I am trying to understand the solution. You suggest that the problem was measurement errors from Amorphous DiskMark but you did eventually get normally results from it on that drive. Then it seems the drive worked better for random writes when you booted off it?
To sum up, I have two 990 Pros from different sources (Best Buy and Amazon). I first installed them as external drives using MacOS Sonoma, Sonoma was running on the internal Apple SSD. I ran Amorphous and got the low random writes. The 9100 tested normally.
I clean installed Catalina on the 9100 and booted from that. Fresh install of Amorphous, still low random writes on the 990's. 9100 still tested normally.
After ruling out the nvme adapter card,
@mattspace recommended a clean install of MacOS onto the 990 and try booting from that. Once I booted directly from the 990, Amorphous finally got normal results. But even when booting from one of the 990 Pros, only the boot drive would test normally. Amorphous would still give low random write results for the storage 990.
ATTO Benchmark measured all drives with expected results no matter the OS or which drive I was booting from. Maybe it's a formatting issue, I originally formatted one drive as APFS and the other as MacOS Journaled. Both tested the same. Again, this suggests something wrong with Amorphous if ATTO tested the drives as normal the whole time.
I've now settled on the 9100 as my main boot drive and put the 990's into use as storage. I've copied millions of tiny files to them and they both are writing extremely fast. So the solution for me is use ATTO Benchmark to begin with. So it makes me think that there was never anything wrong with the drives.
@mattspace I forgot to ask - what made you think to try booting off the 990 Pro? That definitely solved it, just curious how you came up with that.
