The AMD A8-7670K APU Review: Aiming for Rocket League
by Ian Cutress on November 18, 2015 8:00 AM ESTOffice Performance
The dynamics of CPU Turbo modes, both Intel and AMD, can cause concern in environments with variably-threaded workloads. 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 ray traces 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.
All AMD CPUs performed similarly here.
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 2,867 files across 320 folders totaling 1.52GB – 95% of these files are small typical website files, and the rest (90% of the size) are small 30-second 720p videos.
WinRAR is all about threads and DRAM speed, so the CPUs that can support higher DRAM frequencies get a boost.
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 win in the single-thread version, whereas the multithread version has to handle the threads, and loves more cores.
Again, all AMD CPUs seem to perform similarly in 3DPM for single-thread mode, indicating that something more fundamental about the design is a bottleneck.
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.
Single-thread frequency and IPC win here.
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.
Mozilla Kraken 1.1
WebXPRT
Google Octane v2
Facebook
Twitter
RSS
154 Comments
View All Comments
BurntMyBacon - Thursday, November 19, 2015 - link
@Ian Cutress: "It's never an issue of lack of interest or subversion, just procurement (and ensuring we can communicate with the manufacturer at the point of testing)."[sarcasm]I thought Intel and all other computer hardware manufacturers were required by LAW to send their products to Anandtech for prelaunch approval where Supreme Tech Justice Cutress and his colleagues pronounce life or death on potential products. ;' ) [/sarcasm]
Seriously though, I'd say you are doing fine despite the real world procurement and scheduling issues getting in the way.
flabber - Wednesday, November 18, 2015 - link
I can't agree any more with the point that power consumption is the least of my concerns. While there is a significant difference between a 250W TDP CPU and a 50W TDP CPU, where one would have to factor in the cost of cooling and a PSU, 100W is quite manageable. A 500W PSU is more that adequate for just about any current system. However, I am aware that decreased power consumption is an objective in all consumer products and will be seeing in upcoming computer components. (Ironically for mobile components, my 2009 Blackberry with a 1150mAh battery can still run for a couple days before I need charge it.)My rig is equipped with a A10-5800k (2012) and a year old R9-290X (2013). Everyday tasks, such as using a spreadsheet, word processor, citation management or occasional image editing, can't be improved in any noticeably way. With regard to gaming, I can't be bothered to upgrade the motherboard and CPU to a superior Intel alternative. A few more frames per second won't make a game with poor storyline any better, nor will an enjoyable game become any better.
Pissedoffyouth - Thursday, November 19, 2015 - link
If you have 5800 you should look at getting a kaveri for better performance and lower power consumptionGadgety - Thursday, November 19, 2015 - link
"AMD's first talking point is, of course, price. AMD considers their processors very price-competitive"No kidding, I got the A8-7600 for my kid and it's integrated graphics is comparable to the i7 Iris Pro, where the i7 is 5x more expensive. So for 20% of the price we get to enjoy graphics galore. Put it on Asrock's A88X M-ITX motherboard and it outputs 4K cinema. No graphics card means it's compact so we use a tiny chassis, perfect HTPC, and useable for the type of light gaming the kids do.
Gadgety - Thursday, November 19, 2015 - link
@Ian Cutress The performance parity, and sometimes superiority, of the A8-7670k compared to the A10-7850k, and also to the A10-7870k, I guess could be attributable to driver improvements. Did you use the same drivers, or updated versions? If it's improved drivers, this would likely also improve across the APU-range.Ian Cutress - Thursday, November 19, 2015 - link
Same drivers for each. I lock a set of drivers down every test-bed refresh, so in this case it would be 15.4 beta, which is getting a bit old now. Kaveri Refresh does have some minor internal improvements as well I imagine, internal bus frequencies perhaps. There's always a small amount of volatility in the benchmark, depending on what heat density or board issue you might have. Looking back, we haven't always used the same motherboard on the APUs just due to timing (but all A88X), and even though we do some overriding on power profiles it can be difficult to compensate for motherboard manufacturer non-user exposed firmware optimisations on the memory buses.Come the next year test-bed refresh (with DX12 relevant titles hopefully), I'll be going back and redoing them all. That should clear out the cobwebs on the latest drivers and updates, providing a new base.
milli - Thursday, November 19, 2015 - link
Ian Cutress, how is the review of your grandparents laptop coming along? :)I'm waiting for that Carrizo review.
zlandar - Thursday, November 19, 2015 - link
I don't see the point of being so cheap that you are unwilling to spend more for a superior i3-5 and discrete video card. Why would you chain yourself to a dead-end cpu/gpu integrated combo and motherboard that isn't very good to start with?Plenty of people have pointed out how well Intel's cpu have held up since Sandy Bridge. I'm still using a 2600k and have upgraded my video card 3 times. If you are on a tight gaming budget it makes alot more sense to buy 2nd-3rd tier last gen video cards coupled with a good cpu you don't need to upgrade.
BurntMyBacon - Thursday, November 19, 2015 - link
@zlandar "Why would you chain yourself to a dead-end cpu/gpu integrated combo and motherboard that isn't very good to start with?"Aren't just about all laptops deadend with respect to CPU/GPU? (Particularly in the Carrizo price range)
Ton's of laptops are sold without discrete GPUs and no option to upgrade. Why should this matter to a Carrizo review (clearly laptop in this request)?
Ian Cutress - Thursday, November 19, 2015 - link
Something special in the works. After SC15 finishes, I'll be digesting the mountain of data I have. :)