On June 21, 2011 AMD announced it had resigned from the BAPCo organization after failing to endorse the SYSmark 2012 Benchmark.

Press question mark to learn the rest of the keyboard shortcuts Log In Sign Up User account menu 211 Intels Real World Benchmarking: SYSmark 2018 is (far) more in favor of Intel - as Cinebench is in favor of AMD.

XFastest ( Google-translated to english ) showing some of the Intel Real Usage Guidelines presentation slides, mostly interesting is slide 17.

Within this presentation, Intel promote SYSmark 2018 as the best benchmark to test application performance, because SYSmark include some widely used programs like Office Adobe software.

Intel claims as well a performance advantage for Core i7-9700K (3) and Core i9-9900K (7) over Ryzen 9 3900X at the SYSmark 2018, called it a benchmark win at application performance.

On the other side, Intel dismiss the Cinebench benchmark as too much in favor of AMDs CPUs, calling the Cinebench results as outliers in review of other benchmarks.

As we have a bunch of independent benchmarks from the Ryzen 3000 launch, its easy to check Intels claims.

First, I created an application performance index without any of the rendering software, i.e.

D Studio, Blender, Cinebench, Corona, FryBench, Indigo, KeyShot, LuxMark, PCMark rendering test, POV-Ray V-Ray.

The new index include mostly office, browser, packer and encoding software.

On the left side of the following table you see the original index (including the results from rendering benchmarks), on the right side of the table the new index without any of the rendering benchmark.

The base (100) for all results is the Ryzen 9 3900X, so you can compare it to Intels claims.

Yes, the rendering software is in favor of AMD, but the results from all other tests are not so different.

Core i7-9700K and Core i9-9900K are still far away from the performance level of Ryzen 9 3900X at applications.

In fact, even without rendering benchmarks, every single review still see a huge performance advantage for the Ryzen 9 3900X at application performance.

Intels SYSmark 2018 results showing some very much different results, so as second I investigate the deviation of SYSmarks and Cinebenchs results from the new performance index without any rendering software.

To use this new index is clearly in favor of Intel, because usually you would use the original index with all the test results for any performance consideration.

But in this case, only the deviation from the index without SYSmark and without Cinebench (or other rendering software) is in the point of view.

The base (100) for all results is still the Ryzen 9 3900X, the deviation of SYSmark and Cinebench to the overall application performance index is noted as percent point instead of percent (higher percent points means a higher deviation from the overall application performance index):..

Overall Cinebench R20 (M) (deviation) SYSmark 2018 (deviation) Core i9-9900K 76.6 62.1 23 percent points 107 40 percent points Core i7-9700K 63.8 48.9 31 percent points 103 61 percent points.

This is the second clear result: SYSMark 2018 differs far more from the overall application performance index (without rendering software) than Cinebench.

If Cinebench is called for in favor of AMD, then the SYSmark should be called as far more in favor of Intel.

Like it or not: Cinebench is (clearly) nearer on the overall application performance of these CPUs than Intels preferred SYSmark..

Bapco Sysmark 2018 Free To Do

PS: If someone want to create an info graphics based on these numbers, please feel free to do so.

Source: This is a short version of an article from my german website 3DCenter.org 111 comments share save hide report 89 Upvoted This thread is archived New comments cannot be posted and votes cannot be cast Sort by best.

From Wikipedia: BAPCO has suffered criticism for bias in its benchmarking products.

Bapco Sysmark 2018 Series Of CPU

It was found in 2002 that Intel was the sole contributor to a series of CPU tests, tests which heavily favoured their own CPUs vs competitors, where the tests of the year before performed significantly better on non-Intel parts3.

Intel was investigated by the FTC, and eventually fined for this action, among other anti-competitive measures It was around this time that AMD and Nvidia left the consortium.