• 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.

It looks like SiSoft might have fixed the file system bench in Sandra 2003

I got like 29000 on a 80gig 7200 rpm. Maybe is would have showed alittle faster if I restarted. Who knows. Seems about the speed it should be I guess.




Jason
 
Hmmn, I don't seem to be able to use the WebWizard update on my current Sandra to upgrade to 2003...
 
20 GB IBM 34GXP
Overall: 13382
Buff Read: 57
Seq Read: 19
Buff Write: 34
Seq Write: 18
Access Time: 10

60 GB IBM 60GXP
Overall: 16353
Buff Read: 27
Seq Read: 25
Buff Write: 16
Seq Write: 16
Access Time: 9

45 GB IBM 75GXP
Overall: 20040
Buff Read: 31
Seq Read: 29
Buff Write: 28
Seq Write: 24
Access Time: 7

Firewire 120 GB Western Digital WD1200 (Oxford 911 chipset)
Overall: 20252
Buff Read: 39
Seq Read: 29
Buff Write: 28
Seq Write: 28
Access Time: 9

How not to set up a computer by Eug:
34GXP on Promise ATA66 card as primary boot drive
60GXP and 75GXP I believe on ATA33 controller
WD1200 in a Firewire enclosure - fastest drive is usually off. 😛

It does show you the benefit of Firewire and the Oxford 911 chipset. I betcha no USB 2.0 drive will come close to this.
 
Of the six drives in my main system, the lowlevel measurements (access time/STR) weren't accurate for any of them. Sorry Sandra, try again.

Edit: I don't know what this test does exactly, but it involves creating a 767MB temp file on the tested drive which it appears to run the test out of. How accurate can a hard drive test be that can only test a small area of the drive where you have no data? I'd like to know either how fast the whole drive is or how fast it is where my data sits. This would probably also explain why all the lowlevel results are bogus (X15-36LP access time of 3ms? original X15 4ms? don't think so).
 
Edit: I don't know what this test does exactly, but it involves creating a 767MB temp file on the tested drive which it appears to run the test out of. How accurate can a hard drive test be that can only test a small area of the drive where you have no data? I'd like to know either how fast the whole drive is or how fast it is where my data sits. This would probably also explain why all the lowlevel results are bogus (X15-36LP access time of 3ms? original X15 4ms? don't think so).

I don't have a spare X15 to connect to a standard (non RAID) HBA to test the access time. I know the access time of the RAID10 array is about 1 mS quicker than a standard X15. HD Tach shows about 5mS on this array.

I agree it's basically a joke. If you look at the references it shows a RAID0 array with a pair of ATA100 disks with 8MB cache (why don't they just say WD1200JB?) outperforming a single 15k drive. Perhaps in STR but no way in I/O intensive tasks. I must say that the 8MB buffer WD's "feel" as fast as a single X15 on a standard U160 HBA. Two or more X15's on a decent RAID HBA (meaning NOT Adaptec!) with cache is no contest! The system still boots faster than IDE AND there is a 10 second delay where there is no drive activity when booting XP! I found this out by doing a trace with the bootvis utility. If there was a way to eliminate that delay, the system would get from the xp splash screen to press CTRL+ALT+DEL in a few seconds!

Cheers!
 
man.. you got 136 GB of space, and you have used only like 1.8 Gigs??? WTF are you using the system for anywayzz?????
 
man.. you got 136 GB of space, and you have used only like 1.8 Gigs??? WTF are you using the system for anywayzz?????

New array = delete all data. Everything gets moved to another logical drive or to a another SAN device. I can do that on a weekend.

This particular system has several technical tasks. That's all I am at liberty to disclose publicly!

Cheers!
 
Back
Top