AMD’s 5 GHz Turbo CPU in Retail: The FX-9590 and ASRock 990FX Extreme9 Review
by Ian Cutress on August 9, 2014 8:00 AM ESTCPU Benchmarks: Comparing Motherboards
Readers of our motherboard review section will have noted the trend in modern motherboards to implement a form of MultiCore Enhancement / Acceleration / Turbo (read our report here) on their motherboards. This does several things – better benchmark results at stock settings (not entirely needed if overclocking is an end-user goal), at the expense of heat and temperature, but also gives in essence an automatic overclock which may be against what the user wants. Our testing methodology is ‘out-of-the-box’, with the latest public BIOS installed and XMP enabled, and thus subject to the whims of this feature. It is ultimately up to the motherboard manufacturer to take this risk – and manufacturers taking risks in the setup is something they do on every product (think C-state settings, USB priority, DPC Latency / monitoring priority, memory subtimings at JEDEC). Processor speed change is part of that risk which is clearly visible, and ultimately if no overclocking is planned, some motherboards will affect how fast that shiny new processor goes and can be an important factor in the purchase.
Point Calculations – 3D Movement Algorithm Test: link
3DPM is a self-penned benchmark, taking basic 3D movement algorithms used in Brownian Motion simulations and testing them for speed. High floating point performance, MHz and IPC wins in the single thread version, whereas the multithread version has to handle the threads and loves more cores.
Compression – WinRAR 5.0.1: link
Our WinRAR test from 2013 is updated to the latest version of WinRAR at the start of 2014. We compress a set of 2867 files across 320 folders totaling 1.52 GB in size – 95% of these files are small typical website files, and the rest (90% of the size) are small 30 second 720p videos.
Image Manipulation – FastStone Image Viewer 4.9: link
FastStone is the program I use to perform quick or bulk actions on images, such as resizing, adjusting for color and cropping. In our test we take a series of 170 images in various sizes and formats and convert them all into 640x480 .gif files, maintaining the aspect ratio. FastStone does not use multithreading for this test, and results are given in seconds.
The FX-8150 result for the Extreme9 compared to other motherboards shows one of two things - the latest Windows SP1 with core parking updates has an effect, or the BIOS is more efficient at handling turbo modes than our older reviews.
Video Conversion – Handbrake v0.9.9: link
For HandBrake, we take two videos (a 2h20 640x266 DVD rip and a 10min double UHD 3840x4320 animation short) and convert them to x264 format in an MP4 container. Results are given in terms of the frames per second processed, and HandBrake uses as many threads as possible.
Rendering – PovRay 3.7: link
The Persistence of Vision RayTracer, or PovRay, is a freeware package for as the name suggests, ray tracing. It is a pure renderer, rather than modeling software, but the latest beta version contains a handy benchmark for stressing all processing threads on a platform. We have been using this test in motherboard reviews to test memory stability at various CPU speeds to good effect – if it passes the test, the IMC in the CPU is stable for a given CPU speed. As a CPU test, it runs for approximately 2-3 minutes on high end platforms.
Synthetic – 7-Zip 9.2: link
As an open source compression tool, 7-Zip is a popular tool for making sets of files easier to handle and transfer. The software offers up its own benchmark, to which we report the result.
Facebook
Twitter
RSS
146 Comments
View All Comments
TiGr1982 - Saturday, August 9, 2014 - link
No, no FX in the future - at least, not on Bulldozer-derived microarchitectures.Just APUs till 2016 at least.
will1956 - Saturday, August 9, 2014 - link
i've gotten the sabertooth 990FX GEN3 R2.0 and its got pcie 3 x16 with a 8350 and a sapphire 7870 ghz (both OC'ed) and its pretty good although rather greedyTiGr1982 - Saturday, August 9, 2014 - link
It's nice that this kind of boards exist, but, sadly, PCI Express 3.0 is a rarity rather than a norm on AM3+.AMD only has PCI Express 3.0 as a standard on FM2+ with Kaveri APU.
roadapathy - Saturday, August 9, 2014 - link
I see by the comments below that I'm among the rational. AMD is stuck on the 32nm fab process making for a dismal performance experience for us all! My AMD 6-core could boil water. This is great for the cold midwestern American climate but the Summer is unbearable with AMD CPUs. I had waited over 2 years for a 22nm AMD 8 core FX that never appeared. Meanwhile, I'm running the "lower" 95watt CPU. I can't even imagine how it would be with the 220watt. How ridiculous!!RussianSensation - Sunday, August 10, 2014 - link
Exactly!! Someone understands. If AMD could move to 14nm, they could increase the number of modules 50%-100% and lower the power usage at the same time. When you CPU is on 32nm while Intel is soon to launch 14nm Broadwell, the chance of AMD competing in performance or performance/watt is 0%.Death666Angel - Sunday, August 10, 2014 - link
But more cores aren't really the issue for AMD, are they? In multi-threaded stuff they are already doing fine. What they need is better IPC. Even at 5GHz they barely beat i3s of the current generation. Unfortunately the FX-9590 isn't in bench yet, but the FX-8350 even loses to a chip on 32nm (i5-2500k) in most benchmarks except some multi threaded ones. Put an i7-2600k in its place and it loses even more consistently. That is not just a lithography disadvantage, that is a straight up embarrassment from the CPU architecture standpoint. And the fact that they aren't releasing any more FX CPUs based on newer architectures is a slap in the face of any PC enthusiast.TiGr1982 - Monday, August 11, 2014 - link
Indeed; placed my response too.Budburnicus - Wednesday, January 14, 2015 - link
exactly! AMD has HORRIBLE IPC which results in horrible efficiency - ALSO bear in mind that the 3 year old i5 and i7 parts are 32nm fab, and are 95 watt TDP parts - which can easily be clocked at 4.7 Ghz and totally SPANK this CPU in every way!AMD need a totally new architecture to go with a new fabrication, otherwise it will remain meaningless and eat more power. IPC is incredibly important - just because this FX 9590 is much newer than an i7-2600K and therefore has more and newer instruction sets, does NOT mean it performs better! It performs far worse in fact - whilst eating more power - using the same fab size!
TiGr1982 - Monday, August 11, 2014 - link
It's not the manufacturing tech itself - their Bulldozer-derived microarchitecture has drastically slower IPC (Instructions Per Clock). If you you emphasize lithography, then let's compare FX Piledriver from late 2012 on GF 32 nm lithography and Sandy Bridge LGA1155 Core i7 from early 2011 on Intel's 32 nm lithography.Guess what? Sandy Bridge is around 50% faster in single threaded tasks than Pilderiver. At the "same" lithography. Despite the fact that Sandy Bridge i7 has just 9 MB L2+L3 cache, while Piledriver has 16 MB L2+L3 cache. So, AMD's chip has almost twice the amount of cache than Intel's chip and is still 50% slower. So, first, the case with AMD FX is mainly a problem of inappropriate microarchitecture, and only then comes the lithography lag.
So, even if a Cinderella's fairy comes up and magically moves FX Piledriver to Intel's 22 nm or even 14 nm, the resulting tiny Piledriver shrink will still be a Slowpoke in single thread duties - because it is its microarchitecture that prevents it from doing better.
roadapathy - Monday, August 11, 2014 - link
I don't have a complaint about the architecture itself because of the price points. Intel CPU, motherboard and the RAM are all much more expensive! I'd be satisfied with AMD FX series (or the new Kavari) on the 20nm fab process.