Intel Core i9-7960X & i9-7980XE Extreme Edition Sky Lake-X LGA2066 X299 CPU Review: Page 4 of 6
Posted by Damon Bailey on Monday, September 25, 2017 - 9:01am
System Configuration & Performance Testing
ASUS ROG Rampage VI Extreme
GeIL EVO Potenza 32GB (4x8GB) 3200 MHz DDR4 Memory
Samsung 850 Evo 250GB Solid State Drive
Galax GTX 1070 HOF
Custom Liquid Loop
EKWB Supremacy EVO
EKWB PE240 Radiator
Monsoon MMRS Pump/Reservoir with EKWB D5 Pump
2x Corsair ML120 PRO Fans
Testing Operating System and Utilities:
Microsoft Windows 10 Professional
General System Benchmarks:
Cinebench R15 – CPU
Passmark Performance Test 8 – Memory
Grand Theft Auto: V
Metro: Last Light
General Performance and CPU Performance Results
General Performance Results
"Use the PCMark 8 Storage benchmark to test the performance of SSDs, HDDs and hybrid drives with traces recorded from Adobe Creative Suite, Microsoft Office and a selection of popular games. You can test the system drive or any other recognized storage device, including local external drives. Unlike synthetic storage tests, the PCMark 8 Storage benchmark highlights real-world performance differences between storage devices."
Likely due to PCMark 8 needing an update to be compatible with the new CPU’s, we were unable to complete creative test runs, but our Home and Work scores are pretty good considering the comparatively low base clock speeds compared to our other processors in the list. The Creative test favors higher thread counts, so we expect the two CPU’s to lead the pack here once the program is updated.
Super PI Modded 1.5
“In August 1995, the calculation of pi up to 4,294,960,000 decimal digits was succeeded by using a supercomputer at the University of Tokyo. The program was written by D.Takahashi in collaboration with Dr.Y.Kanada at the computer center. This record should be the current world record. (Details are shown in the windows help.) This record-breaking program was ported to personal computer environments such as Windows NT and Windows 95. In order to calculate 33.55 million digits, it takes within 3 days with a Pentium 90 MHz, 40 MB main memory and 340 MB available storage.”
Our Super Pi scores here are actually a bit surprising. Being a single-threaded benchmark, we expected the relatively low clock speeds to drag down the score, but a single core can Turbo Boost well into the 4GHz range on each CPU so the scores are actually pretty good.
CPU Performance Results
CINEBENCH R15 – Multi-Threaded
"CINEBENCH is a real-world cross platform test suite that evaluates your computer's performance capabilities. CINEBENCH is based on MAXON's award-winning animation software CINEMA 4D, which is used extensively by studios and production houses worldwide for 3D content creation. MAXON software has been used in blockbuster movies such as Iron Man 3, Oblivion, Life of Pi or Prometheus and much more.
CINEBENCH is the perfect tool to compare CPU and graphics performance across various systems and platforms (Windows and OS X). And best of all: It's completely free."
Cinebench is the first multi-threaded benchmark in our lineup and the i9-7980XE just decimates everything else in our listing. Its sixteen core sibling is hot on its heels here though. These SKU’s deliver workstation-class performance levels that were only achievable with dual socket systems until recently.
This form allows you to measure the performance of your computer. The benchmark shows a rating in MIPS (million instructions per second). The rating value is calculated from the measured speed, and it is normalized with results of Intel Core 2 CPU with multi-threading option switched off. So if you have modern CPU from Intel or AMD, rating values in single-thread mode must be close to real CPU frequency. There are two tests, compression with LZMA method and decompression with LZMA method. Once the total passes reach 100, the score is taken.
Again we see the massive amount of cores in the i9-7980XE set the grade curve here, besting every other CPU by a small landslide except its smaller 16 core counterpart. Even the powerful ten-core i9-7900X is left in the dust here.
Memory Performance Results
AIDA64 Engineer Edition
“FinalWire Ltd. today announced the immediate availability of AIDA64 Extreme Edition 1.50 software, a streamlined diagnostic and benchmarking tool for home users; and the immediate availability of AIDA64 Business Edition 1.50 software, an essential network management solution for small and medium scale enterprises. The new AIDA64 update implements AVX-optimized benchmarks for the upcoming Intel Sandy Bridge processors, adds a brand new video encoding benchmark, and supports the latest AMD and NVIDIA graphics processors.”
*note* as of this writing, we are using software beta version 5.92.4358 which still does not yet fully support these CPU’s. We receive the below warning on every benchmark run on both the i9-7960X and the i9-7980XE, so scores could significantly be affected.
As we can see here, Both top end chips full in phenomenal amounts of bandwidth from the quad channel memory. Interestingly, both the 7960X and 7980XE manage slightly better write bandwidth than our recently tested Core i7-7820X which also used 3200MHz quad channel memory.
While the bandwidth is increased slightly, it did not seem to significantly impact latency as both CPU’s are within a few nanoseconds of our other Sky Lake-X CPU’s.
Internally to the CPU, Level 1 Cache speeds are significantly raised on this generation, and this scales nearly linearly with core count with our 18 core i9-7980XE chip hitting a blistering 3.7TB/s of reading and copy, and nearly 1.9 TB/s write bandwidth. Our Core i9-7960X trails by a few percent as expected, but still flattens all of our other CPUs quite easily.
We see pretty good Level 2 Cache bandwidth scaling here with the i9-7960X but we seem to be hitting a snag with our eighteen-core Extreme Edition CPU. This likely stems from this version of AIDA64 not being quite ready for this CPU as noted above. We’ll have to revisit this testing once the good folks at Finalwire have a chance to do some updating.
And one final time we are seeing an issue with our testing software. Our originally released i9-7900X shows staggering performance, but our 16 and 18 core CPU’s are not quite reading correctly.
Passmark Performance Test 8 - Memory
We see strong performance from our Passmark Threaded Memory Mark test here. Both CPU’s are within a margin of error of each other while hovering right around the 70,000 mark but due to the ram kit used on our Core i7-7820X test having lower latencies at the same speed, it still holds first place.
“SiSoftware Sandra provides a robust package of diagnostic tools for testing your system and teasing out its problems--or potential headaches.”
Latency doesn’t matter as much here, The i9-7980XE and i9-7960X just edge out the previous champion in our SiSoft Sandra Memory Bandwidth test to hold down the lead. The Extreme Edition just holds the edge here over its sixteen core counterpart.
Written by Alexander J. Lee “From a high-school project that went a little too far...” y-cruncher, (y for gamma) is a number-crunching program that can compute various mathematical constants.
It was originally a small program specialized for computing the Euler-Mascheroni Constant. (Which uses gamma as its symbol, hence the name). It has since gained the ability to compute other constants.
It is the first of its kind that is multi-threaded and scalable to multi-core systems. Ever since its launch in 2009, it has become a common benchmarking and stress-testing application for overclockers and hardware enthusiasts.
We tested using the built-in benchmarks to compute Pi to 1 Billion Digits in both single and multi-threaded mode and recorded the Total Computation time at the end of the run. Y-Cruncher is also able to take advantage of AVX512 instructions incompatible CPU’s. Due to the thermal load of running Intense AVX workloads, Intel has implemented and AVX offset clock that may scale the CPU’s clock speed back, even below stock speeds when running intense AVX loads. For this reason, we note the CPU’s clock speed in both single and multi-threaded test modes.
We have some interesting results here. The i9-7980XE paves the way with 36 AVX512 enabled threads with a blistering 36.7 Seconds to compute 1 Billion digits of Pi, But of our AVX512 capable samples, turns in the lowest single thread performance due to its lower clock speed.
Our thirty-two threaded i9-7960X lags its bigger sibling by only two-tenths of a second in multi-threaded compute but actually, snags the fastest single threaded time by nearly a second and a half over the Core i7-7820X that runs at a higher single threaded clock speed.
Synthetic Game and Non-Synthetic Game Performance Results
Synthetic Game Results
3DMark – Fire Strike
"The new 3DMark includes everything you need to benchmark your hardware. With three all new tests you can bench everything from smartphones and tablets to notebooks and home PCs, to the latest high-end, multi-GPU gaming desktops. And it's not just for Windows. With 3DMark you can compare your scores with Android and iOS devices too. It's the most powerful and flexible 3DMark we've ever created.
Fire Strike is a showcase DirectX 11 benchmark designed for today's high-performance gaming PCs. It is our most ambitious and technical benchmark ever, featuring real-time graphics rendered with detail and complexity far beyond what is found in other benchmarks and games today"
Interestingly the i9-7960X edges out the Extreme Edition here in all but the combined score. It appears the minor thread-count difference is more than offset by the increase in clock speeds. Both CPU’s still turn in the top two overall positions as expected with some of the highest CPU physics scores we have seen, even among overclocked CPU’s.
"Heaven Benchmark immerses a user into a magical steampunk world of shiny brass, wood, and gears. Nested on flying islands, a tiny village with its cozy, sun-heated cobblestone streets, an elaborately crafted dirigible above the expanse of fluffy clouds, and a majestic dragon on the central square gives a true sense of adventure. An interactive experience with fly-by and walk-through modes allows for exploring all corners of this world powered by the cutting-edge UNIGINE Engine that leverages the most advanced capabilities of graphics APIs and turns this benchmark into a visual masterpiece."
Unigen’s Heaven benchmark is not particularly CPU bound, so you won’t notice much change here between CPU’s unless there is a bottleneck, and thankfully we don’t see any here.
Non- Synthetic Game Results
Grand Theft Auto: V
"When a young street hustler, a retired bank robber, and a terrifying psychopath find themselves entangled with some of the most frightening and deranged elements of the criminal underworld, the U.S. government and the entertainment industry, they must pull off a series of dangerous heists to survive in a ruthless city in which they can trust nobody, least of all each other.
Grand Theft Auto V for PC offers players the option to explore the award-winning world of Los Santos and Blaine County in resolutions of up to 4k and beyond, as well as the chance to experience the game running at 60 frames per second."
Grand Theft Auto: V is the newest game in our benchmark suite. The game offers players a huge range of PC-specific customization options, including over 25 separate configurable settings for texture quality, shaders, tessellation, anti-aliasing and more. It has the highest recommended system requirements in our game benchmark suite.
With parts of Grand Theft Auto V being a bit CPU intensive, we do see a noticeable increase in average FPS moving up to the much higher core counts.
Metro: Last Light
"It Is the Year 2034. Beneath the ruins of post-apocalyptic Moscow, in the tunnels of the Metro, the remnants of mankind are besieged by deadly threats from outside – and within. Mutants stalk the catacombs beneath the desolate surface, and hunt amidst the poisoned skies above."
Developed by 4A Games and published by Deepsilver, Metro: Last Light uses the 4A game engine. At its highest settings, the 4A game engine is capable of bringing all but the most extreme gaming systems to their knees.
Metro: Last Light is not programmed to take advantage of a high number of cores, like most games on the market. Due to this, the lower clock speeds hurt benchmark performance more than the thread count helps it.
"Tomb Raider explores the intense and gritty origin story of Lara Croft and her ascent from a young woman to a hardened survivor. Armed only with raw instincts and the ability to push beyond the limits of human endurance, Lara must fight to unravel the dark history of a forgotten island to escape its relentless hold."
Tomb Raider was developed by Crystal Dynamics and published by SQUARE ENIX, Eidos Interactive. It features a modified version of the Crystal game engine and was the first game to integrate AMD's TressFX 2.0, which adds hair, fur and grass physics.
Tomb Raider shows similar results, the lower clock speeds just don’t place these monster chips near the top as the game simply can’t take advantage of the high thread count.