Sign in to follow this  
logicprobe

Is my cheetah crap on win xp pro??

Recommended Posts

Guest russofris

Looking at the atto graph, you are in decent shape. Now run the Winbench99 Disk winmark tests and compare the scores to those in the SR review database for your drive model.

If they are close, then you are running somewhere near peak performance. If not, then we'll continue troubleshooting.

Thank you for your time,

Frank Russo

Share this post


Link to post
Share on other sites

Do any of you guys have a 15K Cheetah and an IDE drive hooked up to the same computer? I'd be interested if taking a large file and doing a same disk copy shows similiar to mine that the SCSI drive chokes. Also, un'raring a huge file (possibly unzipping as well) burns rubber on the IDE compared to my SCSI.

Winbench coming.

Share this post


Link to post
Share on other sites

Ok, look at these sucky scores. The business score is aprox. half what Storage Review got, and the standard score is closer, but still significantly off.

Could any of this be related to the way my disk evolved? It went from a FAT32, then I converted to NTFS through XP.

WEIGHTED SUITE SCORE UNITS

Business Disk WinMark 99 7560 [1,2,4] Thousand Bytes/Sec

High-End Disk WinMark 99 26700 [1,4,5] Thousand Bytes/Sec

TEST SCORE UNITS

Disk Playback/Bus [1,2,3,4]

Overall 7560 Thousand Bytes/Sec

Disk Playback/HE [1,3,4,5]

Overall 26700 Thousand Bytes/Sec

AVS/Express 3.4 23300 Thousand Bytes/Sec

FrontPage 98 170000 Thousand Bytes/Sec

MicroStation SE 31600 Thousand Bytes/Sec

Photoshop 4.0 15100 Thousand Bytes/Sec

Premiere 4.2 21000 Thousand Bytes/Sec

Sound Forge 4.0 37000 Thousand Bytes/Sec

Visual C++ 5.0 24300 Thousand Bytes/Sec

Share this post


Link to post
Share on other sites

There is a very simple answer to your problem. Its not the test you have run or the way you have you computer setup at all.

As I noticed you are running Windows XP. To my knowledge there have been many complaints about people who blew alot of money on 15k rpm hard drives scsi mostly, and found out that 5400 rpm harddrive was whooping their ass.

The reason for this is there is a bug in windows xp similiar in concept but not in logistics to the bug in windows 95 where it is impossible to install on a system with a processor above 1000mhz etc... I have very vague information on this bug, but windows has addressed it and has said they will be coming out with a patch.

If you have the time or the want you should try running this under windows 2000 I am sure it would be blazin'!

Share this post


Link to post
Share on other sites

Flux, I guess we're just dancing around the issue that this is an XP bug. I put in the effort because some of the threads seemed to indicate that some improvement could be made, but apparantly very little.

Another thing is finding any microsoft information on when they plan to fix this. All I could find was that KB article about "slow scsi performance" where they refer to an SP1 fix and file ntfs.sys, which seems to not address the problem. I've also seen rumors back since October about a upcoming "fix", and now we're eating away into December.

Share this post


Link to post
Share on other sites
Guest russofris
There is a very simple answer to your problem.

Thank you for the attempt at helping logicprobe. While we are all aware of the mysterious Windows XP SCSI Boogie Man, it is best to begin troublshooting with the basics. It appears that a number of users are using XP and SCSI without any problem at all. Wouldn't you feel extremely bad if logicprobes problem was poor termination? It's way too easy to simply blame XP and be done with it... without actually attempting to fix or workaround the problem. Actually, It's just plain lazy. And please, easy on the bold face. We're not blind.

Back to the issue at hand....

It went from a FAT32, then I converted to NTFS through XP

Could be. I assume that you have defragged and that the disk not totally full? If not, please do so now and re-run the winmarks. Also post a Winbench99 "Disk Inspection Test" pic so that we can rule out re-mapped sectors and such.

Looking forward to getting this resolved,

Frank Russo

Share this post


Link to post
Share on other sites

I agree and I apologize for my quick remark, I was reading up and heard something of it before and read the forum and did not assume there was another problem that it could be, then again there are millions of problems.

Sorry about the bold I happen to have extremely horrible vision to the point where if I do not have on my glasses I literarly have to get only inches away from the screen. (btw I have a 21' inch monitor). Sometimes I forget people see the world different!

Share this post


Link to post
Share on other sites
Could be. I assume that you have defragged and that the disk not totally full? If not, please do so now and re-run the winmarks. Also post a Winbench99 "Disk Inspection Test" pic so that we can rule out re-mapped sectors and such.

Yeah, the drive is 25% full and 1% fragmentation.

Picture coming...

Share this post


Link to post
Share on other sites
It baffles me.  Imagine if a company dished out big bucks for a winxp server with 15K drives.  I guess nobody has complained because a winxp server is an oxymoron :-).

.Net Server 2003 is of course not released yet. Apparently RC2 will be/has been released recently. I wonder if the SCSI bug has been rectified in this? I only have RC1.

Anyway most companies will stick with NT4/Win2k, at the very least till the first SP comes out, perhaps mid/end 2003.

Share this post


Link to post
Share on other sites

Thanks, I tried the RC1 drivers, and no change. Oh well, I appreciate all the help.

I had wondered if I just discovered a real-world case where the WD drive with the 8 MB cache simply can outperform the scsi drive, but from what people are telling me, that shouldn't be so.

I'll just wait it out a couple more months, and if the MS doesn't release something, then I guess it's back to IDE for me.

Share this post


Link to post
Share on other sites
Guest russofris
I'll just wait it out a couple more months, and if the MS doesn't release something, then I guess it's back to IDE for me.

Have you considered a different OS? I do not know the role that your PC plays, but Win2K may not exhibit this behavior.

I will post if I think of anything else.

Thank you for your time,

Frank Russo

Share this post


Link to post
Share on other sites

Logicprobe, write:

Do any of you guys have a 15K Cheetah and an IDE drive hooked up to the same computer? I'd be interested if taking a large file and doing a same disk copy shows similiar to mine that the SCSI drive chokes.

I have the same as you configuration instead of Via KT333 chipset on Gigabyte 7VRA MB. Also I have Adaptec 19160 SCSI adapter (310 BIOS, latest & previous adaptec drivers) with Seagate 15K 18Gb ST318452LW drive(15% full) attached to 16bit wide SCSI bus and IDE IBM 7200 80Gb 40Gb-platter “Vanquver” (80% full) drive, 2x512MB DDR 3000CL=2 Corsair memory, Athlon 2100XP CPU, RadeOn 8500 video. To narrow (8bit) bus of adapter have attached IOMEGA ZIP drive and TEAC CD-RW.

Two same WInXP pro SP1 system installed on bootable SCSI and IDE disks. Choosing of boot need to be made from BIOS. At first SCSI drive was NTFS formatted and IDE – FAT32. Now both drives have FAT32 partition. SCSI adapter is in PCI2 slot with separate IRQ. In adapter’s BIOS: “domain validation” – dis., “write back cache” – en. for devices “00” and “07”, “parity” – en., “bus scan” – en. only for connected devices. In OS “enable write cache” is checked for both drives. After SP1 have installed last VIA 4in1 drivers, 020B latency patch and last VIA raid patch.

Both disks – basic, don’t want to go for the dynamic. And I have system swap page file – disabled. So detailed description made to eliminate further questions – it’s terribly hard for me to write in English.

ATTO results close to yours for basics disks – “writes” ~ 13.5Mb/s and “reads” ~ 58.5Mb/s

Tried to change SP1 WinXP scsiport.sys to same one from SP2 W2K but with same results.

Want to discuss some “real life” tests. Copying (using FAR file manager) 1Gb (1073741824bite) DVD.vob file from C:! to C:, D:! to D:, C: to D: and D: to C: - booting in turn from SCSI (C:) and IDE (D:).

1. SCSI boot.

C:! to C: = 3m04s (1024M/184s~5.6Mb/s)

D:! to D: = 2m25s (~7.1Mb/s)

D: to C: = 23s (~44.5Mb/s)

C: to D: = 20s (~51.2Mb/s)

2. EIDE boot.

C:! to C: = 3m02s (~5.6Mb/s)

D:! to D: = 1m59s (~8.6Mb/s)

D: to C: = 39s (~26.3Mb/s)

C: to D: = 46s (~22.3Mb/s)

So what we can deduce from this. I will don’t take into account memory paging, swaping (have LargeSystemCache – disabled). We have terrible read-write SCSI performance ONLY when copying the file on SCSI disk (think ought to be ~ 14Mb/s for boot SCSI drive and ~ 11mb/s for non booting SCSI drive). And compare disk to disk results for file transfer for SCSI and IDE boot

we don’t see any drops in read-write SCSI performance under WinXPpro. It will be very interesting to have same file transfer results from one SCSI 15K disk to another SCSI 15K disk attached to two separate same SCSI adapters or in worst case to two independent channels of one adapter. So we have problems with SCSI file management in WinXP and to know made MS repair or not we only need to copy 1Gb file from some folder to the root of SCSI drive. And if we will have time of copying more then 1.5min for 15K disk – problem not fixed.

Share this post


Link to post
Share on other sites

wow thats a lot of information. I recently bought myself SCSI stuff to install in my system. I knew about the problems with WinXP, and was fully prepared for them. Here are the results I got with my Cheetah 34GB 10k fitted to a 19160 / and my older trusty Samsung (ol' faithful) 13GB 5400 rpm in ATTO Disk Benchmark

Cheetah:

1024 - Read - 40 mb/s

1024 - Write - 11 mb/s

Samsung:

1024 - Read - 22mb/s

1024 - Write - 22mb/s

I can appreciate the slow write speeds, but shouldn't it be reading faster? Is it cause my drives only 10k?[/img]

Share this post


Link to post
Share on other sites

I want to add my SCSI drive results test with Winbench99 - Disk plaback~15000 Kb/s and HE ~ 37000Kb/s. With installed program files of winbench on IDE drive/

Share this post


Link to post
Share on other sites

OK here's the results:

TYAN TigerMP

2* AthlonXP 1700

512MB reg ecc

2*X15-36LPs (18GB) in SCA enclosure, both drives on one channel

2*WD1200AB (120GB), both on secondary IDE channel

Tekram U3D controller

WinXP Pro SP1

Copy 1GB .vob, using Windows Explorer

X15-36LP -> X15-36LP (boot drive) 25 secs

X15-36LP ->WD1200AB 35 secs

WD1200AB->WD1200AB 50 secs

WD->X15-36LP 30 secs

WD to itself 1m 42 secs

X15-36LP to itself 2m 19 secs.

Share this post


Link to post
Share on other sites

Could someone interpret the last few posts for me? What is this telling us?

Pradeep doesn't seem to have any problem, though he's running a super high end system! I don't know how that helps determine if others are seeing very poor disk to disk copies for SCSI drives compared to IDE.

Share this post


Link to post
Share on other sites
Guest russofris

The results above show that X15's under XP have extreme difficulty copying files to themselves.

Frank

Share this post


Link to post
Share on other sites
The results above show that X15's under XP have extreme difficulty copying files to themselves.

Frank

That's an interesting observation. Especially when some testing I did here seems to indicate XP has better disk-to-self copy performance. I'd be interested to see anybody's results in Win2k to compare this to.

Share this post


Link to post
Share on other sites

Pradeep,

Your results confirm what I have from my setup. More over file transfer from SCSI to SCSI disk attached to one channel of your adapter don’t have ANY performance hit. We have only drop of WRITE performance copying file from some folder on SCSI disk to same SCSI drive. Of course we need to reboot after each file transfer to clear memory and swap file (tweak in registry, but better to switch off system swapping). We always have worse results of file transfer (IDE or SCSI) made on one drive, but in our case with extreme bad results for SCSI under WinXP. We need to try the same under W2000 SP2 where peoples feel themselves good testing with ATTO.

Logicprobe,

Pradeep have problem with his setup the same as we. Your idea to test the same dvd.vob copying under W2K is good (please do this with SP2 and SP3). We will wait for your results.

Chew,

Your results are seems for raid setup. But now we need to simplify picture - because raid is too many dependable factors task to analyse. And please don’t be so ironic – anyone on each home or work PC can test one disk file transfer (disk-to-self if I understand correct) and find out that results will be 3 times worse then disk to disk file transfer under any windows operation systems. But for SCSI under WinXP we have 5 times worse results. And this is the problem.

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
Sign in to follow this