Since the release of Ryzen, pressure has been mounting in the mid-range segment ($200–$300). With Zen 2, things got even harder for Intel as AMD introduced several offerings in this price range, many just $20 apart. All these parts are six-core with twelve threads, whereas Intel was bringing 6-core, 6-thread CPUs without HyperThreading to the fight. To achieve any sort of competitiveness, Intel had to rethink their policy of reserving HyperThreading to their top SKUs only. This is why the Core i5-10400F in this review has HyperThreading enabled. Unlike the Core i5-10600K, for which Intel pushed the TDP from 65 W to 95 W, the Core i5-10400F remains at 65 W, more on that later. Last but not least, the full L3 cache is available now, 12 MB instead of 9 MB, to make up lost ground against the 32 MB cache AMD is offering on Ryzen.

Averaged over our mix of single/low and multi-threaded applications, the Core i5-10400F ends up pretty much exactly where you would expect it for its MHz ratings. It is clocked considerably lower than the Core i5-10600K (by 11.6%, 100/4300*4800), and average performance in our tests is 11.6% lower, too, a surprising coincidence. I have to admit it felt a lot slower than the Core 10600K when I looked through the benchmarks, but that seems to be a side effect of there being a lot of entries in the bar charts for that range, which makes the performance gap look bigger. Compared to AMD's Ryzen 3600 and 3600X, the 10400F is slightly slower, by 4% and 6% respectively. It depends very much on the workload though, especially tasks that are easy to parallelize, like rendering, are AMD's strongest suit, and Intel has a clear lead in single and low-threaded apps, which are relevant to the majority of consumers today. Performance gains against last generation's Core i5-9400F are impressive because of the added cores and threads; the 10400F enjoys a 15% performance advantage—at similar cost.

For gaming, the Core i5-10400F is a clear winner against AMD. It is faster than any AMD CPU at all resolutions—even the Ryzen 9 3900X is beat by 3%. Against Intel's own lineup, the Core i5 does very well too. It trades blows with last generation's Core i7 and Core i9 models. The Core i9-10900K is merely 5% faster. All these differences are small, barely relevant. As you go up in resolution, the differences shrink even further as the bottleneck moves from the CPU to the GPU. A 6c/12t design is also more future-proof than a 4c/8t CPU, even though I'm personally not a friend of future-proofing as you'll tend to overspend in the long run. As always, if you are mostly a gamer, definitely try to invest as much money as possible in the GPU, as that's what will be driving your framerates. The Core i5-10400F is a great alternative to the i5-10600K—much more affordable, yet with very similar gaming performance, it leaves you with more money for your graphics card.

Just like in all our CPU reviews, we measured the maximum boost clocks of the Core i5-10400F, and I have to say I'm a little bit disappointed. While the CPU is marketed with 4.3 GHz boost clock, sure, "up to," I know, it really doesn't hit that frequency all that often, no matter the load. I have seen individual cores reach 4.3 GHz for a split second, but not nearly for long enough to declare it anything other than a blip. There's probably some sort of observer effect at play, too, due to monitoring software creating a little bit of extra load on the processor. Obviously, the CPU can handle 4.3 GHz single-core stable all day or Intel wouldn't have binned it to be a 10400F, so it seems there's some headroom for Intel to refine their algorithms. AMD faced serious social media drama over their Zen 2 processors not reaching promised boost clocks, let's see if Intel users will be just as demanding.

Power efficiency of the Core i5-10400F is much better than with other Comet Lake processors we've tested. One reason is that clock frequencies are much lower, by roughly 500 MHz compared to the Core 10600K and almost 1 GHz compared to the Core 10900K. This move alone brings with it significant power savings because the silicon is now operating closer to its maximum efficiency point. Another factor could be that our previous two reviews (Core 10600K and Core 10900K) were both based on the 10-core silicon, whereas our Core 10400F SRH3D is based on the 6-core design that's possibly even a reused Coffee Lake die. Single-threaded energy efficiency is better than with other Comet Lakes we've tested and slightly worse than what we saw on Coffee Lake. At 1T, energy efficiency is much better than AMD, too. This changes when looking at multi-threaded workloads. Here, Zen 2 shines; the Core i5-10400F is similar to 9th generation processors, roughly 10% more efficient than the other Comet Lake models in our test group. What really surprises me is that Intel did not include Turbo Boost Max 3.0 and Thermal Velocity Boost on the Core i5-10400F. Turbo Boost Max 3.0 is reserved for the Core i7-10700 series and Core i9-10900 series, and TVB only for the Core i9-10900. Yet again we see the typical Intel segmentation pattern here—AMD does the opposite as all their CPUs have all the features, and those little things add up. I have no idea why Intel would not include their two most advanced turbo modes on the Core i5-10400F; it wouldn't have cost them anything, yet provides free performance.

As denoted by the lack of the "K" suffix, the Core i5-10400F has no support for multiplier-based overclocking, but that's also why it's a little bit more affordable. AMD includes unlocked multipliers with all their processors, and that's the way to go as it's a strong driver for sales because it dangles "free extra performance" in front of potential customers. With Z490, Intel has introduced a new clock tree design that separates the PCIe clock frequencies from the BCLK that drives CPU frequency. I had hopes that this could mean fresh wind for overclocking non-K processors, but unfortunately, it didn't work out. All recent Intel processors have a frequency counter inside that measures the actual BCLK frequency provided by the motherboard. After the launch of 6th generation Intel Skylake processors, ASRock released a BIOS update that enabled BCLK overclocking without artificial limits. This lead to massive unhappiness at Intel because they felt their bottom line threatened—people could now buy non-K CPUs and overclock them up to the physical capability of the silicon without paying extra. For future generations, Intel added code to their ME Management engine to validate BCLK frequency during POST; if above 103 MHz, startup will be refused. Of course, I tested this, and the highest BCLK frequency I could boot and bench at was 102.9 MHz. Anything higher, even 102.95, would significantly increase chances of POST hanging because the CPU's internal measurement isn't 100% accurate.

BCLK overclocking yields around 2–3 % in real-life performance, which is probably not worth the trouble. AMD's Ryzen 3600 series CPUs have unlocked multipliers, but overclocking those doesn't yield any significant gains either, so don't fall for the "unlocked multiplier" lure, you'll be disappointed. In this article, we also tested "stock" performance vs. "Max Turbo", which is the processor running with all its power limits removed. On the Core i5-10400F, these are PL1 = 65 W and PL2 = 134 W. If you scan through our charts, you'll see that there's almost no difference between both results. The reason is that the Core i5-10400F with its 6c/12t configuration and relatively low clock speeds will rarely exceed 65 W in typical applications, and definitely won't hit 134 W unless you're running a stress test. This is good news for motherboard choice, as you don't need a board with extra strong VRM, or other fancy tweaking ability. Just pair the CPU with the cheapest board you can find and you'll be good to go.

With a retail price of around $160, the Core i5-10400F is priced very competitively. It's a little bit cheaper than the AMD Ryzen 5 3600 ($170) and much more affordable than the Ryzen 5 3600X ($205). If you spend most of your time waiting for highly threaded application results, like rendering, a Ryzen will be a slightly better choice. If you're gaming all day, then the 10400F has the upper hand. The differences are quite small either way. Just like AMD, Intel does include a stock cooler with the Core i5-10400F, which definitely helps. The problem is that right now only Z490 motherboards are available for the LGA1200 platform, and these are expensive. While many new motherboards include features like USB-C 20 Gbps and 2.5 Gbps Ethernet, I still feel motherboard pricing has to come down another $20–$40 to be competitive with AMD. At the moment, the platform cost of Comet Lake is too high, AMD definitely has a more cost optimized ecosystem here.

For us, AMD and Intel being neck-to-neck is a great thing as it will result in more innovation and lower prices overall. Ask me now whether I would buy the Ryzen 5 3600 or Core i5-10400F and I wouldn't be able to decide. They are close enough, and the motherboard ecosystem will change soon, too, with the release of AMD B550 and Intel H410. After having thought about it for a little bit longer I feel like I would personally even opt for a Ryzen 3 3300X. It is fast enough for games (I'm not playing 1080p, but 2560x1600), can handle all work-related tasks with ease (I don't do encoding or rendering), and at $120, it's simply too affordable to ignore. I've ordered six more Comet Lake CPUs this morning, so expect more reviews next week.