Wednesday, 29 August 2012

AMD FX-6200 Processor Performance Benchmarks Unlocking core Secrets Review

AMD FX-6200 CPU Review: A Small Bulldozer Refresh

The FX-6200

Ryan and I have covered the Bulldozer architecture in previous articles, so I will not go over the finer details here. I will discuss some of the basics of the chip and the architecture.
The FX-6200 is based off of the same Bulldozer revision as the previous FX series of parts that was released last October. There may have been some minor changes along the way, but they would have more to do with manufacturing rather than any kind of base silicon or extreme metal layer change. The product is still built by GLOBALFOUNDRIES on their 32 nm HKMG SOI process. The chip is a native 4 module/8 core product, but one of the modules has been fused off and is unavailable for unlocking. This leaves 3 modules/6 cores for the processor to work with. Each module features 2 MB of L2 cache to be shared between the two integer units and the single FPU/MMX/SSE/AVX unit, for a total of 6 MB of L2 for the entire CPU. The L3 features the full 8 MB of cache that is available on fully enabled CPUs.

View Full Size

The full meal deal of a retail/boxed processor.  The 3 year warranty and heatsink/fan are likely worth it to most users.


The base clock is now set at 3.8 GHz, which is faster than the 3.6 GHz stock clock for the FX-8150. To achieve this clock AMD raised the TDP to 125 watts, which is up from the 95 watts of the FX-6100. The max Turbo Core speed of the FX-6200 is 4.1 GHz. AMD did raise the Northbridge/L3 cache speed on the FX-6200, and it runs at 2.2 GHz rather than the 2.0 GHz that the FX-6100 was set to. The memory controller is heavily revised from the previous Phenom II generation of parts, and it can handle official speeds up to DDR-3 1866 MHz.
The retail package comes with a pretty hefty heatsink and fan combination that should be able to handle the extra thermal load that the 125 watt CPU provides. The CPU has a standard 3 year warranty for the boxed version, while a tray chip has a 30 day warranty. The boxed product retails for $169 US, but can often be found cheaper with instant rebates and sales.
The FX processors are all unlocked so it makes them very easy to overclock. Most retail motherboards cover all of the settings needed to increase the multipliers on the CPU and overclock the chip. Success of any overclock is not guaranteed and the choice of motherboard will also directly affect the ability of the chip to reach higher speeds.

Setting it Up

This is the first FX processor that I have been able to lay hands on. As such, a few extra precautions must be taken before testing commences.
I am comparing this processor to the previous Phenom II X6 1100T and Phenom II X4 980. I used the same motherboard for testing, but I had to stagger BIOS revisions between them. The 8 mbit BIOS chip only allows so much room for CPU optimizations for different architectures. Earlier BIOS were optimized for the Phenom II series of chips, while the later BIOS revisions focus on improving Bulldozer performance. There is just not enough room in the BIOS to allow for both sets of optimizations to coexist. This is at least the explanation given to me, and it seems to make sense.

View Full Size

The chip in its plastic container with a nice FX sticker thrown in for giggles.  The AMD chip is identical to previous iterations due to the integrated heat spreader and organic substrate that has been featured for years now.

Test Setup

I really wanted to see how the new FX architecture really stood up against the last generation of 45 nm products. Unfortunately, Ryan has all of the Intel based products for comparative purposes, so I was unable to get really accurate benchmarks featuring those chips. The big thing to keep in mind is that for the majority of benchmarks out there the Intel i5 2400 is very close in performance to the AMD Phenom II X6 1100T. Using that as the foundation of understanding Intel’s performance at $179 and up, readers should be able to determine if the FX-6200 is comparable to the lowest end products from Intel.

View Full Size

At first the heatsink does not look all that impressive...

All updates were applied to the Windows 7 install, including the two Bulldozer aimed hotfixes. The recently reviewed MSI R7970 Lightning was used as the video card in the tests and was clocked at 1070 MHz core and 1400 MHz memory. The Catalyst 12.3 drivers were used for testing.
For comparative purposes I used the two highest end Phenom II processors that AMD has released. The Phenom II X4 980 is clocked at 3.7 GHz, and was previously the fastest clocked processor that AMD (or anyone else at the time) had released in the consumer market. The Phenom II X6 1100T was the previous top end processor from AMD with a base clock of 3.3 GHZ for its six cores, or a turbo speed of 3.7 GHz for three cores.
I finally overclocked the FX-6200 from its base 3.8 GHz (4.1 GHz turbo) up to 4.5 GHz with no turbo enabled. I will discuss the actual overclock in the relevant section, but I thought it prudent to provide benchmarks for this speed. The results are interesting, to say the least.

Asus SABERTOOTH 990FX Motherboard
MSI R7970 Lightning Graphics Card
2 x 4GB GSkill DDR-3 1600 @ 9.9.9.24 timings
2 x 4GB GSkill DDR-3 1866 @ 9.10.9.24 timings
WD Caviar Black 1TB SATA 6G HD
Lite-On Blu-Ray Drive
Antec HCP-1200 Power Supply
Windows 7 64 Bit Ultimate
Catalyst 12.3 Drivers

SiSoft Sandra 2012 SP2

These tests look at ALU, FPU, SSE/AVX, and memory performance.

View Full Size

We see some interesting results coming in here. The FX-6200 at stock speed is pushing ahead of the Phenom 1100T in integer based tests, but falls behind in floating point. It is not until we overclock the FX-6200 that the float performance overcomes that of the previous 6 core part.

View Full Size

In multimedia performance we see the new architecture really strut its stuff in AVX/integer performance. Once we get into float type numbers we again see a drop from what the previous 1100T was able to provide. It again takes a pretty significant overclock for the FX-6200 to overcome the disparity.

View Full Size

In terms of memory performance, the new memory controller of the FX takes charge. When overclocked the FX starts putting out some pretty impressive numbers.

3DMark Vantage

3DMark Vantage is actually a good benchmark for CPU performance as the two CPU tests are nicely multi-threaded. The rendering tests are also heavily CPU dependent with the very fast R7970 running at 720P resolution.

View Full Size

Our first surprise is that the FX-6200 only barely exceeds the performance of the X4 980. The 1100T easily outpaces the 6200 at stock speeds. Only when overclocked does the FX move ahead.

View Full Size

In the GPU tests I was honestly surprised by the lower FPS numbers at stock speed. The GPU tests are not as heavily multi-threaded as the CPU tests, but the disparity of numbers here raises a few eyebrows. At 4.5 GHz the FX-6200 barely keeps up with the X4 980.

View Full Size
View Full Size

The two CPU tests are just as interesting as the previous results. Going into these tests I was assuming that the FX part would be cleaning house here with the combination of next generation Bulldozer technology running at significantly higher speeds. I was wrong. In Test 1 the FX barely keeps up with the 1100T, and in Test 2 it frankly just gets destroyed by the 1100T. In fact, it barely trails the 4 thread capable X4 980. Even overclocked the FX cannot reach the same performance plateau as the previous gen X6 1100T.

Cinebench R10

R10 has historically been a little bit more even handed when dealing with AMD processor performance as it has been with Intel. It is able to test single core performance as well as max core. This is based on a real world rendering engine and is happily multi-threaded.

View Full Size

The single core results all fall within expectations. The FX has weaker single core performance at 4.1 GHz Turbo than the Phenom II X4 980 at 3.7 GHz. At 4.5 GHz though, the FX seems to take a nice jump ahead.

View Full Size

In multi-core mode we see the FX barely surpass the X4 980. It is again destroyed by the X6 1100T. We only see the FX take a lead when overclocked to 4.5 GHz. This certainly does not bode well for the rest of the tests coming up.

Cinebench R11.5 and WME 64

Cinebench R11.5

This updated test uses the latest rendering engine technology from Maxon and handles multi-threaded performance in a more efficient way than the previous Cinebench R10. It does not report a single core result, but we get a pretty good idea of that from the previous test.

View Full Size

Ouch. The six thread FX-6200 is barely faster than the four thread capable X4 980. Let us not talk about the FX’s performance against the X6 1100T. Even at 4.5 GHz the FX cannot overcome the 1100T.

Windows Media Encoder 64 Bit Edition

WME 64 bit is a free media encoder/transcoder that is multi-thread aware. I took a full 1080P 240 MB clip and transcoded it to 480P/ 1mbps VBR with a final size of 47 MB.

View Full Size

We are again disappointed by the performance of the FX-6200. It does scale nicely when overclocked, but the results are still not very impressive.

Euler 3D and 7-Zip

Euler 3D

This is a benchmark version of a professional/academic simulation. This program models airflow over an airfoil, and is again nicely multi-threaded. It is very floating point heavy and does not adequately leverage the SSE capabilities of AMD processors (something the authors of the benchmark talk about due to them using a certain Intel compiler for their program).

View Full Size

These results are fascinating. No, seriously. With a single thread at stock speeds the X4 980 is the fastest of the bunch. Throw in another thread and the FX-6200 is the fastest. Take it to four threads and the FX-6200 falls behind the others again. At eight threads it is well behind the others. This of course changes around somewhat when the FX is overclocked, but at eight threads again it fails to overcome the lead held by the X6 1100T. I think the primary reason for this is the shared FP units on the modules. With one and two threads, the decode and execution units can easily handle the workload. Heading to 4 threads causes some issues with the scheduling of the three modules (though the FX-6200 could theoretically handle 6 x 128 bit double precision floating point instructions). At eight threads the contention is even worse, so much so that the native four core 980 is still outperforming the FX-6200. Even heavily overclocked, the eight thread test gives the X6 1100T a lead over the FX-6200. I honestly think the thread handling abilities of this architecture are still not up to snuff.

7-Zip

This free compression program has a nice benchmark that measures the theoretical MIPS of a processor when compressing/decompressing a file. I use the 32 MB dictionary setting for this particular benchmark and 1 through 6 threads.

View Full Size

When going single thread, the FX-6200 keeps up well with its older siblings. Once that goes above one thread, it seems to fall behind. At four and six threads it is well behind the X6 1100T. Only when overclocked does the FX-6200 catch up to the others.

Valve Tests and Skyrim

Valve Particle Map Test

A few years back Valve put out a couple of multi-threaded tests. The first is a particle test that measures multi-threaded CPU performance in a 3D setting. It tackles four scenes which feature heavy particle effects, which are all handled by the CPU.

View Full Size

Even overclocked the FX-6200 is not as fast as the X6 1100T. We could be looking at some compiler issues, but the results are still disappointing.

Valve Map Compilation Test

This multi-threaded test is a map compilation tool which, among other things, determines the static lighting for the entire map. This is a pretty computationally heavy benchmark, and while it is multi-threaded, it may not be as efficient when going above four threads (or non-power of two).

View Full Size

Considering the previous benchmarks, this one is not a surprise. The FX-6200 is again slower than the rest, and even when overclocked it cannot overcome the X6 1100T and X4 980. Very disappointing.

Skyrim

For my true gaming test, I chose Skyrim. This title, while still DirectX 9.0 based, is heavily CPU dependent. It has a goodly amount of multi-threading supported by the engine. It still relies heavily on good, fast, single thread performance as well. I take a manual run through Whiterun captured with FRAPS. Graphics are set at 1680x1050 with 8X AA, 16X AF, and Ultra settings enabled.

View Full Size

When I said this game was CPU bound, I was not kidding. At stock speeds the X4 980 is the fastest offering. The FX-6200 again trails the X6 1100T by around 5%. Once the FX-6200 is overclocked, then performance really perks up. It finally seems to come awake in a gaming situation and thoroughly trounces everything else. To get that performance though, it certainly needs to be overclocked. This could be a situation where the larger per core L2 cache and larger L3 cache trumps that of the older 45 nm Phenom II processors. It is no secret that games such as this really like a lot of cache on the processor.

Overclocking and Power Consumption

All of the latest FX processors are completely unlocked. This is something of a step away from previous generations, as not all Phenom II’s were unlocked. In fact, all of the Athlon II products were fully locked down when it comes to core multipliers. From top to bottom with the Bulldozer series and FX products, they are fully unlocked. And they need to be.

The stock speeds for the FX-6200 are 3.8 GHz core and 2.2 GHz northbridge/L3 cache. For all of the tests I put the core to 4.5 GHz and the northbridge/L3 to 2.4 GHz. I could have gone higher, but I felt this was an easily achievable overclock for most users. To maintain stability at 4.5 GHz, I had to add another 0.025v to the core. I am pretty sure that this particular sample could hit 4.7 GHz with ease, as well as pushing the northbridge to 2.8 GHz with some extra juice.

View Full Size

But when compared to a retail heatsink from the previous generation, the FX unit is a lot beefier.

There have been people reporting that the latest chips can reach upwards of 5.0 GHz on the core. This would make the FX-6200 a pretty quick processor, and certainly people would have no complaints about it running as such. It is very easy to overclock right out of the box when using a good quality motherboard. My results here should be pretty average. Now we get to see what price we pay for hitting 4.5 GHz.

Power

The FX-6200 is built on GLOBALFOUNDRIES relatively new 32 nm HKMG/SOI process. Previous process jumps from AMD have typically been unimpressive, and this is almost no exception. The reason for this is that AMD, and now GLOBALFOUNDRIES, will slowly implement new features into older processes, thereby improving power and thermal performance. The jump to 32 nm should have been a little more impressive due to the introduction of HKMG (High K Metal Gate). We also have a pretty significant improvement with much smaller SRAM cells as compared to previous processes. The Cinebench R10 multi-threaded test was used to load up the processors.

View Full Size
View Full Size

Well that is again pretty disappointing. At idle the FX-6200 leads the pack in energy efficiency. At load it does not differentiate itself from the previous two 45 nm parts. For being slower in overall performance, the FX-6200 pulls 12 more watts at load than the X6 1100T. Once we overclock the FX-6200 so that it finally overtakes the X6 1100T, it is now 104 watts higher under load. This is pretty pathetic. There is most certainly a price for getting any kind of performance out of the FX-6200, and it comes in terms of power consumed.

Unlock-Core - Secrets:
There are no cores to unlock. AMD said months ago that the FX series will not have unlocking features like that on the Phenom II series did. Considering the module design of the chip I would have to agree with that statement.

Conclusion

So in the end is the FX-6200 worth it? Yes and no. It fits the price/performance envelope for a product in its category. It overclocks easily. It is still a 3.8 GHz product that has fair performance across the board. It does not abduct your family or set your house on fire. It just is simply not the budget performer that we were expecting or hoping for. Better luck next time.

FEEDBACK / NEED HELP

I am waiting for your feedback in helping me to improve my Blog. For feedback comment Below. Thanks for reading my Blog.

Q/A
If you need any help about AMD Processors, Graphics cards overclocking, unlocking, maintenance, improving gaming performance write feedback, i reply you as soon as possible.


Tags:
AMD FX-6200 vs Phenom II X4 980 BE vs Phenom II X6 1100T vs FX-6200 4.5 Ghz Tubocore,, comparison, microarchitechture, multicore, Instruction sets sse3 sse4a, desktop, northbrigde hypertransport, power management, Benchmark Score Rating, Gaming Overclocking, unlocking core secrets, contrast FX 6200 specs detail.

4 comments:

  1. Why are you stealing our content you asshat?

    ReplyDelete
  2. This site is one of the best I have found.


    Setiyono
    http://www.jayabhaya.com/2013/09/jayabhaya.html
    princsetiy@yahoo.com
    from : Yogyakarta Special Region

    ReplyDelete
  3. hi..Im student from Informatics engineering, this article is very informative, thanks for sharing :)

    ReplyDelete
  4. Stolen from PC Perspective..

    ReplyDelete