Ahead-looking: Intel’s upcoming Arc flagship has made its first look in a web based benchmark database. An unnamed Intel companion took a GPU engineering pattern for a whirl within the Geekbench OpenCL check suite utilizing a barely dated i5-9600K validation platform and this is what we discovered…
In outright numbers, the GPU does poorly. The 4 runs achieved a median OpenCL rating of about 67,000 factors and its highest rating was 69,000 factors. For comparability, the AMD Radeon RX 6800 XT releases a few yr in the past sits within the 150,000 to 180,000 level ballpark.
However don’t be discouraged. Think about this: in response to the metadata collected by Geekbench throughout the runs, the GPU was hovering at round 1.3 GHz. At that frequency, with 512 EUs, the GPU has a theoretical 11 TFLOPS of efficiency.
AMD’s Radeon RX 6600 XT has comparable 11 TFLOPS and scores within the 60,000 to 70,000 level vary. Therefore, Intel’s structure has the identical per-TFLOPS efficiency in OpenCL as AMD’s does.
So the query is: can Intel match AMD’s high-end TFLOPS numbers?
Based on Geekbench, the GPU had a rated clock pace of two.1 GHz. If it had run at that pace, and never at 1.3 GHz, then it could’ve had 17 TFLOPS, which is on par with the RX 6800 (non-XT). So the reply to that query is, sure, if 2.1 GHz is achievable.
And albeit, 2.1 GHz isn’t that prime. If the rumors are true and Intel is utilizing the TSMC N6 node, then they might probably push the clocks excessive sufficient to interrupt into RX 6800 XT and RTX 3080 efficiency territory.
Do not forget that that is pure conjecture, however we’re not the primary to assert that Intel’s GPUs will land someplace alongside the RTX 3070 or RTX 3080. Intel themselves hinted that their flagship might be priced a contact above the latter at $740 to $790. Nonetheless, as immediately’s OpenCL scores present, Intel should still be arduous at work to get there.