AMD Phenom II X2 560 BE Review

Processors with unlocked multipliers are sort of like the Holy Grail of overclocking. AMD’s Black Edition (BE) chips, similar to the Intel K series, offer this versatility to enthusiasts. Recently, AMD introduced several new processors and one of these is the Phenom II X2 560 BE. This is just a small increase in stock multiplier above the existing Phenom II X2 555 BE processor (16.5 vs. 16) and a small increase in price ($105 vs. $88), but the silicon should be stronger which should result in better overclocking and core unlocking.

Phenom II Logo (picture courtesy AMD)
Phenom II Logo (picture courtesy AMD)

For those of you who aren’t familiar with unlocked multipliers, it is another tool in the overclocker’s arsenal. The important distinction is that you can increase the multiplier, rather than just lower it as on typical processors.  Without the ability to increase the multiplier, the only way to overclock is by increasing the FSB (Base Clock or HT Reference) which can put too much stress on other components.  Having the ability to tweak both the multiplier and the FSB gives overclockers much more control and flexibility.

Specifications

Our victim: the new 560
Our victim: the new 560
  • Product Name: Phenom II™ X2 560 Black Edition Processor
  • Model Number & Core Frequency: X2 560 / 3.3GHz
  • OPN: HDZ560WFK2DGM
  • L1 Cache Sizes: 64K of L1 instruction and 64K of L1 data cache per core (256KB total L1 per processor)
  • L2 Cache Sizes: 512KB of L2 data cache per core (1MB total L2 per processor)
  • L3 Cache Size: 6MB (shared)
  • Total Cache (L2+L3): 7MB
  • Memory Controller Type: Integrated 128-bit wide memory controller *
  • Memory Controller Speed: Up to 2.0GHz with Dual Dynamic Power Management
  • Memory Supported: Unregistered DIMMs up to PC2-8500 (DDR2-1066MHz) and PC3-10600 (DDR3-1333MHz)
  • HyperTransport 3.0 Specification: One 16-bit/16-bit link @ up to 4.0GHz full duplex (2.0GHz x2)
  • Total Processor-to-System Bandwidth: Up to 37.3GB/s total bandwidth [Up to 21.3 GB/s memory bandwidth (DDR3-1333) + 16.0GB/s (HT3)] or up to 33.1GB/s total bandwidth [Up to 17.1 GB/s memory bandwidth (DDR2-1066) + 16.0GB/s (HT3)]
  • Packaging: Socket AM3 938-pin organic micro pin grid array (micro-PGA)
  • Fab. location: GLOBALFOUNDARIES Fab 1 module 1 in Dresden, Germany (formerly AMD Fab 36)
  • Process Technology: 45-nanometer DSL SOI (silicon-on-insulator) technology
  • Approximate Die Size: 258mm2
  • Approximate Transistor count: 758 million
  • Max TDP: 80 Watts
  • AMD Codename: “Calisto”
  • *Note: configurable for dual 64-bit channels for simultaneous read/writes

X2 is the same die as the X4 but only half the cores activated (picture courtesy AMD)
X2 is the same die as the X4 but has only half the cores activated (picture courtesy AMD)

Let’s compare that to the 555:

  • Frequency: 3.2GHz (16*200 MHz)
  • Total L2 Cache: 1MB
  • Packaging: Socket AM3
  • Thermal Design Power: 80W
  • Process Technology: 45nm SOI

The processors are essentially the same except for the stock multiplier. Like the 555 and other Phenom II X2 processors they are the same exact die as the X4 chips but with two of the cores deactivated. They are binned down, which means the two other cores were not strong enough to run at stock specs but does not mean they will not work at all. This is great news for enthusiasts: with certain motherboards like the Biostar TA890FXE used in this review, you can enable the two hidden cores and hopefully (with a little more voltage supplied to the processor) have yourself a quad core (X4) processor for the price of an dual core (X2). Similarly, the stock multiplier bump to 16.5 could mean that these 560’s are better silicon than the 555’s, which could hint at better overclocking performance.

Performance

Marketing can often be deceiving, but benchmarks never lie… or something like that. At any rate, I have put this 560 through several hours of torture to see what it is really capable of. I’ll be using the same Biostar TA890FXE that I previously reviewed which will allow me to compare the numbers this processor achieves to those of the Athlon II X4 640. I’d have liked to see directly how this does compared to the 555, but unfortunately I don’t have one at my disposal. I’ve used the numbers from HWBot for a frame of reference. There is much more diversity in the hardware used in the HWBot database, so ther numbers aren’t directly comparable but will still give a good idea of relative performance.

HWBot data for the Phenom II X2 555 BE:

  • Single Task: Average Super Pi 1M – 16.703 s
  • Multitasking: Average wPrime 32M – 15.124 s
  • Average Overclock (MHz): 2x 3200 (4531)

The table data is from all submissions with all types of cooling, including dry ice and liquid nitrogen, but the averages shouldn’t be too far off from what is possible on air. Filtering the data a little, I found the maximum overclock with air cooling was 4630 MHz with 1.62 V. Knowing that, the 4531 MHz average listed above looks achievable.

Testing platform:

  • Processor: AMD Phenom II X2 560 Black Edition
  • Motherboard: Biostar TA890FXE
  • Memory: G.Skill 2x 1 GB DDR3-2000 Trident
  • Power Supply: NZXT HALE90 850 W
  • Heat sink: Zalman CNPS-9900

Results

When overclocking this chip, I was able to reach a maximum overclock of 4300 MHz, which is just shy of my goal of 4500 MHz.  When running tests, I was only able to complete them all at a maximum of 4133 MHz.  I was able to pass a 1 hour test of OCCT at that speed, but did not have time to test if that was 4 hour Prime 95 stable.  I was also able to reach a maximum HT Reference of 290 MHz, which is higher than I was able to achieve when running the Athlon II X4 640.  What’s even more exciting is that I was able to unlock to 4 cores and run 10 minutes of Prime95 at 4 GHz (20 x 200 MHz).  This is impressive to me since I borrowed a Phenom II X2 555 to test the unlock feature when I first reviewed the Biostar TA890FXE but I could not get that processor stable with 4 cores even at stock speed.

Unlocked to X4 running at 4 GHz
Unlocked to X4 running at 4 GHz

People use their computers for several reasons, many of which can benefit from increased performance. Synthetic benchmarks try to emulate the real world performance but it doesn’t always translate perfectly. At the same time, rendering benchmarks like Cinebench are much better at showing actual performance for the task they are designed to simulate.  Screen shots with stock scores are included in the gallery at the end of the article.

Everest CPU AES at 4133 MHz
Everest CPU AES at 4133 MHz

Everest CPU Photoworxx at 4133 MHz
Everest CPU Photoworxx at 4133 MHz

Everest CPU Zlib at 4133 MHz
Everest CPU Zlib at 4133 MHz

Everest CPU Queen at 4133 MHz
Everest CPU Queen at 4133 MHz

Everest FPU Julia at 4133 MHz
Everest FPU Julia at 4133 MHz

Everest FPU Mandel at 4133 MHz
Everest FPU Mandel at 4133 MHz

Everest FPU SinJulia at 4133 MHz
Everest FPU SinJulia at 4133 MHz

Cinebench R10 at 4133 MHz
Cinebench R10 at 4133 MHz

Cinebench R11 at 4133 MHz
Cinebench R11 at 4133 MHz

Additionally, we have the results that matter to competitive overclockers: SuperPi, PiFast, wPrime, and 3DMark. These may translate in to real world performance, but the boints (bot points on HWBot) matter so much more!  The Athlon II X4 tests were run at different speeds according to what settings would allow the processor to pass the test.  I pulled the numbers straight from my Biostar TA890FXE review, so check that out for more details.

Super Pi 1M, wPrime 32M, and pifast
Super Pi 1M, wPrime 32M, and pifast

wPrime 1024, Super pi 32M, 3DMark Vantage, and 3DMark 06
wPrime 1024, Super pi 32M, 3DMark Vantage, and 3DMark 06

Super Pi 1M at 4133 MHz
Super Pi 1M at 4133 MHz

PiFast at 4133 MHz
PiFast at 4133 MHz

wPrime at 4133 MHz
wPrime at 4133 MHz

Super Pi 32M at 4133 MHz
Super Pi 32M at 4133 MHz

3DMark Vantage at 4133 MHz
3DMark Vantage at 4133 MHz

3DMark 06 at 4133 MHz
3DMark 06 at 4133 MHz

Conclusion

The Phenom II X2 560 BE is a welcome addition to the Black Edition family and is a normal product refresh.  It doesn’t overclock particularly better than the 555, but the silicon is a little more stable allowing you to essentially have a Phenom II X4 on the cheap.  Considering the 560 retails around $105 and the X4 970 retails around $185, this is a perfect example of where overclocking gets more value for your dollar.  If you want a fantastic overclocking chip, you’ll have to look elsewhere, but this processor certainly holds it’s own and shouldn’t be completely overlooked.

splat

More pictures

Loading new replies...

J
JackNSally

Member

810 messages 0 likes

You have the multi for the 555 BE wrong. It should say 16, not 15. Otherwise, nice article.

Reply Like

Avatar of splat
splat

ASCII Moderator

3,956 messages 0 likes

you are correct, thanks for spotting the typo

Reply Like

Q
QuietIce

Disabled

15,734 messages 0 likes

Solid info - thanks for sharing ... :clap:

Reply Like

Z
Zzyzx

New Member

1 messages 0 likes

I just built my new rig around this chip.

I bought the AMD 560 on PR sale @ Fry's for $80! I popped it into my new ASUS CROSSHAIR IV Formula mobo($195)and it unlock a 3rd core, so I have an x3 B60 CPU core. What's great about my piece of silicon is, I've have it at 4.2Ghz @1.40v. Yes, 4.2Ghz @ 1.4 volts running Sandra Lite & Prime 95 no problems for 2 hours (temps 42-45c)!

I must be really selfish, because I wanted the 4th core to unlock, but now, I realize it's more important to have a solid chip that overclocks higher and at lower voltages. I currently have a Zalman copper 9700 which is an okay average cooler, got it brand new for $20 mounted with IC 7 Carat paste. When I get my Lian Li B71 rig($150) complete with additional/better cooling fans and a good CPU cooler, I'll run my OC tests.

I say the deal/value with this AMD Phenom II x2 560 BE is EXCELLENT... would of been OUSTANDING if the 4th core unlock. I know I'll be able to run this 560 running 4.5Ghz in the future.

Very Happy :bday:

Reply Like

click to expand...
Avatar of ScrewySqrl
ScrewySqrl

Member

253 messages 0 likes

"but benchmarks never lie…"

I've alreays been of the opinion that there are lies, damn lies, statistics, and THEN benchmarks.

:D :D :D

Reply Like

Q
QuietIce

Disabled

15,734 messages 0 likes
Avatar of Max0r
Max0r

Member

1,060 messages 39 likes

I've alreays been of the opinion that there are lies, damn lies, statistics, and THEN benchmarks.

:D :D :D

Yea, nothing like getting nice benchmark score increases, then opening up your games and getting the same FPS :rock:

Reply Like

Avatar of hokiealumnus
hokiealumnus

Water Cooled Moderator

16,561 messages 25 likes

Yea, nothing like getting nice benchmark score increases, then opening up your games and getting the same FPS :rock:

Well...benchmark scores are measured in the thousands typically. Games are measured in the dozens, if that. It's harder to increase one FPS than it is 100 points. FWIW.

Reply Like

Q
QuietIce

Disabled

15,734 messages 0 likes

Well...benchmark scores are measured in the thousands typically. Games are measured in the dozens, if that. It's harder to increase one FPS than it is 100 points. FWIW.

That's why I always try to look at them in terms of percent increase/decrease between scores. It keeps me more focused on what is actually happening rather than getting stuck in that numbers game. "Oh, look, those settings gave me 100 points more than those other settings!"

(Yeah, 100 out of 20,000 - big deal. There's more error in the test than that! :rolleyes: )

Reply Like