Drawing Performance Conclusions

As I progressed through the testing for this review, I became aware of trends in two things: absolute performance, in terms of numbers, and generational improvements across platforms. With AMD moving the Bulldozer based architectures from Piledriver in Trinity and Richland to Steamroller in Kaveri, the base CPU design has had a relatively long run in order to be optimized to the limitations on the CPU side. Obviously AMD has also had a chance to change the lithography node in there as well, and as such has optimized for design rather than performance. This matters a great deal when we look at the power banding (45W, 65W, 95W) and performance in each segment.

In order to display just how much Kaveri has grown in relation to the previous generations, I processed our mountain of data to show graphs where percentage gain against the older generations really does matter. As mentioned previously in this review, it all depends on whether the software can take advantage of the new features afforded by the architecture. In this review we have spent a lot of time discussing advantages for Kaveri in terms of compute (HSA, hUMA, hQ) and gaming (Mantle, TrueAudio), but some of these are still a little way off for software integration. So what does that translate to if you are looking for an APU today – does Kaveri still expand its wings with the integration of the GCN architecture, or does it lag behind due to the lithography node no longer being optimized for frequency. The proof is in the results.

IGP Gaming, 1280x1024

For the 100W APUs at 1280x1024, there is almost no movement between the Richland and the Kaveri APUs, except for Company of Heroes which would seem to leverage the extra SPs more than the MHz available.

45W APUs are clearly in vogue at 1280x1024, each of them providing a good bump over Richland.

IGP Gaming, 1680x1050

1680x1050 sees more benefit for the 100W APUs, but the 45W APUs still show big leaps.

IGP Gaming, 1920x1080

At 1080p the effects are more pronounced all around.

CPU Timed Benchmarks

Unfortunately the CPU timed benchmarks at 100W are bit over the place. Here we are dealing with an IPC gain but MHz deficit, and each benchmark has a different way of dealing with it. For Agisoft, MHz wins, but for Xilisoft, the Kaveri improvements work out well for the system.

At 45W the situation is a lot more clearer, and the newer APUs have the advantage, especially in Adobe After Effects.

CPU Score Benchmarks

45W still makes the biggest jump when it comes to score based CPU benchmarks.

IGP Synthetics

I never like putting much thought into the synthetic results – here both the 100W and the 45W range are showing good improvements all around. 3DMark 06 has the most trouble getting improvements as it relies more on the grunt, rather than the features that Kaveri affords.

IGP Compute

Off the bat we can see just how much of a difference the Kaveri GPU Compute can make over previous APUs. This is still without hUMA fully being in effect while we wait for a proper driver and software stack to become available.

Up to a +222% boost from Kaveri is amazing for a 45W part.

AMD vs Intel

For a little tete-a-tete, putting the 65W Iris Pro up against the 65W Kaveri seems like a good idea. To add to the mix, due to the extreme price difference of Iris Pro in this equation, we also add the 95 W Kaveri for comparison.

It is interesting to note that at the lower resolutions the Iris Pro wins on most benchmarks, but when the resolution and complexity is turned up, especially in Sleeping Dogs, the Kaveri APUs are in the lead.

Processor Graphics: Compute and Synthetics Final Words
POST A COMMENT

373 Comments

View All Comments

  • geniekid - Tuesday, January 14, 2014 - link

    Would've been nice to see a discrete GPU thrown in the mix - especially with all that talk about Dual Graphics. Reply
  • Ryan Smith - Tuesday, January 14, 2014 - link

    Dual graphics is not yet up and running (and it would require a different card than the 6750 Ian had on hand). Reply
  • Nenad - Wednesday, January 15, 2014 - link

    I wonder if Dual Graphics can work with HSA, although I doubt due to cache coherence if nothing else.

    While on HSA, I must say that it looks very promising. I do not have experience with AMD specific GPU programming, or with OpenCL, but I do with CUDA (and some AMP) - and ability to avoid CPU/GPU copy would be great advantage in certain cases.

    Interesting thing is that AMD now have HW that support HSA, but does not yet have software tools (drivers, compilers...), while NVidia does not have HW, but does have software: in new CUDA, you can use unified memory, even if driver simulate copy for you (but that supposedly means when NVidia deliver HW, your unaltered app from last year will work and use advantage of HSA)

    Also, while HSA is great step ahead, I wonder if we will ever see one much more important thing if GPGPU is ever to became mainstream: PREEMPTIVE MULTITASKING. As it is now, still programer/app needs to spend time to figure out how to split work in small chunks for GPU, in order to not take too much time of GPU at once. It increase complexity of GPU code, and rely on good behavior of other GPU apps. Hopefully, next AMD 'unification' after HSA would be 'preemptive multitasking' ;p
    Reply
  • tcube - Thursday, January 16, 2014 - link

    Preemtion, dynamic context switching is said to come with excavator core/ carizo apu. And they do have the toolset for hsa/hsail, just look it up on amd's site, bolt i think it's called it is a c library.

    Further more project sumatra will make java execute on the gpu. At first via a opencl wrapper then via hsa and in the end the jvm itself will do it for you via hsa. Oracle is prety commited to this.
    Reply
  • kazriko - Thursday, January 30, 2014 - link

    I think where multiple GPU and Dual Graphics stuff will really shine is when we start getting more Mantle applications. With that, each GPU in the system can be controlled independently, and the developers could put GPGPU processes that work better with low latency to the CPU on the APU's built in GPU, and processes for graphics rendering that don't need as low of latency to the discrete graphics card.

    Preemptive would be interesting, but I'm not sure how game-changing it would be once you get into HSA's juggling of tasks back and forth between different processors. Right now, they do have multitasking they could do by having several queues going into the GPU, and you could have several tasks running from each queue across the different CUs on the chip. Not preemptive, but definitely multi-threaded.
    Reply
  • MaRao - Thursday, January 16, 2014 - link

    Instead AMD should create new chipsets with dual AMU sockets. Two A8-7600 APUs can give tremendous CPU and GPU performance, yet maintaining 90-100W power usage. Reply
  • PatHeist - Thursday, February 13, 2014 - link

    Making dual socket boards scale well is tremendously complex. You also need to increase things like the CPU cache by a lot. Not to mention that performance would tend to scale very badly with the additional CPU cores for things like gaming. Reply
  • 5thaccount - Tuesday, January 21, 2014 - link

    I'm not so interested in dual graphics... I am really curious to see how it performs as a standard old-fashioned CPU. You could even bench it with an nVidia card. No one seems to be reviewing it as a processor. All reviews review it as an APU. Funny thing is, several people I work with use these, but they all have discrete graphics. Reply
  • geniekid - Tuesday, January 14, 2014 - link

    Nvm. Too early! Reply
  • boozed - Tuesday, January 14, 2014 - link

    You must be a hoot at parties. Reply

Log in

Don't have an account? Sign up now