Mantle Showdown: AMD’s A10-7850K

Part 2: Mantle on entry level hardware…

7850K (1 of 3)

In Part 2 of this series we’ll be looking at the benefits of AMD’s Mantle API over Microsoft’s DirectX 11. This is a five part series; in part 2 we’ll be looking at the performance of AMD’s flagship APU the A10-7850K. For information on the Mantle API and DirectX 11 please read part 1 of this series.

Our test system is built around AMD’s A10-7850K which is a $160 chip that has four cores and uses AMD’s socket FM2+. With 16 Gigabytes of DDR3 at 2.133 Ghz and a dual channel memory controller we consider this system to be an entry-level PC. We chose the A10-7850K because it represent the pinnacle of AMD’s on-die graphics solutions and offers midrange CPU performance. It’s also thought to be a memory bandwidth limited chip even with high bandwidth memory like the kit we’re using here.

Test system entry-level mantle

For the sake of transparency we want you to know that AMD provided the A10-7850K and MSI motherboard we’ll be testing today along with codes for all the games we’ll be looking at. AMD also provided the our test system’s RAM. All of the other components we’re using were purchased at retail without the knowledge or consent of those companies. As always you can find our raw testing data on OneDrive. We took no outside input for this article other than the suggestions and support coming from our lovely forums, thanks yet again guys.

7850K (3 of 3)

Our Benchmarks

We’ll be using the same games that we detailed in part 1. Namely Civilization: Beyond Earth, Plants Vs Zombies: Garden Warfare, Thief, Sniper Elite 3, and Battlefield 4. Unlike our testing in part 1 these games will not be benchmarked at their maximum settings. Rather these game were set to their lowest possible quality settings with the aim of achieving as close to 60 frames per second as possible at a resolution of 1080P. Why 60 frames per second? Simply put because our test bench has a 60 Hz monitor. All of these games were playable but most of them were not smooth. The goal of using such a high frame rate is to meet what we consider to be an objective standard for smooth gameplay which is rendering at least as many frames per second as the monitor can display.

A10 APU Mantle Performance

Our first impressions of Mantle running on the A10-7850K weren’t very strong. Performance uplifts range from 0 to 30 percent which sounds good, but we only saw an average increase of 10 percent. Compare this to the 21 percent performance uplift the R9 290 received from Mantle and it looks weak. Whereas Mantle doubled performance on our High-end test system we saw no reliable performance increase from using the Mantle version of Civilization: Beyond Earth on our entry-level PC. We re-ran this test multiple times to confirm our initially confounding results.

APU Mantle Performance Table

Mantle provides a 10 percent average performance uplift on AMD’s A10-7850K. The reality of these results is that they say more about the 7850K than they do about the Mantle API. If your application is GPU bound then there’s not much an API designed to relieve CPU bottlenecks can do for its performance. I’m sure this might come as a surprise to some readers but according to our results the CPU portion of AMD’s APU are not the weakest link for gaming performance. Rather the A10-7850K appears to be largely GPU bound in these applications. Given our prior testing with this chip and the performance increases in games that it sees from bumping its memory speed from 1.333 Ghz to 2.133 Ghz it’s clear that these results are more evidence that AMD’s Kaveri is memory bandwidth limited.

7850K (2 of 3)

Mantle Improves Performance on Entry-level Hardware

Again Mantle offers current APU owners an easy way to see a small improvement in frame rates for free. With Kaveri’s successor on the horizon it nice to see that AMD’s software efforts are still taking into account performance on it’s less prominent hardware. One of the selling points of the GCN architecture is that it allows for software aimed at the high-end products to benefit performance even on comparatively anemic hardware. That’s an assertion that rings true in this comparison even if the gains are modest.S|A

The following two tabs change content below.
Thomas Ryan is a freelance technology writer and photographer from Seattle, living in Austin. You can also find his work on SemiAccurate and PCWorld. He has a BA in Geography from the University of Washington with a minor in Urban Design and Planning and specializes in geospatial data science. If you have a hardware performance question or an interesting data set Thomas has you covered.