AMD A10-4600M ''Trinity'' 3DMark 11 Performance Leaked
AMD has released slides to the Korean market detailing the graphics performance and specifications of its next generation ''Trinity'' APU.
The slides give us a small glimpse of what to expect from “Trinity” based APUs in the near future, specifically from the upcoming A10-4600M mobile APU. On the CPU side, the A10-4600M packs a quad-core powered “Piledriver” CPU clocked at 2.3 GHz with 4MB of L2 cache, a TDP of 35W and features Turbo Core Technology, capable of bringing the clock speeds up to 3.0 GHz. On the GPU side, you'll find an integrated Radeon HD 7660G clocked at 685Mhz with 384 Radeon cores, which can be coupled with a discrete Radeon HD 7670M for Crossfire.
So how well does the APU perform? The folks at Nordichardware extracted the 3DMark 11 scores from the leaked Korean slide and combined it with scores from other notebooks for comparison:

Although the test is strictly focused on graphics performance, we can see the GPU by itself scores 1135 points in 3DMark 11, a 58 percent performance increase over Intel's HD 4000 Graphics. But why stop there? Add in a Radeon HD 7670M to take advantage of the Dual-GPU feature and hit 2083 3DMark 11 points. Not bad for an APU. Unfortunately, the slides lack CPU performance details or even non-synthetic benchmarks, but we'll see all that soon enough. Check out the Korean slides below, for all the leaked goodies.



Does anyone know if the Trinity GPU can overclock itself like the cores can?
Next gen, next year hopefully
Bad Day was talking about vram. It's video ram; usually built onto the graphics board which is usually much faster and higher performance than the ram you put on the motherboard. He seems to be wondering if companies might one day have something like that built right on the die (as part of the chip) rather than use the standard ram on the computer. It would probably be a small amount, but it would maybe run really fast with low latency. I think it's a fine concept.
The issue right now seems to be that there's less of a direct connection between an integrated graphics chip (makes for higher latency) and its associated memory than there would be on a discreet board which has everything built together with high end parts.
Ironically, you're the noob here.
I don't think that would actually help much. The majority of memory is spent storing massive textures, right?