• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

what are some good hard drive benchmarks?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

skidooosl

Member
Joined
Sep 9, 2003
Location
MI
i need some good hard drive benchmarks, i have sandra... i just cant understand why when i ran it last i had like 35000 and when i ran it after i put the system together i had just over 40000... why did i lose 5000? i just wanted to know if there are any better hard disk benchmarks i could try. my disks are the ones in my sig.
 

Sonny

Senior TIFOSI
Joined
Aug 3, 2001
Location
VARIANTE ASCARI
skidooosl said:
i need some good hard drive benchmarks, i have sandra... i just cant understand why when i ran it last i had like 35000 and when i ran it after i put the system together i had just over 40000... why did i lose 5000? i just wanted to know if there are any better hard disk benchmarks i could try. my disks are the ones in my sig.

In THIS THREAD where we posted our RAID/HDD setup scores with ATTO as our benchmarking tool. Sandra & PCMark are useless IMNSHO since they can not produce reliable scores.
 

FTC

Member
Joined
Mar 3, 2002
Location
Barcelona, Spain
Hi,

Let me explain a bit on what you see with Sandra. Sandra is testing the *logical* disk performance, that is, it tries to perform a certain number of *file lever* operations and measures the time it takes... and as such it is heavily affected by things such as which partition is being used to do the test (closer or farther from the fast disk areas, and how big), what is the partition fragmentation level (implying more or less disk head moves), how much full/empty is the partition (meaning the empty space is 'near' or 'far away'), and also for how long you have been working with the system (since some file systems have a circular algorythm for finding empty space).

So, it has been my finding that if you keep these factors contant, you can get reliable results from sandra or such programs. For instance, if you run the test against the same partition, with the same amount of free space (aprox), right after defragging and rebooting, sandra will not deceive you. (beware also of other things that may affect the test such as swapping activity,...)

... But of course, it's always easier to use ATTO or such if you really want to measure your disk's RAW speed and NOT your current system file access speed. This programs *look* like giving more consistent results because they always measure constant things such as 'the time it takes to go to track nnn from track 000, read 1 Mb and transfer it to main memory'. Of course, such direct hard disk tasks are more 'predictable', but do they reflect your system's perfomance ? or just hardware level performance ?.




Hope this helps
Regards
FTC
 
OP
S

skidooosl

Member
Joined
Sep 9, 2003
Location
MI
i fixed the sig... i didnt realize they had to be that small on this forum... where can i download ATTO from? also is HD Tach any good?