Defiler

Terrible SCSI performance in Windows XP

Recommended Posts

The f*ck XP-SP1 has now my cool running , as Dynamic, X15-36LP

lowered to write 8,5MB/s.

Read is ok at 58MB/s

Sure ?

YES I AM SURE :evil: .

Dino

[rocketmanx]

> Microsofts already stated SP1 WON'T fix the problem.

[isochar]

> [sP1] does not fix this problem

Are you SURE?

Share this post


Link to post
Share on other sites
[rocketmanx]

> Microsofts already stated SP1 WON'T fix the problem.

[isochar]

> [sP1] does not fix this problem

Are you SURE?

From http://support.microsoft.com/default.aspx?.../SP1FixList.asp , the list of fixes in SP1, there is:

http://support.microsoft.com/default.aspx?...b;en-us;Q308219

> Hard Disk Performance Is Slower Than You Expect

> The information in this article applies to:

> Microsoft Windows XP Home Edition

> Microsoft Windows XP Professional

> SYMPTOMS

> After you install Microsoft Windows XP, hard disk performance may be slower than you expect. 

> NOTE: Hard disk performance may be even slower when your computer performs many small

> hard disk read/write operations. 

> CAUSE

> This behavior may occur in the following situation: 

> You use Small Computer System Interface (SCSI) hard disks in the computer.

> -and-

> The hard disks are formatted as NTFS.

...

> STATUS

> Microsoft has confirmed that this is a problem in the Microsoft products that are listed at the 

> beginning of this article. This problem was first corrected in Windows XP Service Pack 1.

That sure looks like the issue being described, doesn't it?

I have an XP Pro/X15-36LP/NTFS system (my Northwood), but I won't have physical access to it until Sept. 19, so I can't confirm this firsthand.

--

Stephan T. Lavavej

Share this post


Link to post
Share on other sites

[isochar]

> I'll forgive you since you don't frequent the forums often.

I haven't posted very much in the new forums, as I bought my new computer in January 2002 and haven't been interested very much in hardware since. That doesn't mean that I don't *frequent* the forums.

> As seen in a recent Maximum PC article, which can be seen in this thread, Microsoft has acknowledged

> the problem and know what it is. However,

>> a previous NTFS fix doesn't solve the problem, nor does the upcoming Service Pack 1.

I couldn't really care less about information from an old issue of a magazine. The fact is that SP-1 includes a fix for slow SCSI-NTFS performance, and no one has yet discussed it. I want to know if this is truly the fix for this problem, or not. Experimentation is the only answer.

Share this post


Link to post
Share on other sites
[isochar]

> I'll forgive you since you don't frequent the forums often.

I haven't posted very much in the new forums, as I bought my new computer in January 2002 and haven't been interested very much in hardware since. That doesn't mean that I don't *frequent* the forums.

>  As seen in a recent Maximum PC article, which can be seen in this thread, Microsoft has acknowledged

> the problem and know what it is. However,

>> a previous NTFS fix doesn't solve the problem, nor does the upcoming Service Pack 1.

I couldn't really care less about information from an old issue of a magazine. The fact is that SP-1 includes a fix for slow SCSI-NTFS performance, and no one has yet discussed it. I want to know if this is truly the fix for this problem, or not. Experimentation is the only answer.

The fix you are referring to is a hotfix that has been around for ages. Many of the affected readers of this thread have tried it and found it did not solve their problems. It's actually a similar, but seperate, problem. The article information you have found is saying that the old hotfix is now included in SP1.

Share this post


Link to post
Share on other sites
[isochar]

I couldn't really care less about information from an old issue of a magazine. The fact is that SP-1 includes a fix for slow SCSI-NTFS performance, and no one has yet discussed it. I want to know if this is truly the fix for this problem, or not. Experimentation is the only answer.

That is the August issue. With direct quotes from top-level Microsoft personnel, you definately can put weight behind the content. Question is if Microsoft will have any problems along the way and not be able to release the real fix in the next couple months.

Share this post


Link to post
Share on other sites
I couldn't really care less about information from an old issue of a magazine. The fact is that SP-1 includes a fix for slow SCSI-NTFS performance, and no one has yet discussed it. I want to know if this is truly the fix for this problem, or not. Experimentation is the only answer.

Well, I tried it. It didn't work. I posted my results on the tenth.

Share this post


Link to post
Share on other sites
The fact is that SP-1 includes a fix for slow SCSI-NTFS performance, and no one has yet discussed it. I want to know if this is truly the fix for this problem, or not. Experimentation is the only answer.

It does -not- fix the problem as others have pointed out. I tried it on 2 different machines and had almost identical results pre-SP1 and post-SP1.

Share this post


Link to post
Share on other sites

:(

This rocks.

bench.JPG

Note this is the Benchmark of my Smartarray 3200 with 1x18 gig seagate (15 krpm) and 1x36 gig ( 15 krpm ) in a 36 gig raid-0 set.

Notice the cpu utilization !

Tomorrow i will place this in a Compaq Tasksmart server and compare the benchmarks just to see how much VIA kills the SCSI performance.

I did run the disks on an 39160 and the 36 gig did alone around 60 mb/s in Winbench 99. ( in the same server )

Share this post


Link to post
Share on other sites

The SCSI slowdown on Win XP and Win2000 are due to a write catching problem in XP and 2000's code. You won't notice it in Win2000 unless you turn off write catching. Microsoft expects to release updates withing the next three months. Read "SCSI Slowdown" Maximim PC, Sept 2002, pg. 20 for more information.

Share this post


Link to post
Share on other sites
The SCSI slowdown on Win XP and Win2000 are due to a write catching problem in XP and 2000's code.  You won't notice it in Win2000 unless you turn off write catching.  Microsoft expects to release updates withing the next three months.  Read "SCSI Slowdown"  Maximim PC, Sept 2002, pg. 20 for more information.

Chris,

Is it possible to scan this article and place it here? Thanx

Share this post


Link to post
Share on other sites

I'll be getting my Adaptec 39160 and IBM Ultrastar 36Z15 15krpm drive in a few days and I am wondering which O/S I should install on it to get the maximum performance? Will win2k perform better or does it contain the bug as well?

Share this post


Link to post
Share on other sites
I'll be getting my Adaptec 39160 and IBM Ultrastar 36Z15 15krpm drive in a few days and I am wondering which O/S I should install on it to get the maximum performance? 

Linux.

Will win2k perform better or does it contain the bug as well?

Yes, it has the bug, but it is masked better. Some poeple have mentioned that SP3 is no good.

Share this post


Link to post
Share on other sites

Here is my ATTO benchmark performance with my IDE IBM Deskstart 120GXP under WinXP:

atto-ide-winxp.gif

and here is my ATTO benchmark performance with my new Adaptec 39160 and IBM Ultrastar 36Z15 18GB 15,000 RPM Drive under WinXP:

atto-scsi-winxp.gif

Is this a problem with ATTO because HDTach doesnt show this problem? Is there any fix? Does switching to 2000 help? If 2000 has the bug does that mean all RAID controllers in all servers running Windows 2000 Server have the problem? That would be insane.[/img]

Share this post


Link to post
Share on other sites

This threat is getting reeealy long now, does anyone know "a-quick-n-dirty" solution, if so - mind spelling it out ;)

I've got an Adaptec 19160 + Fujitsu 10K and XP and low write performance that I would like to get fixed.

Share this post


Link to post
Share on other sites
This threat is getting reeealy long now, does anyone know "a-quick-n-dirty" solution, if so - mind spelling it out ;)

I have the same card. the only solution is not to run XP. 2000 has some of the same problems, though they are masked better. New MS doesn't do fast SCSI...

There is always linux....

Share this post


Link to post
Share on other sites

So far in 7 pages, nearly 700 posts, and over 50,000 views we have come to the conclusion that nobody knows exactly what the problem is or how to fix it.

Microsoft has stated that they are working on a fix, but despite the fact that this problem cuts the performance of those with the SCSI bug by at least 50%, they seem content to take months to issue a fix.

So in other words, if you have the problem with XP, your only choices are to reformat with 2k, linux or something else or wait patiently with hardware performing like some cheap 5400 rpm IDE drive.

Share this post


Link to post
Share on other sites

Just a note microsoft is only claiming there was a bug with winxp and scsi and ntfs and supposedly got fixed at SP1 which we all know is bogus and we know the problem exists with win2k as well and with fat32 as well...

%5burl=http://support.microsoft.com/default.aspx?scid=kb;en-us;Q308219%5b/url%5d%5dhttp://support.microsoft.com/default.aspx?...s;Q308219" target="_blank">

]http://support.microsoft.com/default.aspx?...s;Q308219

Share this post


Link to post
Share on other sites

I’ve just gotten around to like the new start bar :D

No way I'll bother reinstalling a 3 year old OS for better write performance. With only 36 gigs in my machine I don't have that many large writes anyway.

Share this post


Link to post
Share on other sites

both small writes and large writes

more so with small writes

(i have the cida same problem with me IDE RAID 2x40gb 7200rpms no benifit from using the raid aray kt7a-raid writes was bad read was ok win2k)

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now