• 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.

Fujitsu 36GB MAU Benchmarks

E

EnderW

[H]F Junkie
2FA
Joined
Sep 25, 2003
Messages
11,459
Finally got around to hooking up my new Fujitsu 36GB MAU SCSI drive, ran a quick test with HD Tach.
How does this look?

hdtachmaud79.jpg


And what else should I test it with?
 
Looks about right. I suggest testing it with... whatever software you use that's I/O intensive ;)

With my MAS, it's fairly flat at 80MB/sec 'til about halfway, then trails off to end around 62MB/sec. Burst speed 102.2MB/sec, CPU utilization 0% +/- 2%, random access 5.5ms, average read 74.4MB/sec

Windows bootup time is a good test (and you'll wipe the floor with RAID-0 Raptor setups)
 
dandragonrage said:
Looks about right. I suggest testing it with... whatever software you use that's I/O intensive ;)

With my MAS, it's fairly flat at 80MB/sec 'til about halfway, then trails off to end around 62MB/sec. Burst speed 102.2MB/sec, CPU utilization 0% +/- 2%, random access 5.5ms, average read 74.4MB/sec

Windows bootup time is a good test (and you'll wipe the floor with RAID-0 Raptor setups)
Click to expand...

I run the 36GB Fujitsu MAS and my bootup time is atrocious =\
 
I changed my boot drive from a diamondmax 9 to a wd740 raptor, and I actually think that the raptor is slower to boot but its loading all apps in windows much faster ... :confused:
 
I think that's an issue with the NF4, my NF4 system takes much longer to boot than my old NF2 system, Methinks there's a driver for the NF4 that takes longer to load than it should.
 
Zorn said:
I run the 36GB Fujitsu MAS and my bootup time is atrocious =\
Click to expand...

That's probably due to the SCSI selection timeout. When a controller is going out on the bus to see what devices are out there it tries to talk to all the 15 possible targets. The way it does this is by trying to select the device and send a command to it. If there's not a device at that particular device ID, the controller waits a full SCSI Selection Timeout period for the device to respond (or not respond, I guess). That's 250ms per device ID that doesn't have something attached.

If you only have one target drive on your SCSI controller and it's a two channel controller, you'll have to sit through 29 selection timeouts, which is 7.25 seconds of nothing. Plus there's all the other initialization that the SCSI bios has to do. On top of this, some OSes go out and do their own bus scan for devices after the BIOS has done one, which easily doubles the waiting time.

One way around this is to disable SCSI channels if there are no devices on them and on the channels with devices, go into the SCSI BIOS and disable the device IDs which don't have any targets attached to them. I know this is possible with LSI BIOSes, but not sure about others.
 
Zorn said:
I run the 36GB Fujitsu MAS and my bootup time is atrocious =\
Click to expand...

Any system with SCSI will take a lot longer to boot than one without. That has to do with the Windows XP SCSI drivers that load during boot up. Even if you are not booting of the SCSI drive and just have the card installed it will slow down your boot. I had a system that was booting off IDE but had a SCSI adapter in for other purposes. Removed the card and the bootup time was around 5-7 seconds quicker.
 
UICompE02 said:
That's probably due to the SCSI selection timeout. When a controller is going out on the bus to see what devices are out there it tries to talk to all the 15 possible targets. The way it does this is by trying to select the device and send a command to it. If there's not a device at that particular device ID, the controller waits a full SCSI Selection Timeout period for the device to respond (or not respond, I guess). That's 250ms per device ID that doesn't have something attached.

If you only have one target drive on your SCSI controller and it's a two channel controller, you'll have to sit through 29 selection timeouts, which is 7.25 seconds of nothing. Plus there's all the other initialization that the SCSI bios has to do. On top of this, some OSes go out and do their own bus scan for devices after the BIOS has done one, which easily doubles the waiting time.

One way around this is to disable SCSI channels if there are no devices on them and on the channels with devices, go into the SCSI BIOS and disable the device IDs which don't have any targets attached to them. I know this is possible with LSI BIOSes, but not sure about others.
Click to expand...

Any idea if this is possible and how with Adaptec SCSI BIOS?
 
You must log in or register to reply here.
Back
Top