To say that the X299 and Core X launch has been bumpy would be an understatement. During Computex, details were announced and the boards were shown off and Youtubers like Linus took to the streets literally and figuratively to voice their opinions. I myself have kept quiet, avoiding a lot of the drama, but mostly because I don’t like to voice an opinion on something from specification sheets and rumors. So I held out waiting to get everything into the office and take a closer look. Well, Intel sent over the new i9-7900X and the i7-7740X and today I’m going to talk a little more about everything and then dive into the performance. When that is all said and done, then we can finally start to look at the overall picture.
Product Name: Intel Core X Series i9-7900X and i7-7740X
Review Sample Provided by: Intel
Written by: Wes
Pictures by: Wes
Amazon Affiliate Link: Core i9-7900X and Core i7-7740X
X299 Testbench also supported by: Asus, Noctua, and Kingston
Intel Core X Series
Okay before getting into the fun stuff like testing and performance numbers we do need to run through some of the details on Intel’s new CPUs. I’m not going to go too far in depth because all of this information has been available for weeks now and everyone and their mom has broken it all down. But anyhow. What they are launching today is what they are calling the New Intel Core X-Series. Now typically they would be more specific with the architecture but this launch is a little different. They are launching both Skylake-X and Kaby Lake-X. For a while now Intel’s enthusiast platform has been a generation behind the mainstream CPUs. So this launch would normally be the Skylake-X launch with Kaby Lake-X next year if that trend continued. One could speculate what with Ryzen being launched and AMD finally having a pulse Intel may have finally had the incentive to work on catching the enthusiast platform back up.
The Kaby Lake based CPUs though are closer to the mainstream lineup than the typical enthusiast lineup though. They cut the integrated GPUs off but didn’t integrate the features that make the enthusiast platform so great. So the two SKUs (i7-7740X and the i5-7640X) only have to channels of memory just like on Z270 and they only have 16 PCIe lanes. For those of you who don’t know what those are, basically, they are communication lanes directly between the CPU and most of the board features. Most are used by the PCIe slots but they can also be used for USB ports, wireless, and other features. With fewer lanes on a CPU, you can’t have as much running at once. In this situation specifically, most X299 boards will have a lot of features that just won’t work when these CPUs are installed. Looking at the rest of the lineup the next two CPUs up are the i7-7800X and the i7-7820X, these are the 6 and 8 core CPUs. So with the Kaby Lake-X CPUs, I get that they would have fewer PCIe Lanes as that could be a carryover from the mainstream Kaby Lake CPUs, but the 6 and 8 core CPUs also have fewer lanes where their equivalent equals last year had more. This is where some people were upset with Intel, intentionally cutting these back in PCIe lanes to make the i9-7900X look better. No one wants to regress or lose features, so I don’t blame anyone for being upset about that one.
The other really unusual thing in the processor listing are the four up at the top with almost no information listed other than cores, names, and prices. I’m told that the 12-core Intel Core i9-7920X shipping in August and the i9-7940X, i9-7960X and 18-core Intel Core i9-7980X Extreme Edition processor, shipping in October. Again, speculating here, but I think these were all in response to AMDs Threadripper announcement. For right now the i9-7900X is the flagship and is one of the two CPUs we are checking out today. The whole i9 thing is another big change, up until now i7 was Intel’s highest designation but with the Core-X introduction, we now have i9’s years after the initial rumors that an i9 was coming.
The Kaby Lake CPUs have impressive base clock numbers with the i7-7740X having a 4.3 GHz base clock that boosts up just slightly to 4.5 GHz with Turbo Boost 2.0. The Skylake-X base clocks are lower with the current flagship i9-7900X being the lowest at 3.3 GHz. The 7900X and 7820X both boost up very well with 4.3 on Turbo Boost 2.0 and they both hit 4.5 GHz with Turbo Boost 3.0. The i7-7800X is a little bit of an odd man out with it not having Turbo Boost 3.0 and having a lower Turbo Boost 2 clock speed at 4 GHz. The pricing of the currently released CPUs mostly follows past Intel pricing with the flagship 7900X being $999. The i7-7800X though is cheaper now to compete with Ryzen. It should be noted that all of the new CPUs have different core/thread counts so there isn’t any chance of snagging a cheaper model and overclocking it for a deal. The 7900X has 10 cores and 20 threads and they move down from there to the 7820X with 8/16, the 7800X with 6/12 and then the Kaby Lake-X CPUs both have 4 with the 7740X being the only one with hyperthreading.
So beyond the basic improvements that the Kaby Lake/Skylake architectures bring to their respective CPUs there is one big improvement. The way cache is handled is now different. MLC is now four times larger than previous X series CPUs while still keeping the shared LLC as well but with just a little less available per core.
Beyond the CPUs, Intel is also introducing the new X299 chipset. They are moving from 2011 R3 to the new LGA 2066 socket. The new socket has the same cooler compatibility as the X99 and X79 before it with the built in mounting points that drop the need for an additional backplate. X299 also gained Intel Optane support. Intel provided a nice breakdown comparing the X99 and X299 platforms with both Skylake-X and Kaby Lake-X CPUs being separate to show the differences. The big thing being the PCIe breakdown that we already spoke about and the dual and quad channel memory as well. The bump in memory clock speeds is nice though, X299 now officially supports 266 MHz over the 2400 MHz on later generation X99 boards. X299 supports up to three PCIe x4 M.2 or U.2 connections on the chipset, up to 10 USB 3.0 ports, up to 8 SATA 3.0’s, and Intel’s I219 NIC.
The last area that has been a big issue as well is Intel’s implementation of RAID with the X299 chipset. So one of the big features is what they call Virtual RAID on CPU or VROC. This only works with the Skylake-X CPUs not the Kaby Lake-X CPU. It allows you to run RAID without having to use a RAID controller. This in it self is awesome, especially with high end PCs like you would be building with CPUs like the i9-7900X. The problem is Intel makes a lot of money from their enterprise hardware and this would be a big hit, so for RAID-1 and RAID-5 they will be forcing the use of a dongle that you will have to buy to gain access to Raid 1 and 5.
Here is a look at the new Socket. Like I said it uses the same mounts as the LGA 2011 cooler support at launch should include everything that supported 2011 before.
Here is a look at the new CPUs, they have the same thick heatspreader as Broadwell-E only on the bottom there are a few more pads to up the pin count to 2066 from 2011.
Test Rig and Procedures
Intel Z270 Test System |
||
Motherboard |
Asus Prime X299 Deluxe |
|
Cooling |
Noctua NH-U12S for cooling Noctua NT-H1 Thermal Paste |
|
Memory |
Kingston 2666 MHz DDR4 Fury quad channel 4x8Gb |
|
Storage |
Kingston HyperX 240GB SSD |
|
Video Card |
AMD RX480 |
|
Power Supply |
Thermaltake 850w |
|
Case |
Microcool Banchetto 101 Test bench |
|
OS |
Windows 10 Pro 64-bit |
CPU Testing Procedures |
|
Team Fortress 2 |
1080p – high, very high, high, reflect world, high, enabled, 8x msaa, x16, multicore on. Testing is then run with this benchmark |
Ashes of the Singularity: Escalation |
1080p – Standard quality setting, DX12, built in benchmark on the CPU-focused setting |
Tom Clancy's Ghost Recon: Wildlands |
Built-in Benchmark, 1080p, High detail setting |
Deus Ex: Mankind Divided |
1080p – Ultra Setting – DX11 – Vsync Turned off using built-in benchmark |
Dolphin 5.0 Benchmark |
For information on configuration, view this thread |
3DMark Fire Strike |
Physics Score – Performance benchmark |
7-Zip |
Built-in benchmark set to 32MB, run 4 passes |
Jetstream |
http://browserbench.org/JetStream/ |
wPrime |
1024M |
X264 HD Benchmark |
Pass 1 and Pass 2 |
Cinebench R15 |
CPU and CPU (Single Core results) |
Passmark 8 |
CPU Mark Score |
PCMark 10 |
Normal test ran, not the express or extended |
AIDA64 |
Memory, Cache, and FLOPS testing done with built in GPGPU and Memory benchmarks |
Power Usage |
Idle and load testing using a Kill-A-Watt and wPrime to put the CPU under load |
Temps |
Aida64 stability test to load and to record temperature readings |
CPU Performance
To test out both of the new CPUs I setup our new X299 testbench with the Asus Prime X299 Deluxe along with a GTX 1080 Ti. Then for tests, I went through our test suite that I updated for the Ryzen launch and made a few changes. Then, of course, I went back and retested for days all of the Ryzen CPUs, the Intel i7-7700K, and the older i7-5960X and the i7-6900K as well. This gave me the chance to retest Ryzen with a higher clock speed memory that it benefits from and to update numbers now that the platform is more up to date. The mix of CPUs covered all of AMDs current offerings, older and newer X99 CPUs, and Intel’s current flagship mainstream CPUs as well. Our test suite includes a variety of CPU specific benchmarks, games, overall PC benchmarks, and a few deep level benchmarks that look at things like the Memory and Cache speeds and latencies.
For the first test, I tested x264 encoding with the older but still solid x264 HD Benchmark 4.0. The results are in frames per second that the CPU could encode. The test is run four times then averaged out to give these results. The Kaby Lake based 7740X performed better here with the 7900X being just behind it. Both were well above the Ryzen CPUs but oddly enough the i7-6900K was also faster as well with the always extremely fast 7700K being the fastest.
The next benchmark might be my favorite test out of all CPU benchmarks. Cinebench R15 is a compute benchmark so it doesn’t apply to everyone but I love it because it actually tests using all cores and then again with just one so we can see the total performance and IPC performance as well. I wish more tests did the same. So the 10 core i9-7900X just dominated in the multicore benchmark where it was able to take advantage of its additional cores. The i7-7740X, on the other hand, was back in the 4 core range with the i7-7700K. In the single core IPC test, the 7900X fell back a little with the two Kaby Lake CPUs at the top. The 7700K was still quicker than the 7740X just slightly. The 7900X did well still though with it having a single core score of 185, well above all of the Ryzen CPUs and the older Intel CPUs.
The next two benchmarks aren’t the same but both handle a similar area. wPrime tests prime number calculation speed and 7zip tests compression speed. In wPrime, the i9-7900X came out ahead with its good core speeds and a high number of cores total. The i7-7740X was again right below the i7-7700K and the Ryzen 7 CPUs and one of the Ryzen 5 CPUs were all in between. The result was exactly the same in 7-Zip where the i9-7900X wasn’t even in the same class as the other CPUs.
For browser benchmarks, I dropped our Google Octane benchmark because it is no longer being updated and the results were all over the place but I kept Jetstream. This test runs three times covering at least 30 tests then combines the tests for a score then averages the three results for one score. Here the Kaby Lake CPUs did well again with the i7-7700K and then the i7-7740X being at the top. I expected the i9-7900X to do well but was surprised that the i7-6900K still did better. None of the Ryzen CPUs were far off, but they were all bunched up with nearly the same results.
For more general CPU benchmarks I took a look at Passmark’s Performance Test 9 and ran their CPU score that runs a mix of CPU focused tests. Here once again the i9-7900X with its 10 cores and good IPC pulled well ahead with Intel’s other 8 core CPUs the next two behind it. Then you get the 8 core Ryzen CPUs. The i7-7740X and i7-7700K are down closer to the bottom with their low core counts.
PCMark 10 covers a wider range of tests all focused on real world usages. The essentials tests are focused on web browsing, video conferencing, and app start-up times. The i7-7700K performed best here with the i7-7740X being next, then working down from there depending on single core performance. The productivity testing tested office applications like spreadsheet and writing workloads with similar scores to the essentials. Content creation, on the other hand, was more dependent on core counts. The tests focused on photo and video editing and rendering and visualization. In that test, the i9-7900X pulled ahead. Then last but not least, the overall PCMark score averages all of the tests out. The i7-7700K and i7-7740X did really well with the 7700K coming out at the top but the i9-7900X was able to edge up into second with just pure core count.
Before getting to actual gaming performance I took a look at 3DMark and Dolphin benchmark. 3DMark is a nice consistent synthetic benchmark and Dolphin tests out emulator performance that is very CPU dependent. In Dolphin, the lower the score the better and the high IPC of the i7-7700K put it at the fastest with the i7-7740X next. The i9-7900X wasn’t far behind though, surprisingly. In 3DMark I used the Fire Strike benchmark set to the 1080p equivalent performance setting. Once again the i9-7900X just stomped the physic scores with the 6900K and the 8 core Ryzen CPUs being next but not close. The i7-7740X was down much lower with the other mainstream CPUs.
For in game testing I tested in four different games. I added Wildlands and TF2 that touch on popular games both new and old then we have the CPU specific Ashes of the Singularity: Escalation tests and Deus Ex: Mankind Divided as another newer game. Deus Ex and Wildlands are both GPU dependent games so the difference between CPUs (especially in Deus Ex) was very small. Wildlands does bring up potential issues with our X99 setup though that I will have to address. TF2 being an older game is very CPU limited and with it still being in the top 10 played games I was excited to add it to the charts. The results were a little off though, especially with the Ryzen results being flipped but that could just be a lack of high core count support. The 7700K and 7740X were up top with the higher core count CPUs below that. Then in Ashes of the Singularity, this was basically a best case scenario for Ryzen where I tested with faster memory in a game that is very Ryzen optimized but the high core count Intel’s still dominated here, even over the always fast i7-7700K.
The last set of performance testing was using AID64. Frankly, there was so much information that I wanted to show that I couldn’t fit it into any graph so I went with a basic table as you can see below. I wanted to show the memory read, write, and latency. This was especially interesting to check out with the quad channel memory on the i9-7900X and the dual channel for the i7-7740X. The i9 did have a higher latency but the bandwidth difference was huge. X299 has more memory bandwidth than past Intel quad channel setups as well though when compared to the X99 CPUs. Ryzen calls in between with more than Kaby Lake but less than Skylake-X. For Cache performance it is interesting to see that the i9-7900X was much faster with its L1 and L2 cache, especially compared to Ryzen but down in the L3 cache numbers it falls off but the 7740X didn’t. Single and Double precision performance favored the Intels but the AES-256 performance of the Ryzen CPUs was extremely impressive. Integer IOPS for the Intels across the board were also way ahead of Ryzen, especially the i9-7900X with almost 5 times the bandwidth at 24 and 32 bits.
CPU |
Memory Read |
Memory Write |
Memory Latency |
Intel Core i9-7900X |
76856 MB/s |
72856 MB/s |
72.8 ns |
Intel Core i7-7740X |
38455 MB/s |
40596 MB/s |
53.2 ns |
Intel i7-6900K |
66786 MB/s |
68130 MB/s |
61.9 ns |
Intel i7-7700K |
38498 MB/s |
40448 MB/s |
52.7 ns |
Intel i7-5960X |
66108 MB/s |
49545 MB/s |
62.8 ns |
AMD Ryzen R7 1800X |
44031 MB/s |
43425 MB/s |
81.8 ns |
AMD Ryzen R7 1700X |
44493 MB/s |
43749 MB/s |
78.3 ns |
AMD Ryzen R7 1700 |
45343 MB/s |
43777 MB/s |
82.6 ns |
AMD Ryzen R5 1600X |
44452 MB/s |
43918 MB/s |
83.7 ns |
AMD Ryzen R5 1500X |
44289 MB/s |
43746 MB/s |
83.1 ns |
|
L1 Cache Read |
L1 Cache Write |
L1 Cache Latency |
Intel Core i9-7900X |
2418.5 GB/s |
709.16 GB/s |
1.0 ns |
Intel Core i7-7740X |
1102.7 GB/s |
561.25 GB/s |
0.9 ns |
Intel i7-6900K |
1963.9 GB/s |
999.59 GB/s |
1.0 ns |
Intel i7-7700K |
1114.6 GB/s |
560.82 GB/s |
0.9 ns |
Intel i7-5960X |
1748.5 GB/s |
874.60 GB/s |
1.2 ns |
AMD Ryzen R7 1800X |
917.46 GB/s |
460.16 GB/s |
1.1 ns |
AMD Ryzen R7 1700X |
867.89 GB/s |
435.33 GB/s |
1.2 ns |
AMD Ryzen R7 1700 |
793.90 GB/s |
398.12 GB/s |
1.3 ns |
AMD Ryzen R5 1600X |
689.47 GB/s |
345.95 GB/s |
1.1 ns |
AMD Ryzen R5 1500X |
446.84 GB/s |
224.01 GB/s |
1.1 ns |
|
L2 Cache Read |
L2 Cache Write |
L2 Cache Latency |
Intel Core i9-7900X |
1037.3 GB/s |
521.94 GB/s |
5.2 ns |
Intel Core i7-7740X |
424.55 GB/s |
304.30 GB/s |
2.7 ns |
Intel i7-6900K |
710.28 GB/s |
310.34 GB/s |
4.1 ns |
Intel i7-7700K |
415.45 GB/s |
285.97 GB/s |
2.7 ns |
Intel i7-5960X |
623.08 GB/s |
272.77 GB/s |
3.5 ns |
AMD Ryzen R7 1800X |
831.95 GB/s |
446.37 GB/s |
4.7 ns |
AMD Ryzen R7 1700X |
809.29 GB/s |
419.38 GB/s |
5.0 ns |
AMD Ryzen R7 1700 |
742.92 GB/s |
389.73 GB/s |
5.4 ns |
AMD Ryzen R5 1600X |
630.99 GB/s |
334.93 GB/s |
4.7 ns |
AMD Ryzen R5 1500X |
306.82 GB/s |
211.58 GB/s |
11.8 ns |
|
L3 Cache Read |
L3 Cache Write |
L3 Cache Latency |
Intel Core i9-7900X |
124.16 GB/s |
106.78 GB/s |
21.2 ns |
Intel Core i7-7740X |
287.01 GB/s |
186.67 GB/s |
21.2 ns |
Intel i7-6900K |
243.96 GB/s |
195.41 GB/s |
14.6 ns |
Intel i7-7700K |
234.50 GB/s |
188.70 GB/s |
10.4 ns |
Intel i7-5960X |
260.72 GB/s |
184.92 GB/s |
15.3 ns |
AMD Ryzen R7 1800X |
388.18 GB/s |
386.01 GB/s |
12.4 ns |
AMD Ryzen R7 1700X |
369.75 GB/s |
350.31 GB/s |
12.9 ns |
AMD Ryzen R7 1700 |
334.36 GB/s |
332.12 GB/s |
14.2 ns |
AMD Ryzen R5 1600X |
386.48 GB/s |
326.86 GB/s |
12.3 ns |
AMD Ryzen R5 1500X |
306.82 GB/s |
211.58 GB/s |
11.8 ns |
|
Single-Precision FLOPS |
Double Precision FLOPS |
AES-256 |
Intel Core i9-7900X |
1148 GFLOPS |
574.5 GFLOPS |
45506 MB/s |
Intel Core i7-7740X |
575.4 GFLOPS |
287.8 GFLOPS |
20503 MB/s |
Intel i7-6900K |
1023 GFLOPS |
511.7 GFLOPS |
36340 MB/s |
Intel i7-7700K |
575.7 GFLOPS |
287.9 GFLOPS |
20519 MB/s |
Intel i7-5960X |
895.5 GFLOPS |
447.7 GFLOPS |
31783 MB/s |
AMD Ryzen R7 1800X |
472.4 GFLOPS |
236.2 GFLOPS |
64016 MB/s |
AMD Ryzen R7 1700X |
446.8 GFLOPS |
223.4 GFLOPS |
61730 MB/s |
AMD Ryzen R7 1700 |
408.5 GFLOPS |
204.3 GFLOPS |
56811 MB/s |
AMD Ryzen R5 1600X |
354.9 GFLOPS |
177.5 GFLOPS |
49340 MB/s |
AMD Ryzen R5 1500X |
229.8 GFLOPS |
114.8 GFLOPS |
31983 MB/s |
|
24-bit Integer IOPS |
32-bit Integer IOPS |
64-bit Integer IOPS |
Intel Core i9-7900X |
515.1 GIOPS |
515.0 GIOPS |
79.76 GIOPS |
Intel Core i7-7740X |
258.1 GIOPS |
258.0 GIOPS |
35.97 GIOPS |
Intel i7-6900K |
255.9 GIOPS |
255.9 GIOPS |
63.96 GIOPS |
Intel i7-7700K |
258.0 GIOPS |
258.0 GIOPS |
35.98 GIOPS |
Intel i7-5960X |
223.9 GIOPS |
223.9 GIOPS |
55.97 GIOPS |
AMD Ryzen R7 1800X |
118.1 GIOPS |
118.1 GIOPS |
59.03 GIOPS |
AMD Ryzen R7 1700X |
111.7 GIOPS |
111.7 GIOPS |
55.84 GIOPS |
AMD Ryzen R7 1700 |
102.1 GIOPS |
102.1 GIOPS |
51.05 GIOPS |
AMD Ryzen R5 1600X |
88.74 GIOPS |
88.73 GIOPS |
44.34 GIOPS |
AMD Ryzen R5 1500X |
57.43 GIOPS |
57.45 GIOPS |
28.72 GIOPS |
Power Usage and Temperatures
For power testing, I hooked each of the test configurations up to a Kill-A-Watt and documented complete idle numbers then loaded the CPU up to 100% with wPrime for the load numbers. Unsurprisingly the 140W TDP i9-7900X was up close to the top of the charts at 226 watts but the idle numbers weren’t too bad at 66.9 watts. The i7-7740X wasn’t as bad with a slightly lower idle number but the load at 123 watts was MUCH lower.
For temperature testing, I retested all of the CPUs in this review using AID64’s stability tester with just an FPU load to heat things up the most. Now remember these numbers aren’t perfect, you have to rely on unreliable software/in CPU results combined with things like AMD adjusting the numbers by 20 degrees as well. That said I included them all just for reference. The i9-7900X didn’t do too bad with a peak of 58 degrees using the Noctua cooler but the i7-7740X, like the i7-7700K didn’t do very well. If I were just using these numbers I would guess that the 7740X was using TIM under the heatspreader and the 7900X was soldered, but everything (pre-launch) is saying that both will have TIM so I’m not sure why it would register so much lower beyond untrustable software based testing. SO take these numbers with a grain of salt right now.
Overall and Final Verdict
So now that we have run through the X299 and Core X features and taken a look at the performance of the i9-7900X and the i7-7740X, where does everything land. Well first off, let me say right out of the hole that you have to keep in mind that X299 and the Core-X series of CPUs are on a higher end platform than the mainstream CPUs. With Z270 only the highest end boards are going to be expensive and the most expensive CPU is closer in price to the low-end Core-X CPUs. So the buy-in cost, even at its lowest it will be closer to a high-end Z270 build, with costs going up from there. This isn’t anything new when compared to X99, X79, and X58 but it always comes as a surprise to anyone who is used to mainstream chipsets.
As for performance, I’ll be honest I came in not expecting too much from either CPU. The i9-7900X is in a lot of ways very similar to last years i7-6950X. The move to Skylake helps but the move from a base clock of 3 GHz to 3.3 and boost clocks going from 4GHz to 4.5 Ghz make a big difference. Unlike past high core count Intel CPUs, the 7900X actually has a decent IPC (instructions per clock) or in other words, the single core performance is really good. Times that by 10 core or 20 threads and in all of the multi-core benchmarks it is a monster with surprisingly good performance in everything else. Now there are still a lot of situations where the i7-7700K and the i7-7740X are faster though. The 7900X actually gets the full 44 PCIe lanes, Turbo Boost 3.0, and four channels of memory where moving down the product stack you slowly lose all of those features. That’s one of the big downsides to the i7-7740X, you only have 16 PCIe lanes and just dual channel memory.
The 7900X had good temps in my testing but when it came to power usage it didn’t do very well. The idle performance was fine but when you load all 10 cores it is going to pull the power. The 7740X, on the other hand, ran hot for me but had great power consumption. The 7740X from what I have heard is actually a great overclocker, I’m really looking forward to diving into that after the launch. It also benefits from Deliding as well.
So with both CPUs, the performance numbers look good, even when compared to Ryzen. The i9-7900X is outdoing Ryzen on the whole high core count gig with even better IPC and the i7-7740X is basically just another i7-7700K so gaming performance is good. What made the Ryzen launch so good was the pricing though and that is where we have to put a little more thought into everything. The i9-7900X is going to rip your wallet to shreds with its $999 MSRP and with that it is twice the price of the Ryzen 7 1800X, so while the performance is much better, it damn well should be at that price. This is the CPU you go for when you want only the best. The problem there is in a few months Intel has four more CPUs that are going to take over that role, not to mention with AMDs Threadripper upcoming. So I’m digging the i9-7900X, but I suspect it only has the main role for a few months unless it gets a big price drop when Threadripper comes out.
As for the i7-7740X, this is an even weirder CPU. You are basically getting an i7-7700K but on the X299 chipset and without the integrated GPU. It does have two things going for it though. If you are trying to build a high-end system but need to save some money now. You can build with this or the i5-7640X and upgrade to a higher end CPU later. The other benefit is that this could have the potential to be a great overclocking CPU. Paul from Pauls Hardware suggested this recently as well on one of his videos and I completely agree. I just don’t see a lot of the normal i7-7700K type of customers spending the extra money on X299 and an i7-7740X to get what is at best the same performance and not even get to use a lot of the features on the boards they buy due to the dual channel memory and the 16 PCIe lanes.
Personally, I’m not even against having the 7740X on X299, I just think it would fit better if you had the full PCIe lanes and memory. Really I’ve always wished the 4770/6700/7700 all were on the enthusiast platform and not on the mainstream like it was with X58. All of the fast CPUs were on the enthusiast platform and that is what you bought if you wanted to game. Mixing the higher end into the mainstream platform muddied everything up and dropped the interest in X79/x99.
So are the Core-X CPUs for you? Well I know I’m excited to build something crazy with them. But the i7-7740X doesn’t seem to be the sweet spot. I think the i7-7820X and the i7-7800X have the potential to compete with the people buying Ryzen R7. The i7-7740 with more PCIe lanes and quad channel ram and crazy overclocking potential though could have been amazing. I hope Intel reconsiders. So to answer my initial question, i7-7740X no go, i7-7800X maybe, i7-7820X, maybe, and i9-7900X yes but only if you don’t like money and you can’t wait for the other CPUs that are upcoming.
Live Pricing: HERE
Live Pricing: HERE