Vista installation fails on first boot-up

[Guest]

Hello all,

Brand new machine:
Gigabyte GA-P35-DS4
Intel Core 2 2.66Ghz
2GB DDR2 800Mhz
Asus nVIDIA 8800GTS 320Mb

Windows being setup on a Raptor 74Gb SATA hard-disk which is set on top in
the boot priority in the BIOS.

After the first boot (right after the phase where the installer hs copied
and extracted all the files), while the system appears to be loading, I get a
blue screen and an immediate reboot.

I tried changing the Controller mode to / from AHCI to RAID, to disabled,
simply using IDE. I tried downloading the SATA driver from the Gigabyte
website and loading it in the setup. The same exact thing happened in all
situations.

I then tried a memory check and see if by using the "repair" option I could
solve this, but the test went fine and no repair is possible (given it is a
setup failure apparently).

I honestly have no idea what could be wrong, so any help would be
appreciated. I set-up literally dozens of machines in the past (9x, 2000,
NT4, XP, SBS) but this is my first Vista machine and I am completely stuck.
No logs, no way to see the blue screen, nothing.

Thanks in advance.

 

[Guest]

Thank you so much for your extensive reply and sorry for the lack of detail
but I spent (same as you I suppose) hours trying to figure out what was wrong
and at 1:30AM decided to ask for help

I will go through your reply carefully when I'll be back from work. Mine is
a clean install of Vista Ultimate (32-bit) and I have two other SATA disks
(SATA-0 is the 74Gb raptor where I'm trying to install).

Again, thank you so much, and I'll be back with some (hopefully positive)
results.

 

[Guest]

Ok... the "one at a time" approach at the BIOS level seems to work. I was
able to proceed with the full installation by disabling EVERYTHING but the
kitchen sink (on-board LAN, sound chip, IEEE1394, RAID, optimizations, etc.).
Now, as you said, the culprit will surface when I start enabling features
again.

Before disabling everything, I updated the BIOS to the latest version. Not
sure what effect this had.

Again, thank you so much for your guidelines. I think from now on the
patient, while still critical, is out of the woods

 

[Guest]

Dear all,

I think it is always worth doing a follow-up to help others in the same
situation. After a couple of weeks I started noticing that the instability I
expected from Vista was simply way out of control. 2-3 BSOD a day, DLLs
getting unregistered, random errors when launching applications, constant
chkdsk's when booting up.

To make a long story short, always, ALWAYS check the memory compatibility
list for your motherboard. I had a couple of 1GB DDR2 800 from Twinmos, which
clearly are not listed in the tested RAMs if you check Gigabyte's website.
Switched to a couple of Corsair, and surprise surprise, with optimized
settings in the BIOS, without disabling ANYTHING, perfect setup, no problems
whatsoever.

Anyway, problem solved (I won't bug you the various tests I did with other
HDs, linux, XP, all of which failed and displayed some sort of instability).

Hope this helps

 

[Richard Urban]

Another confirmation of what I have been stating for years.

If you are experiencing multiple, and varied, unexplained errors - change
out your RAM for known good RAM.

I have see this "fix" a multitude of unexplainable errors over the years.

Many post here, and in the Windows XP news groups, listing all their
problems, and then end their post with:
"and don't tell me it's the RAM. I have tested it and it is fine"!

--


Regards,

Richard Urban
Microsoft MVP Windows Shell/User
(For email, remove the obvious from my address)

 

[Guest]

You're right.
I think, however, this raises the question on how useful it is to do
low-level checks of the RAM with prorgrams like memtest86 or Vista's memory
diagnostics.
Basically if I hadn't dug any deeper and trusted those results I would still
be in the same situation given both of them passed 100%.

On the other hand, better to always play it safe, and take a hard long look
at the memory compatibility list.

Best regards,

Sergio

 

[Richard Urban]

That is the reason I have a "hardware" RAM tester. Sure, it cost more than
$1000 but it will either pass the RAM, fail the RAM or cause the RAM to fail
if it is marginal. When servicing computers a software RAM test does not
give a true indication of the RAM performance or condition.

--


Regards,

Richard Urban
Microsoft MVP Windows Shell/User
(For email, remove the obvious from my address)