SmartKevin logo

A Performance Review: AMD’s Ryzen 5 2600X & Ryzen 7 2700X Processors

AMD Second-gen Ryzen 7 Box

Date: April 19, 2018
Author(s): Rob Williams

Following a deluge of leaks, AMD’s 2nd-gen Ryzen ‘Pinnacle Ridge’ CPUs have landed. These new ‘Zen+’ chips are built on a 12nm process, and bring in a brand-new chipset (even though the older chipset will work just fine!). Read on as we take a look at both the 8-core Ryzen 7 2700X, and 6-core Ryzen 5 2600X.



Introduction, A Look At Second Gen Ryzen

AMD’s second-generation Ryzen CPUs have landed, and if you’re on the lookout for a new chip with a great bang-for-the-buck, you should give a chip like the Ryzen 7 2700X a hard look for your next build.

While I had plenty of time to test and write this review, other things (work and life) prevented me from making this as in-depth as I’d like it to be. I am actually writing the entire thing hours in advance of embargo, after retesting lagged me a bit. So, this article is no work of art, but the point definitely gets across through the variety of performance results.

I’ll say up-front that it’s unwise to look at Ryzen 2 as a revolutionary successor to the original series released last spring. AMD says that there will be around a 3% IPC boost, and of course, the 2700X is clocked higher than the 1800X was, so that contributes to boosts, too. No one on Ryzen 1st gen has reason to upgrade to Ryzen 2; they can safely wait until the actual “Zen 2” architecture arrives. Zen 2 will undoubtedly deliver higher IPC gains than gen 1 to gen 2 has.

Ryzen 7 2700X & Ryzen 5 2600X

That said, this isn’t a mere clock boost. The chips have been shrunk from 14nm to 12nm, built by GLOBALFOUNDRIES. That means they should run cooler, draw less power, and potentially, be better at overclocking. AMD says that with its “Zen+” refinement, cache latencies are much lower this gen, and even DRAM latency should see some improvement.

Those updates are nice, but again are no reason for people to upgrade from first- to second-gen Ryzen. What’s most notable about this launch to me is what’s available for your money nowadays. A couple of years ago, an eight-core CPU was limited to the $1,000 price-point, and the chip available at the time drew a lot more power than the 2700X does.

In other words, AMD broke the mold and released a $500 8-core chip last spring, to much fanfare. This spring, the company has released a slightly faster replacement, and is pricing it at $329. That’s almost half the 8-core X299 counterpart. That latter Intel part would win the performance contest, but from a value standpoint, AMD is a serious force to be reckoned with.

AMD Ryzen 7 2700X Installed In ASUS Crosshair VII HERO

With this launch, AMD is ushering in four Ryzen 2 chips to kick things off. Those include the Ryzen 7 2700X, 7 2700, 5 2600X, and 5 2600. The 2700 chips effectively replace the 1800X, and while AMD hasn’t made any mention about a future 2800X chip, it’s hard to believe that a gap isn’t being deliberately left here.

AMD Ryzen Desktop Processors
CoresClock (Turbo)L2+L3MemoryTDPPrice
Ryzen Threadripper
1950X16 (32T)3.4 GHz (4.0)8+32MBQuad180W$999
1920X12 (24T)3.5 GHz (4.0)8+32MBQuad180W$799
1900X8 (16T)3.8 GHz (4.0)4+16MBQuad180W$549
Ryzen 7
R7 2700X8 (16T)3.7 GHz (4.3)4+16MBDual105W$329
R7 27008 (16T)3.2 GHz (4.1)4+16MBDual95W$299
R7 1800X8 (16T)3.6 GHz (4.0)4+16MBDual95W$349
R7 1700X8 (16T)3.4 GHz (3.8)4+16MBDual95W$309
R7 17008 (16T)3.0 GHz (3.7)4+16MBDual65W$299
Ryzen 5
R5 2600X6 (12T)3.6 GHz (4.2)3+16MBDual95W$219
R5 26006 (12T)3.4 GHz (3.9)3+16MBDual65W$189
R5 1600X6 (12T)3.6 GHz (4.0)3+16MBDual95W$219
R5 16006 (12T)3.2 GHz (3.6)3+16MBDual65W$189
R5 1500X4 (8T)3.5 GHz (3.7)2+16MBDual65W$174
R5 14004 (8T)3.2 GHz (3.4)2+8MBDual65W$169
Ryzen 3
R3 1300X4 (4T)3.5 GHz (3.7)2+8MBDual65W$129
R3 12004 (4T)3.1 GHz (3.4)2+8MBDual65W$109
Ryzen w/ Radeon Vega Graphics
R5 2400G4 (8T)3.6 GHz (3.9)0.5+4MBDual65W$169
R3 2200G4 (4T)3.5 GHz (3.7)0.5+4MBDual65W$99

Also technically part of the “second gen” Ryzen are the APUs found at the bottom of this chart. Those include AMD’s Vega graphics on tap, and become some of the most attractive chips at their respective price-points. 100 clams for a quad-core at 3.5GHz with built-in graphics still impresses me.

Because I am under time constraints, I’m expediting this page, but before moving on, I do want to mention that StoreMI is one new feature worth looking at. It’s built by Enmotus, which creates the FuzeDrive “tiering” storage solution. StoreMI is AMD’s consumer variant, allowing software to automatically move your data between an SSD and hard drive. It’s similar to caching, except that there are no duplicates in data; it’s either in one place or another.

SmartKevin AMD Ryzen 2000 Series PC

Ultimately, the data you access most often will be put on the SSD. I talked to Enmotus a few weeks ago about its solution, and it claims that its solution can in some cases improve performance even over a straight SSD. One example given was with installing a fresh game to Steam. Often, that process lags; you’ll sit there waiting for the initial process to finish. Through its magic, Enmotus apparently makes processes like these even faster than using a straight SSD, although the real-world differences are not going to be dramatic. It’s interesting nonetheless, and does warrant some future testing.

Another thing worth stressing is that while AMD has released a new chipset for the new Ryzen series, it’s not explicitly required. You can still use X370 for a chip like the 2700X, but you’ll miss out namely on the updated XFR2 feature. You’ll still be able to use the StoreMI feature on the older boards, as I understand it (which makes sense, because the FuzeDrive solution isn’t specific to Ryzen). Likewise, you can use the old chips in the new boards. How sweet is it to have the same socket from one generation to the next?!

Test Systems & Methodologies

Benchmarking a CPU may sound like a simple enough task, but in order to deliver accurate, repeatable results, and not to mention results that don’t favor one vendor over another, strict guidelines need to be adhered to. That in turn makes for rigorous, time-consuming testing, but we feel that the effort is worth it.

This page exists so that we can be open about how we test, and give those who care about testing procedures an opportunity to review our methodology before flaming us in the comments. Here, you can see a breakdown of all of our test machines, specifics about the tests themselves, and other general information that might be useful.

Let’s start with a look at the test platforms, for AMD’s TR4 and AM4, along with Intel’s LGA2011-v3, and LGA1151. We’re still in the process of benchmarking every relevant current-gen chip we have over, so for now, only nine CPUs are included in these results.

To prevent unexpected performance results, the “Multi-Core Enhancement” optimizations (effectively overclocking all cores to max turbo, instead of just two cores) offered by ASUS and GIGABYTE on their respective motherboards is disabled. There is, however, an exception. For Ryzen 2 testing, I had to use the DOCP profile on the motherboard to take proper advantage of the memory kit.

Since the other platforms were tested using 32GB of 3200MHz memory at 16-18-18, I had hoped that I’d be able to do the same here. Not so, at least with the kit I have here, and to be fair, that’s not a huge surprise since it’s not explicitly touting Ryzen support. That said, the Threadripper rig managed to use the kit just fine.

After much testing, I ultimately could run this same kit at the required speeds only if I had two sticks installed. As soon as the others were introduced, the system would simply not boot. Perhaps I could have worked around this by overclocking, but work had to get done, so I couldn’t spend time fiddling with that.

Just before I decided to use that configuration as-is for testing, a good friend reminded me that Ryzen kits are generally optimized for Ryzen for a reason, and that I should stick to an approved kit. That being the case, I decided to stick with the same kit AMD supplied with the Ryzen 2 reviewer kit, even though it bumps the speeds from 3200 to 3400. As much as I’d love to keep things apples-to-apples, it’s simply not possible that often where memory’s concerned.

That all said, here’s the test platforms:

SmartKevin’s CPU Testing Platforms

AMD TR4 Test Platform
ProcessorAMD Ryzen Threadripper 1950X (3.4GHz, 8C/16T)
MotherboardGIGABYTE X399 AORUS Gaming 7
CPU tested with BIOS F3g (Oct 13, 2017)
MemoryCorsair VENGEANCE RGB (CMU32GX4M4C3200C16) 8GB x 4
Operates at DDR4-3200 16-18-18-36 (1.35V)
GraphicsNVIDIA TITAN Xp (12GB; GeForce 388.13)
StorageCrucial MX300 525GB (SATA 6Gbps)
Power SupplyEnermax RevoBron 80+ Bronze (600W)
ChassisEnermax Equilence
CoolingEnermax Liqtech TR4 AIO (240mm)
Et ceteraWindows 10 Pro (Build 16299), Ubuntu 17.10 (4.13 kernel)
NVIDIA GeForce GTX 1080 Ti - SLI Configuration NVIDIA GeForce GTX 1080 Ti - SLI Configuration
As tested configuration: AMD Ryzen Threadripper 1950X (Zen)
NVIDIA GeForce GTX 1080 Ti - SLI Configuration
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
As tested configuration: AMD Ryzen Threadripper 1950X (Zen)
Sony Ps4 Pro Angled View

There’s not too much to say here, which is a good thing. While we didn’t find a huge difference to be made with the Core optimization left on (default), we disabled it to keep things as “reference” as possible. I truly dislike the EFI GIGABYTE has equipped on this board, as it’s pretty clunky, and in my opinion, disorganized. But, where it lacks in EFI polish, the board has made up for in stability.


AMD AM4 Test Platform
ProcessorsAMD Ryzen 7 2700X (3.7GHz, 8C/16T)
AMD Ryzen 5 2600X (3.6GHz, 6C/12T)
MotherboardASUS Crosshair VII HERO Wi-Fi
CPU tested with BIOS 0508 (Apr 13, 2018)
MemoryG.Skill (F4-3400C16-8GSXW) 8GB x 2
Operates at DDR4-3400 16-16-16-36 (1.35V)
GraphicsNVIDIA TITAN Xp (12GB; GeForce 388.13)
StorageCrucial MX300 525GB (SATA 6Gbps)
Power SupplyEVGA Bronze 600B1 (600W)
ChassisFractal Design Define C
CoolingNoctua NH-U12S SE-AM4 (1x120mm)
Et ceteraWindows 10 Pro (Build 16299), Ubuntu 17.10 (4.13 kernel)
NVIDIA GeForce GTX 1080 Ti - SLI Configuration NVIDIA GeForce GTX 1080 Ti - SLI Configuration
As tested configuration: AMD Ryzen 7 2700X (Zen)
NVIDIA GeForce GTX 1080 Ti - SLI Configuration
NVIDIA GeForce GTX 1080 Ti - SLI Configuration NVIDIA GeForce GTX 1080 Ti - SLI Configuration
As tested configuration: AMD Ryzen 7 2700X (Zen)
NVIDIA GeForce GTX 1080 Ti - SLI Configuration
NVIDIA GeForce GTX 1080 Ti - SLI Configuration NVIDIA GeForce GTX 1080 Ti - SLI Configuration
As tested configuration: AMD Ryzen 5 2600X (Zen)
NVIDIA GeForce GTX 1080 Ti - SLI Configuration

Nothing special had to be done on the Crosshair VII HERO to get up and running quickly. As covered above, I had to run with the DOCP Standard profile to get the memory to run at optimum settings. I’m not entirely sure if this means the CPU is slightly overclocked, because really, it’s hard to even tell nowadays given the tricks both motherboard and processor vendors employ.

Ultimately, my Cinebench scores came close to the reference results AMD supplied, so ultimately I feel pretty good that this platform is running as it should for any regular owner.


Intel LGA2011-3 Test Platform
ProcessorsIntel Core i9-7980XE (2.6GHz, 18C/36T)
Intel Core i9-7960X (2.8GHz, 16C/32T)
Intel Core i9-7900X (3.3GHz, 10C/20T)
Intel Core i7-7820X (3.6GHz, 8C/16T)
Intel Core i7-7740X (4.3GHz, 4C/8T)
MotherboardASUS ROG STRIX X299-E GAMING
CPU tested with BIOS 1004 (Nov 14, 2017)
MemoryCorsair VENGEANCE RGB (CMU32GX4M4C3200C16) 8GB x 4
Operates at DDR4-3200 16-18-18-36 (1.35V)
GraphicsNVIDIA TITAN Xp (12GB; GeForce 388.13)
StorageCrucial MX300 525GB (SATA 6Gbps)
Power SupplyCorsair Professional Series Gold AX1200 (1200W)
ChassisCorsair Carbide 600C
CoolingNZXT Kraken X62 AIO (280mm)
Et ceteraWindows 10 Pro (Build 16299), Ubuntu 17.10 (4.13 kernel)
NVIDIA GeForce GTX 1080 Ti - SLI Configuration NVIDIA GeForce GTX 1080 Ti - SLI Configuration
As tested configuration: Intel Core i9-7980XE (Skylake-X)
NVIDIA GeForce GTX 1080 Ti - SLI Configuration
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
As tested configuration: Intel Core i9-7980XE (Skylake-X)
Sony Ps4 Pro Angled View
NVIDIA GeForce GTX 1080 Ti - SLI Configuration NVIDIA GeForce GTX 1080 Ti - SLI Configuration
As tested configuration: Intel Core i9-7960X (Skylake-X)
NVIDIA GeForce GTX 1080 Ti - SLI Configuration
CyberPowerPC AMD VR Gaming PC - Keyboard Switches CyberPowerPC AMD VR Gaming PC - Keyboard Switches
As tested configuration: Intel Core i9-7960X (Skylake-X)
CyberPowerPC AMD VR Gaming PC - Keyboard Switches
NVIDIA GeForce GTX 1080 Ti - SLI Configuration NVIDIA GeForce GTX 1080 Ti - SLI Configuration
As tested configuration: Intel Core i9-7900X (Skylake-X)
NVIDIA GeForce GTX 1080 Ti - SLI Configuration
NVIDIA GeForce GTX 1080 Ti - SLI Configuration NVIDIA GeForce GTX 1080 Ti - SLI Configuration
As tested configuration: Intel Core i9-7900X (Skylake-X)
NVIDIA GeForce GTX 1080 Ti - SLI Configuration
AMD Ryzen 7 1600X (CPU-Z & GPU-Z) AMD Ryzen 7 1600X (CPU-Z & GPU-Z)
As tested configuration: Intel Core i7-7820X (Skylake-X)
AMD Ryzen 7 1600X (CPU-Z & GPU-Z)
AMD Ryzen 7 1600X - As Tested EFI Overview AMD Ryzen 7 1600X - As Tested EFI Overview
As tested configuration: Intel Core i7-7820X (Skylake-X)
AMD Ryzen 7 1600X - As Tested EFI Overview
AMD Ryzen 7 1500X (CPU-Z & GPU-Z) AMD Ryzen 7 1500X (CPU-Z & GPU-Z)
As tested configuration: Intel Core i7-7740X (Kaby Lake-X)
AMD Ryzen 7 1500X (CPU-Z & GPU-Z)
AMD Ryzen 7 1500X - As Tested EFI Overview AMD Ryzen 7 1500X - As Tested EFI Overview
As tested configuration: Intel Core i7-7740X (Kaby Lake-X)
AMD Ryzen 7 1500X - As Tested EFI Overview

As with the TR4 platform, the Core “Enhancements” option is disabled on the LGA2011-v3 platform.

Windows Benchmarks

For the bulk of our testing, we use Windows 10 build 16299 with full updates as the base. After installation, LAN, audio, and chipset drivers are installed even if they are not explicitly needed (because Windows can use generic driver versions). Our basic guidelines beyond that are:

Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Adobe Lightroom Classic CC: RAW to JPEG Export
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Autodesk 3ds Max 2015: (SPECapc 3ds Max 2015)
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Autodesk 3ds Max 2018: Fish Bowl Render (Arnold Renderer)
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Adobe Premiere Pro CC 2018: Blu-ray Concert Encode
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Adobe Premiere Pro CC 2018: 4K RED Encode
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Adobe Premiere Pro CC 2018: 8K RED Encode
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Blender: Pavillon Render
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Cinebench R15.038
Sony Ps4 Pro Angled View
dBpoweramp - Convert FLAC to MP3
dBpoweramp R15.1
dBpoweramp - Convert FLAC to MP3
Intel Core i7-6700K (CPU-Z & GPU-Z) Intel Core i7-6700K (CPU-Z & GPU-Z)
SiSoftware Sandra 2017 SP3
Intel Core i7-6700K (CPU-Z & GPU-Z)

All of the tests shown above are used in their stock configuration. If you’re a Blender user and wish to compare your system’s performance to ours, you can download the project files for free here.

Gaming Benchmarks

Because the biggest bottleneck in a game is the graphics card, the workload needs to be put on the CPU as much as possible in order to better understand the raw performance scaling. As such, some of the games tested here were run at 1080p and 4K, with moderate detail levels.

Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Ashes of the Singularity Escalation: Benchmark Screenshot
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Ashes of the Singularity Escalation: Game Settings
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Grand Theft Auto V: Benchmark Screenshot
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Grand Theft Auto V: Game Settings
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Grand Theft Auto V: Game Settings
Sony Ps4 Pro Angled View
Futuremark 3DMark
Grand Theft Auto V: Game Settings
CyberPowerPC AMD VR Gaming PC - Keyboard Switches CyberPowerPC AMD VR Gaming PC - Keyboard Switches
Grand Theft Auto V: Game Settings
CyberPowerPC AMD VR Gaming PC - Keyboard Switches
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Total War 2: WARHAMMER II: Benchmark Screenshot
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Total War 2: WARHAMMER II: Game Settings
Sony Ps4 Pro Angled View
Intel Core i7-6700K (CPU-Z & GPU-Z) Intel Core i7-6700K (CPU-Z & GPU-Z)
Watch_Dogs 2: Benchmark Screenshot
Intel Core i7-6700K (CPU-Z & GPU-Z)
Intel Core i7-6700K (CPU-Z & GPU-Z) Intel Core i7-6700K (CPU-Z & GPU-Z)
Watch_Dogs 2: Game Settings
Intel Core i7-6700K (CPU-Z & GPU-Z)
Intel Core i7-6700K (CPU-Z & GPU-Z) Intel Core i7-6700K (CPU-Z & GPU-Z)
Watch_Dogs 2: Game Settings
Intel Core i7-6700K (CPU-Z & GPU-Z)

Because Ashes offers the ability to act only as a CPU benchmark, it makes sense to use that for a CPU performance article. GTA V isn’t very GPU intensive, leading us to see similar results at both 1080p and 4K. As such, only 1080p is used for that title. WARHAMMER II and Watch_Dogs 2 are tested at 4K, as well as 1080p.

Linux Benchmarks

Ubuntu 17.10 is the OS of choice for our test bed, as it’s both simple to set up, and so de facto that everyone reading the results should feel at home. The OS is left as stock as possible, with minor software added, and everything updated.

Before testing begins, we take the Phoronix Test Suite suggestion of enabling the “performance” power profile; something that actually improved the encode test by 15%-ish. The command run (sudo):

echo performance | tee /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor

Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Blender: Pavillon Render
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Phoronix Test Suite 7.0
Sony Ps4 Pro Angled View

Both the Blender and HandBrake tests are shared with the Windows testing. Phoronix Test Suite is used for the bulk of our Linux testing, as it supplies the tests we need, and makes them easy to use.


If you think there’s some information lacking on this page, or you simply want clarification on anything in particular, don’t hesitate to leave a comment.

Rendering: (3ds Max 2015 & 2018), Blender, Cinebench, POV-Ray & V-Ray

(All of our tests are explained in detail on page 2.)

Design and rendering is one of the greatest areas of computing to benchmark to highlight the benefits of faster hardware, whether it be a CPU, GPU, memory, and even storage. On a low-end system, a production render might take hours, for example, whereas on a high-end system, that render could be shaved down to the tens of minutes.

With these results, it’s up to you to gauge where the best value can be found. In some cases, it might be beneficial to go with more modest hardware if the time-to-render isn’t of a great concern; in other cases, spending more on faster hardware might actually save you money in the long-run.

For our rendering tests, we use Autodesk’s 3ds Max (2015, for SPECapc, and 2018, for our real-world model render), the popular open source design suite Blender, as well as Cinebench, POV-Ray, and V-Ray Benchmark for some quick-and-dirty results.

Autodesk 3ds Max 2015 & 2018

AMD & Intel 16-core CPU Performance - Autodesk 3ds Max 2018

As of the time of writing, Arnold is only available as a CPU renderer, but that’s going to soon change with a revamped plugin that will tackle both CPU and GPU duties. Fortunately, for a CPU review, a GPU renderer doesn’t matter at all, so naturally, we see some very clean scaling with our render here.

AMD & Intel 16-core CPU Performance - SPECapc 3ds Max 2015

SPEC’s outlook on things paints a similar picture, with almost identical scaling across the board. In both 3ds Max tests, Intel’s lowly quad-core 7740X sits at the bottom, but that’s not a surprise given what it brings to the table. That said, what’s strange about that chip is that it falls so far behind in performance, yet costs generally the same as the new 2700X thanks to the fact it’s destined for the beefy X299 platform. A parallel would be AMD releasing a 1900X Threadripper with 8 cores to match the 1800X or 2700X, but eight cores is actually suitable for power users.

Blender

AMD & Intel 16-core CPU Performance - Blender Renders

Based on these results, it seems like IPC matters quite a bit with Blender, but additional cores can help tighten the deltas between one chip and another. Nonetheless, in both cases here, AMD’s new 2700X beat out Intel’s six-core 8700K. If you’ve been familiar with Ryzen first-gen performance, this probably won’t surprise you too much. 8 AMD cores are faster than 6 Intel cores. Where Intel will strike back is with its optimizations for things like media, and in many cases, simply with its stronger IPC.

Another angle to look at things at is with the 2600X, and how it also performs against the 8700K. It carries an SRP of $219, at least $100 less than the 8700K, but despite that, it doesn’t lag too far behind – about 10 minutes on an 80~90 minute render.

Synthetic Renderers: Cinebench, POV-Ray, V-Ray

AMD & Intel 16-core CPU Performance - Cinebench
AMD & Intel 16-core CPU Performance - POV-Ray
AMD & Intel 16-core CPU Performance - V-Ray Benchmark

AMD continues to clean house across these synthetic renderers, with the 2700X outperforming the 8700K in every single one. I can honestly say that after spending a lot of time testing the real V-Ray plugin, that the CPU renderer works much better on AMD than the GPU renderer does. This renderer that’s existed since well before AMD was a serious competitor in the market uses the architecture so well, it allows the 2700X to really impress here (not just in V-Ray, but the others, too).

So far, AMD is really dominating at its respective price-points, but we will see a few examples coming up of where that’s not quite the case. Ryzen 2 doesn’t fix every Ryzen 1 wrong, but some of those wrongs are tied directly to optimizations not entirely in AMD’s control. Where raw compute is concerned, AMD delivers an explosive value, and these results prove it.

Media: Adobe Premiere Pro, Adobe Lightroom, dBpoweramp & HandBrake

(All of our tests are explained in detail on page 2.)

As seen on the previous page, rendering can take amazing advantage of even the biggest processors, but video encoding is not that far behind – if at all. Even the free conversion tool HandBrake can take advantage of our sixteen-thread processors to significantly decrease encode times. For our video encoding purposes, we use Adobe’s Premiere Pro, as well as HandBrake.

To a lesser degree, music conversion and image manipulation can also see benefits on beefier chips, so Adobe’s Lightroom and dBpoweramp will be used to help us gauge that performance.

Adobe Premiere Pro

AMD & Intel 16-core CPU Performance - Adobe Premiere Pro (Blu-ray & 4K RED)

While Intel generally dominates the multi-media performance landscape, straight-forward encodes that can take advantage of many cores can still see great benefit on AMD hardware. We’re yet again seeing an example of the 2700X beating out the 8700K pretty handily, and the 2600X barely falls that far behind the much more expensive Coffee Lake chip.

AMD & Intel 16-core CPU Performance - Adobe Premiere Pro (8K RED)

With 8K encodes, not a thing changes on the scaling front (that I can see). The 2700X continues to deliver great performance, and it’s clear that if you have a good GPU, sometimes that itself might take over the lion’s share of the work, but faster CPUs still help more work get done. This article is about Ryzen 2, but it’s impossible to ignore the value proposition of a chip like the 1950X, seeing it slot just under the much more expensive Intel competition.

For more information about CPU+GPU with Premiere Pro, I’d recommend checking out this section from an article earlier this year which shows that a CPU+GPU can lead to much better CPU efficiency than if you just use the CPU itself. It’s as if the CPU chokes itself, but the GPU helps things keep running smooth. So, whenever you’re in doubt, you definitely want to introduce GPU acceleration when it’s available (CUDA or OpenCL) in PP.

HandBrake

AMD & Intel 16-core CPU Performance - HandBrake

Here, we finally begin to see Intel’s real strengths in multi-media. The x265 encodes perform far better on Intel than AMD, while x264 scales pretty much as expected. Here, the 2700X manages to match the 8-core 7820X perfectly with the x264 encode, but it of course falls behind with x265.

Adobe Lightroom & dBpoweramp

AMD & Intel 16-core CPU Performance - Adobe Lightroom and dBpoweramp

With Lightroom, we see another example of Intel’s media prowess. Again, this isn’t much of a surprise, since Intel directly works with Adobe on optimizations. While Lightroom Classic 7.1 is used for testing here, 7.2 would shave another ~10 seconds off each export on Intel (I have not tested AMD’s gains yet).

Ultimately, because AMD can’t match certain Intel strengths in media, it falls well behind in Lightroom. But thanks to its overall brawn, it strikes back with dBpoweramp, a straight-forward FLAC to MP3 process. With that test, the 2600X doesn’t keep far behind the 8700K.

SiSoftware Sandra: Computation, Memory & Cache Tests

(All of our tests are explained in detail on page 2.)

SiSoftware’s Sandra needs no introduction, but I’ll give one anyway. It’s been around for as long as the Internet, and has long provided both diagnostic and benchmark features to its users. SiSoftware keeps on top of architectural updates as they’re revealed, and often, the software supports a specific processor feature or design before consumers can even get their hands on the product.

As a synthetic tool, Sandra can give us the best possible look at the top-end performance from the hardware it can benchmark, which is the reason we use it to test CPUs, memory, motherboards, and even graphics cards (for compute). It also allows us to benchmark very specific tests, such as inter-core bandwidth and latency, financial and scientific scenarios, as well as cache performance.

Arithmetic & Multi-Media

AMD & Intel 16-core CPU Performance - SiSoftware Sandra 2016 Arithmetic & Multi-Media

With this set of results, we’re seeing a continuation of the media dominance Intel started on the previous page. Sandra helps prove that Intel’s strengths there are undeniable, but for straight math, AMD’s extra muscle for the dollar can make all of the difference. Here, the 2600X falls far behind the 8700K in multi-media, but not so much in arithmetic. Meanwhile, the 8-cores of the 2700X help it edge a bit closer to the Coffee Lake chip.

Cryptography

AMD & Intel 16-core CPU Performance - SiSoftware Sandra 2016 Cryptography

With both its Ryzen CPUs and Vega GPUs, AMD delivers some extremely good cryptographic performance. That’s easily seen here, with the 2700X yet again trouncing the 8700K. An even funner result is the 2600X, which also trounces the 8700K. This is a perfect example of where it pays to know your workload, and your hardware. This is a case where if crypto is largely your bag, it could make sense to go AMD. It’s no wonder so many sites recently talked about mining with Threadripper with this kind of strong hashing performance.

Financial & Scientific Analysis

AMD & Intel 16-core CPU Performance - SiSoftware Sandra 2016 Financial & Scientific Analysis

Like media, Intel has strong advantages in some scientific tests, such as this one. Even with its additional cores, the 2700X falls short of catching the 8700K, but again, the roles are reversed with the other test which relies more on additional cores than other general optimizations.

Memory & Core Bandwidth / Latencies

AMD & Intel 16-core CPU Performance - SiSoftware Sandra 2016 Memory Bandwidth
AMD & Intel 16-core CPU Performance - SiSoftware Sandra 2016 Memory Latency

These results become our first “odd” ones, because despite using the same memory speed, the 2700X placed just behind the 2600X (across multiple tests that involved reinstalling each chip). As you’ll see on the Linux page later, the same kind of effect was seen even worse in the Stream memory benchmark. This isn’t an issue, per se, but it definitely stands out. There’s something about our configuration or the 2700X in particular that allows the 2600X to best its bandwidth.

As for memory latency, it’s pretty abysmal to look at on the AMD side here, but as most of the performance in this article can attest, those latencies don’t matter a great deal. If you are to measure the minutae of single-threaded operating system performance, you’re going to see a stronger result from Intel, but more often than not, Ryzen will help you get many substantial jobs done quicker, dollar for dollar.

Gaming: 3DMark, Ashes, GTA V, TW: WARHAMMER 2 & Watch_Dogs 2

(All of our tests are explained in detail on page 2.)

It’s been easy to highlight the performance differences across our collection of CPUs on the previous pages, since most of the tests used take advantage of every thread we give them. But now, it’s time to move onto testing that’s a different beast entirely: gaming.

In order for a gaming benchmark to be useful in a CPU review, the workload on the GPU needs to be as mild as possible; otherwise, it could become a bottleneck. Since the entire point of a CPU review is to evaluate the performance of the CPU, running high detail and high resolutions in games won’t give us the most useful results.

As such, our game testing revolves around 1080p, and sometimes 4K, with games being equipped with moderate graphics detail (not low-end, but not high-end, either). These settings shouldn’t prove to be much of a burden for the TITAN Xp GPU. For those interested in the settings used for each game, hit up page 2 (a link is found at the top of this page).

In addition to 3DMark, our gauntlet of tests includes four games: Ashes of the Singularity: Escalation (CPU test only), Grand Theft Auto V (Fraps), Total War: WARHAMMER II (built-in benchmark), and Watch Dogs 2 (Fraps).

Futuremark 3DMark

AMD & Intel 16-core CPU Performance - Futuremark 3DMark Overall Scores
AMD & Intel 16-core CPU Performance - Futuremark 3DMark Physics Scores

With the memory result on the last page, we saw that the 2600X somehow managed to beat out the 2700X. With 3DMark, we don’t see quite the same thing, but the fact that both chips perfect extremely similarly definitely stands out a little bit. Extra retesting had to be done to make doubly sure these results are accurate, and they are, but because these results are so close, variation can sometimes mean that the 2600X will outperform the 2700X on occasion; multiple runs will smooth things out.

That all said, gaming remains best on Intel, but that doesn’t mean gaming on Ryzen 2 is poor. As with the first-gen chips, these second-gen offerings still fall behind at 1080p versus Intel, but as the resolution grows higher, the delta between the cards tightens quite a bit.

From a straight-forward physics perspective, the extra brawn of Ryzen helps keep it ahead of the pack, but I haven’t really seen how this kind of performance result correlates with real gaming, since most physics nowadays are not going to be designed to take advantage of many-core chips. If more games start to take advantage of 6+ cores, we’d be able to see some more interesting results.

Ashes of the Singularity: Escalation

AMD & Intel 16-core CPU Performance - Ashes of the Singularity Escalation

Like the physics test above, the CPU test in Escalation removes the GPU from the equation effectively entirely, so the scaling should technically be similar. However, it’s not, leading me to believe that this test favors IPC as well as extra cores.

Grand Theft Auto V

AMD & Intel 16-core CPU Performance - Grand Theft Auto V

For the final 1080p-only test, GTA V runs very well on Ryzen, and it’s not because of its overall framerate, but rather its minimums. The 2700X fell 19 FPS short of the 8700K, but few people are going to scoff at 122 FPS. In another comparison, though, the 8700K hit 57 FPS minimum, which was improved by 21 FPS with the 2700X. Even the 2600X managed to match that. There’s really not much to say here: for minimums, Ryzen delivers more than the competition in GTA V.

Total War: WARHAMMER II

AMD & Intel 16-core CPU Performance - Total War WARHAMMER 2

WARHAMMER II seems to be a game that benefits greatly from both IPC and additional cores, but likely the former more than the latter. Here, the new Ryzen chips score very well, being bested only by Intel’s highest-clocked chips. At 1080p, even the 18-core Intel chip falls short of the 8700K. At 4K, I think it’s safe to call performance equaled across the board.

Watch Dogs 2

AMD & Intel 16-core CPU Performance - Watch Dogs 2 (1080p)
AMD & Intel 16-core CPU Performance - Watch Dogs 2 (4K)

Watch_Dogs 2 seems to be a pretty decent CPU benchmark, outside of the fact that it doesn’t run too well on AMD’s hardware. Both Ryzens fall to the bottom at 1080p, but deliver performance more on par with the competition at 4K.

Before moving onto the next page, I will say that the games on this page are not by any stretch an ideal set. Gaming is one thing in particular I haven’t had much time to dedicate to improving for CPU reviews, but once the Spectre dust finally settles, and the suite can be updated, I’ll be scouring for more appropriate benchmarks. And likely more of them, since it’s hard to appreciate the overall picture with a mere four titles. Or, ultimately, I may just keep the choice simple for articles like this, but plan on doing a dedicated article looking at the grander picture in more depth. If you have interest in our take on that, please leave a comment.

Linux: Blender, HandBrake & Phoronix Test Suite

(All of our tests are explained in detail on page 2.)

To wrap up our look at performance, we have a bunch of Linux test results to pore over. These include two of the same test found in the Windows suite (HandBrake, Blender).

The OS used in testing is Ubuntu 17.10, using the stock 4.13 kernel. As with the Windows tests, the Linux OS is kept as minimal as possible, with only required software packages installed on top of the stock software. HandBrake is procured through Ubuntu’s repository, while Blender is grabbed from the official source. Version 7.6 of the Phoronix Test Suite is also used.

Blender & HandBrake

AMD & Intel 16-core CPU Performance - Blender & HandBrake (Linux)

It’s performance results like these that make me think one of the reasons Ryzen chips don’t cost $100 more than they do is because of their weaker IPC performance. In so many tasks, outside of things like gaming, the extra cores on a chip like the 2700X helps it stomp the 6-core i7-8700K, even though that chip carries an SRP price tag about $20 higher.

Even the $219 2600X impresses. It finds parity on the core/thread front with the 8700K, but doesn’t fall as far behind it as the 8700K falls behind the 2700X. Seeing as how Threadripper’s 16-cores weren’t enough to beat out Intel’s 10-core, Intel has a clear advantage in these tests, but AMD’s perf-to-$ ratio is strong enough to keep it super-competitive.

Phoronix Test Suite

AMD & Intel 16-core CPU Performance - Compiler Performance (Linux)

Like some other tests strewn throughout this article, compiling can take great advantage of both strong IPC performance and as many cores as I’ve ever personally been able to throw at it. In the case of a chip like Threadripper, it sits behind the 16-core Intel equivalent, but not to the degree you’d expect between $1,699 and $999 offerings.

A general rule of thumb, then, is that cores-for-cores, Intel is going to be faster, and offer the “ultimate” level of performance to those who demand it, but based on the chips in this particular list, AMD’s value proposition appears to me to be better. Look at the $219 2600X in comparison to the $319 7740X, an oddball chip that requires an enthusiast platform to be used.

AMD & Intel 16-core CPU Performance - Ray Tracing (Linux)

With ray tracing tests, we see very similar scaling, and explicit scaling in terms of where each GPU ranks. However, there are a couple of result discrepancies that are easy to miss. Threadripper, for example, performs exceptionally well in the smallpt test, beating out everything else in the lineup. Intel strikes back with tachyon, but even falls a bit short with cray.

The 2600X continues to be a force to be reckoned with at the low end, delivering great performance for its price-point, once again beating out the 7740X in every single test. It also manages to best the 8700K in smallpt and cray, so AMD definitely has a few tricks up its sleeves.

AMD & Intel 16-core CPU Performance - SciMark (Linux)

SciMark’s tests are entirely single-threaded, so IPC matters an awful lot here, as evidenced by the six-core 8700K sitting at the top. AMD performance is lacking pretty substantially in the FFT test, but the Ryzens perform fairly in the others. Aside from Threadripper 1950X, that is; its lower clock speed is a death sentence for a single-threaded suite like this.

AMD & Intel 16-core CPU Performance - OpenSSL (Linux)

In some tests, IPC matters just as much as core count, and OpenSSL is one such test. Core for core, Intel comes well ahead of AMD with the 16 core options, but dollar for dollar, AMD can keep ahead. This is also an example of where extra cores can help substantially, as the 2700X again trounces the 8700K. And, in case you missed it: so does the 2600X. Phreaking great performance.

AMD & Intel 16-core CPU Performance - HMMer Search (Linux)

It’s HMMer time, with another single-threaded test that ekes every little bit of IPC performance you can give it. As such, Intel reigns supreme here, with all three of the Ryzen chips occupying the bottom end of the chart. Not the kind of hammering AMD would like to see, I’m sure.

AMD & Intel 16-core CPU Performance - 7-Zip (Linux)

For straight-forward compression that can take advantage of many cores, it really does boil down to the fact that the more cores you have, the better the performance. But, Intel clearly has a strong performance advantage here, really notable when you compare the two 16-core chips. However, looking at the value proposition again, the 2700X still managed to place ahead of the 8700K by a few percentage points.

AMD & Intel 16-core CPU Performance - Stream (Linux)

Considering the fact that the Ryzen 2 chips here were running the fastest memory of any one of the other configurations (3400MHz vs. 3200MHz), these results surprise me. Well, there’s actually more than one reason for that: the 2600X outperforms the 2700X. With Sandra, covered earlier, the 2600X keeps ahead as well, but barely – not to this kind of degree.

I upgraded the Linux kernel from 4.13 to 4.17 in hopes it’d give more expected scaling here, but not so. This is not to make it seem like 38GB/s vs. 35GB/s is actually a big deal, but it’s definitely more so of one on Ryzen than it is on Intel. Fortunately, the performance results in the multi-threaded tests don’t seem to care about that discrepancy too much.

John The Ripper (Encryption)
BlowfishMD5DES
Intel Core i9-7980XE31.3K395.0K101.6M
Intel Core i9-7960X29.5K373.1K95.7M
AMD Ryzen Threadripper 1950X23.2K363.0K26.0M
Intel Core i9-7900X20.5K259.1K66.9M
Intel Core i7-7820X16.5K207.5K53.7M
AMD Ryzen 7 2700X14.1K462.2K61.0M
Intel Core i7-8700K12.7K159.7K41.6M
AMD Ryzen 5 2600X10.4K346.8K45.7M
Intel Core i7-7740X8.9K112.0K26.7M

On all fronts, AMD’s 8-core 2700X delivers enough oomph to rank above the 8700K in this list; and, in another repeat, the lowly 6-core AMD outpaces the 4-core 7740X, despite costing $100 (SRP) less. Results like any of those seen on this page definitely adds some intrigue to Zen 2, which is surely going to add more of a substantial IPC boost than these second-gen Pinnacle Ridge chips do.

Power Consumption & Final Thoughts

To generate power-draw results for our collection of CPUs, we plug the test PC into a Kill-a-Watt for real-time monitoring, and stress the CPU with the help of our CLI-based HandBrake test (which can peak 100% of course in our tests). Idle power consumption is measured about 5 minutes after boot, once Windows decides to calm itself, and the wattage reading keeps stable.

Because AMD and Intel measure temperatures very differently, and there’s never a guarantee that software applications are reporting accurate readings, we forgo that testing. The only reliable method for capturing CPU temperatures is to go the hardware route, which is both very time-consuming, and expensive.

I will also add that HandBrake has recently been deemed a bit lacking to gauge the peak power draw, so with the next suite iteration, I’ll likely introduce something a bit more demanding. However, a video encode test which takes full advantage of the CPU is hardly a poor example of power draw, since it represents a typical demanding scenario.

AMD & Intel 16-core CPU Performance - Power Consumption

With results all over the board like this, it’s hard to appreciate the real power efficiency differences between the platforms, but this still gives us a decent overall picture. Even with 4 sticks of DDR4 on the 8700K platform versus AMD’s 2, Intel delivers the best result here.

It’s hard to ignore the massive difference in idle draw between the X299 platform and the others. That platform is in fact beefier than the others (in terms of motherboard features), but a difference of 50W almost seems extreme. Nonetheless, what impresses me here is that a super-fast chip like the 2700X can operate at just over 200W. Some overclocked CPUs would require that much alone.

Final Thoughts

I admit that despite all of the testing I’ve been involved with the past few weeks, it feels like I rushed through Ryzen second-gen testing, and honestly, I’m surprised to be making the embargo lift. Beyond this point, I’m thinking I’ll probably wait for the Spectre dust to completely settle before continuing on with more testing. In particular, it might be smart to wait for the next Creators Update to land, which is apparently going to be happening very soon. With that build of the OS, I am hoping we can safely rebenchmark with the assurance that we won’t have to rebenchmark again a few weeks later.

That all aside, for as little time as I feel like I’ve spent with Ryzen 2, my initial thoughts are very good. Let’s be real, though: this is not a totally new product. It’s an iteration, so if you’re familiar with recent Ryzen experiences, this second-gen lineup is not going to introduce any surprises. Again, it’s an iteration, but a good one.

I’ll bring back an example I made earlier. Last year, the 8-core 1800X cost $499. This year, the 8-core 2700X costs $319. That’s in comparison to Intel’s $339 6-core Coffee Lake-based 8700K. Across all of the benchmarks here, AMD and Intel traded blows, but ultimately, AMD can usually dominate better than Intel can, thanks to its extra horsepower.

Ryzen 7 2700X & Ryzen 5 2600X

I sound like a broken record at this point, I’m sure, but I really do love Ryzen as a workstation chip. For the “ultimate” in performance, especially when we’re talking about every minor interaction done with a PC, the higher IPC of Intel is going to deliver the better performance. But if you’re willing to accept that caveat (one that you may not even notice, admittedly), you can gain handsomely when it comes to engaging in actually demanding scenarios, like video encoding, rendering, et cetera.

In 3ds Max, for example, the 2700X came well ahead of the 8700K, and the 2600X didn’t lag far behind. The same was repeated for Blender, Cinebench, V-Ray, and POV-Ray. Things change a bit in some media tests; Intel dominates Adobe Lightroom and x265 encodes, but again, AMD’s extra brawn does help it tighten the gap a little bit in those cases.

For gaming, Ryzen’s second-generation still leaves a bit to be desired. At 1080p, the differences remain noticeable for the most part, whereas at higher resolutions, like 4K, the differences are largely nonexistent between any one of the CPUs (except those that have strange performance in some cases, like Threadripper with WARHAMMER II).

The compromises people need to make with the second-gen Ryzen remain unchanged from the first. Slightly weaker gaming performance, lacking performance in certain areas where Intel tends to dominate (eg: media), but better performance overall in any scenario that simply requires more raw horsepower.

SmartKevin AMD Ryzen 2000 Series PC

To that end, I personally couldn’t imagine choosing the 8700K over the 2700X if I’m a workstation user. I’m not a designer, but I benchmark enough 3D projects to feel like one, and it’s obvious the kinds of performance gains extra cores can make. As a hardcore gamer, I’d go Intel, because while higher resolutions seem to stabilize performance, Intel is still clearly dominant in this space, and gives the assurance that you’re going to get the best gaming performance across the board.

By now, you probably know enough about this chip, and admittedly, you probably knew enough before reading. Whereas the first-gen Ryzen launch was undeniably rough with regards to desperately needed updates that came out continually, I haven’t experienced any real issue with Ryzen second-gen to speak of, outside of the one where my Corsair 3200MHz kit wouldn’t run as a kit of four (even though it did on Threadripper). What that tells us is that you don’t want to haphazardly choose your memory kit for Ryzen – go with something that’s known to work well.

And with that, hopefully by now you know whether or not Ryzen’s second-gen is for you. If one of your questions hasn’t been answered, or you simply want to leave your two cents, please leave a comment.

Copyright © 2017 SmartKevin