rec.autos.simulators

Help! T-bird hard drive problems

Alison Hin

Help! T-bird hard drive problems

by Alison Hin » Thu, 03 Oct 2002 00:45:35

I recently upgraded the BIOS in my main racing computer, a T-bird 1.3 in
an Abit KT7, so I could use the 1.3 at full speed (the original BIOS
only went to 1.2).

However, now the machine is screwed up.  Hard drive writing is
excruciatingly slow (15 seconds to write 9 mb!) and brings the machine
to its knees all the time it's writing.  

So every time GPL writes out some replay bits or anything to virtual
memory, the screen freezes, the controls stop working, and all Internet
communication stops.  Sometimes this lasts for several seconds.  It's
dreadful!  If I save a replay after a race, it takes so long (and kills
communications so comprehensively) that the server typically disconnects
me.

Any suggestions are welcome.  I've done just about everything I can
think of, including:

- installed the latest VIA 4 in 1 drivers and the 1.10 USB driver
- installed the George Breeze PCI latency patch
- disabled serial port 2 to free up an IRQ
- removed both USB controllers from the Device Manager to let Windows
find them again
- set the jumper on the hard drive to cable select
- disabled PCI#2 Access Retry in the BIOS
- enabled Fast Write in the BIOS
- set PNP OS to No in the BIOS
- cut GPL replay length to 20 mb

Nothing helped.

One thing I haven't tried is resetting the BIOS using the jumper on the
motherboard.  Also, oddly, Windows Device Manager thinks I have two USB
controllers and two USB Root devices.  I don't know if this is a
problem, but I don't remember it being this way before the BIOS upgrade.

Any suggestions?  I'm at my wits' end, and can't see too many other
options short of reformatting and reinstalling Windows - and I hate
doing that!  

I would be terribly grateful for any suggestions that might help me fix
this machine and get it running as well as I know it should!

Computer specs:

Abit KT7
Tbird 1.3 Ghz
256 mb PC133 memory
3Com Ethernet card
Voodoo5
SB Live! Value
Logitech Wingman FF
CH Pro USB Pedals
Act Labs GPL USB Shifter
Windows 98

Alison



Remove the spam blocker NOSPAM to email me.
http://www.racesimcentral.net/

Ian Riche

Help! T-bird hard drive problems

by Ian Riche » Thu, 03 Oct 2002 00:58:04



<rest snipped>

Is DMA enabled for the hard drive?

Ian
--
Ian Riches
GPL Rank +1.76 Monsters of GPL +284.19

frederickso

Help! T-bird hard drive problems

by frederickso » Thu, 03 Oct 2002 01:19:56

I dont know specifically but here is the KT7 FAQ...

http://www.sudhian.com/faqs.cfm/fid/2.sud

GTX_SlotCa

Help! T-bird hard drive problems

by GTX_SlotCa » Thu, 03 Oct 2002 01:53:43

Hi Neighbor,

I hardly ever do it myself, but they do recommend resetting the bios after
you flash it. Probably won't help, but with all the time you've got in this
problem already, what's a couple more minutes? If you haven't done it
before, turn the computer off, move the jumper for a second and put it back.
You probably remember all your bios settings anyway.

I had this happen on my kid's computer. I think it took a bios flash to fix
it.

Gary

--
Slot

Tweaks & Reviews
www.slottweak.com

massivechicke

Help! T-bird hard drive problems

by massivechicke » Thu, 03 Oct 2002 13:47:57

I had the same problem, turned out it was indeed that DMA was not turned on
in the device manager for the hard drives. Not sure why it turned itself
off.




> >I recently upgraded the BIOS in my main racing computer, a T-bird 1.3 in
> >an Abit KT7, so I could use the 1.3 at full speed (the original BIOS
> >only went to 1.2).

> >However, now the machine is screwed up.  Hard drive writing is
> >excruciatingly slow (15 seconds to write 9 mb!) and brings the machine
> >to its knees all the time it's writing.

> <rest snipped>

> Is DMA enabled for the hard drive?

> Ian
> --
> Ian Riches
> GPL Rank +1.76 Monsters of GPL +284.19

Joachim Trens

Help! T-bird hard drive problems

by Joachim Trens » Thu, 03 Oct 2002 19:26:25

Hi Alison,

Like the others, I'd suggest first resetting the BIOS (or reloading the BIOS
defaults as it's called on Asus motherboards), as the new BIOS may have a
somewhat different internal structure (I learnt that from Asus' Readme's),
and if you don't reset it, the values you set in the BIOS' UI may get
assigned to parameters they're not meant for.

After that, I'd check DMA and PIO modes.

Any other BIOS settings should really be tried after you've checked the
above three things, but although I don't know your motherboard's BIOS, I
don't think any of them could - if simply set to their defaults - cause such
a severe speed drop. Smaller ones, yes, but not that substantial.

I'm using WinME and for me, the VIA drivers have always been slower and less
reliable that Windows's own - I've consequently never used VIA drivers other
than for a short test. But I don't know how Win98 is in this respect.

In the device manager, is one of your IDE controllers or Harddisks marked as
having errors (yellow question mark), or as running in 'Compatibiliy' or
'DOS' mode (not sure about the English names, I'm using a Spanish version of
Windows)? You'd see that under the properties tab.

And you didn't change any hardware, like hooking the HD up with another
device on the same cable? I never used Cable Select btw, I always define a
clear master and slave myself. Not sure this could cause your prob though,
but I assume you've tried master/slave anyway before you went to CS.

Have you tried msinfo32.exe? I think Win98 should have that as well, and it
gives you a lot of info about potential conflicts etc. If it's not the BIOS
but something Windows can sense, you'd probably see it there. It's started
through the 'Run' command line in the start menu, and it self-explanatory.

When you restarted Windows after the BIOS flash, did it detect any new
devices (or re-detect old ones)?

Good luck!

Achim

Steve Smit

Help! T-bird hard drive problems

by Steve Smit » Thu, 03 Oct 2002 19:57:59

First, make sure it's not a Windows prob.  I have an intermittant HD prob
that slows everything down like pulling the boards out of a HAL 9000.  You
must have a DOS sys burn-in pgm., no?  (Or I'll send you one.)  Run some DOS
benchmarks and make sure it isn't something the Evil Dr. Gates hasn't
initiated.


> I recently upgraded the BIOS in my main racing computer, a T-bird 1.3 in
> an Abit KT7, so I could use the 1.3 at full speed (the original BIOS
> only went to 1.2).

> However, now the machine is screwed up.  Hard drive writing is
> excruciatingly slow (15 seconds to write 9 mb!) and brings the machine
> to its knees all the time it's writing.

> So every time GPL writes out some replay bits or anything to virtual
> memory, the screen freezes, the controls stop working, and all Internet
> communication stops.  Sometimes this lasts for several seconds.  It's
> dreadful!  If I save a replay after a race, it takes so long (and kills
> communications so comprehensively) that the server typically disconnects
> me.

> Any suggestions are welcome.  I've done just about everything I can
> think of, including:

> - installed the latest VIA 4 in 1 drivers and the 1.10 USB driver
> - installed the George Breeze PCI latency patch
> - disabled serial port 2 to free up an IRQ
> - removed both USB controllers from the Device Manager to let Windows
> find them again
> - set the jumper on the hard drive to cable select
> - disabled PCI#2 Access Retry in the BIOS
> - enabled Fast Write in the BIOS
> - set PNP OS to No in the BIOS
> - cut GPL replay length to 20 mb

> Nothing helped.

> One thing I haven't tried is resetting the BIOS using the jumper on the
> motherboard.  Also, oddly, Windows Device Manager thinks I have two USB
> controllers and two USB Root devices.  I don't know if this is a
> problem, but I don't remember it being this way before the BIOS upgrade.

> Any suggestions?  I'm at my wits' end, and can't see too many other
> options short of reformatting and reinstalling Windows - and I hate
> doing that!

> I would be terribly grateful for any suggestions that might help me fix
> this machine and get it running as well as I know it should!

> Computer specs:

> Abit KT7
> Tbird 1.3 Ghz
> 256 mb PC133 memory
> 3Com Ethernet card
> Voodoo5
> SB Live! Value
> Logitech Wingman FF
> CH Pro USB Pedals
> Act Labs GPL USB Shifter
> Windows 98

> Alison



> Remove the spam blocker NOSPAM to email me.
> http://eaglewoman.maximumspeed.net

Alison Hin

Help! T-bird hard drive problems

by Alison Hin » Fri, 04 Oct 2002 00:56:24



Thanks, Achim, and thanks to everyone who responded with suggestions!
It's so great to have a group like this available to help when we've got
problems!

Also, thanks to fredrickson for the link to the FAQ.  I had already read
this, plus two of the BIOS FAQs that it points to.  Lots of great
information!

Ok!  After reading the posts in this thread, I tried some things and I
did have some success, although this Abit KT7-based machine's hard drive
writes are still not quite as good as my other racing machine, a 1 Ghz
Celeron in an ASUS P2B.  

[The P2B machine writes out GPL replay chunks to virtual memory almost
totally transparently; there's virtually no interruption in screen flow
and no impact on the controls.  If only I could get the P2B machine's
GeForce2 Ultra to behave!]

Anyway, here's what I did:

1. Checked Device Manager once again to make sure DMA was turned on for
the hard drive.  It was.

2. Reset the BIOS using the jumper on the motherboard, and went in and
reset the BIOS to 1.3 GHz.  I also set DRAM Clock to HCLK+PCICLK to run
my memory at 133, and disabled Video BIOS Shadow.

This made no difference.

I recalled that once in the past, Windows reported something as being
turned on when it actually wasn't.  So I went into the Device Manager
and turned off DMA for the hard drive, and rebooted.  I figured I'd try
it just to see how slow the hard drive was with DMA off, and then turn
DMA back on.

Presto!!  With DMA *off*, the hard drive got waaaay faster!  It writes a
9 mb file in a couple of seconds.  Better yet, there's hardly any impact
on the rest of the system when it's doing this.  In GPL, I get a little
bit of screen chop when replay is being written out while I drive, but
the controls remain perfectly functional, so I assume the Ethernet
controller will also keep communicating.

I decided to try turning DMA back on for the hard drive, but when I did,
Windows gave a scary warning that my hardware may not function properly
with DMA on, so I changed my mind and left it off for the moment.

In the BIOS, all IDE devices are set to AUTO for PIO and Ultra DMA.  Is
this correct?

The hard drive is a standard Western Digital 450AA, and the BIOS
recognizes it as a WDC WD450AA-OOBAAO.  Windows says it's a Generic IDE
Disk Type 47.

Does this drive have Ultra DMA or even DMA?  I thought it had at least
the latter.

The FAQ advises installing the VIA drivers right after installing
Windows (which I did).  Since this motherboard was built after Win98 was
released, I think it's desirable to use the VIA drivers.  

No.  All devices look cool in DM.

I did have a slave drive on the same cable, but it was malfunctioning,
so I disconnected it.  This was after the BIOS reinstall and the very
slow write problem began, however, so it didn't cause the slow write
problem.  Prior to that, the main drive jumper was set to Master and the
second to Slave.  

After disconnecting the second drive, I found I had to either remove the
jumper or put it on CS in order for the BIOS to see the main drive.
Since one of the FAQs I read recommended CS for the KT7 motherboard, I
left it there.

Thanks!!  I didn't think about msinfo32.  I don't do this stuff enough,
I guess!

msinfo32 doesn't report any problems under Conflicts/Sharing, and the
hard drive and CD and CD-ROM drives (on IDE2) all show up in
Components/Storage.

Also, it does show two USB controllers.  I do have four USB ports, two
next to the PS/2 port and two in a plate connected to the motherboard,
so I guess maybe I should show two controllers as John Hilsman says.

Yes, it re-detected the motherboard chipset, hard drive controller, and
USB controller.  Unfortunately I had not installed the latest VIA 4 in 1
drivers before flashing the BIOS, and for some reason when Windows
re-detected the devices it tried to find a file that wasn't in the
existing drivers.  

I had to skip the file to get the Windows boot to complete (I tried
several times but couldn't find the file and couldn't find another way
around the problem) and then install the latest 4 in 1 drivers.  Later I
also installed the latest USB and IDE drivers.

I don't know if this skipped file/belated 4 in 1 driver install caused
problems or not.  I suppose I could remove the 4 in 1, USB, and IDE
drivers, and then reinstall them.  Maybe I should drop into Safe Mode
first, to make sure I get rid of all the files installed by the drivers?

Since the machine works a lot better with DMA turned off in Windows, it
seems like maybe I just need to optimize the BIOS settings for the
drive, and then maybe I can turn DMA back on.  

Does this sound reasonable?  If so, what should the BIOS settings be for
the drive?

Or, maybe I should just leave well enough alone!

Thanks again, Achim, and everyone who replied!

Alison


        ________
       /        \
     _/__________\_
   /_ \  '===='  / _\
  |(_)    ____    (_)|
  |\o [] (____) [] o/|
  |   ---______---   |
  '---'          '---'

Alison Hin

Help! T-bird hard drive problems

by Alison Hin » Fri, 04 Oct 2002 00:58:40

Steve-

No, I don't have a DOS sys burn-in program.  I'd be glad to try yours!

BTW, I did try George Breese's hard drive test program, and it reports
that my hard drive is working fine and is quite fast.  Here are the
results:

TST disk-speed test
32-bit edition
Copyright 1999-2000, George E. Breese.  Rev. 12/26/00
Write rate is 34.482 megabytes per second.
Read rate is 19.230 megabytes per second.
Testing random reading for five seconds...
Random read rate is 14.677 megabytes per second.

From these results and comments in the readme, I think this program
bypasses Windows.  It was originally a DOS program, and was adapted to
the Windows environment but still runs in a DOS prompt.

Alison

On Wed, 02 Oct 2002 10:57:59 GMT, "Steve Smith"


>First, make sure it's not a Windows prob.  I have an intermittant HD prob
>that slows everything down like pulling the boards out of a HAL 9000.  You
>must have a DOS sys burn-in pgm., no?  (Or I'll send you one.)  Run some DOS
>benchmarks and make sure it isn't something the Evil Dr. Gates hasn't
>initiated.

Alison



Remove the spam blocker ARGLEBARGLE to email me.
http://eaglewoman.maximumspeed.com

Andy

Help! T-bird hard drive problems

by Andy » Fri, 04 Oct 2002 01:06:31


[snip]

You've got a similar setup to my machine!
And I had a mare of a problem when I updated the bios, again similar.
I realised that the problem was to do with AGP acelleration, which is
different from your issue, but anyway, running the DirectX diags caused the
machine to hang.
I tried everything, from Fdisk (I nearly cried when I found out my backup CD
was corrupt!!) through to ancient versions of drivers for everything...
What it turned out to be was that somehow I'd managed to introduce another
couple of IRQs to the PC causing a clash with the G-card. Sorting it was
easy. I simply disabled the printer and serial ports on the PC giving me
several further IRQs. Everything has been stunning since!

The m***of all this is to disable the hardware you're not using I think.

AndyC

Joachim Trens

Help! T-bird hard drive problems

by Joachim Trens » Fri, 04 Oct 2002 03:52:04

Hi Alison,

I'm afraid I must start guessing, but here are a few ideas what I'd look
into.

- if Windows redetected devices and then a file was missing, it is possible
that one such device didn't get installed properly. When you reinstalled it,
it is possible that a second instance of that device got installed, but
Windows is now using the wrong one. Or, the two are conflicting (the latter
is a guess, the former could indeed happen). An indication for this is
perhaps that turning DMA off made things faster - apparently the DMA driver
did not work properly.

I think your taking out one HD could _not cause something like the above.
This looks like something different. But...

- my second guess is something that happened to me just last week. I thought
my second HD was defective because it didn't get recognised, but after still
having some problems with the first HD after having taken out the second
one, I noticed that it was actually my first HD which had given me the
problem.

However, I don't really think that this is what happened on your machine, as
it all happened in the context of flashing the new BIOS.

In my device manager, all HD's are called Generic IDE as well, this seems to
be correct. I am sure your disk is DMA capable, I think nowadays they all
are and have been for years. I also think users like you and I wouldn't
notice a difference between UltraDMA and DMA; I think that's more noticeable
on servers.

As for the PIO and DMA, I think both can be set in the BIOS as well. Maybe
your BIOS showing wrong values to Windows caused Windows to detect new
devices. Have you checked the options for DMA and PIO in the BIOS? I think
selecting the second-to-highest PIO and DAM modes should work for your HD,
to be safe.

My personal guess is what I described above, a hard drive controller and/or
its driver got installed incorrectly, something got garbled, and now Windows
isn't using the resources correctly. If btw such a file is missing, it's
often already somewhere in the System or Options directories (or another
subdirectory of Windows) despite the fact that Windows doesn't find it, and
simply pointing Windows to it (you can just open the directories in the
search dialogue, if the file is in there it'll be visible there) is enough
to make Windows use it.

Maybe install all the driver software, also unzip them into a directory in
C: (to have any files that might be needed handy), then just delete your HD
and USB controllers from the device manager and restart. You might as a
precaution save system.dat and user.dat from the Windows directory, so you
could under DOS copy them back in if something goes really wrong in the
process). Maybe the device detection would then work properly.

Good luck, and please let us know how it's going!

Achim

Steve Smit

Help! T-bird hard drive problems

by Steve Smit » Fri, 04 Oct 2002 03:14:07

Can such a pgm. test 32-bit under DOS?  Anyway, if even *one* pgm. sez it's
working fine, then it seems unlikely that it could be a BIOS
program...unless you've checked/unchecked the part where the BIOS knows it's
running a PnP OS.  One of my ME 'putas was having gawdawful problems until I
told it that it *wan't* looking at a PnP OS...and now it's fine (although
the list of IRQs is frighteningly long).  If yer BIOS is set for a non-PnP
OS, then you should check for IRQ conflicts.


> Steve-

> No, I don't have a DOS sys burn-in program.  I'd be glad to try yours!

> BTW, I did try George Breese's hard drive test program, and it reports
> that my hard drive is working fine and is quite fast.  Here are the
> results:

> TST disk-speed test
> 32-bit edition
> Copyright 1999-2000, George E. Breese.  Rev. 12/26/00
> Write rate is 34.482 megabytes per second.
> Read rate is 19.230 megabytes per second.
> Testing random reading for five seconds...
> Random read rate is 14.677 megabytes per second.

> From these results and comments in the readme, I think this program
> bypasses Windows.  It was originally a DOS program, and was adapted to
> the Windows environment but still runs in a DOS prompt.

> Alison

> On Wed, 02 Oct 2002 10:57:59 GMT, "Steve Smith"

> >First, make sure it's not a Windows prob.  I have an intermittant HD prob
> >that slows everything down like pulling the boards out of a HAL 9000.
You
> >must have a DOS sys burn-in pgm., no?  (Or I'll send you one.)  Run some
DOS
> >benchmarks and make sure it isn't something the Evil Dr. Gates hasn't
> >initiated.

> Alison



> Remove the spam blocker ARGLEBARGLE to email me.
> http://eaglewoman.maximumspeed.com

Joachim Trens

Help! T-bird hard drive problems

by Joachim Trens » Fri, 04 Oct 2002 07:59:55

One more thought, Alison. I think Win98 makes backups of its Registry
(system.bak and user.bak). If you could find copies of these two files made
before your BIOS flashing, maybe that would cure the problem already.

I am not sure if that's Win98, but some version of Windows I think created
up to 5 backups of these Registry files. If I'm right, is there anyone in
the knowing where these files are located?

Achim

Don

Help! T-bird hard drive problems

by Don » Fri, 04 Oct 2002 08:15:00

Actually it's "user.da0" and "system.da0".

Allison, make sure you also have an 80-pin cable going to the hard drive.
The lines on it will be half as thick as a "standard" IDE cable, but there's
still only 40 pins on the end. This will ensure you're getting at least
ATA-66 transfers. The 40pin cables don't support that speed.
Also, I'd ignore (at least temporarily) the warnings from Windows about DMA.
Don't forget: it's Microsoft warning you there! lol.

Gunner


REDLINE42

Help! T-bird hard drive problems

by REDLINE42 » Fri, 04 Oct 2002 10:57:16


Alison, You may need to install (reinstall?) this driver, hpt201019.exe. I
believe that this is the Highpoint? Ultra DMA100 driver, read on....

<snip>

Yes this is correct.

<snip>

Here's some info on your hard drive;

Specifications for the WD Caviar WD450AA

Physical Specifications

Formatted Capacity      45,020 MB

Interface 40-pin EIDE

Rotational Speed 5400 RPM (nominal) << slow by today's standards.

Data Transfer Rate (maximum)
- Buffer to Host

66.6 MB/s (Mode 4 Ultra ATA)  << Yes your Hard Drive can do Ultra ATA with
the right cable.
33.3 MB/s (Mode 2 Ultra ATA)
16.6 MB/s (Mode 4 PIO)
16.6 MB/s (Mode 2 multi-word DMA)

Buffer Size 2 MB

Found at; http://www.wdc.com/products/legacy/Legacy.asp?r=3

Abit KT7, ???

All I could find on Abit was the KT7A at here,

http://www.abit-usa.com/pt_main_back29be.html?pPRODUCT_TYPE=MotherBoa...
EL_NAME=KT7A

AMD Socket A Based ATX Mainboard With Ultra DMA100 & SoftMenu III Technology

If this is your motherboard, it has Ultra DMA100 (Mode 5 Ultra ATA 100.0
MB/s).
It will also run Mode 4 Ultra ATA 66.6 MB/s. Backwards compatible.

You should get rid of the Generic IDE Disk Type 47 by installing the right
Ultra DMA100 drivers.

To get the most speed out of your motherboard a recommendation of a ATA100
7200 rpm drive would be the best, as the motherboard can handle it.

Did you get the right bios, some of these motherboards have Raid, but I read
the bios update will detect if you have Raid or not.

Latest bios is,
KT7A/KT7A-RAID (A9 BIOS Compile Date: 7/11/2002)
or,
KT7A V1.3/KT7A-RAID V1.3 (9R BIOS Compile Date: 6/11/2002).

I'm not at all familiar with the VIA chipset, but know that there are many
drivers for this chipset.

And did you load the right drivers for the Ultra DMA100 chipset?
Abit used to use the Highpoint Ultra DMA66 controller which on a Intel
chipset was a separate install, if I recall on my old BE6 board.

Are these Abit VIA drivers or,,,?

This is the Abit driver download page;

http://www.abit-usa.com/download_content9c5d.html

When Windows asks for a file that is most likely already installed and there
is no browse button to find the file, just type in c:\windows, or
c:\windows\system and it may find the file already installed. Sometimes
there will be a "info" button that gives you a clue as to what directory the
file is to be installed (found). This is more of a problem on Win95/98 than
XP.

The driver for Ultra DMA100 chipset is not installed, thus the "Generic IDE
Disk Type 47".

You need to get rid of the  "Generic IDE Disk Type 47" by installing the
Ultra DMA100 drivers. And then you will have a grayed out DMA check box, as
DMA will always be active.

You may need to install (reinstall?) this driver, hpt201019.exe found here
at the bottom of the page;
http://www.abit-usa.com/download_content0792.html

Hope this helps you Alison.

--
Daytona Beach - Winter Speed Carnival 1903 - 1935

1919 - Ralph DePalma Driving Champion and winner of hundreds of races, sets
new 1 mile record driving 905 cubic inch V-12 Packard on beach at 149.875
mph. DePalma also sets World Records for distances of 2 to 20 miles and
standing mile. His standing mile record of 92.713 mph is unbroken for 36
years.


rec.autos.simulators is a usenet newsgroup formed in December, 1993. As this group was always unmoderated there may be some spam or off topic articles included. Some links do point back to racesimcentral.net as we could not validate the original address. Please report any pages that you believe warrant deletion from this archive (include the link in your email). RaceSimCentral.net is in no way responsible and does not endorse any of the content herein.