The Intel Broadwell-E Review: Core i7-6950X, i7-6900K, i7-6850K and i7-6800K Tested
by Ian Cutress on May 31, 2016 2:01 AM EST- Posted in
- CPUs
- Intel
- Enterprise
- Prosumer
- X99
- 14nm
- Broadwell-E
- HEDT
Linux Performance
Built around several freely available benchmarks for Linux, Linux-Bench is a project spearheaded by Patrick at ServeTheHome to streamline about a dozen of these tests in a single neat package run via a set of three commands using an Ubuntu 11.04 LiveCD. These tests include fluid dynamics used by NASA, ray-tracing, OpenSSL, molecular modeling, and a scalable data structure server for web deployments. We run Linux-Bench and have chosen to report a select few of the tests that rely on CPU and DRAM speed.
C-Ray: link
C-Ray is a simple ray-tracing program that focuses almost exclusively on processor performance rather than DRAM access. The test in Linux-Bench renders a heavy complex scene offering a large scalable scenario.
NAMD, Scalable Molecular Dynamics: link
Developed by the Theoretical and Computational Biophysics Group at the University of Illinois at Urbana-Champaign, NAMD is a set of parallel molecular dynamics codes for extreme parallelization up to and beyond 200,000 cores. The reference paper detailing NAMD has over 4000 citations, and our testing runs a small simulation where the calculation steps per unit time is the output vector.
Redis: link
Many of the online applications rely on key-value caches and data structure servers to operate. Redis is an open-source, scalable web technology with a strong developer base, but also relies heavily on memory bandwidth as well as CPU performance.
[words]
Facebook
Twitter
RSS
205 Comments
View All Comments
RussianSensation - Tuesday, May 31, 2016 - link
It's a slap in the face when 6850/6900/6950X are also crap overclockers and will get owned hard in games and every day tasks by a $310 6700K. The only CPU that even remotely makes sense is the 6800K. For workstation use case, dual Xeons will smash the 6950X. Heck, it's better to build a 6900K + 6700K in the same case allowing one to be productive and game at the same time. Phanteks makes such cases now. 6950X is just a way to show your status, nothing more.mapesdhs - Thursday, June 9, 2016 - link
Something usually missing from reviews now is an oc'd 4820K, which is annoying because a 4c IB-E on X79 allows for quite a lot of oc headroom given the high rating of the socket and the beefy power delivery available on boards like the R4E, etc. I bet it would give many of the newer CPUs a serious pelting.Drazick - Tuesday, May 31, 2016 - link
Could we have Extreme Edition with Iris Pro + 128MB eDRAM?That would be a great addition (Even only for the 6 Cores Part).
Eden-K121D - Wednesday, June 1, 2016 - link
Iris pro would be useless but i agree with the eDRAM acting as L4 Cachebarleyguy - Tuesday, May 31, 2016 - link
Great review.One possible omission though: You mentioned that the Xeon E-2640 is a better deal as far as price/performance, but there are no Xeons on the benchmark charts. Do you plan to review the E-2640 at some time in the future?
Thanks.
ShieTar - Tuesday, May 31, 2016 - link
That might indeed be a statement which needs to be proven by tests?The 2640 has 10 Broadwell-Cores at a 2.8GHz All-Cores-Turbo, the 6950X should have a 10-core-Turbo of 3.2GHz, so you might argue you get 87.5% of the Performance for ~60% of the CPU cost. But the 2640 does have a slower verified memory speed, which may have a little impact. And its turbo boost settings are defined to hit a 90W TDP, and I don't think you can change that even in a workstation with plenty of cooling available. Add to that the fact that you can overclock to improve the performance of the 6950X, and that the 700$ price difference should be considered relative to the overall system cost, and you probably end up with very similar price-to-performance ratios.
I think the stronger challenger to the 6950X price-to-performance figures is the 2687W v4, which can be had for just over 2k$, and gets you annother 2 cores at almost the same clocks. That's ~16% more performance for ~16% more CPU cost, which translates into less than 10% higher system cost.
samer1970 - Tuesday, May 31, 2016 - link
Hello ,we all know that games dont use more than 8 threads today ...
so to take advantage of an 8 cores or 10 cores CPU in Gaming you should Disable HT (Hyperthreading) and run the gaming test again to compare it against the 4 cores i7 6700K .
and test it with SLI as well to reach the i7 6700k bottleneck !
let me put it more simple ,
The i7 6700K has 4 cores and can oc to 4.4 ghz easy . this CPU will give us 8 Virtual cores comparable to 2.2 GHZ clock for each virtual core .
However the 8 Coresi7 6900K , With the HT Turned OFF , will give us 8 cores @ 4.4 ghz EACH !
Thats double the speed of the 4 cores i7 ! if the game uses 8 threads .
EVEN if we dont OC the 8 cores , it would be 3.2GHZ VS 2.2 GHZ !!!
if you ask why Disable HT ? simple because the game will never use 16 Virtual cores !!! and the advantage is LOST .
Please run the test again for games with HT turned off .
and to stress the CPU more , TEST SLI as well , we want the i7 6700K to bottleneck !
THANKS
oh and Intel Should release i5 Broadwel-E CPU , 8 cores without HT , CHEAPER and BETTER for GAMERS
RussianSensation - Tuesday, May 31, 2016 - link
Nice try but no cigar. 6700K @ 4.8ghz + HT is the optimal gaming CPU. No current game even scales linearly across 6 cores + HT. 8-10 core CPU with a slower architecture would lose badly to an i7 6700K @ 4.8Ghz + DDR4 4000.There is not a chance 6950X @ 4.4Ghz can keep up with 6700K OC.
http://www.techspot.com/article/1171-ddr4-4000-mhz...
By the time games use 8-10 cores, we'll be on PS5/XB2 generation in 2020-2021 and Icelake-E. Broadwell-E 8-10 cores will be outdated.
adamod - Wednesday, June 1, 2016 - link
http://www.pcworld.com/article/3039552/hardware/te...look at the ashes bench
mapesdhs - Thursday, June 9, 2016 - link
Oh grud not here aswell! You've been banging on about this HT thing on toms for ages.