• 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.

WARNING! for people with IBM 75gxp and WinME. (and 98se)

Page 6 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
took me a day to do the low level format and install win98se but i would've done it anyway...winme turned out to be more headache than its worth

all is back to normal w/ win98se
 
I'm an interested party since I use Win98 occassionally and I just bought a 75GXP on Friday and should receive it tomorrow). Here are the facts as I understand them in an attempt to summarize this thread :

Problem summary:

This affects Win98, Win98SE and WinME - possibly Win95 - but not WinNT or Win2k and not Linux.

It does not specifically affect only IBM 75GXP drives, but any drive on the ATA100 interface with a large (2MB or more) local cache and a really fast processor >733MHz or >933MHz depending on who you talk to.

It is caused by the fact that on really fast CPU's with really fast interfaces (ATA100) these OS's shut down the computer and turn off the power without waiting for the cache on the drive to be written back to the drive. When the power is turned off, the cache is lost and so updated data may not be written. So this ends up being the equivalent of shutting down Windows while it is running by powering down hard (with the power supply switch).

It should not cause a drive failure (as in a mechanical failure) since the drives should still autopark correctly. But there is a potential data corruption issue.

Microsoft and IBM are aware of the issue.

Solution:
There are several ways around it that all involve prolonging Win98's shutdown time. You can play a long .wav file, you can turn off power management (so you'll get the "It's safe to turn off your computer now" message), you can download a patch or you can hack the registry.

Other Issues
This is a completely different issue from the problems of running IBM 75GXP's from a Highpoint ATA100 controller. There is apparently a well known issue involving IBM 75GXP's running in ATA100 on the HighPoint controller. The summary of this issue is: use another HD other than the 75GXP, use another controller other than the HighPoint, or run the Highpoint in ATA33 mode.


I pieced this together over the 100 postings. Correct me if I'm wrong on any of these points.
 
Thats the way I see it. A problem thats not really a problem if you shut down manually. The "unsupported" patch seems to do more harm than good.

Here's a question that hasnt been asked. Why arent SCSI drives affected? I would guess its because of the interface but if MS had a instruction line written into Windows to wait for the cache to be purged, wouldn't they have included it with IDE support also?

 
Hey pm,
Thanks for the great summary...
I think you made great sense out of a big mess that had a lot of people (including myself) a little bit worried. I just hope there is a definitive fix for this issue sometime soon.
 
Trishkin,

I defrag Win Me on a regular basis using Norton Utilites 2000 (with updates). Never had a problem with my ten gig IBM hard drives. Always make sure the NU program you use says it is for Win98 SE or later. Trying to use NU for Win95 will destroy your Win98 SE and ME installation.

I use two NU programs. DiskDoctor and SpeedDisk.

Every now and then it is a good idea to run NU WinDoctor to clean up your Registry. Run it whenever you install a new program or uninstall an old program. First time you run WinDoctor you will be surprised at all the problems it finds in the Registry and will fix for you.

A friend of mine recently bought a new computer. First time he ran WinDoctor it found over 100 problems in the Registry that it fixed easily. His computer then ran faster and a few problems disappeared.

JohnL

 
There is a problem with Win98 and any flavor of ATA100 controllers that has absolutely nothing to do with drives connected to it. Micron has a patch, and I'm sure somethings available from Microsoft as well. If you don't use the patch with your ATA100 hardware and Win98, your guaranteed problems. Only took me three days for four brand new Micron's to come down with hard drive problems. Replacement drives did the same thing and they weren't IBM's.

Granted there may be other issues with high speed drives, but this one I can personally attest to.
 
What exactly is the problem with the 75GXP and HighPoint ATA100 controllers? I just want to know if I should look to check for a problem or not. While my 75GXP was running, I didn't notice any problems unless I was unaware I had them in the first place.
 
I've read this entire thread but it's still very unclear.

Does this only apply to OVER 733 systems? I'm running a 733 now so it would be nice to escape the problem (barely) having just ordered a new 45 gigger...

What if you're using the drive for data only and not running your OS on it? It's strange that IBM would let such a bug get loose in their drives.

I've just hacked my registry to slowdown my shutdown as sometimes I had a problem where my drives would start crunching and then just shut off seemingly before the time was right. I also have had to run scandisk too often.

BTW to all you people having lockups, watch out for heat problems... I have to leave my case cover off because my case doesn't keep cool enough for my GeForce. Blah!
 
Well, I'm running (was running) on a P2 400, Win ME with the 30gig 75GXP on UDMA 33, and the exact thing as described in the first post happened to me. I didn't see any signs of damage until I suddenly heard the clunking on bootup. Thankfully, my other HDD salvaged the data and I backed it up before sending the 75GXP back for a replacement.

So this problem certainly doesn't only apply to ATA100 and 733/933Mhz processors. I recommend anyone using WinME with a fast HDD, at whatever CPU speed or ATA level, use a different OS. If your processor/memory is good enough, go for Win2000.:frown:
 
just wondering about this cache flushing thing. if i turn off all my applications and save all my data. then let my cmoputer sit idle for like 5 seconds before shutting the computer down, should the problem still arise?

i'm running windows98se.... so far after about 1 month no problems with corrupted drives or anything. my chip is a 533 celeron @ 800... ibm75gxp drive (running at UDMA66). so it meets to over 733mhz and the large cache cautions.

thanks
 
I think I have the same problem but possibly not.

When my hard disk is not being accessed at all I can hear relatively high pitch loud noises coming from the PC - either hard drive or PSU (not certain). This only started occurring after installation of Win ME. If it is the HDD it could be the same problem others here are describing. Also, it occurs more when case temperatures are up.

However, my PC is not particularly fast: PII 350 MHz, and HDD is a Seagate 20 GB 7200 rpm UDMA 66 (connected to a UDMA 33 controller).

Perhaps the problem is with Win ME in that the OS is much faster at shutting down than previous versions due to lack of Dos or something? Anyway, I will try manual shutdowns to see if it stops the noise and I'll check back here later for any progress which has been made by people.

Thanks Ulkesh for telling me about this thread - very interesting.
 
I have seen this question asked many times so I am going to asked it again too, will this problem affect a 75gxp running at ata66 or slower?
 
I hope I'm not stating the obvious here, because this wasn't obvious to me as I skimmed through this thread.

Three possible solutions have been posted here:
- A patch from Micron (MEata100.exe)
- A patch from IBM (IBMPatch.zip)
- A link to Microsoft's Knowledge Base article Q273017

I just want to say that these are all the same solutions. If you extract the Micron patch and the IBM patch, the files are the exact same size, version, etc. The IBM patch is called "273017USAM.EXE", which is the same number as Microsoft's Knowledge Base article.

Q2730137 refers to a file call ifsmgr.vxd that must be present for the registry key to work. This did not exist on my WinME system until I applied the patch. Applying the patch added the ifsmgr.vxd file, and set the registry key.

Searching the internet, I found that this patch is in "test" status, and that Microsoft does plan to release the patch when it is finalized. A search in dejanews for "Q273017" will give a few hits on what some Microsoft support people have said about the patch.

Only problem is, this seems more like a workaround rather than a proper fix. It delays the shutdown so the cache can be written. It still does not verify that the cache was written before shutting down.

Sorry if I'm stating the obvious, but maybe this will clear up at least a little of the confusion.
 
I went to disabled the fast shutdown, and it was already disabled for some reason.. oh well.

I also talked to MS and all their knowledgable tech support could sudgest was to disable fast shutdown, and he only said that was a good idea after I mentioned it to him.

Ha ha HA.
 
I applied Micron patch to my Windows 98 SE. So far no problems, just takes a little longer to shut down my computer now, which perfectly makes sense
 
I just wanted to thank everyone for posting--I recently bought a 30GB 75GXP--haven't installed it yet.

Hopefully this(thread) will help others avoid the problems you guys had.
 
did you use the micron patch right after a fresh installation of windows98? or is this patch applied on an existing windows98se in the computer? thanks.
 
I applied patch to existing installation of Windows98SE (it would be too painful for me to reinstall every single program in Windows)
 
I have found something interesting after installing the Win98SE patch. There are some files, to make changes in the registry, in this directory:
C:\WINDOWS\INF\QFE\273017

They are called:
Delay2.reg
Delay4.reg
Delay8.reg
Nodelay.reg

No prizes for guessing what they do 😉
From what I have seen checking through the registry, "Delay2.reg" is the default shut down speed for the patch. Looks like we are beta testing with this patch.

Overclockers Australia Forum
 
as many have suggested looking at this page:

http://support.microsoft.com/support/kb/articles/q273/0/17.asp

i was looking at the version of ifsmgr.vxd they are referecing to. it says version:

"The English version of this workaround should have the following file attributes or later:

Date Time Version Size File name Operating System
--------------------------------------------------------------------
9/21/2000 2:46p 4.10.2225 186,438 Ifsmgr.vxd Windows 98 Second
Edition
9/21/2000 2:46p 4.10.3002 185,910 Ifsmgr.vxd Windows Millennium
Edition
"

however, in Windows98SE taht i am using, my version of ifsmgr.vxd is 4.10.2223

after reading that sentence "The English version of this workaround should have the following file attributes or later: ", does this shutdown thing even apply to me since i'm using an older version of the file? i would assume that it does, but the wording of the articles leaves me wondering.

for now, i'm just going to install the micron patch..... if that doesnt work too well, i guess i gotta get Win2k. i really don't want to reinstall OS's....... :frown:
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top