AMD Brings Bulldozer To The YardToday is a huge day in terms of significance for AMD, who just moments ago launched the first major redesign of their processor architecture since 2003. AMD has not been able to keep the new architecture much of a secret as the new 'Bulldozer' core architecture is widely known amongst enthusiasts and gamers. Another reason that so many people know about Bulldozer-based CPUs is because AMD already announced Valencia and Interlagos processors for the server market. Today is significant in the sense that AMD is launching the first new Bulldozer-based desktop CPUs that are codenamed Zambezi and feature up to eight cores.
The Bulldozer die is built by GlobalFoundaries using the 32nm DSL SOI (silicon-on-insulator) technology process and measures in at ~315mm2 in size with a transistor count of ~2 billion.
The AMD Bulldozer lineup consists of seven high-end consumer desktop processors that are clearly aimed at performance minded individuals.
This is made evident by the fact that the AMD FX CPU series does not feature an integrated GPU (unlike Llano and Ontario/Zacate), so you'll have to use discrete graphics with one of these processors. The entire AMD FX CPU series is designed for use with socket AM3+ motherboards, so they will work best with 990FX, 990X and 970X chipsets by AMD. Older AM3 motherboards will also support the Zambezi FX processors after a BIOS update. AMD's new FX processor family consists of the following models: FX-4100, FX-B4150, the FX-4170, FX-6100, FX-8100, FX-8120 and FX-8150 The nomenclature of the AMD FX CPU series is fairly easy to remember. It is broken down into FX-81XX, FX-61XX and FX-41XX model groups, with the first number being the number of cores that the processor has. The second number is a '1' on all of the processors and that stands for single CPU socket configurations and then the final two numbers are to show how fast the processor clock speed is.
Although the AMD FX series of unlocked processors brings back the FX branding from a era when AMD was able to compete with any of Intel's processors that is not true for this generation. The flagship processor from this series is designed to compete with the Intel Core i5-2500K and the Core i7-2600K. AMD isn't going after the super high-end LGA1366 platforms here, which obviously means that when the Intel LGA2011 platform comes out with Sandy Bridge-E processors next month that Intel will once again dominate that market segment. That said, the AMD FX-8150 does stack up well on paper compared to the Intel Core i5/i7 series for the LGA1155 platform.
AMD sent us this image of how their entire desktop processor lineup compares to what Intel is offering. The Intel LGA1366 platform is missing here as we'd expect to see the Intel Core i7-990X at the very top of the Intel processors.
AMD has for years been marketing themselves as being a cost effective platform for consumers. Nothing has changed there as AMD still claims they offer equal, if not better gaming performance, on the FX-8150 when compared to the Intel Core i7-980X. Those readers that know how to overclock might laugh at a slide like this as you can buy an Intel Core i7-950 processor for $199 and overclock it and the platforms would be at the same cost.
The processor that AMD sent to Legit Reviews for testing is the FX-8150. This processor is priced at $245 and AMD said that it was priced to compete against the Intel Core i5-2500K. We feel this is a stretch as the Intel Core i5-2500K can be found for under $220 on PriceGrabber.com or $179.99 in-store at Micro Center. The Intel Core i7-2600K is priced at under $315 on PriceGrabber.com or $249.99 in-store at Micro Center. We'll let you determine which processors line up to what!
Let's take a closer look at the new Bulldozer architecture and then get to benchmarking!
AMD Bulldozer CPU Architecture
The L1 and L2 BTB (Branch Target Buffer), Prediction Queue, Ucode ROM, ICache, Fetch Queue, and the 4 x86 Decoders are present on each of the Bulldozer modules. These particular components of the module are shared between the two cores.
Each of the cores within the Bulldozer module has its own Integer Scheduler, Instruction Retire, L1 Data Cache and L1 DTLB. These components, among others make up the Integer unit.
Each of the Bulldozer modules has 2MB of 16-way unified L2 Cache. Depending on the number of active modules on your particular AMD FX processor, you can have anywhere from 4MB L2 to 8MB L2 Cache.
AMD FX processors will support the latest instruction set extensions, SSE 4.1 and 4.2; AVX, -256-bit YMM registers, -Non-destructive source operand capability, -AES subset, -FMAC subset (AMD 4-operand form); XSAVE state space management; XOP Instructions. The FX series of processors will also include Light Weight Profiling with Low-overhead user-level profiling; they use XSAVE state space; and will store records for configured events.
The new processors feature 128KB of L1 Data Cache (16KB per core), 256KB of L1 Instruction Cache (64KB per module), and 8MB of L2 Cache (2MB per module). When asked about the latencies of the various cache, AMD opted for no comment. The Northbridge onboard the AMD FX "Bulldozer" controls the 8MB of L3 Cache, two 72bit wide DDR3 memory channels and four 16bit HyperTransport links.
Above we can see the simplified block diagram of the Northbridge on the AMD FX processors.
One of the key points of the new AMD "Bulldozer" is the power efficiency. AMD has minimized the silicon by sharing components between the cores. Throughout the design of the FX processors there is extensive flip-flop clock-gating, as well as having ciruits that are power-gated dynamically. There is also a number of power saving features that are controlled by firmware or software.
The AMD Turbo Core Technology has a couple of different levels. First we have the processor at the base frequency. When it's needed we have a turbo boost for all eight cores, the AMD FX processors also have the ability to disable half of the cores and hit the maximum processor frequency.
The AMD FX-8150 Press Kit & Water Cooler
Last week we took delivery of the AMD FX-8150 press kit for this launch article. The box was rather huge as press kits aren't usually 23" by 17" in size.
Included inside the box was an ASUS Crosshair V Formula motherboard, an AMD 8-core FX-8150 processor, an empty metal tin and a large Texas style belt buckle.
The first thing that we had to check out was the massive belt buckle that AMD made for the FX series processor launch. The metal belt buckle has a scorpion on it, which is the mascot for the AMD 990FX platform. Prior to scorpion you had the dragon platform, which was the 890FX.
The empty metal tin that was inside the press kit was actually a retail boxed processor packaging. It looks like AMD is doing away with the cardboard box and going with these sweet metal tins. They'd make for great storage if it didn't have the big hole in the side for where the processor is shown. AMD obviously spent some time designing this tin and it looks way better than the box art that we saw way back in March 2011.
When we opened up the ASUS Crosshair V Formula motherboard we found the AMD FX-8150 processor was already installed in the board. This is great news as for a second we thought they forgot the processor! AMD did not include the reference cooler with the FX-8150, but did say that they are not introducing a new cooler and that if we have the retail cooler for the Phenom II X6 1100T that it would be the same. The problem is that AMD doesn't ever send us retail coolers, so we went up to Micro Center and bought a retail boxed 1100T just to get the cooler out of it for testing.
The stock cooler that comes with the Phenom II X6 1100T features a copper base with four copper heat pipes that attach to aluminum cooling fin. The cooler is topped off with a 80mm PWM fan that uses a 4-pin fan header. This particular heatsink said Cooler Master on it, so the source of this one was obviously them.
This morning we heard a knock on the door and we got an overnight package from AMD! It appears that AMD feels guilty for sending this to us on the day of launch as they tied a ribbon on it. This AMD Liquid CPU Cooling System will be available only with the FX-8150 processor in select regions. AMD informed us that this water cooling kit will not be available as a standalone part and must be bought with the FX-8150 processor.
The cooler is made by Asetek and closely resembles the Antec KÜHLER H2O 920 Liquid Cooling System that we have already reviewed. Since AMD got this cooler to us just 12 hours before launch, there was no way to give it a fair shake in time for this article. AMD also said that it would add around $100 to the price of the FX-8150, so we can assume this cooler to be available with the FX-8150 for $345. We have begun testing this cooler and will post a review of it on the FX-8150 later this week to see how the system performs and overclocks with the factory air cooler and then this water cooler.
AMD has said that the Zambezi-FX series of CPUs will be very overclocking friendly and that you can expect to reach 5.0 GHz on air and 5.2 GHz on water when run with dual cores. If you want to break the 8GHz mark you'll have to be using extreme cooling in the likes of liquid nitrogen or liquid helium. AMD shows only a 200-300MHz overclock gain from going from the default AMD air cooler to a water cooler, so it will be interesting to see if it is worth the extra $100 cost.
The Test SystemBefore we look at the numbers, here is a brief glance at the test system that was used. All testing was done on a fresh install of Windows 7 Ultimate 64-bit. All benchmarks were completed on the desktop with no other software programs running.
AMD AM3+ Platform
The AMD AM3+ platform that we used to test the AMD Phenom and FX processors were run on the ASUS Crosshair V Formula motherboard with BIOS 9901 that came out on 9/26/2011. The Corsair Vengeance 8GB 1866MHz memory kit was set to 1866MHz with 1.5v and 9-10-9-28 1T memory timings on the AMD FX-8150 and then at 1600MHz with 1.5v and 9-10-9-28 1T memory timings on the Phenom II X4 and Phenom II X6 processors.
Intel Z68 Platform
The Intel Z68 platform that we used to test the Intel Pentium Processor G850 was running the Gigabyte GA-Z68X-UD3H-B3 motherboard with BIOS F8 that came out on 7/26/2011. The G.Skill Ripjaws 8GB 1866MHz memory kit was set to 1866MHz with 1.5v and 9-10-9-28 1T memory timings.
The AMD Test System:
The test system was run in dual-channel memory mode at 1866MHz with 9-10-9-28 1T memory timings. The Gigabyte GA-A75-UD4H motherboard was running BIOS F3, which was the latest build available when testing was completed. You can see all of the CPU-Z and GPU-Z information for this test platform below.
The integrated graphics comes up as Radeon HD 6530D according to GPU-Z and we used AMD CATALYST 11.6 drivers was used for testing.
|AMD Test Platform|
A8-3850 / A6-3650
|8GB GSKILL Ripjaws 1866|
|AMD Radeon HD 6950|
|Crucial C300 256GB SSD|
|AMD Retail Boxed|
|Windows 7 Ultimate 64-Bit|
AMD Turbo Core Details and Windows Index Score
Here is a look at the AMD FX-8150 Bulldozer processor sitting at an idle state on the desktop of our Windows 7 Ultimate SP1 64-bit test system. Notice that the processor is only running at 1.4 GHz thanks to the AMD Power Manager that noticed the CPU was and an idle and put it in a low power state for power saving reasons. This is key as it will have lower idle power consumption and generate less heat. The base clock of the processor is 200MHz and our ASUS Crosshair V motherboard is running the base clock at 200.7MHz, which isn't too far off from default.Each Core is equipped with a 16 KB Level 1 Data cache, a 32-entry fully associative DATA TLB, and a fully out of order load/store – capable of two 128-bit loads per cycle or one 128-bit store per cycle. Each dual Core module includes a 2 MB 16-way unified L2 cache with an L2 TLB capable of 124 entry, 8 way that services both instruction and data requests. Bulldozer supports up to 23 outstanding L2 cache misses for memory system concurrency. Finally AMD has designed a shared 8 MB L3 cache with 64 way associativity for both cores in a Bulldozer module.
Next we want to look at the various power states as with AMD Turbo technology that can get very confusing if you don't know what is going on. The AMD FX8150 that we have for benchmarking today has a base clock of 3.6GHz, a CPU Turbo Core clock of 3.9GHz and a CPU Max Turbo clock of 4.2GHz. Depending on the number of threads and the temperature state of the processor, the clock frequency will adjust to better handle the tasks you are asking it to do.
When running a single threaded application the AMD FX-8150 can jump up to the rated clock frequency of 4.2GHz (as shown above) and it will not go any higher than this as this is the fastest the processor will do out of the box without manually overclocking it.
When running two threads the AMD FX-8150 was observed running at 3900MHz, which just happens to be the standard 'Turbo Core' clock speed.
When running all eight threads at 100% when using a application like handbrake the AMD FX-8150 was observed running at 3600MHz, which just happens to be the standard base clock speed for this processor.
The interesting thing about the Bulldozer architecture is that each of the cores can run at different clock speeds, so this is a truly dynamically clocked processor unlike one that we have ever seen before.
We get people asking about the Windows Experience Index Score all the time, so we thought we'd include it. According to Windows 7 the lowest performing component on the system was the processor, which would be the AMD FX-8150 CPU, as it had a subscore of 7.6 and that is the base score for the entire system.
SiSoftware Sandra 2011 SP5
The Sisoft Sandra 2011 SP5 benchmark utility just came out a few weeks ago and we have started to include it in our benchmarking. Sandra 2011 comes with support for Virtualisation (Virtual PC/Server, Hyper-V, VMware) and GPGPU (OpenCL, DirectX 11 DirectCompute), but today we will be using the program to look at memory and CPU performance!
Results: All of test systems used the identical Corsair Vegenance Series 8GB (2 x 4GB) memory kit that ran at 1866MHz with 9-10-9-28 1T timings with the exception of the AMD Phenom II systems that were run at 1600MHz with CL9 timings. As you can see the Intel processors rule the roost here, but the FX-8150 does show more memory bandwidth than the A8-3850 Llano APU and both are running in dual-channel memory mode at 1866MHz.
The Sandra Processor Multi-Media benchmark has been a long time favorite of ours to look at floating point performance on processors.
Results: The AMD FX-8150 processors loves the Int x16 iSSE4.1 test and fairly respectable floating point performance.
Cryptography has become an important part of our digital life: it allows us to conduct safe transactions online, certify programs and services, keep our data secure and much more. Sandra 2010c has a dedicated benchmark built-in that measures cryptographic performance, which is important on the new Intel 32nm processors like the Core i7 980X. It includes features like AES-NI! Intel AES-NI is a new set of Single Instruction Multiple Data (SIMD) instructions that are going to be introduced in the next generation of Intel processors, as of 2009. These instructions enable fast and secure data encryption and decryption, using the Advanced Encryption Standard (AES), defined by FIPS Publication number 197. The architecture introduces six instructions that offer full hardware support for AES. Four of them support high performance data encryption and decryption, and the other two instructions support the AES key expansion procedure. Let's take a look at how this feature impacts Cryptography performance.
Results: The processors with the AES-NI instruction set are very evident in this benchmark. The AMD FX-8150 now includes AES-NI instructions and the performance is now much closer to what Intel is offering with the Core i7-2600K.
x264 HD EncodingThis application did fairly well when run on 4 threads, as you can see from the screen shot above. The first pass was not using all of the processing power available on the four cores, but on the second pass all 4 threads were at ~98% load.
Benchmark Results: The x264 HD benchmark is very CPU intensive and it showed that AMD FX-8150 doesn't do that hot on the first pass, but it does do very good on the second pass, which is the most time consuming. On the second pass the AMD FX-8150 was able to beat out the Intel Core i7-2600K processor and lost to only the Intel Core i7-990X.
HandbrakeHandBrake is an open-source, GPL-licensed, multiplatform, multithreaded video transcoder, available for MacOS X, Linux and Windows. It is popular today as it allows you to transcode multiple input video formats to h.264 output format and is highly multithreaded.
This workload is a 43-second HDV. The input file is encoded in Mpeg format. Video encode parameters are 23.9mbps, 1440*1080, 29.9fps. Audio encode parameters are 384 kbps, 48kHz, 2channel, mpeg audio version 1 layer 2. File size is 128MB. The workload is encoded into h.264 output format using the preset - high profile. HandBrake version 0.9.5 was used for benchmarking.
POV-Ray 3.7 RC3
Once rendering on the object we selected was completed, we took the elapsed time from the dialog box, which indicates the exact time it took for the benchmark to finish the benchmark. A lower time indicates faster system performance.
Benchmark Results: The Intel Core i7-2600K and the AMD Bulldozer FX-8150 were once again trading blows in POV-Ray and the FX-8150 came out on top by just a hair.
Cinebench R11.5 was able to put a 100% load across all the cores on all of the processors, which makes this a great benchmark to look at multi-core platforms.
Benchmark Results: When it comes to multi-core performance the AMD FX-8150 loses out to the Intel Core i7-2600K by nearly a full point and was just barely ahead of the AMD Phenom II X6 1100T processor.
Benchmark Results: When looking at single core performance you can see that the FX-8150 processor doesn't do too hot as it's single core performance is actually lower than the previous generation Thuban and Deneb cores.
This is our first use of Futuremark PCMark 7 for a processor review. PCMark 7 which is specifically designed for Windows 7. It measures the performance of the latest PC hardware across a variety of common scenarios.
PCMark Vantage 7 supports both system level and component level benchmarking and comprises several different test suites but for the purposes of this review, we employed the standard test suite. The nice thing about it is that you can submit your scores online and compare against others.
AMD FX-8150 Detailed Benchmark Results:
From the looks of things the main difference between the FX-8150 and the 1100T is the data decrypting test suite.
Benchmark Results: PCMark 7 showed that the AMD FX-8150 was able to score 4351 points, this meant that it was able to beat all of the system that use a comparably priced processor!
Resident Evil 5
Resident Evil 5 is a survival horror video game developed and published by Capcom. Resident Evil 5 features similar gameplay to Resident Evil 4, with context-sensitive controls and dynamic cut scenes also making a return. The player can control Chris Redfield or Sheva Alomar in a similar fashion to Leon S. Kennedy in Resident Evil 4, with the same over-the-shoulder perspective.
This game title runs on up to four threads as you can see from the task manager shot above that was taken on a test system that was running the Intel Core i7-2820QM processor.
Benchmark Results: The AMD FX8150 was easily able to out perform the Intel Core i7-2600K by 8.6% at 1024x768 and 9.4% at 1920x1080. The Intel Core i7-990X crushed the AMD FX-8150 by 27% at 1024x768, though realistically who is going to run a processor that retails for a thousand bucks at 1024x768? When we look at the performance of the Intel Core i7-990X and the AMD FX-8150 at a much more realistic gaming resolution (for processors of this caliber) the difference is only 2 frames per second (1.3%) in favor of the Intel Core i7-990X. For the $750.00 price difference, a 1.3% gain, well we'll let you decide on that one.
H.A.W.X. 2 Benchmark
Aerial warfare has evolved. So have you. As a member of the ultra-secret H.A.W.X. 2 squadron, you are one of the chosen few, one of the truly elite. You will use finely honed reflexes, bleeding-edge technology and ultra-sophisticated aircraft - their existence denied by many governments - to dominate the skies. You will do so by mastering every nuance of the world's finest combat aircraft. You will slip into enemy territory undetected, deliver a crippling blow and escape before he can summon a response. You will use your superior technology to decimate the enemy from afar, then draw him in close for a pulse-pounding dogfight. And you will use your steel nerve to successfully execute night raids, aerial refueling and more. You will do all this with professionalism, skill and consummate lethality. Because you are a member of H.A.W.X. 2 and you are one of the finest military aviators the world has ever known. H.A.W.X. 2 was released on November 16, 2010 for PC gamers.
The H.A.W.X. 2 PC game title runs on what looks like five threads if the processor being tested has that many available.
Benchmark Results: The AMD FX-8150 was able to put up some respectable numbers in H.A.W.X. 2. At 1024x768 the AMD FX-8150 was able to average 145 frames per second, that's 4% behind the Intel core i7-990X and 9% behind the Intel Core i7-2600K. Once again though, once we increased the resolution to 1920x1080 we saw the differences all but disappear. When we look at the previous generation Thuban and Deneb cores, the AMD FX-8150 out performed them at 1024x768. While at 1920x1080 the margins once again shrunk to almost nothing.
Total System Power Consumption
Since power consumption is a big deal these days, we ran some simple power consumption tests on our test beds. The systems ran with identical power supplies, Solid-Sate Drives, Memory kits and motherboards from the same company. To measure idle usage, we ran the system at idle for one hour on the desktop with no screen saver and took the measurement. For load measurements, Prime95's in-place large FFT's were run on all cores to make sure each and every processor was at 100% load for maximum power consumption and heat. Curious about other test scenarios, we decided to HAWX 2 the PC game for testing. We also tested a video transcode using HandBrake 0.95 to see how the power draw on that was.
Benchmark Results: The AMD FX-8150 has really impressive idle test results as the processor consumes just 84 Watts at idle. When transcoding on all eight cores or playing a game the power numbers start to jump up though and the Intel Core i7-990X and the AMD FX-8150 processor were consuming roughly the same about of power in Prime 95 and HandBrake benchmarking. We had hopes that the 32nm process would help this 8-core processor cut down on load power use, but it uses more power than 45nm processors like the Phenom II X6 1100T.
AMD FX-8150 CPU Temperature TestingFor temperature testing we used the AMD retail boxed CPU cooler and placed the system into the Thermaltake Chaser PC case and let it idle for 30 minutes and then ran it at full load with Prime 95 64-bit.
The ambient air temperature in the room was 70F or 21C. We used Core Temp to monitor the temperatures and as you can see from the image above the low was 6C and the high was 46C. AMD says the maximum safe temperature to run any FX-series processor at is 61C, so the standard air cooler that comes with the retail processors is more than enough to keep it nice and cool.
We'll have an article posted later this week that shows off the new AMD water cooler and will have idle/load temperatures compared with other cooling solutions.
AMD FX-8150 CPU OverclockingOverclocking greatly varies due to what hardware is being used and who is doing the overclocking. Always remember that no two pieces of hardware will perform the same, so our results will differ from what you might be able to get.
AMD said that you should be able to hit 4.6GHz on the AMD retail boxed CPU cooler and 4.9GHz on all of the cores. To try this out we gave it a shot was used for overclocking, so we'd expect this overclock to be easily reached by anyone!
The AMD FX-8150 starts off at 4200MHz in 'CPU Max Turbo' mode, so it should be that hard to get to 4.6GHz or 4.9Ghz right? The FX-8150 starts off with a base clock of 200MHz with a multiplier of 21. All AMD FX series processors feature an unlocked multiplier, so you can both lower and raise the multiplier in the BIOS. You can also adjust the base clock frequency to better dial in the overclock. Since we were time limited we overclocked by just using the multiplier to see how far we could push our processor.
To overclock the FX-8150 processor we wanted to put it under water, so we broke out the Corsair Hydro H100 liquid water cooling kit to see what this processor could do. AMD says that the FX series processor has a max suggested temperature of 61C and we all know that will be hard to keep under. AMD also suggested not going over 1.55V on the VCore when running water cooling.
With everything left to default in the BIOS and by just raising the multiplier up to 24.5 and then increasing the VCore up to 1.45V we were able to reach 4916.6MHz with Turbo disabled. This means that all eight cores were running this fast.
The benchmarks at 4900MHz were impressive as we were able to hit 8.04 pts on Cinebench R11.5 when running all eight cores. When run at stock speeds we saw a score of 6.01 pts, so this is an improvement of 33%! In x264 HD we averaged 148 FPS on pass 1 and then 49 FPS on pass two, which blew away the 121 FPS on pass 1 and 37 FPS that we saw in pass 2 with the FX-8150 running in stock form.
We tried our very best to get 5 GHz stable one afternoon, but we were unable to get it stable across all of the cores. We were able to run short benchmarks like SuperPi 1M or things like that, but Prime 95 and Cinebench would throw errors. We tried to run the FX-8150 at 1.60V, but the platform did not like that and never got past the post screen. To our surprise we were able to reach 4900 Mhz on the FX-8150 with the aid of the Corsair Hydro series H100 water cooler and that is exactly what AMD said all cores would likely hit with water cooling! With more time and experience with this platform we are certain that it has more left to offer, but with just an afternoon dedicated to overclocking that is the best that we could come up with.
Reaching 4900 MHz with full stability on the FX-8150 is a great overclock though considering the standard clock speed of this processor is 3600MHz. If you look at it like that, this overclock is 1300MHz!
The AMD FX-8150 offers solid performance and is competitive with the Intel 'Sandy Bridge' series of processors.
Unlock-Core - Secrets:
AMD FX-8150 has fully Eight core enabled processor. There are no chances to unlock cores. 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.
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.
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.
AMD FX-8150 vs Intel i7-2600k vs Intel i7-990X vs Intel core i3-2120 vs AMD Phenom II X4 980 BE vs Phenom II X6 1100T vs A8-3850, Gaming Performance, Overclocking Benchmarks score FPS, FX-8150 Box opening, Box containing, comparison, microarchitechture, multicore, first eightcore, Guiness world record eightcore at 8.4ghz, L2 L3 cache details, 8150 Bulldozer Feature, Instructiond sets sse3 ssse3 4.1 4.2 FMA4 AVX XOP AESNI AES, Core roadmap, zambezi desktop, northbrigde hypertransport, power management, AMD Turbocore Technology, Watercooler Press Kit, AMD 990FX, Air Water Liquid Nitrogen Helium Overclocking.