The Intel Core i9-9900K at 95W: Fixing The Power for SFF
by Ian Cutress on November 29, 2018 8:00 AM ESTCPU Performance: Legacy Tests
We have also included our legacy benchmarks, representing a stack of older code for popular benchmarks.
All of our benchmark results can also be found in our benchmark engine, Bench.
3DPM v1: Naïve Code Variant of 3DPM v2.1
The first legacy test in the suite is the first version of our 3DPM benchmark. This is the ultimate naïve version of the code, as if it was written by scientist with no knowledge of how computer hardware, compilers, or optimization works (which in fact, it was at the start). This represents a large body of scientific simulation out in the wild, where getting the answer is more important than it being fast (getting a result in 4 days is acceptable if it’s correct, rather than sending someone away for a year to learn to code and getting the result in 5 minutes).
In this version, the only real optimization was in the compiler flags (-O2, -fp:fast), compiling it in release mode, and enabling OpenMP in the main compute loops. The loops were not configured for function size, and one of the key slowdowns is false sharing in the cache. It also has long dependency chains based on the random number generation, which leads to relatively poor performance on specific compute microarchitectures.
3DPM v1 can be downloaded with our 3DPM v2 code here: 3DPMv2.1.rar (13.0 MB)
x264 HD 3.0: Older Transcode Test
This transcoding test is super old, and was used by Anand back in the day of Pentium 4 and Athlon II processors. Here a standardized 720p video is transcoded with a two-pass conversion, with the benchmark showing the frames-per-second of each pass. This benchmark is single-threaded, and between some micro-architectures we seem to actually hit an instructions-per-clock wall.
101 Comments
View All Comments
GreenReaper - Saturday, December 1, 2018 - link
I'm not so sure it's as big a win in most cases as it's cracked up to be, especially now that new security measures are required to prevent threads on the same core from being able to use Spectre-class attacks to divulge secrets from timing based on data accessed by other thread.stux - Thursday, November 29, 2018 - link
Apparently a post with an ‘at mark’ in it is spam...Harry_Wild - Thursday, November 29, 2018 - link
I thinking about replacing my proprietary Lenovo Thinkstation P300 motherboard. It is so limited and Lenovo does not update their PC BIOS like other manufacturers to keep the PC up to date with new hardware. Lenovo answer is to buy a new Lenovo PC! Just have to find a new one about the same size and I will jerry rig it in.bairlangga - Thursday, November 29, 2018 - link
So, basically with the power limit in place an 8C-16T i9-9900k is an 8C-16T i7-7820x in a different tshirt riding a different cars, being uncapped is like giving it the pass to run on the autobahn.While we've know that Ryzen 1800x had blown the 7820x out of the water. No more IPC increment for Intel, it seems.
SanX - Thursday, November 29, 2018 - link
IanThe most amazing thing you have revealed in your benchmarks over few last months was the crazy speedup of avx512 on 3D Particle movement which put all recent HEDT incarnations both from Intel and AMD deep into the mud. And in this paper you removed this test. Where is 7900x in the second plot? Or your test was buggy showing these crazy 5x improvements even 7th gen over 9th gen when avx512 was on?
HollyDOL - Friday, November 30, 2018 - link
That would be interesting to see. While AVX can do wonders if the workload is suitable, it IS power hungry. I guess you would still end up with better performance ("tasks per kWh") even with the power limit, but hard to say by how much.I can see on my 8700 how much power at wall and core temp rises when it gets loaded with something AVX heavy.
xTRICKYxx - Friday, November 30, 2018 - link
Once the 9900k is at 95w, the 2700X is looking far more competitive.sharath.naik - Friday, November 30, 2018 - link
You missed come nebench scoresDeath666Angel - Friday, November 30, 2018 - link
Wouldn't mind some tuned undervolted tests for the top consumer processors out right now. :)Consumer1 - Friday, November 30, 2018 - link
Would you be so kind as to change the price of the 9900K in your graphs to the list prices for which it can actually be bought at Amazon and Newegg? Those prices are $579 and $569 respectively when not on sale. It is deceptive to keep listing it at $488.