When my 512GB 950 Pro was new, I ran CrystalDiskMark tests first with the Microsoft driver and then the Samsung (NVMe v.1) driver:
Today, after installing v.2 of the Samsung NVMe driver I decided to re-test using CDM:
While running the CDM tests a second time, I wondered about possible thermal throttling and HWMonitor reported the SSD temp as 77*C. This was in an un-air conditioned room with an estimated ambient temp of around 30* or 31*C. (I don’t know what the starting SSD temp was since I didn’t think about temps until the second CDM tests were running.)
Then, I turned on the air con, let it run for 45 minutes, and took the computer case cover off. With an estimated ambient temp of around 25* or 26*C I re-ran the CDM test starting with a SSD temp of 60*C. I ran each test separately with a pause between them to let the SSD temp to return to 60*C because it ramped up to 74*C during Seq Q32T1 read test (even with ambient temp cooler):
I didn’t think that the throttling would be such a factor in my home PC installation even here in the tropics, but I guess it is? Or, is there some other factor in play?
The Seq write speed now is even better than back in August when the SSD was new. Do you suppose that was due to thermal throttling at the time?
FYI, I ran the CDM tests today twice without air con and twice with air con. Results within each pair were very similar. My motherboard has the m.2 slot on the underside so airflow is not good (and why I removed the case cover for the second pair of tests.)
PS: While composing this post, the SSD temp dropped to 58*C.

Today, after installing v.2 of the Samsung NVMe driver I decided to re-test using CDM:

While running the CDM tests a second time, I wondered about possible thermal throttling and HWMonitor reported the SSD temp as 77*C. This was in an un-air conditioned room with an estimated ambient temp of around 30* or 31*C. (I don’t know what the starting SSD temp was since I didn’t think about temps until the second CDM tests were running.)
Then, I turned on the air con, let it run for 45 minutes, and took the computer case cover off. With an estimated ambient temp of around 25* or 26*C I re-ran the CDM test starting with a SSD temp of 60*C. I ran each test separately with a pause between them to let the SSD temp to return to 60*C because it ramped up to 74*C during Seq Q32T1 read test (even with ambient temp cooler):

I didn’t think that the throttling would be such a factor in my home PC installation even here in the tropics, but I guess it is? Or, is there some other factor in play?
The Seq write speed now is even better than back in August when the SSD was new. Do you suppose that was due to thermal throttling at the time?
FYI, I ran the CDM tests today twice without air con and twice with air con. Results within each pair were very similar. My motherboard has the m.2 slot on the underside so airflow is not good (and why I removed the case cover for the second pair of tests.)
PS: While composing this post, the SSD temp dropped to 58*C.