Intel Core i9-7980XE & 7960X Evaluation

The Skylake-X series is finally finished. After hastily announcing the 18-core Core i9 at Computex 2017 last May, we now have the new 16- and 18-core processors from Intel at hand four months later.

We have no doubt that Intel had planned to release Skylake-X all the time, but did they plan to offer anything higher than a 12-core part before they got the wind from AMD's Threadripper?

They are also unlikely to announce the X299 platform in late May and plan to release it in June, as evidenced by the rushed motherboard development and resulting chaos.

We received the first Intel Core i9 part in late June – the $ 1,000 10-core 7900X – along with some Core i7 models with 8, 6, and even 4 cores, but it wasn't until August that we learned the official specs for Intel's new ones Processors with 12 cores for $ 1,200, 14 cores for $ 1,400, 16 cores for $ 1,700 and 18 cores for $ 2000, of which we will review the last two today.

Before we delve into the Core i9-7980XE and 7960X, we should remember that during the mayhem of Intel's latest desktop platform release on August 10th, AMD launched its Ryzen Threadripper series including the 1950X and 1920X brought CPU, which shamed Intel's then flagship 7900X.

As we'll see shortly, more should come from Intel, but at the time we were wondering whether the $ 1,200 7920X with 12 cores could even beat the 1950X and suspected the $ 1,400 7940X with 14 cores probably with AMD's 16-core CPU. While we don't have 12-core or 14-core models, Intel has provided the extremely expensive 16-core and 18-core chips to give us an idea of ​​where the 12- and 14-core Versions will be located. and of course we are always excited to see what the flagship parts have to offer.

To summarize briefly: The Core i9-7980XE contains 18 cores and, with its support for Hyper-Threading, advertises crazy 36 threads. Each core has its own 1 MB L2 cache, so there is a total of 18 MB L2 cache. There is now an L3 cache with a value of 24.75 MB, and due to the restructuring of the cache for the Skylake-X architecture, the 7980XE only has a little more L3 than the previous generation 6950X.

The 7980XE operates at a base frequency of 2.6 GHz with a Turbo Boost 2.0 frequency of 4.2 GHz and a Turbo Boost Max 3.0 frequency of 4.4 GHz. The 16-core 7960X has a base clock speed of 2.8 GHz with the same turbo boost frequencies. It has a slightly smaller 22 MB L3 cache, while there is a total of 16 MB for the L2. Both CPUs offer the full 44 PCIe lanes, quad-channel memory support and a TDP power of 165 watts.

Both CPUs use the LGA2066 socket and are supported by existing X299 motherboards. For testing, I use the Gigabyte Aorus X299 Gaming 9 with 32 GB G.Skill TridentZ RGB DDR4 memory that is clocked at 1600 MHz, i.e. DDR4-3200 with CL14 timings. The GTX 1080 Ti was used for all tests, while we used Vega 64 Liquid Cooled for some gaming tests. The focus will of course be on the productivity tests, after all, it is 18 and 16 core CPUs.

Benchmark time

Storage and application performance

First, let's look at the memory bandwidth performance. These DDR4 four-channel memory controllers appear to be suitable for a memory bandwidth of approximately 62 to 64 GB / s when using 3200 memory. Please note that all configurations have been tested with the same DDR4-3200 CL4 memory.

The multithread test from Cinebench R15 gives us a good idea of ​​what the productivity tests will look like when all cores and threads can be fully used. Here we see that the 7960X is only 4% faster than the 1950X for the multithreaded workload, both of course are 16 core parts. The 7980XE is 10% faster than the 1950X with an impressive score of 3317 points, although it costs twice as much at $ 2,000 – we'll surely talk more about pricing at the end of the test.

When it comes to single-thread performance, the higher clocked Skylake-X parts can achieve the best thread ripper.

Before we move on to the more serious benchmarks, I just wanted to see how these extreme CPUs exist in PCMark 10, which deals with office-type workloads (usually not core-heavy). To my surprise, Threadripper did well in this test and performed significantly better than the Skylake-X CPUs. The 7980XE looked particularly weak here. Of course, none of these CPUs is slow for these more general office tasks, so this is somewhat irrelevant.

Excel offers us an office benchmark that can use many threads, especially when running the extreme Monte Carlo simulation. Here the Threadripper 1950X impressed with only 1.66 seconds. However, the 16-core and 18-core Skylake-X parts are better by one because they drop in the 1.4 second range, so the 18-core part can do the workload 14% faster.

For those of you who don't know, VeraCrypt is an open source utility used for on-the-fly encryption that offers optimized implementations of cryptographic hash functions and ciphers that increase performance on modern CPUs . It also supports parallelized encryption for multi-core systems as well as hardware accelerated AES to further improve performance.

In short, it takes full advantage of the many cores that these CPUs offer, and that's what we see here. However, the Threadripper 1950X delivered the best results, and even the 7980XE couldn't beat it – a disappointing result for Intel.

Next we have unencrypted compression and decompression performance with 7-Zip. While hyper-threading has a similar efficiency when compressing and decompressing, SMT is much more efficient for decompression work.

For decompression, the Threadripper 1950X is again able to outperform the 7980XE, although it was much slower for the compression test. If you do a lot of compression, it looks like Intel CPUs with a high number of cores are king here.

Leave a Reply

Your email address will not be published. Required fields are marked *