Windows 7 - Frozen BIOS Boot screen

eMachine 6216 being reimaged, rebooted to frozen eMachine splash screen, won't POST,
No F keys respond. CMOS battery pulled, replaced: same thing.

Update: Pulled HD power, can enter BIOS setup, even boot a disk. This supposedly means HD is bad, however HD will start up when slaved to another PC.

Swapped cables with known-good OD which had just booted disk, HD still blocks POST, freezing at Emachine screen.

Solution: Next I swapped in a known-good HD and it started up so I knew prob had to be limited to the HD. But since it started up in the other machine it didn't seem to be a bad HD. Could it be that HD (boot) code might be blocking this BIOS from even POSTing? I wanted to test this so slave the HD again to the other machine and Cleaned it with Diskpart.

Sure enough when returned to the eMachine it allowed BIOS to post, so I booted installer and reinstalled 7. Since it was during reimaging that this boot code apparently futzed the BIOS the image is getting deleted. And I thought I'd gotten a copied image to reimage over the network - wrong!

Fascinating stuff Greg.

So the question that pops into my head is: what is the bios reading from the hard drive that could be affected by a disk image?
I was unaware that the BIOS could/would read any data from the drive. I was under the impression it only read the info from the firmware.

To the laboratory Igor!

Bob the only code I can think of is boot code somehow being queried or accessed by BIOS even before post. But this is only based on familiarity with how corrupt boot code can block installs, remedied by Cleaning the HD which wipes the boot sector, and how boot often needs repair after reimaging which I'd just done.

If BIOS doesn't see code at that stage then wouldn't it have to be mechanical?

I wonder if BIOS refuses to recognize HD because it's an image copied to another network computer and then reimaged over the network, changing the SID or HID. Brink warns about this in System Image Recovery but I had heard it could now be done and wanted to test it:

Update: HD just passed Hitachi Drive Fitness Test.

That EMachines wouldn't be running a UEFI BIOS, would it?

No it's way older, maybe 7 years.

It's very interesting. I was able to play hookey a little yesterday and do a little research (not much, my customer actually wants me to finish their project).

I have a hypothesis: Every hard drive has a controller chip on it's logic board. That is the device that actually communicates to the PC. The first thing the BIOS does after performing the POST is enumerate the attached devices. During enumeration the devices are queried to see which, if any, are boot devices. If there is a bootable DVD in the optical drive it's controller chip will tell the BIOS "I'm here, and I have bootable media". If the hard drive is blank the chip says "I'm here, but not bootable". If there is an OS installed then the chip says "I'm here, and bootable".

That would mean that during OS installation or use something gets written to the controller chip. Perhaps there is a kind of "checksum" that happens when the drive starts up. If the bad image contained a wrong "checksum" then the controller might simply fault. And that fault may make it electrically dead. And that may explain the no POST situation.

And if it is an IDE drive then that may explain why it can be recognized when it is jumpered to SLAVE. "I'm here, but don't try and boot me"

So you wiped the drive, possibly resetting whatever "checksum" code is written the the controller chip and viola: the dead come back to life. If this is true it would be huge, as we get posts regarding mysteriously dead hard drives often.

What do you think?

It sounds like a good theory but does it even POST when it freezes at the eMachine spash screen which gives the F keys (none of which respond?).

I also think the HD was somehow invalidated, likely because I was trying to use a copied image from the network which had orginally been stored onboard.

I don't think you will get the splash screen unless the BIOS completes the POST.

Thanks, I wondered about that. Several search results had referred to this same frequent issue with eMachines as occurring before the POST.

Yeah the worst idea they ever had was getting rid of the POST Beep. And beep codes in general. They encourage folks to throw it out and buy a new one!

I think they replaced the POST beep for the big E. You see the E and you know yo POSTed.