The AMD A8-7650K APU Review, Also New Testing Methodology
by Ian Cutress on May 12, 2015 10:00 AM ESTOffice Performance
The dynamics of CPU Turbo modes, both Intel and AMD, can cause concern during environments with a variable threaded workload. There is also an added issue of the motherboard remaining consistent, depending on how the motherboard manufacturer wants to add in their own boosting technologies over the ones that Intel would prefer they used. In order to remain consistent, we implement an OS-level unique high performance mode on all the CPUs we test which should override any motherboard manufacturer performance mode.
All of our benchmark results can also be found in our benchmark engine, Bench.
Dolphin Benchmark: link
Many emulators are often bound by single thread CPU performance, and general reports tended to suggest that Haswell provided a significant boost to emulator performance. This benchmark runs a Wii program that raytraces a complex 3D scene inside the Dolphin Wii emulator. Performance on this benchmark is a good proxy of the speed of Dolphin CPU emulation, which is an intensive single core task using most aspects of a CPU. Results are given in minutes, where the Wii itself scores 17.53 minutes.
Dolphin likes high IPC and clock frequency, which indicates that the highest clocked APUs perform the best out of AMD here. Because the benchmark is single threaded, even a dual core Intel wins though.
WinRAR 5.0.1: link
Our WinRAR test from 2013 is updated to the latest version of WinRAR at the start of 2014. We compress a set of 2867 files across 320 folders totaling 1.52 GB in size – 95% of these files are small typical website files, and the rest (90% of the size) are small 30 second 720p videos.
The varied-thread workload of WinRAR seems to vary between the dual core Intels and the dual module AMD chips, showing that threads matter.
3D Particle Movement
3DPM is a self-penned benchmark, taking basic 3D movement algorithms used in Brownian Motion simulations and testing them for speed. High floating point performance, MHz and IPC wins in the single thread version, whereas the multithread version has to handle the threads and loves more cores.
In this nieve benchmark, the compiler prefers x87 style commands which AMD's Bulldozer based architectures isn't too fond of. This benchmark is meant to be a representation of crude scientific code, similar to that used in a research lab. Ultimately, Bulldozer architectures such as Kaveri prefer specific commands, especially when dealing with basic math.
FastStone Image Viewer 4.9
FastStone is the program I use to perform quick or bulk actions on images, such as resizing, adjusting for color and cropping. In our test we take a series of 170 images in various sizes and formats and convert them all into 640x480 .gif files, maintaining the aspect ratio. FastStone does not use multithreading for this test, and results are given in seconds.
Web Benchmarks
On the lower end processors, general usability is a big factor of experience, especially as we move into the HTML5 era of web browsing. For our web benchmarks, we take four well known tests with Chrome 35 as a consistent browser.
Sunspider 1.0.2
Mozilla Kraken 1.1
WebXPRT
Google Octane v2
Facebook
Twitter
RSS
177 Comments
View All Comments
akamateau - Tuesday, May 12, 2015 - link
Your comments assume that ANAND provided benchmarks using DX12 and they did not. ALL of the GRAPHIC benchmarks were with either synthetic benchmarks or game benchmarks using DX11.DX11 cripples the performance of ALL APU's, IGP and dGPU. Draw calls ARE a measure of CPU-to-GPU "bottleneck" or elimination thereof. You can not render a polygon until you draw it.
DX12 enables CPU core scaling; basically increased draw calls are a function of the amount of multithreaded cpu cores. DX11 does not allow multithreeaded gaming.
DX11 may be current but why should I base hardware purchases from testing based on obsolete software AND benchmarks?
DX12 will be in widespread use by game developers by Christmas.
Anand has psent quite a bit of time and money testing hardware on obsolete benchmarks TO WHAT END?
Starswarm and 3dMArk API Overhead Test are available but ANAND ignore them.
Why?
AMD's APU was designed to FLY using Mantle and DX12. It is not AMD's fault that Intle IGP is so poorly designed. That is Intel's problem.
Test Intel IGP using the latest API and you will see. Comaparatively test AMD and Intel using obsolete benchmarks with DX11 and ANAND is lying to the consumer and can not be trusted.
AN unbiased and well balanced piece should use legacy benchmarks, they should also use the very latest available. ANAND di not do this.
rp1367 - Tuesday, May 12, 2015 - link
"Starswarm and 3dMArk API Overhead Test are available but ANAND ignore them.Why?"
Because they want to hide the truth. "It is hard for a person to wake if he is asleep because he pretends to be asleep but infact he is not. He just want to fool you because of his stupidity"
The refusal to support the upcoming DX12 give as hint that the review is biased and something fishy going at the backdoor. I am not an IT guy and new on this site but i could easily detect what is the difference between biased and unbiased review.
The reviewer and Anadtech guys for sure are all intellegent guys but they allowed themselves to be succumed by their own personal interest.
Gigaplex - Tuesday, May 12, 2015 - link
Or its just something as simple as DX12 not being released yet, the performance is likely to change, so is an invalid test for comparing hardware at this time. The benchmarks you refer to are only valid as a preview for potential gains.akamateau - Tuesday, May 12, 2015 - link
Windows 10 with DX12 will be released in less than 2 months. Mantle is final and DX12 is final. Anand has it but ignores it.By Christmas ALL new games released will be DX12.
Gigaplex - Tuesday, May 12, 2015 - link
DX12 may not be final. API probably is, runtime is likely close, drivers likely won't be.And you're delusional if you think all new games released at the end of this year will be DX12. It takes years to develop a AAA game, so they would need to have started before DX12 was available. The market for DX12 will be tiny by Christmas as DX12 will be Windows 10 only. Not everyone will be willing or able to upgrade the OS. Not all hardware even supports DX12. You're completely ignoring the history of previous DirectX roll outs.
akamateau - Tuesday, May 12, 2015 - link
Windows 10 with DX12 will be released in less than 2 months. Mantle is final and DX12 is final. Anand has it but ignores it.By Christmas ALL new games released will be DX12.
V900 - Tuesday, May 12, 2015 - link
Ah, so basically the choice is buy an AMD APU and get shoddy performance now, and great performance in a year, or buy an Intel/Intel-Nvidia solution and get great performance now and great performance in a year!So theres really no reason to get the AMD is what you're saying?
JumpingJack - Tuesday, May 12, 2015 - link
I will bet you what ever you make in a year that not all games released between DX12 release and christmas of this year will be DX12 native.akamateau - Tuesday, May 12, 2015 - link
@Ian Cutress"as well as having parallelizable code"
ARE YOU NUTS?
You really need to cut the crap.
Mutlthreaded gaming will come as a result of DX12 and Asynchronous Shader Pipelines and Asynchronous Compute Engines.
galta - Tuesday, May 12, 2015 - link
I don't think he is nuts, but you seem a bit angry.From a CPU perspective, multithreaded games need not wait for DX12. They could have been written before.
Anyway, we have a clear statement of you: DX12 will make AMD shine. We should talk again on Christmas.
Just keep in mind that the same was said when DX11 was about to be released, with known results...