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 Legacy Tests
Our legacy tests represent benchmarks that were once at the height of their time. Some of these are industry standard synthetics, and we have data going back over 10 years. All of the data here has been rerun on Windows 10, and we plan to go back several generations of components to see how performance has evolved.
All of our benchmark results can also be found in our benchmark engine, Bench.
3D Particle Movement v1
3DPM is a self-penned benchmark, taking basic 3D movement algorithms used in Brownian Motion simulations and testing them for speed. High floating point performance, MHz and IPC wins in the single thread version, whereas the multithread version has to handle the threads and loves more cores. This is the original version, written in the style of a typical non-computer science student coding up an algorithm for their theoretical problem, and comes without any non-obvious optimizations not already performed by the compiler, such as false sharing.
CineBench 11.5 and 10
Cinebench is a widely known benchmarking tool for measuring performance relative to MAXON's animation software Cinema 4D. Cinebench has been optimized over a decade and focuses on purely CPU horsepower, meaning if there is a discrepancy in pure throughput characteristics, Cinebench is likely to show that discrepancy. Arguably other software doesn't make use of all the tools available, so the real world relevance might purely be academic, but given our large database of data for Cinebench it seems difficult to ignore a small five minute test. We run the modern version 15 in this test, as well as the older 11.5 and 10 due to our back data.
x264 HD 3.0
Similarly, the x264 HD 3.0 package we use here is also kept for historic regressional data. The latest version is 5.0.1, and encodes a 1080p video clip into a high quality x264 file. Version 3.0 only performs the same test on a 720p file, and in most circumstances the software performance hits its limit on high end processors, but still works well for mainstream and low-end. Also, this version only takes a few minutes, whereas the latest can take over 90 minutes to run.
Facebook
Twitter
RSS
222 Comments
View All Comments
mapesdhs - Friday, October 6, 2017 - link
Any idea what that optimisation is? Seems odd that adding extra pure cores would harm performance, as opposed to adding HT which some games don't play nice with. Otherwise, are you saying that for this test, if it was present, the i3 8100 would come out on top? Blimey.Ian Cutress - Saturday, October 7, 2017 - link
They're either doing something to align certain CPU tasks for AVX, or it's bypassing code. You'd have to ask the developers on that.mapesdhs - Monday, October 9, 2017 - link
I doubt they'd explain what's happening, might be proprietory code or something.WickedMONK3Y - Thursday, October 5, 2017 - link
You have the spec of the i7 8700K slightly wrong. It has a base frequency of 3.7GHz not 3.8GHz.https://ark.intel.com/products/126684/Intel-Core-i...
Ian Cutress - Thursday, October 5, 2017 - link
Mistake on our part. I was using our previous news post as my source and that had a Typo. This review (and that news) should be updated now.Slomo4shO - Thursday, October 5, 2017 - link
Ian, this is probably your worst review to date. Lackluster choice of CPUs, mid-grade GPU, and lack of direct competition in the product stack... Why would you not use a GTX 1080 Ti or Titan XP?Ian Cutress - Thursday, October 5, 2017 - link
All the CPUs we've ever tested are in Bench. Plenty of other data in there: the goal was to not put 30+ CPUs into every graph.Our benchmark database includes over 40 CPUs tested on the GTX 1080, which is the most powerful GPU I could get a set of so I can do parallel testing across several systems. If that wasn't enough (a full test per CPU takes 5 hours per GPU), the minute I get better GPUs I would have to start retesting every CPU. At the exclusion of other content. Our benchmark suite was updated in early Q2, and we're sticking with that set of GPUs (GTX 1080/1060/R9 Fury/RX 480) for a good while for that reason.
Note I had three days to do this review.
crimson117 - Thursday, October 5, 2017 - link
Good job! More people need to know about the bench...Slomo4shO - Thursday, October 5, 2017 - link
To be fair the R5 1600 was added to the benches after the fact. In addition, your othwr reviews tend to be much more detailed and data driven with relevant products and multiple GPUs.Why would I read your review if you expect me to dig through your benchmark to obtain relivant data?
I can understand and appreciate the time crunch but it is a poor excuse for some of the decisions made in this review.
Take it with a grain of salt, this was not your best work.
mapesdhs - Friday, October 6, 2017 - link
Ooohhh the effort of examing the data in Bench! :D First world problems. Sheesh...Run your own tests then, see how you get on with having a life. It's insanely time consuming.