Problems reflashing BIOS

All,

I have a Gateway 7422gx. In the process of trying to fix the performance issues with XP x64, like an idiot I flashed the NSF04.P0 BIOS. While the flash worked, it doesn't seem to have fixed the problem, so I was trying to flash back to NS.00.02, which I *think* is newer?

Well now no matter what image I try to flash (my BIOS.BAK, NS.00.02 from the resource page, NSF04.P0 again, etc) using either WinPhlash or Phlash16 from DOS, after it backs up the current BIOS, verifies everything, etc, as soon as it gets to the point where it's supposed to start writing the BIOS, the machine shuts down. I have it plugged into AC, so that's not the problem. It always does it at the exact same spot.

Does NSF04.P0 have some wierd virus protection setting I don't know about or something?

Thanks!

Tried that option. Even tried reflashing NS04. WinPhlash doesn't error out, in fact it goes through all of the verifiying and backing up, etc. It's just as soon as it gets to actually writing the new BIOS, something literally shuts the system down. It's really wierd.

Okay, I spent the last 4 hours playing with this, and this is what I've found so far:

The M6805 updater put BIOS NSF04.P0 on my laptop (7422gx). I can reflash that same version (NSF04P0E.ROM) as many times as I want, no problem. But as soon as I try to flash any other BIOS (my original, an older M6811 BIOS, etc), the system immediately powers off as soon as Winphlash/Phlash16 try to write the new BIOS.

Almost sounds like the BIOS doesn't like being overwritten with something else? Very wierd..

If anybody else has NSF04.P0 on their machines can they try to reproduce this please?

Thanks!

Yep. I've unchecked everything except for the options to clear/zero the blocks before writing, verifiy the block after it's written, and to clear the CMOS checksum afterwards. I've even unchecked those, but I figured with those 3 I ought to at least see that it's TRYING to write.

The wierd part, like I said, is it goes through all of the steps (verifying, backing up, etc) until it gets to the very last step, which is writing the new BIOS. If I reflash the existing version it writes fine. If I try to flash anything else (I even found a copy that's technically newer, so I know it's not a version/date thing), it says "Writing new BIOS..." and then the system immediately powers off. No errors, no warnings, just off. Happens in both WinPhlash in XP and Phlash16 on a DOS bootdisk.

Okay, this is getting really curious... Came home and was bound and determined to figure this thing out. I opened the machine up and found what I thought was the flash chip: An SST SST39VF040 512kx8 CMOS flash. I downloaded UNIFLASH and sure enough, that's exactly what it shows.

Going back into WinPhlash I loaded the NSF04.P0 .ROM up, went into Advanced Settings, went to Devices, and indeed SST39VF040 was listed (w/ a 8x64K block size). I loaded up both my original backup .ROM and the NS.00.02 .ROM file. Under the Devices tab for those two, a multiple devices were listed (even some 8x64K SST's), but the SST39VF040 was not listed.

I'm wondering if that's why WinPhlash is having problems loading the other .ROMs, because they doesn't seem to support my flash chip? It seems wierd that the backup of my original BIOS says that it doesn't support the chip, but I have a feeling it worked for Gateway since I'm sure they used a flash programmer to slap it on the chip.

Verified my wife's 7422gx has the same flash chip as mine. If I'm understanding this correctly, I have 5 choices:

1) Try to find another Arima .ROM somewhere that is compatible with both the laptop and the flash chip (good luck).
2) Try to find a way to either get the .ROM to 'support' my chip, or force it on myself (good luck).
3) Buy a new flash chip that the other .ROMs support (will this work? I know I'd have to hot-swap the chips to get to the point where I could write to it, but will another chip work in the laptop?).
4) Buy my own flash programmer (ya, right).
5) Stick w/ NSF04.P0

Am I understanding this correctly? Does this make sense? Does anyobdy else have any other ideas?

Thanks!

Something interesting I just noticed...

The one SST 8x64K flash chip listed in my backup image is a SST49LF040 (instead of the SST39VF040 in my laptop). According to the SST web page, the SST49LF040 has been EOL'd. I'm wondering if Gateway ran out of those, ordered the SST39VF040 to replace it, and never bothered to update the BIOS image to handle the new chip?

I think it's because even the original backup image doesn't have my flash chip listed under supported devices.

As a hunch I email'd my backup image to badflash and ordered one of the SST chips that does show to be supported. I should get that chip in the next couple of days, and I'll see what happens when I try that one

If you look in the PHLASH.LOG file (I think that's what WinPhlash calls it), or boot from a DOS disk and run UNIFLASH, I'd be curious to know if it says your flash chip is a SST39VF040. If so, then I have a feeling that's the problem.

On a side note I finally decided to send my laptop in to get the hinges fixed. It came back with the NS.00.05 BIOS flashed on it, which seems to have fixed most (but I don't think quite all, maybe 95%) of my microfreezing issues with XP x64. The funny part is I did a backup of the BIOS in WinPhlash, loaded hte backup, and the SST39VF040 chip still isn't listed in the list of supported devices.

Thought about popping in the SST49LF040 chip I got from badflash.com and seeing if I have the same problems with that. The SST49LF040 is listed as a supported flash chip.

If it is then it should have stopped at that point. Might want to check elsewhere in the logfile. UNIFLASH will also tell you when you first run it what kind of flash chip it found. The only catch is UNIFLASH won't run from within Windows.