When it comes to graphics processing, Intel has been known for…integrated. Meaning the bottom-of-the-line performance, what’s needed when you can’t afford to game on anything except that three year old laptop. Even with steady improvements to integrated graphics, there’s no comparison to the power of a dedicated card.
Which is why Intel’s still unconfirmed DG2 (“Discrete Graphics” series 2; series 1 hasn’t seen a public release) is an interesting development. The longstanding battle between NVIDIA and AMD for graphical supremacy may see a new warrior enter the ring, and not just in the lighter classes.
WCCFTech, reporting on documents found by known tech leaker Komachi (originally spotted by Hardwareluxx.de) describes some of the technical aspects of the GPU architecture. WCCF noticed a few interesting tidbits, including variants on the drivers that seem to indicate high EU (Execution Unit) counts.
Think of EUs as NVIDIA’s Streaming Multiprocessors; the more you have, the more powerful the GPU is. As WCCF puts it so eloquently:
Three parts were mentioned with 128, 256 and 512 EUs (these could also be taken as bus widths, but previous GPUs used this number to refer to the EU count, not bus width). Considering that DG1 with 96 EUs sits at around 2-3 TFLOPs, a 128 EU chip could end up around 4-5 TFLOPs with 256 EUs offering around 5-10 TFLOPs and 512 EUs offering around 10-15 TFLOPs of FP32 compute output if clocks scale really well on the higher-end GPUs.
Which…is pretty impressive. If these numbers are accurate, we may be looking at possible mid- to high-range graphics cards. A full range of options for both workstations and consumers is a surprising move. If Intel pushes this selection out to the public, combining that performance with existing EUs from integrated graphics may be huge.