AMD did provide some interesting numbers that are worth discussing though. Please keep in mind that these benchmarks were all run by AMD and as such should only be used as a placeholder until media and consumers can verify them. Tests were run with a combination of two graphics cards and two processors to represent the scale of potential scalability. The percentage improvements displayed below are in the average frame rate of a pre-set test run in BF4.
Core i7-4960X CPU + R9 290X GPU
1080p, Ultra Preset, 4xAA: 9.2% improvement with Mantle
1600p, Ultra Preset, 4xAA: 10% improvement with Mantle
Core i7-4960X CPU + R7 260X GPU
1080p, Ultra Preset, 4xAA: 2.7% improvement
1600p, Ultra Preset, 4xAA: 1.4% improvement
A10-7700K CPU + R9 290X GPU
1080p, Ultra Preset, 4xAA: 40.9% improvement
1600p, Ultra Preset, 4xAA: 17.3% improvement
A10-7700K CPU + R7 260X GPU
1080p, Ultra Preset, 4xAA: 8.3% improvement
1600p, Low Preset: 16.8% improvement
So what can we make of these results until we can run our own? Clearly the advantage of Mantle shows itself most dramatically when you are in a heavily CPU limited environment. Take a look at the results when using the Kaveri A10-7700K APU as the primary processor; with the R9 290X you are seeing more than 40% average frame rate increase at 1920x1080 and 17.3% at 2560x1600 resolutions! However, pairing that same R9 290X graphics card with the Core i7-4960X Ivy Bridge-E processor, probably the highest performing consumer CPU, and that drops to 9.2% and 10% at each resolution. Those results are much more modest but are still pretty compelling if they turn out to be true.
With the lower cost graphics card, the R7 260X, scaling takes another twist. Coupling it with the high performance Core i7-4960X results in just a 2.7% and 1.4% increase in average frame rate. Nothing worth raving about there for sure. With the A10-7700K though the R7 260X is able jump up performance levels by 8.3% at 1080p and by as much as 16.8% at 2560x1600 (though at lower quality settings).