVX-512 support, because we found it. If you’re prepared to disable the E-cores, and use specific motherboards, it works. After Intel spent time saying it was fused off, we dug into the story and found it still works for those that need it. It’s going to be interesting to hear how this feature will be discussed by Intel in future.

Overall though, it’s no denying that Intel is now in the thick of it, or if I were to argue, the market leader. The nuances of the hybrid architecture are still nascent, so it will take time to discover where benefits will come, especially when we get to the laptop variants of Alder Lake. At a retail price of around $650, the Core i9-12900K ends up being competitive between the two Ryzen 9 processors, each with their good points. The only serious downside for Intel though is cost of switching to DDR5, and users learning Windows 11. That’s not necessarily on Intel, but it’s a few more hoops than we regularly jump through.

 

Gaming Performance: Intel vs AMD
Comments Locked

474 Comments

View All Comments

  • mode_13h - Sunday, November 7, 2021 - link

    I'm not even going to say this is a bad idea. The problem is that it's a big change and Intel normally prepares the software developer community for big new ISA extensions a year+ in advance!

    Again, what you're talking about is an ABI change, which is a big deal. Not only that, but to require code to handle dynamically switching between AVX2 and AVX-512 paths means that it can't use different datastructures for each codepath. It even breaks the task pre-emption model, since there need to be some limitations on where the code needs to have all its 512-bit registers flushed so it can handle switching to the AVX2 codepath (or vice versa).

    This adds a lot of complexity to the software, and places a greater testing burden on software developers. All for (so far) one CPU. It just seems a bit much, and I'm sure a lot of software companies would just decide not to touch AVX-512 until things settle down.
  • GeoffreyA - Sunday, November 7, 2021 - link

    My view on this topic is that Intel made a sound decision disabling AVX512. Some of the comments are framing it as if they made a mistake, because the tech community discovered it was still there, but I don't see any problem. Only, the wording was at fault, this controversial "fused off" statement. And actually, the board makers are at fault, too, enabling a hidden feature and causing more confusion.

    On the question of whether it's desirable, allowing one core with the instructions and another without, would've been a recipe for disaster---and that, too, for heaven knows what gain. The simplest approach was bringing both cores onto the same footing. Indeed, I think this whole P/E paradigm is worthless, adding complexity for minimal gain.
  • Oxford Guy - Monday, November 8, 2021 - link

    ‘Intel made a sound decision disabling AVX512’

    That’s not what happened.
  • O-o-o-O - Sunday, November 7, 2021 - link

    Really? Our tech guys tried out Xeon Phi but couldn't make use of it. Years later, Xeon Phi was abruptly discontinued due to lack of demand. GPGPUs are much easier to handle.

    Yeah, coding cost and risks aside, it's interesting to see complex work of art in the modern CPU. But I'd rather wish for expansion of GPU support (like shared memory and higher band-width).
  • kwohlt - Sunday, November 7, 2021 - link

    My understanding is that Raptor Lake's change is replacing Golden Cover P cores with Raptor Cove P cores, doubling Gracemont E-Cores per SKU, and using the same Intel 7 process. Granted, it's all leaks at this point, but with Gracemont being reused for Raptor Lake, I don't expect AVX-512 next year either.
  • mode_13h - Monday, November 8, 2021 - link

    > Raptor Lake's change is ... doubling Gracemont E-Cores ... using the same Intel 7 process.

    I was merely speculating that this *might* just be a transient problem. If they're using the same process node for Raptor Lake, which seems very plausible, then it's understandable if they don't want to increase the size or complexity of their E-cores.

    However, there's some precedent, in the form of Knights Landing, where Intel bolted on dual AVX-512 pipelines + SMT4 to a Silvermont Atom core. And with a more mature Intel 7 node, perhaps the yield will support the additional area needed for just a single pipe + 512-bit registers. And let's not forget how Intel increased the width of Goldmont, yet simply referred to it as Goldmont+.

    So, maybe Raptor Lake will use Gracemont+ cores that are augmented with AVX-512. We can hope.
  • GURU7OF9 - Saturday, November 6, 2021 - link

    The is by far the best review I have read so far.

    A great comparison I would love to see just out of curiouslty would be to see P core only benchmarks and then e core only benchmarks! We could gain a much better understanding of the capabilities and performance of both .
    This would bring a little bit of familiarity back to benchmarking .
  • nunya112 - Saturday, November 6, 2021 - link

    the only info provided was its on intels new process 7 node. what does that mean? are they using TSMC and at 7nm? or did they finally crack 7nm at Intel?
  • mode_13h - Sunday, November 7, 2021 - link

    "Intel 7" is the process node formerly known as "10 nm ESF" (Enhanced SuperFin), which is the 4th generation 10 nm process, counting by the revisions they've introduced between the different products based on it. They like to pretend that Cannon Lake didn't happen, but that's why Ice Lake was actually 10 nm+ (2nd gen).

    They rebranded 10 nm ESF as "Intel 7" for marketing reasons, as explained here:

    https://www.anandtech.com/show/16823/intel-acceler...
  • Hossein - Sunday, November 7, 2021 - link

    It's funny that most reviewers are conveniently silent about the fact that there are quite a 'few' games which are incompatible AL.

Log in

Don't have an account? Sign up now