The AnandTech Coffee Lake Review: Initial Numbers on the Core i7-8700K and Core i5-8400
by Ian Cutress on October 5, 2017 9:00 AM EST- Posted in
- CPUs
- Intel
- Core i5
- Core i7
- Core i3
- 14nm
- Coffee Lake
- 14++
- Hex-Core
- Hyperthreading
Benchmarking Performance: CPU Office Tests
The office programs we use for benchmarking aren't specific programs per-se, but industry standard tests that hold weight with professionals. The goal of these tests is to use an array of software and techniques that a typical office user might encounter, such as video conferencing, document editing, architectural modeling, and so on and so forth.
All of our benchmark results can also be found in our benchmark engine, Bench.
Chromium Compile (v56)
Our new compilation test uses Windows 10 Pro, VS Community 2015.3 with the Win10 SDK to compile a nightly build of Chromium. We've fixed the test for a build in late March 2017, and we run a fresh full compile in our test. Compilation is the typical example given of a variable threaded workload - some of the compile and linking is linear, whereas other parts are multithreaded.
For our compile test, it would appear that the extra memory width afforded by the quad-channel memory of Skylake-X can have a direct benefit in compile performance.
PCMark 10
PCMark 10 is the 2017 update to the family favorite, PCMark 8. PCMark 8 has been part of our test bed since the latest update in Q1. For the most part it runs well, although for some processors it doesn’t recognize, some tests will not complete, leading to holes in our benchmark data (there’s also an odd directory quirk in one test that causes issues). The newest version, PCMark 10, is the answer.
The new test is adapted for more 2016/2017 workflows. With the advent of office applications that perform deeper compute tasks, or the wave of online gamers and streamers, the idea behind PCMark 10 is to give a better ‘single number’ result that can provide a comparable metric between systems. Single metrics never tell the whole story, so we’re glad that Futuremark provides a very detailed breakdown of what goes on.
Ganesh’s article on PCMark 10 goes into more detail than I will here, but the ‘Extended Benchmark’ runs through four different sets of tests: Essential, Productivity, Creation and Gaming. Each of these have sub-test results as well, including startup performance, web performance, video conferencing, photo/video editing, spreadsheets, rendering, and physics, which you can find in Bench.
PCMark8: link
Despite originally coming out in 2008/2009, Futuremark has maintained PCMark8 to remain relevant in 2017. On the scale of complicated tasks, PCMark focuses more on the low-to-mid range of professional workloads, making it a good indicator for what people consider 'office' work. We run the benchmark from the commandline in 'conventional' mode, meaning C++ over OpenCL, to remove the graphics card from the equation and focus purely on the CPU. PCMark8 offers Home, Work and Creative workloads, with some software tests shared and others unique to each benchmark set.
Facebook
Twitter
RSS
222 Comments
View All Comments
DigitalFreak - Thursday, October 5, 2017 - link
Anyone having an issue with Bench? I'm trying to compare my i7-3770k to the i7-8700k and it comes back with no data. Same with trying the Threadripper 1920xmkaibear - Friday, October 6, 2017 - link
CPU tests changed so benchmarks weren't comparable. Latest processor tested on the old tests was the 7700K iirc, and not everything is tested on the new tests.I'd compare results for the 3770k and the 2600K to get a baseline then you can compare 2600K to the 8700K. It's a bit fiddly, I have to do the same with my 4790K.
Ian Cutress - Saturday, October 7, 2017 - link
We updated our CPU testing suite for Windows 10 in Q1. Regression testing is an on-going process, though it's been slow because of all the CPU launches this year. Normally we have 1/2 a year. We're so far at what, 6 or 7 for 2017?mczak - Thursday, October 5, 2017 - link
Doesn't look to me like the die size actually increased at all due to the increased gate pitch.The calculations in the article forgot to account for the increase of the unused area (at the bottom left) - this area is tiny with 2c die, but increases with each 2 cores added significantly. By the looks of it, that unused area would have grown by about 2 mm^2 or so going from 4 to 6 cores, albeit I'm too lazy to count the pixels...
jjj - Thursday, October 5, 2017 - link
Your conclusion is weirdest thing ever, you fully ignore the 8359k and AMD.In retail, the 8350k will do very very well and retail is what matters for most readers
And ignoring AMD is not ok at all, it's like you think that we are all idiots that buy on brand.You do think that, your system guides make that very clear but you should not accept, support and endorse such an idiotic behavior.
AMD got hit hard here, Intel takes back the lead and it's important to state that. Sure they might have Pinnacle Ridge in a few months and take back the lead but buyers that can't wait should go with Intel right now, for the most part. AMD could also adjust prices ofc.
Tigris - Thursday, October 5, 2017 - link
Really confused why the pricing listed in this review isn't consistent- for Intel you were posting prices you found online, but for Ryzen you appear to be posting MSRP.The truth is- you can find 1700x for $298 right now EASILY (Amazon), yet Microcenter is selling the 8700k for $499.
If you factor this information in, the AMD solutions are still far more valuable per dollar.
wolfemane - Thursday, October 5, 2017 - link
I really can’t belive the amount of flak Anandtech takes these days. I find it un-earned an unwarrented. Out of all the tech sites and forums I manage to read in a given week, Anandtech is the most often quoted and linked to. Hell I use it as my go to for reference and comparison (and general reading). My only big complaint is your ads, and I’d gladly pay a sub to completely remove that nonsense and directly support the site!Ian, you and your staff deserve far more credit than you get and that’s an injustice. Each piece is pretty thorough and pretty spot on. So for that thank you very much.
This article is no exception to the rule and is superb. Your graph layouts are a welcome feature!!!!! I look forward to your ever expanding tests as new chips roll in. I think the 8600k is going to be a game changer in the i5 vs i7 performance category for these hexacore cpus. I think that’s why almost all the reviews I’m reading today are with the 8700k and 8400.
Agin, thank you and your staff very much for the work you put into publishing amazing articles!!
vanilla_gorilla - Thursday, October 5, 2017 - link
Personally I buy whatever is best at the time. Right now I'm typing this on a 1700x and I can see a 4770k build on the desk next to me. So it's always funny to see the bias. Intel review gets posted, AMD fanboys come out of the wood works to trash them as paid shills. But it works exactly the same on any positive AMD reviews. Intel fans come in trashing them. It's really odd. Anandtech is one of the most unbiased sites I've found and I trust their reviews implicitly.mkaibear - Saturday, October 7, 2017 - link
> Anandtech is one of the most unbiased sites I've found and I trust their reviews implicitly.Yep. Anyone who looks at AT and sees bias needs to examine their own eyesight.
SeannyB - Thursday, October 5, 2017 - link
For the H.264 encoding tests, you could consider using the "medium" preset or better. The "very fast" preset has a tendency to use fewer cores.