Promise IDE controllers do NOT improve performance...

[Madcowz]

in fact they decrease performance... at least in my case. You see, I was doing some benchmarking & came across an interesting discovery. I have two identical 5400RPM ATA100 hard drives & I used the latest version of Sandra to benchmark these two drives. One of them is on a Promise Ultra100 controller, & the other one is onboard (ATA100) my CUSL2 motherboard. I fire up the drive benchmark and came up with interesting results:
Ultra 100 controller - Drive index 11688
onboard - Drive index 13698

What gives? I thought the Promise controller is supposed to boost your hard drives performance? Of course I don't expect the "40% faster than onboard ATA100" results they tell you on the box, but people on the forums confirmed that it is indeed faster, and even more so with the latest drivers. Well I'm running the latest drivers & bios for both the controller and motherboard, & I don't see why I'm getting bad results. Any ideas?

 

[bigbootydaddy]

is the OS on the hdd that is on the promise?

 

[Noriaki]

Actually I always thought the promise was one of the slower controllers.

HighPoint and Via both perform better to (not much...not like you are getting), and I'm sure Intel would as well.

But not to the extent you are getting.

Just becuase two drivers are the same size & model doesn't make it an even test. I would test both controllers with the same drive.

Hell look at the 45gig 75gxp, they aren't even all made in the same country.

Also data quantity and fragmentation could have an effect (depending on how exactly Sandra tests)

Is the OS or Virtual memory on either of these drives?

 

[cbuchach]

Yeah, you need to bench the same drive, with the same amount of data on it in order to make a fair comparison. As I am sure you know, it makes a big difference where on the platter the data is being read from.

 

[Madcowz]

Cool, you guys are probably right. I'll go try testing the same drive on both!