discussion AMD FX-9590 Review; Piledriver at 5GHz

[SKYMTL]

720P Gaming: Deus Ex: HR / Dirt 3 / Skyrim

720P Gaming Benchmarks

Processors play a huge roll in gameplay performance since they have to process copious amounts of information for the GPU. In the following tests, we use a simple 720P resolution and the lowest possible detail settings in an effort to remove the graphics processor from the equation and place additional pressure on the CPU. 720P was used since it is a resolution that is extensively used by gamers sporting lower end HDTVs and it doesn’t put as much stress upon the GPU as 1080P.

For every one of the following titles, a simple 1 minute gameplay walkthrough was used and the average frames per second was logged via FRAPS.

RESULTS: At lower resolutions, each processor’s gaming ability is put into stark contrast as the graphics card takes a back seat. Unfortunately the FX-9590 just can't keep up here, which is to be expected.

 

[SKYMTL]

720P Gaming: Super Street Fighter IV / Torchlight

720P Gaming Benchmarks (pg.2)

Processors play a huge roll in gameplay performance since they have to process copious amounts of information for the GPU. In the following tests, we use a simple 720P resolution and the lowest possible detail settings in an effort to remove the graphics processor from the equation and place additional pressure on the CPU. 720P was used since it is a resolution that is extensively used by gamers sporting lower end HDTVs and it doesn’t put as much stress upon the GPU as 1080P.

For every one of the following titles, a simple 1 minute gameplay walkthrough was used and the average frames per second was logged via FRAPS.

RESULTS: The results here reflect those seen on previous pages with one noteworthy exception: the graphics card becomes the bottleneck in Street Fighter.

 

[SKYMTL]

1080P Gaming: Deus Ex: HR / Dirt 3 / Skyrim

1080P Gaming Benchmarks

While lower resolution gaming highlights processor bottlenecks, most people use slightly higher resolution monitors and want to play with increased detail settings. In these situations, the CPU tends to take a back seat to the graphics processor but even at 1080P (ie: 1920x1080) a slower CPU can still have a drastic impact upon in-game performance. In order to illustrate this, we have carried over the games from our previous tests, pumped detail levels to their max and used the increasingly popular 1080P resolution standard.

RESULTS: As the GPU moves into the driver's seat, we can see the FX-9590 nearly catching some of the higher end Intel processors as its performance becomes less important. However, without any mods installed, Skyrim seems oddly CPU-bound on AMD's processors.

 

[SKYMTL]

1080P Gaming: Super Street Fighter IV / Torchlight

1080P Gaming Benchmarks (pg.2)

While lower resolution gaming highlights processor bottlenecks, most people use slightly higher resolution monitors and want to play with increased detail settings. In these situations, the CPU tends to take a back seat to the graphics processor but even at 1080P (ie: 1920x1080) a slower CPU can still have a drastic impact upon in-game performance. In order to illustrate this, we have carried over the games from our previous tests, pumped detail levels to their max and used the increasingly popular 1080P resolution standard.

RESULTS: Our last set of standard gaming benchmarks shows the FX-9590 posting respectable results and improving upon its predecessor's performance. However, what gamers really want to see is how this CPU performs in newer titles....so onto the next page we go!

 

[SKYMTL]

4770K vs. FX-9590 Gaming (Current Games)

4770K vs. FX-9590 Gaming (Current Games)

While our standard in-game testing consists of slightly older but still popular titles, we decided to throw in a number of additional, newer games. Each of these puts a significant amount of load on the CPU but many are also multi-core optimized.

All of the standard testing guidelines apply here as well though there is one major difference: instead of using FRAPS, we are utilizing the FCAT tool. This captures frames in real time before analyzing them, allowing for a much more accurate depiction of what will eventually be displayed onscreen. Once again, a GTX 670 has been used.

RESULTS: The first group of five games displays some interesting characteristics. While Intel’s i7-4770K is able to maintain a slight lead in most situations, a gamer really won’t be able to tell the difference between it and AMD’s FX-9590.

There are some oddballs here like Dirt: Showdown which still seems slightly CPU-bound and Far Cry 3 which favors the FX-9590 but for the most part, average framerates are very much similar. This is what happens when the onus is taken off of the CPU; in many cases, the GPU becomes a potentially limiting factor.

 

[SKYMTL]

4770K vs. FX-9590 Gaming (Current Games) pg.2

4770K vs. FX-9590 Gaming (Current Games) pg.2

While our standard in-game testing consists of slightly older but still popular titles, we decided to throw in a number of additional, newer games. Each of these puts a significant amount of load on the CPU but many are also multi-core optimized.

All of the standard testing guidelines apply here as well though there is one major difference: instead of using FRAPS, we are utilizing the FCAT tool. This captures frames in real time before analyzing them, allowing for a much more accurate depiction of what will eventually be displayed onscreen. Once again, a GTX 670 has been used.

The results in this round of games are interesting as well since the 4770K and FX-9590 trade blows with the Intel processor winning handily in Max Payne and Metro Last Light. Meanwhile, AMD’s processor dominates in the newest Company of Heroes game, a title that is shockingly well optimized for multi core systems.

 

[SKYMTL]

Power Consumption / Temperature Testing

System Power Consumption

Our power consumption numbers are broken down into two categories: one which simply stresses all of the CPU cores with WPrime and another which puts a high amount of load on both the CPU cores and the IGP. The latter will only be included if a given processor includes a dedicated internal graphics sub-processor.

For the CPU power consumption test, we use the standard testing system (with an NVIDIA GTX 670 installed) and wait until the system and discrete GPU are at idle speeds in order to log the idle power consumption. After this, WPrime 1024M is looped for 15 minutes while the power consumption is logged with a calibrated power meter to determine the peak watts.

Please note that after extensive testing, we have found that simply plugging in a power meter to a wall outlet or UPS will NOT give you accurate power consumption numbers due to slight changes in the input voltage. Thus we use a Tripp-Lite 1800W line conditioner between the 120V outlet and the power meter.

With a TDP of about 220W, it the FX-9590’s chart-topping performance shouldn’t come as any surprise. Pushing the 32nm Piledriver architecture to 4.7GHz (or 5GHz in some situations) takes a fair amount of voltage and no small amount of excess PSU capacity. As a result, the FX-9590 requires about 70W more power than an FX-8350 while the current generation Haswell processors aren’t even in the picture.

Unfortunately this leads to AMD’s latest processor having an extremely poor performance per watt ratio, even when compared against Sandy Bridge-E competitors.

Temperature Testing

Click on image to enlarge​

The FX-9590 is a hot running processor and we don’t mean hot in any conventional meaning of the word either. This thing is like having a miniature nuclear reactor strapped to your motherboard; it will thoroughly overwhelm mid-tier heatsinks and AIO water coolers alike. Since it doesn’t come with an included heatsink we’re told that retailers will endeavor to bundle the FX-9590 with high end Corsair, Cooler Master or NZXT water cooling units in an effort to ensure customers won’t damage their new processors with sub-par cooling solutions.

With the potential for astronomical heat output, one would hope for an adequate way to measure temperatures. That just didn’t happen. RealTemp and CoreTemp routinely showed overly low readings and even AMD’s vaunted Overdrive utility was completely out to lunch. It claimed the chip idled at 19°C (ambient temperature was 23°) while load temperatures supposedly hit 46.7°C under load even though our Noctua NH-U14S was hot to the touch.

Only ASUS’ AI Suite II (which takes its temperature readings directly from the BIOS) was somewhat accurate with its reading of 65°C under load but we had reasons to doubt this too since, as you see in the screenshot above, our FX-9590 began throttling some cores down to the 4.515GHz mark after 20 minutes or so of continual full-load testing. Another possibility is that AMD has set Turbo Core 3.0 to begin throttling downwards when core temperature hits that 65°C mark in an effort to cap thermals and power consumption.

The lack of accurate temperature logging software poses a large problem for anyone with one of these 220W TDP chips: they have no way of knowing how hot (or cool) their processor is running. Not only will this play havoc when trying to dial in overclocks but it makes trouble-shooting stock issues all that much harder.

 

[SKYMTL]

Overclocking Results

Overclocking Results

In order to achieve a Base Clock of 4.7GHz and the potential for Turbo frequencies to hit the 5GHz mark, AMD chose only the highest leakage Piledriver CPUs for the FX-9590. As we saw in the previous pages, this results in a significant amount of heat being produced at default clocks. Add overclocking on top of that and things start to get a bit dicey, especially on air cooling.

Since our Corsair H80 and Noctua NH-U12S were both overwhelmed whenever Turbo Core was disabled and the system was left at idle (yes, you read that right: at IDLE), we decided to step things up a notch. Bringing in a Noctua NH-U14S with dual fans as a pinch hitter resulted in more acceptable temperatures and allowed for some overclocking. Though not all that much.

With Turbo Core disabled and a bit more voltage, we were able to achieve a constant frequency of 5.016GHz on ALL cores, regardless of the situation. Unfortunately, we couldn’t get this 300MHz to a point that we would consider 24/7 stable due to the elevated temperatures it caused but it was more than capable of running a few benchmarks.

The benefits were of course tangible since Turbo Core no longer stepped in to lower clock speeds or keep them at 4.7GHz. Unfortunately, even the U14S struggled to keep temperatures in the safe zone. In order to get any higher, the only solution would be to move to exotic forms of cooling, none of which we have in-hand right now but expect a follow-up article sometime in the future.

Throwing caution into the wind, we decided to push things even further and eventually hit nearly 5.2GHz. Unfortunately, while the system would boot into Windows, any multi-core load would shut the whole works down. No BSOD or OS freeze, just a complete reboot. It could be that at that point, we had reached the motherboard’s power limits but testing continues.

 

[SKYMTL]

Conclusion

Conclusion

Typically, by the time writing the conclusion comes around, I know exactly how to sum things up. Not this time. The FX-9590 is an impressive tour de force from AMD but one which ultimately caused an epic battle between my enthusiast mindset and the side of me that appreciates efficiency. In the end, the little red speed demon won out over the halo-totting tree hugger. What AMD has created may not be the best processor around and it isn’t unassailable in every situation but it is pure, unadulterated fun for anyone who appreciates technological achievements.

The FX-9590 is a CPU that says “why the hell not?” to those who question its existence and keeps with an FX-series tradition first charted years ago. It brings us back to the glory days of AMD’s FX lineup when the FX-60 and its ilk were the fastest things around. Back then, power consumption was routinely pushed aside in an effort to cater to enthusiasts’ need for leading edge performance and significant overclocking headroom. While this latest iteration doesn’t quite beat Intel’s high end offerings in every benchmark, it sure makes things interesting. On the other hand, seeing AMD go back to their enthusiast roots is nothing short of priceless.

What AMD did here was push their current architecture to its absolute limit and that deserves some credit. However this quest for the highest possible out-of-box frequencies brings forth a number of issues as well. Power consumption was nothing short of stratospheric and the amount of heat produced by the chip will leave all but the best heatsinks begging for mercy.

The extreme levels of thermal output put a damper on overclocking unless you opt for exotic cooling methods like a TEC cold plate, dry ice or LN2. We achieved a constant speed of 5GHz (with Turbo disabled) in multi-threaded applications which represents a 300MHz increase over reference frequencies. For air cooling this isn't all that bad. But then again, the FX-9590 is all about out-of-box performance since there are plenty of other FX options which overclockers can tweak to their heart’s content.

Is a 4.7 / 5GHz AMD processor ultimately enough to compete against Intel’s high-end offerings? The answer to that is multi-faceted. In properly optimized applications, the FX-9590 is a dominating presence which often runs just neck and neck with Intel’s comparatively priced Sandy Bridge-E processors. However, due to the disappointing single thread performance of the Piledriver architecture, in-game framerates in some titles tend to lag behind. There’s also just no looking past Intel’s ability to achieve similar or better performance without pushing their architecture to its absolute limit.

While actual retail pricing is a bit of an unknown at this point, if a $750 to $850 bracket remains in place AMD may have a hard time moving the few FX-9590’s they produce. You can buy a 4770K along with a fairly high end motherboard for the price of a single 5GHz processor and still achieve better gaming performance than AMD can offer.

The FX-9590 certainly isn’t for everyone, nor is it a practical solution for current AM3+ users since quite a few FX-8350 processors can reach this level of performance with some judicious overclocking. Rather, this is an achievement catering directly to the AMD enthusiasts who will appreciate the FX-9590 for what it is: a successful attempt by AMD to throw caution into the wind by building the fastest processor they possibly can.

For daring to be different and injecting some much-needed excitement into a stagnant CPU market, AMD may not have broken any performance barriers but they have certainly earned my respect.