Why is NT4 slow with U160 disks?
I threw together some SCSI perfomance benchmarks noting differences between past/current versions of Microsoft's premier OS. NT4 really shined with my Ultra2 disk, but struggled with the newer Ultra160 disks.
I threw together some SCSI perfomance benchmarks noting differences between past/current versions of Microsoft's premier OS. NT4 really shined with my Ultra2 disk, but struggled with the newer Ultra160 disks.
Because of NT4's poor performance with my Ultra160 disks, I was wondering if anyone can tell me if the Ultra160 standard was released before or after the final iteration of NT4 (i.e., SP6a)?
The link below shows comparisons between NT4, 2K, & XP. Because there is around 30 images (with close to 20 benchmarks shown), it will take a while to load. I apologize to the dial-up folk
Ron_Jeremy's SCSI Comparison
Because of NT4's poor performance with my Ultra160 disks, I was wondering if anyone can tell me if the Ultra160 standard was released before or after the final iteration of NT4 (i.e., SP6a)?
The link below shows comparisons between NT4, 2K, & XP. Because there is around 30 images (with close to 20 benchmarks shown), it will take a while to load. I apologize to the dial-up folk
Ron_Jeremy's SCSI Comparison
Participate on our website and join the conversation
This topic is archived. New comments cannot be posted and votes cannot be cast.
Responses to this topic
Okay, as for the slow XP benchmarks. It's a documented bug. It's supposedly going to be fixed in SP1. Basically it's ALOT slower than 2K in SCSI.
As for the NT4 benchmarks it could very well be a driver problem. Dunno what the drivers are dated at but as the years progress expect less and less NT4 performance. Just the way it 'tis I'm afraid.
It's ridiculous too because hardware is hardware. All a developer has to do is load up identical machines with different OS's and then develop the drivers. 98SE/ME/2K/XP? WDM. NT4 well..not WDM but the other 4 frickin' OS'S are and take 9x out of the question anyway for SCSI. So that leaves 2 OS's for WDM and 1 for not. Considering that NT4 is still HIGHLY used in corporate environments and that XP is not...well geesh. NT4 drivers should still be produced as well as their 2k counterparts...the only excuse is a terrible company.
Check out www.storagereview.com about the XP bug and ask around about your card.
*As for the Basic vs Dynamic: It's probably an MS Conspiracy. Every since they intro'd Dynamic in 2K people have still been using Basic for Compatibility and MS is probly doing as usual (making thing's bugging, slowing things down to push their new standard...even if it's not really needed)
As for the NT4 benchmarks it could very well be a driver problem. Dunno what the drivers are dated at but as the years progress expect less and less NT4 performance. Just the way it 'tis I'm afraid.
It's ridiculous too because hardware is hardware. All a developer has to do is load up identical machines with different OS's and then develop the drivers. 98SE/ME/2K/XP? WDM. NT4 well..not WDM but the other 4 frickin' OS'S are and take 9x out of the question anyway for SCSI. So that leaves 2 OS's for WDM and 1 for not. Considering that NT4 is still HIGHLY used in corporate environments and that XP is not...well geesh. NT4 drivers should still be produced as well as their 2k counterparts...the only excuse is a terrible company.
Check out www.storagereview.com about the XP bug and ask around about your card.
*As for the Basic vs Dynamic: It's probably an MS Conspiracy. Every since they intro'd Dynamic in 2K people have still been using Basic for Compatibility and MS is probly doing as usual (making thing's bugging, slowing things down to push their new standard...even if it's not really needed)
Quote:
Okay, as for the slow XP benchmarks. It's a documented bug. It's supposedly going to be fixed in SP1. Basically it's ALOT slower than 2K in SCSI.
DosFreak, I was really concerned with NT4's poor showing with my U160 disks. However, as for XP, when Dynamic disks were tested XP posted faster numbers with the U160 disks than 2K. Similar results with the U2 disk. XP was also faster than 2K utilizing software RAID0 too.
What puzzles me is the poor showing with Basic disks for XP, & poor U160 performance with NT4. I can understand maybe why NT4 struggled with a SCSI standard that was released after it's last Service Pack. But, why XP rocks with Dynamic disks & blows chunks with Basic disks is baffling to me. I mean, it's still the same physical disk. Basic disks are after all the industry "standard", while I am told Dynamic disks is proprietary Microsoft voodoo. I dunno.....
Looks like XP + Dynamic disks are the way to go?
Okay, as for the slow XP benchmarks. It's a documented bug. It's supposedly going to be fixed in SP1. Basically it's ALOT slower than 2K in SCSI.
DosFreak, I was really concerned with NT4's poor showing with my U160 disks. However, as for XP, when Dynamic disks were tested XP posted faster numbers with the U160 disks than 2K. Similar results with the U2 disk. XP was also faster than 2K utilizing software RAID0 too.
What puzzles me is the poor showing with Basic disks for XP, & poor U160 performance with NT4. I can understand maybe why NT4 struggled with a SCSI standard that was released after it's last Service Pack. But, why XP rocks with Dynamic disks & blows chunks with Basic disks is baffling to me. I mean, it's still the same physical disk. Basic disks are after all the industry "standard", while I am told Dynamic disks is proprietary Microsoft voodoo. I dunno.....
Looks like XP + Dynamic disks are the way to go?
