AMD A10-5800K vs. Intel Core i3 3220 Head to Head

Introduction

Choosing hardware for a computer can be a daunting task. For most, choosing the right processor can be as equally challenging. For those not familiar with processors and performance, Intel has held the king of performance for many years but AMD has shaken up the competition as recent.

In 2011 AMD introduced Zacate and Llano which used AMD’s Fusion technology and they called these new processors APU’s (Accelerated processing Units). Fusion was the world first on die graphics processor that gave consumers a low priced solution for an budget OEM, custom desktop or laptop namely in the $300 to $600 range. The pride of AMD’s Fusion technology was not only price but more than adequate video capabilities. Last year, AMD rolled out Trinity and took their “Bulldozer” cores and they evolved into the new “Piledriver” cores which also incorporated Fusion technology. This resulted in a much better APU replacing Llano in their line-up. AMD’s strategy with the new processors is: Bobcat (good), Trinity (better) and Vishera (best).


In the meantime, Intel keeps the train rolling with their “Tick Toc” strategy. Every other year, Intel’s plan is new architecture (Tock) and on the flip side, increasing transistor density (Tick). This year, Intel rolled out IvyBridge and uses 22nm fabrication and increased the transistor density with their new “Tri-Gate” transistors resulting in a more efficient processor. Intel’s strategy remains core i3(good), core i5(better) and core i7(best).



Today, we are going to take a look at AMD’s and Intel’s lower priced offerings namely Intel’s core i3 3220 and AMD’s A10-5800K and put them head to head in competition. Since both IvyBridge and Trinity’s introduction, many still lean toward Intel not realizing AMD has stepped up their game. Will Intel still take the cake in this arena? Let’s roll both processors to the bench and fire them up and see who hits the finish line first.

Intel IvyBridge

Intel continues its business plan as mentioned with their “Tick Tock” strategy. SandyBridge brought to the table 32nm fabrication process and IvyBridge shrinks the die to 22nm along with increasing transistor density with the new “Tri-Gate” transistors. Here we see a diagram of the i3 3220. The architecture does not change from SandyBridge and Intel’s next generation due out mid 2013 is named Haswell. Haswell will bring new architecture to the table.

Here is the Ivy Bridge 22nm die. Along with Intel using up to 4000 graphics with Ivy, it has DX 11 support along with Open GL and Open CL support. There are also some modifications to Ivy Bridge that will have better single threaded performance dedicating all the resources to the single core. In multi threaded applications the resources are used simultaneously. Intel says a 5-15% performance boost on the processing side over Sandy Bridge but a 20-50% increase on the GPU.



Intel’s model structure stays the same as SandyBridge with “K” models being unlocked for overclocking but non “K” models are locked have no overclocking. Here is the current Intel line-up including SandyBridge processors that will eventually be phased out. As we see in the diagram, Intel’s architecture has an on die graphics processor on all IvyBridge CPU’s. As the models egress, so do the graphics processors. The 3770 models feature Intel’s 4000 GPU’s and bottom out on the i3 models with Intel’s 2000 graphics. Today’s head to head competition features a dedicated GPU but Intel’s on die GPU’s really do not stand up to AMD’s solution as we will see later.




AMD Trinity

AMD Trinity APU’s use AMD’s new Piledriver cores that evolved from Bulldozer cores. Bulldozer, in many persons opinion, was a flop for AMD. AMD did not have the instruction set nailed and performance was very lack luster. That changed with Piledriver.

The main focus from AMD with Piledriver was to maximize the resources with their newly designed cores and they have done just that. Each module is made from two of AMD’s “Four Star” x86 cores and they use two modules making four cores. In addition, each core is connected through pipelines sharing the resources with the principle of efficiency. AMD still dedicates a large amount of the die to the on-die GPU but as we see, the cores change to AMD’s new Piledriver cores, which come with 2MB L2 cache (Llano had 1MB L2 cache). AMD also carries on their 32nm fabrication and each module and has two integer cores & a floating point. A four core Trinity APU has 2 modules (4 integer cores and 2 floating points per core).



If you didn’t notice and as I mentioned, AMD dedicates more die space than Intel for the GPU and with Trinity, they top out with AMD’s 7660D GPU which is far superior to Intel’s 4000 graphics package. The result in my initial Trinity review was far superior graphics package than Intel and to be frank, Intel was choppy at best with most games. With YouTube, videos and movies being at an all time high, AMD takes the crown here over Intel’s 3220 core i3 in graphics.



Graphics part of computing has become to some as important to regular computing nowadays. The UVD (Unified Video Decoder) that AMD uses is based on an ATI video processor. This is incorporated into the same die of the GPU for hardware decoding videos, along with the Advanced Video Processor (AVP). UVD handles decoding of video codecs entirely in hardware. Moving back in time, we saw UVD+ in the Radeon 3000 days, UVD 2 in the Radeon 4000 days and UVD 2.2 in RV770 and RV730 series of GPUs. Today, AMD uses UVD 3. UVD 3 (Universal Video Decoder 3) is a new version of the graphics for smoother and better video.



Here is AMD’s Trinity line-up. The Athlon series was an afterthought and requires a dedicated GPU but still an inexpensive solution for most. Trinity tops out with AMD’s 7660D with the A10-5800K on die GPU and bottoms with the A4-5300K and 7480D graphics.



Next, is a look at the motherboards.

ASRock FM2A85X Extreme 6

For the A10-5800K, the board selected is the ASRock FM2A85X Extreme 6.



MSI Z77 MPower

And for the Intel core i3 3220, we see the MSI Z77 MPower.



Test Bench

Processors: Intel core i3 3220/AMD A10-5800K
Motherboards: MSI Z77 MPower/ ASRock FM2A85X Extreme 6
PSU: Nexus 1100 Watt
Memory: 8G G.Skill 2400MHz Trident tuned at 1600MHz Cas 9
Graphics: Sapphire 7970
Hard Drive: 240G OCZ Revo Drive/ 120G OCZ Vertex 3
Cooling: Custom Water
Case: Phobya WayCoolit Test Bench





Benchmarking Mindset

There are countless benchmarks that can be performed in the head to head competition. We narrowed them down to a manageable amount. I will test CPU, platform specifics, real world, synthetics and some gaming using a dedicated Radeon 7970. This should give a well rounded look at performance.

Let’s start with platform/motherboard specific benchies

Memory


Aida 64

AIDA64 implements a set of 64-bit benchmarks to measure how fast the computer performs various data processing tasks and mathematical calculations. Memory and cache benchmarks are available to analyze system RAM bandwidth and latency. Processor benchmarks utilize MMX, 3DNow!, SSE, XOP, FMA, and AVX instructions, and scale up to 32 processor cores. For legacy processors all benchmarks are available in 32-bit version as well. AIDA64 Disk Benchmark determines the data transfer speed of hard disk drives, solid-state drives, optical drives, and flash memory based devices.

There is no secret here. Intel owns AMD with memory read/write and copy speeds along with Latencies. We will see if the additional bandwidth equates to better performance shortly.






Disk Performance

ATTO

The ATTO Disk Benchmark performance measurement tool is compatible with Microsoft Windows. Measure your storage systems performance with various transfer sizes and test lengths for reads and writes. Several options are available to customize your performance measurement including queue depth, overlapped I/O and even a comparison mode with the option to run continuously. Use ATTO Disk Benchmark to test any manufacturers RAID controllers, storage controllers, host adapters, hard drives and SSD drives and notice that ATTO products will consistently provide the highest level of performance to your storage.

In disk performance between the platforms using the Asmedia and Intel Controller: Linear speeds trade blows in read and write performance.

We see the Intel core i3 leading AMD’s A10-5800K slightly in read speeds and overclocking does help AMD here.



Similar results in write speeds using ATTO.



USB 3.0/ CrystalMark

CrystalMark is a useful benchmark suite, that lets you to test your CPU, memory, HDD and video (GDI, Direct Draw, OpenGL) subsystems. The program also shows some information about your system. The report can be saved in a text or HTML.

I also wanted to test USB 3.0 performance between the platforms. For today’s test, I used a USB 3.0 doc station with a 240G Verbatim Solid State Drive. Again, AMD and Intel trade blows but Intel was slightly better in 4K and 4KQD32 which is representative of saturated lanes. Both performed very well.

INTEL core 13 3220




AMD A10-5800K Default 3.9GHz




AMD A10-5800K Overclock 4.6GHz

Overclocked, The A10 performs pretty equal to the 3220 trading blows.



Ok, let’s take a look at CPU performance next.

CPU Performance

Sandralite

Specifically designed to provide users with detailed information regarding their computer, SiSoftware Sandralite offers an impressive array of information concerning every single hardware component that is hidden under the hood of the machine. CPU performance is measured by GFlops ((Giga)FLoating-point Operations Per Second and Gips (Giga instructions per second.)

Dhrystone tries to represent the result more meaningfully than MIPS (million instructions per second) because instruction count comparisons between different instruction sets (e.g. RISC vs. CISC) can confound simple comparisons. For example, the same high-level task may require many more instructions on a RISC machine, but might execute faster than a single CISC instruction. Thus, the Dhrystone score counts only the number of program iteration completions per second, allowing individual machines to perform this calculation in a machine-specific way. Another common representation of the Dhrystone benchmark is the DMIPS (Dhrystone MIPS) obtained when the Dhrystone score is divided by 1757 (the number of Dhrystones per second obtained on the VAX 11/780, nominally a 1 MIPS machine).


With Sandralite, The Intel core i3 3220 beats the AMD A10-5800K at default speed in Gips but Intel’s 3220 pulls the victory in GFlops.



The Whetstone benchmark originally measured computing power in units of kilo-Whetstone Instructions Per Second (kWIPS). This was later changed to Millions of Whetstone Instructions Per Second (MWIPS).

With Sandralite’s Whetsone, Intel’s 3220 beats AMD in measuring floating point and arithmetic performance. Floating point and arithmetic are considered by some to be a good measuring stick for a processor but some are not convinced this leads to better real world performance. We will soon see the results.



LinX

Linx is stress test software and a form of Linpack that measures CPU performance in GFlop measurements.

Using Linx(which is a form of Linpack) AMD squeaks out a victory with the A10-5800K overclocked.



SuperPi

Super PI is a computer program that calculates pi to a specified number of digits after the decimal point—up to a maximum of 32 million. It uses Gauss–Legendre algorithm and is a Windows port of the program used by Yasumasa Kanada in 1995 to compute pi to 232 digits. SuperPi is used to measure single threaded performance.

Again, no surprise with SuperPi, Intel rules the roost. Some think SuprPi is an outdated benchmark but there is still a following with it.



wPrime

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. wPrime is used in measuring multi-threaded performance.

Intel beats AMD in wPrime as suspected at stock frequencies but uh oh! AMD takes the lead with multi-threading here.

Synthetic Benchmarks

3DMark 11

3DMark 11 is a DirectX 11 video card benchmark test for Windows that is designed to measure your PC's gaming performance. 3DMark 11 makes extensive use of DirectX 11 features including tessellation, compute shaders and multi-threading. Trusted by gamers worldwide to give accurate and unbiased results, 3DMark 11 is the best way to consistently and reliably test DirectX 11 under game-like loads.

In 3DMark 11, we see a tight race: about 1% difference in total score. Intel wins by a close margin at stock speeds but AMD pulls ahead overclocked. Looks like we are seeing a trend.



3DMark 2013 (Fire Strike)

The new 3DMark includes everything you need to benchmark your hardware. With three all new tests you can bench everything from smartphones and tablets, to notebooks and home PCs, to the latest high-end, multi-GPU gaming desktops. And it's not just for Windows. With 3DMark you can compare your scores with Android and iOS devices too.

Ditto with FireStrike: The A10-5800K overclocked takes a close race.



PCMark 7

PCMark 7 is a complete PC benchmarking solution for Windows 7 and Windows 8. It includes 7 tests combining more than 25 individual workloads covering storage, computation, image and video manipulation, web browsing and gaming. Specifically designed for the full range of PC hardware from netbooks and tablets to notebooks and desktops, PCMark 7 offers complete Windows PC performance testing for home and business use.

The trend continues with PCMark7. Many like this benchmark as it scores the total system with a slew of tests. AMD pulls a very short victory. Very little separation with PCMark 7.



PassMark

Passmark is a fast, easy to use, PC speed testing and benchmarking. PassMark PerformanceTest ™ allows you to objectively benchmark a PC using a variety of different speed tests and compare the results to other computers.

PassMark turns the table back to Intel with a 5% lead at default but AMD passes Intel again. Intel fan boys may be worried?



Maxon Cinebench 11.5

CINEBENCH can measure systems with up to 64 processor threads. ThIS test scene contains approximately 2,000 objects which in turn contain more than 300,000 polygons in total, and uses sharp and blurred reflections, area lights, shadows, procedural shaders, antialiasing, and much more. The result is displayed in points (pts). The higher the number, the faster your processor.

Again, about a dead heat rendering with Cinebench at stock. Overclocked, AMD pulls away.



Unigine Heaven

Heaven Benchmark with its current version 4.0 is a GPU-intensive benchmark that hammers graphics cards to the limits. This powerful tool can be effectively used to determine the stability of a GPU under extremely stressful conditions, as well as check the cooling system's potential under maximum heat output. It provides completely unbiased results and generates true in-game rendering workloads across all platforms, such as Windows, Linux and Mac OS X.

The trend continues.



So far, both processors look to be pretty even with Intel slightly ahead at stock core speeds but AMD is beating Intel when overclocked. Remember, these are synthetic benchies: some do not put much credence in synthetics. None the less, they are measuring tools.

Real World Benchmarks

Excel Trader.net

Trader Net Excel benchmark is an excellent real world tool. It takes multiple Excel spreadsheets and uses them in building OHLC data, Tick conversion, charts, price changes and calculations.
In our first real world benchmark, AMD’s A10-5800K beats out Intel’s 3220 by 10%.

Overclocking hinders AMD compared to stock setting which is confusing. Perhaps it could be AMD workload distribution between the cores?



WinRar

WinRAR is a lightweight, flexible, and easy-to-use archiving utility that can unpack most archive formats, as well as compress to both RAR and ZIP.

Using WinRar, I compressed a 230Mb file and Intel and Intel wins in file compression hands down.



Photoshop

Adobe® Photoshop® CS6 software delivers even more image editing power, new creative options, and the Adobe Mercury Graphics Engine for blazingly fast performance. Retouch with new Content-Aware features, and create superior designs as well as movies using new and reimagined tools and workflows.

Using Photoshop, I took a photo and used Radial Blur and timed the render. AMD beats Intel handily here.



Cyberlink Media Espresso

Media Espresso is a blazingly fast media universal converter that can transcode your videos, photos and music files and out put them to a huge range of portable devices including mobile phones, portable media players and even game consoles.

Using Media Espresso, I took the new James Bond Skyfall trailer in 1080p and converted a MP3 file to a .wav file. The results show Intel better again at stock but behind the A10 at 4.6GHz.

Gaming Benchmarks

Battlefield 3

Battlefield 3 (commonly abbreviated BF3) is a first-person shooter video game developed by EA Digital Illusions CE and published by Electronic Arts. It is a direct sequel to 2005's Battlefield 2, and the twelfth installment in the Battlefield franchise.

In BF3, Intel beats AMD using our Radeon 7970 in both stock and overclocked. As a side note: I used 13.2 Beta drivers for the testing.



Call of Duty

Call of Duty: Black Ops II is a first-person shooter video game, developed by Treyarch and published by Activision (Square Enix for Japan). The single-player campaign features two connected storylines, with the first set from 1986 to 1989 during the final years of the first Cold War, and the other set in 2025 during a second Cold War. The protagonist of Black Ops, Alex Mason returns as the protagonist in the first Cold War section, and chronicles the rise to infamy of the game's primary antagonist, Raul Menendez,[14] a Nicaraguan narco-terrorist and the leader of "Cordis Die", a populist movement celebrated as the champion of victims of economic inequality.

C.O.D shows equal performance at 93FPS but overclocked was a tad slower. When running multi-player and scenes that are not programmed, these things do happen. None the less, looks like relative equal performance with Call of Duty.



Dirt 3

Dirt 3 (stylised DiRT 3) is a rallying video game and the third in the Dirt series of the Colin McRae Rally series, developed and published by Codemasters.

With Dirt 3, we see a reversal. AMD pulls one out here.



Alien vs. Predator

Test your PC's DirectX 11 performance with this free, standalone benchmark test based upon Rebellion's 2010 inter-species shooter Aliens vs. Predator. The test shows xenomorph-tastic scenes using tessellation and other such fancy DX11 features.

AvP also shows equal performance.



Resident Evil 6

Resident Evil 6 is a third-person shooter video game in the Resident Evil series, developed and published by Capcom. Capcom defines the game's genre as "dramatic horror", however there is disagreement among reviewers whether this installment belongs in survival horror genre.

Lastly, Intel’s core i3 3220 again, squeaks out a victory here by 400 points at stock but AMD kicks in at overclocked.




AMD 7660D vs. Intel 2500

I ran a few quick gaming benchmarks to pit AMD’s 7660D against Intel’s 2500 graphics. Let’s take a peek.









In each game, AMD owns Intel for on die GPU performance. Gaming is choppy at best in some games with Intel so they are not even on the same planet against AMD. And to top it off, AMD offers their top on die GPU with the A10-5800K where as Intel does not with the core i3 3220.


Power Consumption

Intel has always been very good with Power Consumption. AMD’s Bulldozer cores were not very good but as we see, AMD idles lower than Intel but still draws much more at load. Remember, AMD still uses 32nm fabrication and next generations are working on 22nm fab. 32 nm will pull more power so we will see what shakes out as AMD evolves. Power consumption was done with the Radeon 7970 FYI.

Summary


The debate may continue on what processor is better but facts are facts. Dissecting the various levels of CPU’s can be confusing as many assume Intel makes the best processors. Intel’s SandyBridge “E” processors (on the high end) continue to be the cream of the crop but dropping $500 to $1000 can be a stretch for most. Intel’s 3770K continues to lead AMD but as most saw with the release of Piledriver (FX-8350), AMD is not far behind. In a matter of fact, FX-8350 is a close competitor to Intel’s 3570K and in many multi-threaded applications; the FX processor beats the 3550K. That debate can live another day.

In the budget sector, we saw today that Trinity (A10-5800K) competes well with Intel’s IvyBridge core i3 3220 and they pretty much trade blows out of the box with no clear winner. Intel does offer more memory bandwidth and slightly faster SSD speeds but these did not equate to the 3220 beating the A10-5800K. Some may argue Intel’s core speed is 3.3GHz and AMD’s 3.9GHz, so core for core would be the fairest comparison but that is not true here. The core i3 3220 cannot be overclocked at all. Here is where AMD shines and wins. In almost every benchmark, AMD rose to the top and beat the 3220.

Intel has dropped the price on its core i3 3220 to $129 equaling the price of the A10-5800K at NewEgg so both platforms would set you back equally. One thing that separates the A10-5800K for those not looking for a dedicated GPU is its 7660D on die GPU. The 7660D handily stomps Intel’s 2500 Graphics. As a bonus, AMD’s 7660D can be Cross-fired with a Radeon 7500 GPU or lower. This gives AMD’s APU’s a huge advantage as it can be used 3 ways: standalone, Crossfire or with a dedicated GPU.

Both processors do shine in cost. With either Intel’s i3’s or AMD Trinity, one can build a system for as low as $400 to $600 and get “decent” performance(decent is interpreted many ways). With the entire package looked at, AMD’s Trinity is a much better choice that Intel’s core i3 3220 hands down for performance and versatility.



Recently, news has been abuzz that AMD’s Jaguar is better than Intel’s core i3. The latest news have a 5.9W quad-core Temash outperforming a 17W Core i-3 Sandy Bridge but admittedly, it is too early to tell. If this is true Intel may have some worries. AMD may be closer in bringing x86 technology to the mobile and ultra mobile arena plus rumor has it; Apple is looking heavily into AMD. Maybe AMD was smarter than most thought developing APU’s and looking into efficiency and leaving high end performance for the time being. Rumor also has Haswell not being the hype some think and Steamroller may give AMD some more life at the high end sector catching Intel. Interesting how things change.

Fourstaff said:

Pretty close in almost all benchmarks, with 5800K having an edge. Based on this I think the FX4300 would be the recommended piece of hardware in the budget range for now. I am surprised the 5800K didn't pull ahead in BF3.

Click to expand...

I ran the benchmarks and did the Head to head because some people think core i3's are better than Trinity and that is not the case as seen. When jumping to the FX 4300, the platform changes as to AM3+. FM2 motherboards can be bought very cheap as the A10-5800K at $129. For budget conscious users, the A10-5800K Trinity or even 5600K for less may be the call. I wasn't trying to raise awareness yo the FX-4300 but not a bad point.

camoxiong said:

Thanks for the benchmarks, this means a lot because i'm going to build a budget gaming pc soon.

Click to expand...

Anytime

eidairaman1 said:

5800 Lacks an L3

Click to expand...

Your point? We already know this

eidairaman1 said:

im just saying for comparison to the i3 as to why they are both neck and neck often

Click to expand...

Ahhh. Good point.

newtekie1 said:

They are both $129.99, how many free GPUs do you know that would outperform the iGPU on the A10?

And I'm pretty sure the 7970 was used to try to put the most strain on the CPU as possible(within reason of course).

Click to expand...

Used the 7970 cause:

1- its all I had ATM besides a 6770 Radeon



Wanted to give a more CPU/APU performance approach

There are plenty of reviews of the 3220 and A10-5800k naked too BTW