Game Load Time
We used the same settings for the game load test as our standard reviews on a desktop platform. Instead of compiling the average of three tests, we used the best score of three runs to show the Optane Memory advantage. The average time with the Optane Memory H10 was just over 34 seconds. The third pass, with the data in the Optane Memory cache, shaved 6 seconds off the H10 QLC only test.
These scores are still significantly higher than our desktop system. The difference comes from the video card, or in this case, the onboard video attached to the CPU. Our desktop system uses a discrete video card that loads the data much faster allowing the drive to be the bottleneck rather than the system.
PCMark 8 Total Storage Bandwidth
We expected stronger performance out of the 512GB 760p but remembered this drive had some issues with thermal throttling in our synthetic tests. The lower power 660p with QLC memory and a 4-channel controller didn't have the same issue, and neither did the Optane Memory H10.
SYSmark 2018 System Responsiveness and Power Tests
This is the first time we've used the SYSmark 2018 test. Part of the suite measures the system responsiveness. This is one area where Optane technology usually shows an advantage. We didn't see that in our tests where the default test runs three times back to back.