For the unaware, IPC (instructions per cycle) provides a adept indicator of how fast a processor is and having both a high IPC with a high operating frequency is the best combination for maximum operation. Such is the example for Intel'south eighth-Gen Coffee Lake CPUs, and although AMD is conspicuously trailing when information technology comes to frequencies, the visitor appears to have really closed in on Intel's IPC functioning. That's likely the reason why and so many of you accept been asking for this kind of test.

To come across how much headway AMD has made here, we're going to limit as many variables every bit we can, while as well keeping things as realistic every bit possible. The outset and most obvious step is to remove core frequency from the equation and to exercise this we've locked all of the CPU cores at 4GHz. Any type of boost applied science has been disabled and the cores cannot go past 4GHz.

The second-gen Ryzen CPUs were tested on the Asrock X470 Taichi Ultimate and the Coffee Lake CPUs were on the Asrock Z370 Taichi. Both configurations used the same M.Skill FlareX DDR4-3200 memory with the 'Xtreme' memory profile and the same MSI GTX 1080 Ti Gaming X Trio for all the testing.

We can say upfront that this article is in no style buying advice, but we're testing purely for the science of information technology.

The Coffee Lake CPUs have a clear clock speed advantage. For real-world functioning, please refer to our contempo Ryzen 5 2600, 2600X and 2700X reviews.

For this exam we've included results for the Intel Core i7-8700K, 8600K, Ryzen 7 2700X, 2600X and Ryzen vii 1800X along with the 1600X. Now, the 1600X, 2600X and 8700K all have the same CPU resources: 6 cores with 12 threads.

The 1800X and 2700X take an advantage being that they are 8-core/16-thread CPUs while the 8600K is at a disadvantage as information technology'southward a half dozen-cadre/6-thread CPU, so please keep all that in mind as we continue. Let'southward become to the results.

Benchmarks

Starting with the sustained memory bandwidth exam nosotros run across that the first and second-gen Ryzen CPUs are similar with a bandwidth of about 39GB/due south. Meanwhile, using the exact same memory, the Coffee Lake CPUs are limited to around 33GB/s and this is a xv% reduction in bandwidth when compared to the Ryzen CPUs.

Moving to Cinebench R15 we see that the 2600X scores 4% college than the 1600X for the multi-threaded exam and 3% higher for the unmarried thread score. Then as we look at the 8700K we run into that information technology's iv% faster than the 2600X for the single thread score only four% slower for the multi-threaded score.

As you might take expected, clock-for-clock the 8-cadre/sixteen-thread Ryzen CPUs easily beat out the multi-threaded score of the 8700K. I included them only considering I had the results. Depending on demand, I could update this examination with the Core i7-7820X for example.

Next up we take the PCMark 10 video editing scores and this is a more lightly threaded test though we did previously run into a noteworthy deviation between the 1600X and 1800X. Every bit a consequence nosotros run across a solid 10% leap from the 1600X to the 2600X and this places AMD on par with Intel in terms of IPC performance for this test.

Like what we saw with Cinebench R15, when maxed out SMT appears more efficient than Intel's HT technology. Here the 1600X was faster than the 8700K by a 3.5% margin while the 2600X was 8% faster and that'southward a noteworthy margin right there.