The Intel Kaby Lake-X i7 7740X and i5 7640X Review: The New Single-Threaded Champion, OC to 5GHz
by Ian Cutress on July 24, 2017 8:30 AM EST- Posted in
- CPUs
- Intel
- Kaby Lake
- X299
- Basin Falls
- Kaby Lake-X
- i7-7740X
- i5-7640X
Power Consumption
With the two Kaby Lake-X processors, the main comparison for power consumption is going to be against their Kaby Lake-S counterparts on the mainstream platform. For an extra 100 MHz, Intel is moving from a 95W TDP to a 112W TDP, indicating that these processors will need better cooling – but when we tested KBL-S we were rather surprised by the power, especially on the Core i5.
The Core i5-7640X almost perfectly matches up with its Core i5 counterparts, even beating the Core i5-7600K in every situation but especially at load. For users who want to overclock near to the official TDP, it looks like there’s some headroom (other circumstances permitting). The Core i7-7740X is noticeably more power hungry at full load, hitting 79W, but lower than the Core i7-7700K at 87W.
Overclocking to 5.0 GHz
One of the elements promoted at the launch of Basin Falls was overclockability: Intel ran a private invite-only extreme overclocking workshop with sub-zero coolants that week, which was then followed by ASUS and GIGABYTE both stepping it up another gear by using liquid helium rather than liquid nitrogen. For some of these chips, the colder the better, and liquid helium is going to be limit for cooling (and cost) as there is inherent difficulty in using sub-zero coolants for records.
Those sub-zero events for records are just on single benchmarks, or single frequency records, and are not meant to be run at those speeds all day every day. This makes the air and water overclocking potential of the CPUs more important for regular users and enthusiasts.
When Kaby Lake-S was launched, we saw good overclocking performance on the Core i7-7700K: our set of 4.2 GHz base processors reached a range between 4.6 to 4.8 GHz, arguably giving a 8-15% overclock and given some of the numbers we had seen in the wild, the consistency between the processors meant that most of the 7700K CPUs could reach that amount. One of the reasons for such good performance on these CPUs was that Intel had incorporated a new feature for overclocking, the AVX offset. This allows overclockers to reduce the frequency that AVX code runs at, as this can be a limiting factor when it comes to a stable overclock. This feature is in Kaby Lake-X as well.
The end result is that the Core i7-7740X went one further in our testing, and hit 5.0 GHz all-cores at 1.25V, and completed our CPU testing suite with a -10 AVX offset. We went for 5.1 GHz all-core, up to 1.35 volts, but the temperatures were getting high and the system was still not stable. 5.0 GHz is a very respectable result, even if we only have a sample of one.
The results at 5.0 GHz are as follows. Gain is absolute for the benchmark for the 5.0 GHz value, with anything within 2.0% listed in orange. Green indicates a win for the faster chip.
| Comparison: Intel Core i7-7740X at Stock vs 5 GHz | |||
| 4.3 GHz Base 4.5 GHz Turbo Auto AVX |
5.0 GHz Base 5.0 GHz Turbo 4.0 GHz AVX |
||
| Score / Time | Web Benchmarks | Score / Time | Gain |
| 135 ms | Sunspider | 133 ms | +1.6% |
| 622 ms | Kraken | 642 ms | -3.1% |
| 45485 | Octane | 45279 | -0.5% |
| 561 | WebXPRT15 | 570 | +1.5% |
| System Tests | |||
| 2229 ms | PDF Opening | 2055 ms | +8.5% |
| 46.172 s | FCAT | 45.656 s | +1.1% |
| 1287 | 3DPM 2.1 | 1429 | +11.0% |
| 282 s | Dolphin v5 | 280 s | +0.7% |
| 891.56 s | Agisoft Total | 830.06 s | +7.4% |
| Rendering | |||
| 2474593 | Corona | 2660002 | +7.5% |
| 436 s | Blender | 485 s | -10.1% |
| 195 | Cinebench 15 ST | 215 | +10.6% |
| 986 | Cinebench 15 MT | 1084 | +10.0% |
| Encoding | |||
| 27433 | 7-zip | 29433 | +8.9% |
| 48.698 s | WinRAR | 46.633 s | +4.4% |
| 5.5 GB/s | AES | 6.2 GB/s | +12.7% |
| 25.2 FPS | Handbrake HEVC | 22.6 FPS | -10.2% |
| Office Tests | |||
| 4878 | PCMark8 Creative | 5084 | +4.2% |
| 4844 | PCMark8 Home | 5071 | +4.7% |
| 4110 | PCMark8 Work | 3984 | +3.2% |
| 19.45 per day | Chrome Compile | 17.70 per day | -9.0% |
| GTX 1080 | |||
| 72.56 FPS | 1080p Ashes Av | 76.19 FPS | +4.8% |
| 53.95 FPS | 1080p Ashes 99% | 55.45 FPS | +2.7% |
| 62.30 FPS | 4K Ashes Av | 63.46 FPS | +1.8% |
| 41.24 FPS | 4K Ashes 99% | 40.79 FPS | -1.1% |
| 103.66 FPS | 1080p RoTR-1 Av | 106.9 FPS | +3.0% |
| 85.14 FPS | 1080p RoTR-1 99% | 85.88 FPS | +0.9% |
| 61.60 FPS | 4K RoTR-1 Av | 61.33 FPS | -0.4% |
| 50.25 FPS | 4K RoTR-1 99% | 50.43 FPS | +0.3% |
There are a few obvious trends, and a couple of red flags. Most benchmarks split themselves into two groups: benchmarks with a heavy CPU component tend to get a 4-12% gain when overclocked, and those that require a lot of DRAM involvement or AVX tend to be in the 2% region. The red flags are the three big 9-10% loss values for CPU heavy benchmarks, Blender, Handbrake HEVC and Chrome Compilation.
All three benchmarks are sizable workloads with sustained high frequencies throughout, and have heavy AVX components as well. The 4.0 GHz setting for AVX could account for most of that difference, however other AVX benchmarks in our testing (Corona, Cinebench, elements of PCMark, Agisoft) do not have the same effect. The only real difference here is time: the amount of time each benchmark stays at a sustained load.
The three that had the biggest dips are our longest benchmarks: Blender at 8 minutes, Handbrake HEVC at 25 minutes, and the Chrome Compile at over an hour. In this case it seems we are hitting thermal limits for the power delivery, as explained by Igor Wallossek over at Tom's Hardware. He tested an upcoming theory that the early X299 boards are not up to the task for cooling VRMs at heavy load, and through analysis he determined that this was likely to occur in heavily overclocked scenarios. His data showed that the Skylake systems he tested, when overclocked, would hit thermal limits, come back down, and then ramp up again in a cyclical manner. He tested Skylake-X, which draws a lot more power overclocked than our KBL-X setup here, so it likely isn't affecting our setup as much, but still enough for certain benchmarks. I fully suspect we will see second-wave X299 motherboards with substantial heatsinks on the power delivery to overcome this.
More testing is needed, especially in upcoming X299 motherboard reviews. We used an early X299 prototype for our CPU testing here, and a Gaming 9 motherboard on the GPU tests (which showed little variation and no problems). We did take power numbers for this setup but when looking back on the results, they seemed off so we need to go back and re-test those.
Facebook
Twitter
RSS
176 Comments
View All Comments
Santoval - Tuesday, July 25, 2017 - link
That is not how IPC works, since it explicitly refers to single core - single thread performance. As the number of cores rises the performance of a *single* task never scales linearly because there is always some single thread code involved (Amdahl's law). For example if your task has 90% parallel and 10% serial code its performance will max out at x10 that of a single core at ~512 cores. From then on even if you had a CPU with infinite cores you couldn't extract half an ounce of additional performance. If your code was 95% parallel the performance of your task would plateau at x20. For that though you would need ~2048 cores. And so on.Of course Amdahl's law does not provide a complete picture. It assumes, for example, that your task and its code will remain fixed no matter how many cores you add on them. And it disregards the possibility of computing distinct tasks in parallel on separate cores. That's where Gustafson's Law comes in. This "law" is not concerned with speeding up the performance of tasks but computing larger and more complex tasks at the same amount of time.
An example given in Wikipedia involves boot times : Amdahl's law states that you can speed up the boot process, assuming it can be made largely parallel, up to a certain number of cores. Beyond that -when you become limited by the serial code of your bootloader- adding more cores does not help. Gustafson's law, on the contrary, states that instead of speeding up the boot process by adding more cores and computing resources, you could add colorful GUIs, increase the resolution etc, while keeping the boot time largely the same. This idea could be applied to many -but not all- computing tasks, for example ray tracing (for more photorealistic renderings) and video encoding (for smaller files or videos with better quality), and many other heavily multi-threaded tasks.
Rickyxds - Monday, July 24, 2017 - link
I just agree XD.Diji1 - Wednesday, July 26, 2017 - link
"Overall speed increase 240%."LMAO. Ridiculous.
Alistair - Wednesday, July 26, 2017 - link
No reason to laugh. I compared the 6600k vs the Ryzen 1700. 1 year speed increase of 144 percent (2.44 times the speed). Same as this: 1135 vs 466 points.http://cpu.userbenchmark.com/Compare/Intel-Core-i5...
Dr. Swag - Tuesday, July 25, 2017 - link
I disagree, best value is 1600 as it oces as well as 1600x, comes with a decent stock cooler, and is cheaper.vext - Monday, July 24, 2017 - link
Interesting article but it seems intended to play down the extremely bad press x299 has received which is all over the internet and Youtube.Once you get past Mr. Cuttress' glowing review, it's clear that the I5-7640x is not worth the money because of lackluster performance, the I7-7740X is marginally faster than the older 7700k, and the I7-7800x is regularly beaten by the 7740X in many benchmarks that actually count and is a monstrously inefficient energy pig. Therefore the only Intel CPUs of this batch worth buying are the 7700k/7740x, and there is no real advantage to x299. In summary, it doesn't actually change anything.
It's very telling that Mr. Cutress doesn't comment on the absolutely egregious energy consumption of the 7800x. The Test Bed setup section doesn't list the 7800x at all. The 7840x and 7740x are using a Thermalright True Copper (great choice!) but no info on the 7800x cooler. Essentially, the 7800x cameo appearance is only to challenge the extremely strong Ryzen multi-threaded results, but its negative aspects are not discussed, perhaps because it might frighten people from x299. Tsk, tsk. As my 11 year old daughter would say "No Fair." By the way, the 7800x is selling for ~ $1060 right now on Newegg, not $389.
Proudly typed on my Ryzen 1800x/Gigabyte AB350 Gaming 3. # ;-)
Ian Cutress - Monday, July 24, 2017 - link
You may not have realised but this is the Kaby Lake-X review, so it focuses on the KBL-X parts. We already have a Skylake-X review for you to mull over. There are links on the first page.mapesdhs - Monday, July 24, 2017 - link
Nevertheless, the wider picture is relevant here. The X299 platform is a mess. Intel is aiming KL-X at a market which doesn't exist, they've locked out features that actually make it useful, it's more power hungry, and a consumer needs a lot of patience and plenty of coffee to work out what the heck works and what doesn't on a mbd with a KL-X fitted.This is *exactly* the sort of criticism of Intel which should have been much stronger in the tech journalism space when Intel started pulling these sorts of stunts back with the core-crippled 3930K, heat-crazy IB and PCIe-crippled 5820K. Instead, except for a few exceptions, the tech world has been way too forgiving of Intel's treading-on-water attitude ever since SB, and now they've panicked in response to Ryzen and released a total hodgebodge of a chipset and CPU lineup which makes no sense at all. And if you get any disagreement about what I've said by anyone at Intel, just wave a 4820K in their face and say well explain this then (quad-core chip with 40 PCIe lanes, da daa!).
I've been a big fan of Z68 and X79, but nothing about Intel's current lineup appeals in the slightest.
serendip - Tuesday, July 25, 2017 - link
There's also the funny bit about motherboards potentially killing KBL-X CPUs if a Skylake-X was used previously.What's with Intel's insane product segmentation strategy with all the crippling and inconsistent motherboard choices? It's like they want to make it hard to choose, so buyers either get the cheapest or most expensive chip.
Haawser - Tuesday, July 25, 2017 - link
'EmergencyLake-X' is just generally embarrassing. Intel should just find a nearby landfill site and quietly bury it.