• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Sandra Drive Index - WTF????

M

metroplex

Golden Member
IBM Deskstar 75GXP, 75GB
UDMA 5 Mode
Windows 2000 SP-2

I get an index of roughly 19731.

66 MB/s burst read
blah blah

Does this sound right?

The reference drive in Sandra (ATA100/72000 RPM 30GB) got an index of 24,000.

Is it because my drive is bigger?
The 75GXP is 7200 RPM as well, with 2MB cache.
 
I've seen 10K RPM SCSI drives that Sandra scores under 20,000 and 5400RPM IDE drives w/12ms+ seek times that Sandra scores over 28,000.... SiSoft Sandra Disk benchmark is extremely poor. It varies dramatically between runs and it tends to be not even remotely consistent with reality.

 
yeah, HDtach is much better,

i'll post my scores in both in a minute (30GB 75GXP 7200RPM 2MB cache)

dave
 
HDTach is by no means perfect, bit it is fairly decent. At the very least it is consisten in it's results unlike Sandra.
One thing with HDTach, you MUST remember to enable "Advanced" Size Check" on order to get accurate results!
 
I have Win2k SP2 and HD Tach ran fine.

Why do I need the aadvanced size check feature?
I enabled it for my test:

IBM 75GXP 75GB
mounted on RAID IDE channel 3 (#1 for RAID)

Random access time: 18.2ms
Read Burst Speed: maxed out - no reading!! (it went off the scale)
Sequential Read: Max 38541, avg = 28639
7.7% CPU utilization

with the same drive on the IDE channel:
Random access time: 15 ms
Read Burst Speed: 67 MB/s
Sequential Read: Max 38563, avg = 29337
7.4% CPU utilization

In Sandra (lastest version)
With it mounted on the IDE, I got a index of 19731
Mounted on the RAID, I get 21405.
 


<< I have Win2k SP2 and HD Tach ran fine.

Why do I need the aadvanced size check feature?
I enabled it for my test:
. >>



If you disable advanced size checking then HDTach will ONLY measure performance across the first 10GB of the drive, hence strictly measuring performance on the inner tracks of the HDD which are by definition the fastest area of the drive and will also as a consequence measure seek times that are entirely unrepresentative of the drive as a whole.
 


<< hence strictly measuring performance on the inner tracks of the HDD which are by definition the fastest area of the drive >>


actually the outer edge of the hdd is the fastest part of the drive! 😱
 


<<

<< hence strictly measuring performance on the inner tracks of the HDD which are by definition the fastest area of the drive >>


actually the outer edge of the hdd is the fastest part of the drive! 😱 >>



True, my terminology was in error. I should have stated that it would measure performance from the beginning of the drive, and the first platter- which is the fastest portion of the drive.... though not the inner tracks as I inferred above.

In any case, for mentions of inaccuracies with testing drive performance with Advanced Size checking disabled you merely need to peruse Storage Review, their constantly ranting in their "Elsewhere" section about how many sites are forgetting to enable it and hence measuring inaccurate results.
GamePC and Tom's Hardware specifically have been guilty of this numerous times in their testing of IDE drives and as a consequence have inaccurately "measured" Disk Access that are below even the theoretical capabilities of a 7200RPM IDE drive. By measuring access over a small portion of the drive and misrepresenting it as overall access time they've on occasion amusingly obtained results that are not even theoretically possible given the spindle speed of the drives tested.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top