AMD Kaveri Review: A8-7600 and A10-7850K Tested
by Ian Cutress & Rahul Garg on January 14, 2014 8:00 AM ESTCPU Performance: Continued
Xilisoft Video Converter 7 - link
The XVC test I normally do is updated to the full version of the software, and this time a different test as well. Here we take two different videos: a double UHD (3840x4320) clip of 10 minutes and a 640x266 DVD rip of a 2h20 film and convert both to iPod suitable formats. The reasoning here is simple – when frames are small enough to fit into memory, the algorithm has more chance to apply work between threads and process the video quicker. Results shown are in seconds and time taken to encode. XVC also offers acceleration via CUDA and AMD APP, so if these are available on the CPU we offer results with and without.
With large frame data, the IGP on Kaveri does not particularly help much.
For smaller frames however, there is an advantage to enabling the AMD APP function.
HandBrake v0.9.9 - link
For HandBrake we do the same files as XVC but convert them into the default format Handbrake offers upon loading the software. Results shown are in Frames Per Second.
Handbrake loves cores, threads and MHz
Adobe After Effects 6
Published by Adobe, After Effects is a digital motion graphics, visual effects and compositing software package used in the post-production process of filmmaking and television production. For our benchmark we downloaded a common scene in use on the AE forums for benchmarks and placed it under our own circumstances for a repeatable benchmark. We generate 152 frames of the scene and present the time to do so based purely on CPU calculations.
7-Zip 9.2 - link
As an open source compression tool, 7-Zip is a popular tool for making sets of files easier to handle and transfer. The software offers up its own benchmark, to which we report the result.
PovRay 3.7 - link
PovRay historically loves threads, MHz and IPC. The standard benchmark from PovRay is what we use to test here.
TrueCrypt 7.1a - link
TrueCrypt is an off the shelf open source encoding tool for files and folders. For our test we run the benchmark mode using a 1GB buffer and take the mean result from AES encryption.
Facebook
Twitter
RSS
380 Comments
View All Comments
jjj - Tuesday, January 14, 2014 - link
lol now that AMD is paying you,you jumped at the other extreme , what do you get over Richland and how does that deserve a positive conclusion? (and maybe a reminder is needed A10-6800K is 140$)Drumsticks - Tuesday, January 14, 2014 - link
The author recognized that 100W TDP isn't really any different...? The 45W more mainstream range is where things are a lot more interesting, but please, ignore the 30+% gains.Conduit - Tuesday, January 14, 2014 - link
I have been waiting almost 2 years for a mobile Kaveri telling myself the wait will be worth it. Looks like it actually may be as Kaveri performs better at lower TDP's than Trinity and Richland.Hubb1e - Tuesday, January 14, 2014 - link
This guy actually gets it. These should make very nice cheap laptops able to actually play a game or two without vomiting all over themselves. At 95W the gains are not there, but at 45W it's impressiveJDG1980 - Tuesday, January 14, 2014 - link
What an odd choice of benchmarks. Why not use Photoshop (which supports OpenCL and is an important real-world application) instead of Agisoft (who?) I can't help but wonder if Agisoft paid for their inclusion in Anandtech's benchmark suite.ddriver - Tuesday, January 14, 2014 - link
You may notice that the entire review is not exactly OpenCL compute rich... I mean even for gaming cards AT usually includes several OpenCL benchmarks, but not for this product, where it is supposed to be its strongest point???Conduit - Tuesday, January 14, 2014 - link
I have to say, the 45W A8-7600 kicks ass. It's competitive with the 100W A10-5800K, I think that's where the benefit of Kaveri lies, in the low TDP department.UtilityMax - Tuesday, January 14, 2014 - link
Indeed. However, I personally would like to see benchmarks that measure the power draw. The 65watt TDP Richland A10s were nearly as power thirsty as the +100watt TDP parts. A8 does look interesting.thegreatjombi - Tuesday, January 14, 2014 - link
Any chance we can get HD5200 and HD5400 benches? The 5000 series has started to catch up with AMD in terms of GPU Compute. I am curious if AMD has left Intel in the dust once again, or has Intel actually made a big enough leap to stay in the game.thegreatjombi - Tuesday, January 14, 2014 - link
Oh! It seems the 5000 series is just omitted in the CPU Performance benchmarks? Why is that?