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
xprojected - Tuesday, May 12, 2015 - link
Mid-range:- MSI GTX 770 Lightning 2GB ($245-$255 on eBay/Amazon, $330 new)
- MSI R9 285 Gaming 2GB ($240)
Hold on. You suggest that the 770 and the 285 are nearly the same price, but you list the used/refurbished price for the 770 first. That opens up a Pandora's box, doesn't it? If it's too hard to find a card new, pick a different one, like the 970, or 960, which is actually close in price to the 285 (at least a couple go for $200 on Newegg). Even though you say you split the GPUs based on price ranges, rather than similar prices, people are going to compare ATI to NVidia and you have an unfair used-vs-new price comparison.
Ian Cutress - Wednesday, May 13, 2015 - link
Ideally the tests are meant to show comparisons within a GPU class, not between GPU classes. Ultimately I work with the cards I have, and on the NV side I have a GTX 980 and a GTX 770, whereas on the AMD side is an R9 290X and R9 285 (the latest Tonga). In comparison to what I have, the 980/290X are high end, and the 770/285 are a class below. The 770 Lightning is also hard to source new, due to its age, but is still a relevant card. If I could have sourced a 960/970, I would have.Navvie - Wednesday, May 13, 2015 - link
That is one of the things that really puts me off AT these days. The attitude of "this is what I have available, this is what I'll test against." If it's not relevant don't use it. Get out the AT credit card and buy some new hardware.milkod2001 - Wednesday, May 13, 2015 - link
if they want to spend more money they will have to make more money first.To make more money they would have to spread million annoying ads and this site would quickly turned into Toms super boring "best SSD for the money" articles like, with millions amazon, newegg links, ads and all that crap
They better use what they have in disposal and try to maintain still decent enough articles, reviews etc.
MrSpadge - Tuesday, May 12, 2015 - link
"Despite the rated memory on the APUs being faster, NPB seems to require more IPC than DRAM speed."Guys.. the Intel chips have better memory controllers since many years. They extract much higher performance and lower latency if you compare them at similar DRAM clock & timings. Lot's of AT benchmarks showed this as well, back when such things were still included (e.g. when a new architecture appears).
rp1367 - Tuesday, May 12, 2015 - link
Here we go again, this is another pro Intel review. The crooked company who paid AMD Billion $ settlement case due to unfair competition practices are still being supported by lots of rotten people based on their comments here is disturbing. I find these people scum on this earth as they continue to support the scammers. Shame on you guys!By the way the review is biased as you benchmark it using DX11. This is another manipulative benchmark trying to hold on the past and not on the future. Read my lips "WE DONT WANT DX11 review only! DX12 is coming in a few months why not use the MS Technical review version so people can have a glimpse instead of repeating to us reader, then you redo the benchmark when MS has released its new OS (MS 10). If you dont have full version DX12 now (as we all know) then do not benchmark it bacause it is the same for the past 5 years. You are just wasting your time if your intent is neutral to general consumers as if you are trying to sway us from the truth.
shadowjk - Friday, May 15, 2015 - link
Speak for yourself. What use to me are benchmarks of APIs I don't have and can't get, of games I don't have and don't play. With this latest review method change, the last game I had disappeared, so now I can't compare the results with my own system anymore. This makes it harder to decide whether this product is a good upgrade over my current system or not.Teknobug - Tuesday, May 12, 2015 - link
Looks like A8 7600 with R7 240 dual gfx is the combination here. I have an A8 7600, so now just need that card.UtilityMax - Tuesday, May 12, 2015 - link
Am I the only poster who is impressed with the performance of the Kaveri parts in the gaming benchmarks?For one, the Kaveri parts virtually eliminate the need for a $70 discrete GPU. If you were thinking of that kind of low-end GPU, might as well buy an APU. Next, the is very little difference or none at all in average FPS under many settings if you use a $240 dedicated GPU, which means that and $200 GPU is still the bottleneck in a gaming system. Only once the benchmarks are run with very high end GPUs, we finally see the superiority of the Haswell parts.
Of course, the business, web, compression, and conversion benchmarks are another story. Except for a few special cases, the APUs struggle to catch a Core i3.
meacupla - Wednesday, May 13, 2015 - link
Yeah, these APUs are certainly a lot better on the CPU front than what they used to be.I think the APU's only downfall is that $72 Pentium G3258 and availability of cheaper (sub $100) H97 motherboards to overclock them on.
A88X FM2+ boards are around the same price as H97 boards, but that $30 saving from cheaper CPU can go into a decent cooler for overclocking.