Puget Systems print logo

https://www.pugetsystems.com

Read this article at https://www.pugetsystems.com/guides/588
Share:


Synthetic & Scientific Benchmarks

Haswell-E 5960X 5930K 4960X benchmark - Geekbench 3 single core

Starting with the single core portion of GeekBench 3, we see the first evidence of performance issues with the 5960X. Since it only has a clock speed of 3.0-3.5 GHz, it has a lower GeekBench score than either the 5930K or 4960X. What is interesting is that the 5930K outperforms the 4960X even though it's maximum turbo frequency is .4 GHz lower. This is a good sign for the Haswell-E architecture as a whole and suggests that the 5930K may outperform both the 5960X and 4960X in single-threaded applications.

One thing we want to point out is the low memory score on both the 5960X and 5930K. These CPUs use the new DDR4 memory running at 2133MHz versus the DDR3 memory running at 1600MHz we used with the 4960X. One benchmark is obviously not conclusive, but this does suggest that DDR4 memory might not have much of a performance advantage over DDR3 quite yet.

Haswell-E 5960X 5930K 4960X benchmark - Geekbench 3 multi core

Moving on to the multi core portion of GeekBench 3, we see where the 5960X really shines. With eight cores, it easily outperforms both the 5930K and 4960X. Interestingly, the 5930K performs about the same as the 4960X in this test even though its single core performance was better.

Unlike the single core benchmark results, the memory score with DDR4 on the Haswell-E CPUs is actually higher than the 4960X.

Haswell-E 5960X 5930K 4960X benchmark - CineBench R15 CPU

CineBench R15 is a great rendering benchmark based on MAXON's CINEMA 4D software that measures single and multi core performance as well as OpenGL graphics performance. In this benchmark, all three CPUs get effectively the same score on the single core portion of the benchmark, but the 5960X really pulls ahead in the multi core portion. 

Haswell-E 5960X 5930K 4960X benchmark - CineBench R15 OpenGL

The OpenGL portion of CineBench is supposed to measure the performance of the GPU so we didn't expect to see much difference between the three CPUs. Oddly, both of the Haswell-E CPUs outperformed the 4960X by a healthy margin. This may just be an aberration, or it may be an indicator that GPU intensive tasks might see a benefit when using a Haswell-E CPU.

Haswell-E 5960X 5930K 4960X benchmark - Linpack

Linpack is a standard benchmark in the scientific community that measures a system's floating point computing power. This benchmark can take advantage of the AVX2 instruction set that is included in Haswell-E which is clearly shown by the large increase in performance. The 5930K is anywhere from 27-75% faster than the 4960X and the 5960X is anywhere from 45-114% faster depending on the problem size.

This is amazing for the scientific community, but we want to be clear that for the average user this is not really an indicator that Haswell-E will double your performance. To find out how much of a performance increase you could expect in a variety of situations, lets move on to some real-world benchmarks. 

Application Benchmarks

Haswell-E 5960X 5930K 4960X benchmark - Adobe Photoshop CC

To benchmark the performance of Photoshop, we applied a number of effects to both a 110MB and 1GB image and recorded how long it took for each effect to finish. These effects include CMYK Color Conversion, RGB Color Conversion, Ink Outlines, Dust and Scratches, Watercolor, Stained Glass, Lighting Effects, Mosaic Tiles, Extrude, Smart Blur, UnderPainting, Palette Knife, and Sponge.

The 110MB image shows roughly a 5% drop in performance with the Haswell-E CPUs compared to the 4960X. The 1GB image also shows a drop in performance, but the 5960X is a bit better than the 5930K. What this means is that if you use Photoshop a lot and already have a 4960X CPU, you probably should not upgrade to a Haswell-E CPU.

Haswell-E 5960X 5930K 4960X benchmark - Adobe Premiere Pro CC

Unlike Photoshop, Premiere Pro appears to like the new Haswell-E CPUs. The parts of the benchmark that uses GPU acceleration doesn't show much difference between the three CPUs, but the portion that is CPU only shows almost a 30% performance increase with the 5960X compared the the 4960X. The 5930K is also better than the 4960X, but only by about 5%.

Haswell-E 5960X 5930K 4960X benchmark - AutoDesk AutoCAD 2015

The Cadalyst benchmark for AutoCAD 2015 is interesting because it shows a slight performance increase with Haswell-E on the CPU and 2D Graphics tests, but a larger performance increase in the 3D Graphics portion. This increase in 3D performance is reminiscent of the OpenGL performance in CineBench we saw earlier and is something we did not expect to see.

Encoding Benchmarks

Haswell-E 5960X 5930K 4960X benchmark - Handbrake Encoding

Handbrake is an open source video transcoder that reports performance in the number of frames it is able to process per second. As you can see, the 5930K is almost identical in performance to the 4960X. The additional cores on the 5960X, on the other hand, give it a nice ~10% performance boost over the 4960X.

Haswell-E 5960X 5930K 4960X benchmark - Lame Encoding

Unlike Handbrake, Lame is not multithreaded so the number of CPU cores you have does not affect performance. Because of this, both the 5930K and 5960X (both of which have a lower frequency than the 4960X) show about a 6-9% drop in relative performance.


< Previous Next >
Tags: X99, Haswell-E, X79, Ivy Bridge-E, 5960X, 5930K, 4960X, Benchmark
Stark

A processor with 8 cores is good enough for most of people. So hopefully next gen will boost its clock to improve single core performance instead adding more cores.

Posted on 2015-07-03 19:12:56
John Og Elisabeth Andersen

I am pretty sure that the way it will go is more cores slower speeds for multiple reasons. One of the main reasons being, that we can make transistor smaller.

Visit my build here: http://computer.bazoom.com/dk/...

Posted on 2015-10-17 10:27:58
Daniel Padia

they did the exact opposite of what you said, lol. 6950x

Posted on 2016-06-01 23:41:54
xostrowx1991

Well to be fair they DID improve single core IPC performance by roughly 10%. A 4.4ghz i7 6800K is easily equal to a 4.7ghz i7 5820K. I've seen some people on overclock.net saying their ~4.3ghz i7 6900K 8 cores are as fast as their golden silicon lottery winning i7 5960X running at 4.7ghz. That's a "400mhz increase" in performance so to speak. (i.e. The broadwell-e gives identical performance to the haswell-e despite running roughly 400mhz lower max clock speed) And since, for example, the i7 6900K runs at a stock speed of 3.7ghz we can find out that my 10% estimate is pretty accurate, for example, 10% of 3700mhz/3.7ghz = 370mhz. If we add 370mhz to 3700mhz we get 4070mhz or 4.07ghz. That's DAMN close to the actual "400mhz" performance difference i just said people are seeing with Broadwell-E (technically its 10.8%, but that's close enough. i did say ROUGHLY lol)

So if we combine that knowledge of 10.8% increased performance on Broadwell-E and then factor in that the 8 core i7 6900K runs 200mhz faster than 5960X (3.7ghz boost clock speed compared to the 3.5ghz boost on 5960X) we can actually see the difference of what a "stock" 6900K would have over a stock 5960X.

3700mhz * 110.8% = Speed that a 5960X would have to run at to be as fast as a stock 6900K

3700mhz * 110.8% = 4099mhz, (i.e. 4.1ghz)

This means that a stock 3.7ghz 6900K is as fast as a 4100mhz or 4.1ghz 5960X. And since most Broadwell-E chips overclock to 4.3ghz or 4.4ghz if you push voltage (good samples can reach 4.4ghz at safe ~1.3v ranges sometimes) we can compare overclocked too

4300mhz * 110.8% = Speed that an overclocked 5960X would have to run at to be as fast as an overclocked 6900K

4300mhz * 110.8% = 4764mhz (i.e. 4.764ghz) That's a 476mhz increase!

So lets round the number and just say that a 4.3ghz Broadwell-E 6900K is as fast as a 4.75ghz 5960X!!

That's not a bad performance increase frankly, at least considering Intel's slow crawl of CPU gains lately. If you think about it, this is BROADwell, NOT SKYLAKE! But Broadwell-E is having as big of a performance increase over Haswell-E as we got moving from Haswell 4790K to Skylake 6700K!!

And if you can get a "gold silicon lottery" chip you're in an even BETTER situation comparing Broadwell-E to Haswell-E. Let's try a 4.4ghz which we know is possible at just over 1.3v on "good lottery chips", and then since HWBot has already shown a record of 4.6ghz on a 6800K we can assume 4.5ghz is possible at ~1.4v which is still "relatively" safe, so we'll use that too.

4.4ghz Broadwell-E * 110.8% = 4,875mhz!!! So a 4.4ghz Broadwell-E 6900K is equal to a whopping 4.875ghz 5960X!!! I don't think i've seen a SINGLE 5960X hit over 4.7ghz even on water cooling, only with phase change or liquid nitrogen!

Now, the finale. The "golden lottery winner" 4.5ghz ~1.4v chip

4.5ghz Broadwell-E * 110.8% = 4,986mhz!!!!! So a 4.5ghz Broadwell-E 6900K is equal to a fucking MASSIVE ~5GHZ 5960X!!!!!

Posted on 2016-06-20 09:00:05

Ok, now I feel bad about the used 5960x I just bought, thanks. :D

Posted on 2016-07-20 09:22:22
Mike

Which one of these benchmarks is most like calculating formulas in a massive Excel 2010 32-bit spreadsheet? If I had to guess, many upgrade taking place in the work place are inspired by the frustration of watching a slow calculation every time you open, make changes to or save a massive spreadsheet.

Posted on 2015-07-08 16:28:08
nashathedog

A good review with interesting results, Thanks.
It would of been nice to see some overclocking data too though. :)

Posted on 2015-07-22 11:55:37