AT Mobile Kaveri CPU Performance Preview

[jdubs03]

PCMark 8 uses OpenCL acceleration, hence why it favours faster IGPs and why its a constant in AMD's most recent comparison marketing slides.

Oh alright, I wasn't sure if it came down to just OpenCL, or there were further tweaks. Thanks.

 

[AtenRa]

From what I have seen notebookcheck does a real good job in determining CPU performance aggregation, but when devices are compared against one another it is harder to get a clear picture.

For instance in the Acer Aspire review with the A10-7300 it has PCMark 7 comparing a i5-4200U only scoring 2533, but on Anandtech all the PCMark 7 scores are over 4000. And in the Surface Pro 3 review, they don't even include PCMark 8, but they did with their latest review of the HP Elitebook Revolve (in which the Acer A10 score higher in home, and slightly lower in creative, yet the PCMark 7 was only 4131, but SP3 is 5076 with the same processor - i5-4300U). And from Anandtech the core i5 SP3's score is lower in PCMark 8 then the A10 in both home and work.

Very contradictory. And my question is why is there such a large difference between PCMark 7 and 8 results?, where clearly PCMark 7 favors the i5-4300U by almost triple (in the case of SP3) over the A10-7300.

Because PC MARK 7 is very depended on HDD/SSD. If one device has SSD and the other device has HDD, the final score difference is huge.

 

[Phynaz]

Because PC MARK 7 is very depended on HDD/SSD. If one device has SSD and the other device has HDD, the final score difference is huge.

As it should be.

 

[artivix]

Very nice! It is matching or beating intel's equivalent in standard workloads and obliterates intel's equivalent in 3D, gaming and OpenCL.

No wonder OEM's are lining up with new mobile Kaveri designs: Acer, Asus, HP, Lenovo, Samsung, Toshiba, and more.

Many people interested low mid cost gaming laptops and with FX Kaveri potential for good crossfire performance. Recent test with AMD A8-7600 Kaveri and low end R7 240 running games at 1920x1080 point at very good boost with crossfire.

http://pcgamerhome.com/amd-a8-7600-and-r7-240-in-crossfire-mode-to-the-test/

Now FX Kaveri models (FX-7500) showing with R7 M255/M265 graphics --

http://www.pc21.fr/fiche/80ec003kfr-z50-75-amd-fx-7500-1tb-8gb-15-6in-dvdrw-w8-1-fr-i565864.html


http://www.notebooksbilliger.de/lenovo+z50+75+80ec001fge

After testing at 1920x1080 resolution around 40-50fps, AMD FX-7500 Kaveri with M255/M265 notebooks with 720p med will reach 60 fps in most games.

First time in history for around $500-650!

 

[Erenhardt]

^yes, that is what we are waiting for. Laptops with crappy GPUs made little sense. You had to see what is playing the game better: IGP, or GPU and disable the slower one. Now you will just enable xfire and boost you fps to high-end level (sort of)

 

[frozentundra123456]

Looks promising, but still skeptical until we see a more complete test with frame pacing and microstutter evaluation and a wider variety of games (edit: including cpu demanding ones). Similar fps gains were seen with Richland but the experience was marred by microstutter.

 

[monstercameron]

Looks promising, but still skeptical until we see a more complete test with frame pacing and microstutter evaluation and a wider variety of games (edit: including cpu demanding ones). Similar fps gains were seen with Richland but the experience was marred by microstutter.

AMD has sorted out framepacing on kaveri and gcn 1.1 so I have high hopes it will be decent.

 

[TrulyUncouth]

Yah, I guess the question is whether they will use the xdma or whatever engine to do the crossfire like the 290/x... could be an interesting value gaming laptop

 

[lefty2]

A lot of games don't support dual graphics

 

[frozentundra123456]

A lot of games don't support dual graphics

Interesting point, but I thought most games did now. Could be a problem with some older games, or indie games, etc., but those might run adequately on the IGP.

The two main potential problems I see are whether microstutter has been eliminated, and how much cpu performance they can get out of 19 watts, when part of that TDP has to be shared with the igp. Hopefully we will get some benchmarks, as I said, with a wider variety of games, especially cpu demanding ones, including on-line multiplayer and mmos.

 

[USER8000]

Looks promising, but still skeptical until we see a more complete test with frame pacing and microstutter evaluation and a wider variety of games (edit: including cpu demanding ones). Similar fps gains were seen with Richland but the experience was marred by microstutter.

With some earlier drivers:

http://www.pcper.com/reviews/Graphi...0-Dual-Graphics-Testing-Pacing-Fixed/Battlefi

 

[hoorah]

I feel like I'm coming in with a "GO BANANA!" comment after all of this in depth technical discussion.

Looking at all of the benchmark numbers for the current games, it seems like most of them have not great performance even on low settings (for all of the iGPUs).

Typically the performance I want out of a laptop to is play one or two of my frequently played, low end games like CS/SourceMod/RTCWET(dating myself)/L4D at a good framerate and leave the big games for my desktop. As long as I had one game I could jump into on my laptop that was usually good enough.

I'm not sure how to place this Kaveri/HD4400/4600 stuff. Its right in that middle ground where its plenty fast for the older games but not really there to be a desktop replacement for a really light gamer like me.

I also wonder if, since all I'm going to do on my next laptop (gaming-wise anyway) is play CS or L4D, I might as well go with the intel setup since its adequate for source games but better for everything else.