With their 11th generation processors, Intel is finally introducing a new architecture to replace the Skylake arch that has been dragged along since forever. At the same time, they're adding support for PCI-Express 4.0 and new instructions, like DLBoost for Deep Learning and AVX-512 for SIMD calculations. At just $390, the Core i7-11700KF is a very reasonably priced 8-core option. Thanks to the K suffix it also supports overclocking, yet lacks integrated graphics.

With our Rocket Lake reviews, we're introducing our new CPU Test Suite comprised of 38 (!) benchmarks that cover the whole spectrum of workloads, from consumer to scientific and content creation to enterprise. The applications tested are a healthy mix of single-threaded, lightly-threaded, and fully multi-threaded workloads—just like you would encounter in real life. Averaged over all these tests, the eight-core Core i7-11700KF beats last generation's ten-core Core i9-10900K flagship. That is a huge achievement and shows that the new Cypress Cove architecture offers significant improvements. Another interesting match-up is that the i7-11700KF ties the AMD Ryzen 9 3900X, a 12-core design. AMD hasn't been idle though, and their new Zen 3 CPU cores are much faster, with the i7-1700KF's application performance roughly in the middle, between the Ryzen 5 5600X and Ryzen 7 5800X. This generation's Intel flagship, the Core i9-11900K, is only marginally faster than the i7-11700KF because it is 8-core, too, with the same TDP, and it only has a clock speed advantage, but is much more expensive.

Rocket Lake introduces a new memory controller, which has the ability to run in two modes: Gear 1 has the memory controller clock match memory frequency and Gear 2 has the memory controller clock equal half the memory frequency. The problem is that Gear 1 is only officially supported until 2933 MHz—a laughable speed given today's standards. There's some headroom in Gear 1, but not nearly enough to run memory speeds that we enthusiasts care about. My i7-11700KF retail CPU couldn't even run DDR4-3733 in Gear 1, so I had to settle for DDR4-3600 in Gear 1 or pick Gear 2. On a 125 W CPU, Gear 1 is the better option; on lower powered 65 W CPUs, the opposite is often true. I really keep wondering how this memory controller design got approved at Intel. It's a huge step backwards and complicates things for no apparent reason. Don't get me wrong, if Gear 1 were able to achieve something like 3800 MHz consistently and Gear 2 would be needed for 5000 MHz+, it would all be good, but 2933 MHz? C'mon.

Overall gaming performance is excellent on the Core i7-11700KF. Differences are very small, a few percent, really nothing worth worrying about. What is surprising is that the last-generation Comet Lake CPUs are clearly faster at gaming than the new Cypress Cove design. While the differences are tiny, I'm yet again surprised: didn't anybody at Intel test gaming performance? AMD, on the other hand, has made huge architectural improvements with Zen 3, improving both application performance and gaming. Do take a hard look at our gaming benchmarks and go down the performance rankings a bit. You don't need a high-powered, high-core-count CPU to play games—the graphics card is almost always the bottleneck. That's why I strongly suggest you consider options like the 10400F/11400F/5600X; these are fast enough for gaming, and every dollar saved on the CPU brings you a bit closer to maybe being able to afford one of the new graphics cards.

What could bring big wins for Intel is the newfound love for AVX512 and DLBoost—extensions that have been available for years, but never made it to the desktop. At this time, software support for either of those instruction sets is extremely limited, and they are not useful. I am convinced that they can offer tangible benefits once adoption rates go up, though. Remember AVX—everybody said it's a useless technology that's not needed as we already have SSE; today, a lot of apps and games use AVX, also owing to excellent compiler support. Using these new instructions is often as simple as checking a box that tells the compiler it may optimize with AVX instructions—that's it. The hard work will be done by the compiler; you don't have to mess with hand-coded assembly instructions. Today, all this doesn't matter as consumers won't need AVX-512 for a couple of years at least. That said, Rocket Lake can be a cost-effective option for researchers and industry professionals who want to use these new instructions to speed up their calculations, but don't want to pay up for the expensive Xeons.

Just like their other recent desktop processors, Rocket Lake is fabricated on Intel's 14 nanometer process. The company does have a working 10 nm process, which is in production for Ice Lake, and everybody is wondering why they didn't use it for Rocket Lake. Silicon manufacturing has extremely long lead times, so I suspect they decided to continue using 14 nm well before AMD's Zen 3 came out to make sure their older fabs continue to see full use. The excellent yields on the mature 14 nm process certainly played a role in this business decision, too. The drawback of this aging technology is that power consumption is much higher than what AMD offers.

Energy efficiency has definitely suffered with Rocket Lake even compared to Comet Lake, which wasn't impressive either. Looking at our power consumption results, we have a total energy of 16.5 kJ to complete one Cinebench run for the i7-11700KF, against 10 kJ for the Ryzen 5800X and 7.8 kJ for the Ryzen 5900X—these are HUGE differences. AMD's Ryzen 9 5950X is almost a third of that, with 6.4 kJ. This "total energy consumed for Cinebench" metric is important because it takes the processor's speed into account, not only momentary power usage in watts.

While the Core i9-11900K has received Intel's full arsenal of boosting algorithms to increase the processor's clock frequencies, the Core i7-11700KF only gets Turbo Boost 2.0 and 3.0. Unlike other Rocket Lake CPUs, Turbo Boost 3.0 doesn't really do much for the i7-11700KF. It promises a maximum boost of 5.0 GHz—our testing shows that the CPU runs 4.9 GHz most of the time, even with low-threaded loads applied. With some light loads, I can see blips of 5.0 GHz on some cores, but it's not sustained. In all my other Rocket Lake reviews, the various turbo boosts worked perfectly fine and yielded the expected results. Maybe it's just my sample, but this reminds me a bit of Comet Lake and some Ryzens where promised boost clocks were barely or never reached. I also wish Intel would have included their more advanced boost technologies on lesser SKUs—it could be enough to make a difference on top of the psychological value of "Intel giving it all they've got."

Rocket Lake finally brings with it PCI-Express 4.0 support, which is an essential capability gamers demand these days even though it doesn't make that much of a difference, neither for graphics nor storage. With the much more modern platform, AMD is obviously capitalizing on Intel's shortcomings, even if it's just for marketing. While many pieces of the Rocket Lake puzzle are now running at PCI-Express 4.0, the chipset still puts out Gen 3 lanes, which means you're limited to one PCIe Gen 4 M.2 NVMe SSD; the other slots support Gen 3 only. This is certainly not a dealbreaker, but it's still worth mentioning.

What really displeases me is how tacked-together and unfinished the whole Rocket Lake platform feels. The BIOSes have numerous bugs that are completely obvious to anyone using them for more than 10 minutes. Maybe this is not Intel's fault, but since AMD introduced AGESA, a common-base software stack, things have gotten much better for the red team. POST times have always been good with Intel, but I'm now sometimes sitting at A2 (VGA) for 20 seconds with an occasional double boot when changing a BIOS setting, which we criticized AMD for in the past. This whole experience reminds me of the first generation of Ryzen.

If you've read this far, you'd think the Core i7-11700KF is a terrible CPU—it's not, there's only bad pricing. On the contrary, Intel has priced the i7-11700KF quite competitively, it's available at just $390, which is $60 cheaper than the Ryzen 7 5800X and just $40 more than the Ryzen 5 5600X. Since they gained a leadership position, AMD has raised their prices quite a bit and only offers expensive "X" models for the DIY market, while OEMs get the more cost-effective non-X SKUs. At $390, I feel like I'd be willing to overlook the increased energy usage of the i7-11700KF compared to the Ryzens if I wanted higher application performance than the 5600X, or spend less than on the 5800X. For gaming, all three of those CPUs will be awesome choices, with only minimal FPS differences. If gaming is your focus, do check out our Core i5-11400F review, and also consider the 10400F, if the price is right.

Compared to the Core i9-11900K, which Intel wants $550 for, I'd pick the Core i7-11700KF any day of the week, and invest the savings in a better graphics card. Last but not least, an ultra-immensely popular option these days is the Core i9-10850K, which is discounted to only $385 and offers performance comparable to the i9-10900K, making it a faster gaming CPU than both the i9-11900K and i7-11700KF. In applications, the 10-core i9-10850K will also shine, especially when you use a ton of threads. In lightly threaded apps, the i7-11700KF has the upper hand. Since multiplier-based overclocking wasn't that impressive on the i7-11700KF, and you can get pretty much the same results by increasing just the power limit, the locked Core i7-11700F is a great option, too. Once availability improves, it'll be only slightly more expensive than $300, but you must absolutely raise its power limit from 65 W to at least 125 W, better 180 W+.