• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Nvidia vs Microsoft Driver: CPU Usage

G

gerbiaNem

2[H]4U
Joined
Mar 6, 2005
Messages
2,169
So I performed a very rudamentary test to see which driver used less cpu for the same exact file copy test. I basically copied my steam folder which contains a mix of large and small files from one sata drive to another. Then deleted the files created and redid the copy with a different driver.

Observations:

Nvidia Driver - CPU utilization between 13-17% on a dual-core...

Microsoft - CPU utilization between 3-8%...

I think the results speak for themselves. No copy was noticeably faster than the other.
 
Did you reboot between tests?

This would nix the effect of caching, as well as makign sure there were not any artifacts from the driver switch affecting the results.
 
Ok, so what's the point? Did you check the performance of each drivers which in my opinion will show that the nVidia driver adds NCQ support and full performance as in transfer rate and seek time. It's almost certain that nVidia uses some scheme to increase performance like the Intel IDE/SATA drivers do, for maximum performance under different scenarios.

Here, we do not have all the facts. If you can prove that the performance is the same, the feautures are the same then the cpu load will mean something.
 
Actually NCQ does jack squat for desktop application level performance. According to SR's measures, it raises throughput in one scenario on Maxtor drives. On Maxtor drives, two more of the tests break even with NCQ, while NCQ hurts performance on two more. On drives from other makers, NCQ is a total loss and does not increase performance in any of the measures.

Also, regardless of drive or maker, neither NCQ nor special drivers will improve the mechanical performance of the drive.
 
I tend to agree with you but without hard number, I am not talking about NCQ, on all the scenarios like best seek time, average seek time, transfer rate, multitask operations, command flooding and more, it makes no sense comparing cpu usage alone.

The mechanical part is surely important but like Intel has proven with their IDE drivers (especially in the recent versions), in some cases it really does make a difference, usually when HD is under high demand.

Show me comparable performance tests and then we'll have something.

(BTW, NCQ does make a difference for what its purpose is: http://www.anandtech.com/weblog/default.aspx?bid=195)
 
You must log in or register to reply here.
Back
Top