AMD Kaveri Review: A8-7600 and A10-7850K Tested
by Ian Cutress & Rahul Garg on January 14, 2014 8:00 AM ESTProcessor Graphics: Compute and Synthetics
Our GPU Compute tests are designed to stress the integrated graphics of the system using old and new GPU tools. This is where the benefits of Kaveri should come in – if a benchmark is programmed to take advantage of the GCN architecture then the relevant speed up should be observed. In none of the GCN benefits have any use then we are relying on pure grunt to get the top scores.
CompuBench CL Fluid GPU + Computer Vision GPU
CompuBench CL is an OpenCL compute benchmark designed to test some common in-use algorithms. Due to the way it is designed, any OpenCL compatible device can run the benchmarks, and as such we have run both the Fluid Dynamics and Computer Vision portions on the CPU and integrated graphics on each processor we have tested. Here are the GPU results.
GCN really wins out here, showcasing how important OpenCL compliance can be to extract potential compute.
Facebook
Twitter
RSS
380 Comments
View All Comments
ABR - Thursday, January 16, 2014 - link
Kind of funny to hear "Iris" mentioned here given that the SGI O2 was an early example of a workstation with HUMA back in the mid-90's. :) There was no such thing as GPGPU back then (it might have helped save SGI if there was), but it provided a big help for memory-intensive things like texture-mapping.vinayshivakumar - Thursday, January 16, 2014 - link
Why are the iGPU codenames mentioned in the CPU benchmark and the CPU is in brackets ?? What is the reason to mention the GPU at all in a CPU test ? To the untrained eye , it might look as is HD3000 is faster than the AMD R7... Which is obviously not true...vinayshivakumar - Thursday, January 16, 2014 - link
Am I missing something here , apart from a bias ?Th-z - Thursday, January 16, 2014 - link
I agree the labels can be misleading, in CPU benchmarks other than the second table "Agisoft PS v1.0 Mapping IGP" on page 10, all other are done with CPU only. Some tables got it right while some have iGPU names leading the labels and model names in the brackets. What should be in the brackets are iGPUs. Consistency in labelling is needed. It's probably not intentional, just a result of messing up the labels and get the review out ASAP.ryrynz - Thursday, January 16, 2014 - link
You should be listing the GPU first in the Winrar results. The GPU is obviously not important in this benchmark and really shouldn't be specified at all.sonofsanta - Thursday, January 16, 2014 - link
"...the benchmark follows Jenson Button in the McLaren who starts on the grid in 22nd place, with the field made up of 11 Williams cars, 5 Marussia and 5 Caterham in that order."You sure a 2013 McLaren could make it through that field? I reckon it'd still fail to make the podium.
HaryHr - Thursday, January 16, 2014 - link
What's the story with Iris Pro and huge difference between average FPS and minimum FPS?In BI performance test: 78 average and 6 minimum, quality test: 20 average and 1 minimum.
If this is happening during gameplay this makes game almost unplayable. AMDs lineup has much smaller variance.
It would be nice to see time frame tests.
synce - Thursday, January 16, 2014 - link
What a disappointment... I thought the 7850K would be the first rig I've built in years but it's barely an improvement over the 6800 and still no match for an i5 :(Rogatti - Thursday, January 16, 2014 - link
Drivers still immature ... I will wait to form an opinionSolMiester - Thursday, January 16, 2014 - link
WOW, looks to me that the A10 6800k is the better bang. I saw only 1 game where the R7 bettered the 6800K where any of them were playable....