AMD Phenom II X6 1055T & 1090T Six-Core Processors Review

[MAC]

Two six-core processor reviews six weeks apart? We must be in heaven. Well as you all have guessed AMD's very highly anticipated Phenom II X6 processors have now launched. The media embargo is over and we can finally reveal what you probably already know courtesy of the rumor mill.

The condensed specifics of the Phenom II X6 1055T are as follows: 45nm SOI manufacturing process, 2.80Ghz core clock, Turbo CORE up to 3.3Ghz across three cores, 9MB of total cache, dual-channel DDR3-1333 memory interface, 125W TDP, and a $199 USD price tag. The 1090T Black Edition is clocked at 3.2Ghz, can Turbo up to 3.6Ghz, also has 9MB of total cache, a dual-channel DDR3-1333 memory interface, 125W TDP, and a $285 price tag. Being a Black Edition model the 1090T also has fully unlocked multpliers.

Unlike the only other six-core processor on the market, the Intel Core i7-980X, these chips are targeted at your average enthusiast, someone who is looking for a lot more multi-threading power than their current AM2+/AM3 processor can deliver. Yes, to make full use of these processors' capabilities you need to have software that supports more than four threads, but that is becoming more and more common nowadays. As you will see in the coming pages, these new Thuban chips will absolutely appeal to those who take their multi-threading performance serious, yet who don't have an unlimited budget, nor a desire to upgrade their whole system. Speaking of which, the best part is that these new processors are compatible with over 160 AM2+ and AM3 motherboards, with a mere BIOS update.

These new chips should prove popular among more than just the AMD faithful though. Because AMD really haven't had any way to challenge the high-end/upper-mainstream dominance of Nehalem-based chips like the Core i7 series, most enthusiast really haven't had their sights on AMD processors in far too long. However, after having extensively tested the Phenom II X6, we strongly believe that this product launch might be the turning point that converts a large swathe of price-conscious powers users over to AMD's side.

For all you AMD owners though, this is definitely the upgrade path that you have been patiently waiting for...

 

[MAC]

Codename 'Thuban' - Phenom II X6

Codename 'Thuban' - Phenom II X6 1055T & X6 1090T

Thuban Die - Click on image to enlarge​

Thuban is a codeword that any self-respecting computer enthusiast should be familar with by now, at least those who yearn for excellent multi-threading performance without having to sell a kidney (*cough* Core i7-980X *cough*). The Phenom II X6 processors that we are reviewing today are the new Top of the Line models in AMD's CPU roster, and the very first affordable/mainstream six-core processors on the market. Although we are only reviewing the Phenom II X6 1055T and X6 1090T models at the moment, if you have been keeping up with internet chatter you know that there are other models right around the corner, but our lips are contractually sealed for now.

So you want the juicy details right off the bat? Here they are:

Despite still being based on the venerable 45nm SOI manufacturing process, AMD have not held back with regards to clock speeds, with the 1055T and 1090T coming in at 2.8Ghz and 3.2Ghz respectively. This is just 200Mhz off of the company's highest clocked quad-core model, the 3.4Ghz Phenom II X4 965 BE. What's even more impressive is that both of these new hexa-core models are rated at 125W thermal design power (TDP), which is identical to the higher-end quad-core models. How have they accomplished this while adding two additional cores? Well GlobalFoundries have obviously managed to further improve the 45nm process that they inherited from AMD's manufacturing division, but we will get into that later.

Aside from the six cores, high clocks and cheap price, what really has got people interested in these Phenom II X6's is the new Turbo CORE feature. Every model ending with a T will have this new feature, but it's not equal on every model. The models ending in xxx5T will have Turbo CORE capabilities up to 500Mhz, but the xxx0T models top out at 400Mhz. Not a massive difference in the grand scheme of things.

Now it is time to take a closer look at AMD's new processors themselves, starting with the packaging of course.

​

Instead of showing you the usual media sample white box, we used our friends in high places to get our hands on a retail Phenom II X6 1090T Black Edition model. This model's packaging is effectively identical to that of previous Phenom II Black Edition processors. It's quite a bit smaller than even Intel's redesigned LGA1156 boxes.

Click on image to enlarge​

Inside the retail box you have the carefully packaged processor, the manual with serial number sticker pre-affixed, and a smaller box containing the stock CPU cooler. As far as we can tell this stock CPU cooler is identical to the one that was packaged with the Phenom II X4 965 Black Edition, which makes sense since both processors are rated for the same TDP.

Pre-Retail 1055T/ Pre-Retail 1090T / Retail 1090T - Click on image to enlarge​

Forgive the noticeable Photoshop blurring on all three processors, but we had to edit out some details in order to protect our much beloved sources.

As you can see the two pre-retail chips have a slightly different stepping code than the retail one. What this usually indicates is that there was a change/revision in the CPU manufacturing process. In practice what we noticed was that the retail processor had a lower operating voltage when Turbo CORE kicked in (1.428V vs. 1.440V). This shouldn't concern those who will being purchasing their processors in the retail channel, since they should all be the newest & freshest chips.

Aesthetically speaking these new Phenom II X6's are identical to the Phenom II X2 and X4 models. Speaking of which...

X6 1090T / X6 1055T/ X4 945 - Click on image to enlarge​

This picture represents the biggest feather in AMD's hat. All three chips share the same AM3 form factor, a total of 938 pins each, and compatibility with approximately 160 AM2+ and AM3 motherboards. By not switching sockets all the time AMD has ensured that their newest processors will work on older motherboards with a mere BIOS update, giving users a simple and direct upgrade path. Intel could learn a lesson here...

Click on image to enlarge​

As always, here are the obligatory CPU-Z screenshots. Now these screenshots were taken with our pre-retail chips which had higher voltages than our retail sample did. If we are to trust CPU-Z, all three chips idled at approximately 1.212V. However, between 4 to 6 cores were in use the pre-retail chips ran at 1.272V, whereas the retail chip only needed 1.248-1.260V. When 1 to 3 cores were loaded and Turbo CORE kicked in, the vCore on the pre-retail chips would increase to 1.440V, while the retail one only needed 1.428V. AMD claims that the voltage range of the retail chips is 1.125-1.40V, but as you can see here our retail X6 1090T certainly went a little above that. Despite these seemingly high voltages, these new processors have a 62C max temp, so they will undoubtedly run cool, just like previous Phenom II's.

As you can see from these screenshots, Thuban processors have same 6MB of L3 cache as the Phenom II X4 models, but 50% more L2 cache. This is because while each core has its own L2 cache allocation, the L3 cache is unified and shared across all 6 cores. While the northbridge/integrated memory controller (IMC) frequency remains at 2000Mhz, and official DDR3 support tops out at DDR3-1333, we do know that AMD have tweaked the X6's to have significantly greater memory overclocking headroom than previous chips.


Next let's take a quick look at the new enthusiast chipset that AMD have launched to round out the high-end of their enthusiast 'Leo' platform.

 

[MAC]

AMD 890FX Chipset Revealed

AMD 890FX Chipset Revealed

AMD’s new range of 800-series motherboards spans nearly every single pice range but even the high-end 890FX deftly avoids a price range that will earn it a “very nice but too expensive” comment from consumers. But what makes this chipset tick?

Yes, we know this is the customary and somewhat cliché block diagram that will surely be seen in an untold number of articles but it does go a long way into explaining the functionality aspects of the 890FX / SB850 combination. We have commented that the overall layout of the 890FX was introduced and later refined. Let’s start at the top of this diagram and make our way down.

One of AMD’s most distinguishing features over the last few product generations is their commitment to offering backwards compatibility for their motherboards and forwards compatibility on the processor end of things. Basically, even though the 890FX series boards will feature an AM3 socket making them compatible with all current processors as well as upcoming Phenom X6 products. This will in effect also allow older AM2+ processors to be used on these boards for those of you who may only want to upgrade one component at a time.

While Intel has transferred most of their Southbridge functionality to their processor die, AMD has staunchly adhered to a two-part chipset configuration consisting of a Northbridge and a Southbridge. While this may change in the future, for the time being all AMD AM3 processors only feature on-chip DDR3 memory controllers while all the I/O and expansion slot functions are handled through the chipset. With the 890FX and AM3 processors, we also get native support for DDR3 1333Mhz memory while the processor and Northbridge communicate through a Hypertransport 3.0 link at 5.4 GT/s which should eliminate any bottlenecks.

The Northbridge of AMD’s two-chip solution acts as a controller hub for most of the board’s PCI-E 2.0 lanes and facilitates any communication between the SB850 and the processor. The 32 dedicated graphics card lanes are split up into either two 16x slots or can be evenly dispersed for up to a quartet of 8x slots which gives the 890FX incredible flexibility for Crossfire setups. Meanwhile, the remaining ten PCI-E 2.0 lanes are divided up into one grouping of four lanes while an additional six lanes 1x lanes can be dispersed as needed for integrated components like audio and networking controllers.

Moving on down to the shiny new SB850, we see that AMD has updated the chipset interconnect and is now using a 2GB/s interface dubbed “Alink Express III”. Although we’re not sure exactly what has changed, the older Alink Express II was essentially a 4x PCI-Express 1.1 lane, so bandwidth appears to have been increased – likely to a 4x 2.0 lane - for improved chipset to chipset communication performance.

Speaking of the Southbridge, the most significant new feature that is has been brought to the table is 6Gbps SATA support. That’s right, those lucky enough to own one of the new Sandforce 1500 based SSDs can now enjoy Read/Write performance well beyond 300MB/s. Aside from updated SATA support, the remainder of the Southbridge is consistent with the older SB750. We unfortunately don’t get to enjoy integrated USB 3.0 support at this point in time, as the SB850 remains a USB 2.0 controller.

On the topic of USB 3.0, we should note that AMD was very careful to point out that USB 3.0 controllers can be interfaced to the chipset using the 1x PCI-Express 2.0 lanes for a maximum theoretical throughput of 500MB/s. Coincidentally, Intel’s new H55 and H57 are limited to half bandwidth lanes and a maximum of 250MB/s to off-chip USB 3.0 and SATA 3.0 controllers. This likely won’t be of concern for USB 3.0, but having on-chip SATA 3.0 support is certainly a benefit as the only bottleneck is the 2GB/s Alink interface between the chipsets and the 1x component interface lanes don’t need to be used at all.

 

[MAC]

Test Setups & Methodology

Test Setups & Methodology

For this review, we have prepared four different test setups, representing all the popular platforms at the moment, as well as most of the best-selling processors. As much as possible, the four test setups feature identical components, memory timings, drivers, etc. Aside from manually selecting memory frequencies and timings, every option in the BIOS was at its default setting.

AMD Phenom II AM3 Test Setup​

​

Intel Core i3/i5/i7 LGA1156 Test Setup​

​

Intel Core i7 LGA1366 Test Setup​

​

Intel Core 2 LGA775 Test Setup​

​

For all of the benchmarks, appropriate lengths are taken to ensure an equal comparison through methodical setup, installation, and testing. The following outlines our testing methodology:

A) Windows is installed using a full format.

B) Chipset drivers and accessory hardware drivers (audio, network, GPU) are installed followed by a defragment and a reboot.

C)To ensure consistent results, a few tweaks were applied to Windows Vista and the NVIDIA control panel:

• Sidebar – Disabled
• UAC – Disabled
• System Protection/Restore – Disabled
• Problem & Error Reporting – Disabled
• Remote Desktop/Assistance - Disabled
• Windows Security Center Alerts – Disabled
• Windows Defender – Disabled
• Screensaver – Disabled
• Power Plan - High Performance
• NVIDIA PhysX – Disabled
• V-Sync – Off

D) Programs and games are then installed & updated followed by another defragment.

E) Windows updates are then completed installing all available updates followed by a defragment.

F) Benchmarks are each run three times after a clean reboot for every iteration of the benchmark unless otherwise stated, the results are then averaged. If they were any clearly anomalous results, the 3-loop run was repeated. If they remained, we mentioned it in the individual benchmark write-up.

Here is a full list of the applications that we utilized in our benchmarking suite:

• 3DMark06 Professional v1.2.0
• 3DMark Vantage Professional Edition v1.0.2
• Cinebench R10 64-bit
• Cinebench R11.5 64-bit
• Crysis v1.21
• Far Cry 1.02
• HyperPi 0.99b
• wPRIME 2.03
• Lame Front-End 1.0
• Lavalys Everest Ultimate v5.50.2109 Beta
• Left 4 Dead
• PCMark Vantage Advanced 64-Bit Edition (1.0.1)
• Photoshop CS4 Extended (64-bit)
• ScienceMark 2.0 Build 21MAR05
• Street Fighter 4 Demo
• Valve Particle Simulation Benchmark
• WinRAR 3.8.0
• World in Conflict v1.010
• x264 HD Benchmark v1.0

That is about all you need to know methodology wise, so let's get to the good stuff!

 

[MAC]

Feature Test: Turbo CORE Technology

Feature Test: Turbo CORE Technology

So you want to know how AMD's implementation of Turbo CORE works? Well let us direct you to the following chart:

Confused? Don't worry, we were too. Here is a very basic explanation.

Processor performance states (P-States) are effectively the supported operating frequencies and voltages that the processor can switch between in order to modulate power consumption and lower heat output. These states are controlled by the ACPI function in the operating system depending on idle/load status. Processors can move in and out of these P-states very quickly, seamlessly to the user. The lower the P-state number, the higher the processor speed. P1 in this chart is the 'stock' frequency.

With the Phenom II X6 processors, AMD have implemented a new P-state, Turbo/P0, that is only switched to under single, dual, or triple-threaded workloads. In those cases, the 3 or 4 or 5 idle cores go into a low-frequency halt state (800Mhz), the voltage is increased for across all of the cores, and the additional power and thermal headroom is used to increase the frequency of the 1 or 2 or 3 cores that are currently loaded. If your workload requires four or more cores, the processor will simply run at default clock speeds.

Although great to have one way or another, we could definitely make the case that AMD's Turbo implementation is at least one generation behind Intel's. With Intel Turbo Boost, the unused cores are shut off, this means that they aren't drawing any power whatsoever. As a result, a larger percentage of power envelope can used allocated towards boosting the loaded cores. AMD's solution merely put the cores into idle, in which case they are still drawing power, thus reducing the theoretical Turbo capabilities of chip.This is arguably why AMD can't have the Turbo feature enabled on more than 3 cores.

Nevertheless, here is how the Turbo CORE implementation works in real-time:

1055T on the left, 1090T on the right - Click on image to enlarge​

The 1055T 2.8Ghz will Turbo up to 3.3Ghz when up to 3 cores are loaded, the 1090T 3.2Ghz will Turbo up to 3.6Ghz when up to 3 cores are loaded. Why only enable Turbo on 3 cores? Well as mentioned above that was a likely decided based on thermal and power considerations, since they do have to say within that 125W power envelope. Having said that, we would have liked to see a mid-way Turbo point for four-threaded workloads.

Since this is a performance-oriented feature, let's see how the Turbo CORE function affects real applications:

As you can see there are worthwhile gains in apps that use less than four-cores, but that's becoming less and less common nowadays. Thankfully, if you do encounter a highly threaded application you do have those wonderful six cores are your disposal. On a side note, we do have to point out that we did encounter a few occasions where the Turbo performance was not as high as expected. We suspect that this issue was caused by the operating system shifting the workload from core-to-core, but that's just conjecture at this point.

 

[MAC]

Synthetic Benchmarks: Everest CPU & Memory

Synthetic Benchmarks: Everest CPU & Memory

Lavalys Everest Ultimate

Everest Ultimate is the most useful tool for any and all benchmarkers or overclockers. With the ability to pick up most voltage, temperature, and fan sensors on almost every motherboard available, Everest provides the ability to customize the outputs in a number of forms on your desktop. We selected two of Everest's seven CPU benchmarks: CPU Queen and FPU Mandel. According to Lavalys, CPU Queen simple integer benchmark focuses on the branch prediction capabilities and the misprediction penalties of the CPU. It finds the solutions for the classic "Queens problem" on a 10 by 10 sized chessboard. At the same clock speed theoretically the processor with the shorter pipeline and smaller misprediction penalties will attain higher benchmark scores. The FPU Mandel benchmark measures the double precision (also known as 64-bit) floating-point performance through the computation of several frames of the popular "Mandelbrot" fractal. Both tests consume less than 1 MB system memory, and are HyperThreading, multi-processor (SMP) and multi-core (CMP) aware.

Lavalys Everest Ultimate

Everest Ultimate is the most useful tool for any and all benchmarkers or overclockers. With the ability to pick up most voltage, temperature, and fan sensors on almost every motherboard available, Everest provides the ability to customize the outputs in a number of forms on your desktop. In addition to this, the memory benchmarking utility provides a useful tool of measuring the changes to your memory sub-system.

.


Will ScienceMark 2.0 paint a different picture? Let's find out.

ScienceMark v2.0

Although last updated almost 3 years ago, and despite its rudimentary interface, ScienceMark v2.0 remains a favorite for accurately calculating bandwidth on even the newest chipsets.

These are just synthetic numbers though, real-life applications and games are what count. Let's check those out next.

 

[MAC]

System Benchmarks: SuperPI 32M / wPRIME 1024M

System Benchmarks: SuperPI 32M / wPRIME 1024M

SuperPi Mod v1.5

When running the SuperPI 32MB benchmark, we are calculating Pi to 32 million digits and timing the process. Obviously more CPU power helps in this intense calculation, but the memory sub-system also plays an important role, as does the operating system. We are running one instance of SuperPi via the HyperPi 0.99b interface. This is therefore a single-thread workload.

wPRIME 2.03

wPrime is a leading multithreaded benchmark for x86 processors that tests your processor performance by calculating square roots with a recursive call of Newton's method for estimating functions, with f(x)=x2-k, where k is the number we're sqrting, until Sgn(f(x)/f'(x)) does not equal that of the previous iteration, starting with an estimation of k/2. It then uses an iterative calling of the estimation method a set amount of times to increase the accuracy of the results. It then confirms that n(k)2=k to ensure the calculation was correct. It repeats this for all numbers from 1 to the requested maximum. This is a highly multi-threaded workload.

 

[MAC]

System Benchmarks: PCMark Vantage / Photoshop CS4

System Benchmarks: PCMark Vantage / Photoshop CS4

PCMark Vantage x64

PCMark Vantage Advanced 64-bit Edition (1.0.0.0)
PCMark Suite / Default Settings
Comparison: Generated Score

The main focus of our General Tasks category lies with the most recent installment of the PCMark series, Vantage. While still classified under the description of a Synthetic benchmark, PCMark Vantage uses many of Vista's (Note - Vantage is Vista-only) built-in programs and features along with its own tests, so it is "real-world" applicable in regards to CPU performance. The following is a general list of the tests in the PCMark suite, very much in line with tasks of an average user: Data encryption, Data compression, CPU image manipulation (compression/decompression/resize), Audio transcoding, Video transcoding, Text editing, Web page rendering, Windows Mail, Windows Contacts, and CPU game test.

Photoshop CS4

For the image editing portion of this review, we will use Photoshop CS4 in coordination with Driver Heaven’s Photoshop Benchmark V3, which is an excellent test of CPU power and memory bandwidth. This is a scripted benchmark that individually applies 15 different filters to a 109MB JPEG, and uses Photoshop’s built-in timing feature to provide a result at each test stage. Then it’s simply a matter of adding up the 15 results to reach the final figure.

 

[MAC]

System Benchmarks: Cinebench R10 / Cinebench R11.5

System Benchmarks: Cinebench R10 / Cinebench R11.5

Cinebench R10

Cinebench R10 64-bit
Test1: Single CPU Image Render
Test2: Multi CPU Image Render
Comparison: Generated Score

Developed by MAXON, creators of Cinema 4D, Cinebench 10 is designed using the popular Cinema software and created to compare system performance in 3D Animation and Photo applications. There are two parts to the test; the first stresses only the primary CPU or Core, the second, makes use of up to 16 CPUs/Cores. Both are done rendering a realistic photo while utilizing various CPU-intensive features such as reflection, ambient occlusion, area lights and procedural shaders

Cinebench R11.5

Cinebench R11.5 64-bit
Test1: CPU Image Render
Comparison: Generated Score

The latest benchmark from MAXON, Cinebench R11.5 makes use of all your system's processing power to render a photorealistic 3D scene using various different algorithms to stress all available processor cores. The test scene contains approximately 2,000 objects containing more than 300,000 total polygons and uses sharp and blurred reflections, area lights and shadows, procedural shaders, antialiasing, and much more. This particular benchmarking can measure systems with up to 64 processor threads. The result is given in points (pts). The higher the number, the faster your processor.

 

[MAC]

System Benchmarks: LFE / x264 HD / WinRAR

System Benchmarks: Lame Front End / x264 HD / WinRAR

Lame Front End

Lame Front End v1.0 is a single-threaded application, which means that it only utilizes a single processor core. This will obviously limit performance but it will allow us to gauge a processor's single-threaded performance as well as test any turbo feature that it might have. We will be encoding a WAV rip of Santana’s Supernatural album and converting it to MP3 using the highest fidelity VBR 0 quality preset.

x264 HD Benchmark

x264 HD Benchmark v1.0
Test: MPEG-2 HD 720P Video Clip Conversion to x264
DVD Video Length: 30 Seconds
Comparison: FPS of Second Pass

x264 is quickly becoming the new codec of choice for encoding a growing number of H.264/MPEG-4 AVC videos. Think of it as the new Divx of HD and you can understand why we felt it critical to include. Tech Arp's recent development of the x264 HD Benchmark takes a 30 second HD video clip and encodes it into the x264 codec with the intention of little to no quality loss. The test is measured using the average frames per second achieved during encoding, which scales with processor speed and efficiency. The benchmark also allows the use of multi-core processors so it gives a very accurate depiction of what to expect when using encoding application on a typical full length video.

WinRAR

WinRAR 3.8.0
Test: Compression of 1GB of Assorted Files
Comparison: Time to Finish

One of the most popular file compression/decompresion tools, we use WinRAR to compress a 1GB batch of files and archive them, timing the task until completion.