ECSsem v1.0 badflash

[walwal]

Another day, another badflash.
Wanted to upgrade to an Athlon Xp 1800 which needs a bios update from v1.1 dated 5.31.01 to v1.1e. (Or does it?)
Started badly - save present file not possible under Uniflash 1.28. Went ahead and after flash got verification error.
Does this mean the chip is fried or is it a bad file? Any ideas how to progress from here guys. Is a hot flash with v1.1 bios possible or is the chip beyond recovery?
Thanks in advance.

[NickS]

I find the best place for an ECS opinion is alt.comp.sys.periphs.mainboard.elitegroup (or something like that).
Do you actually get any startup message now, or is it totally dead ?

[walwal]

CD lights up and HDD starts but thats it. Classic BSOD.
I just wondered if there was a known bug with Uniflash and this Windbond chip or whether there was some way to overwrite the badflash and start again. I'm running out of spare bios chips

[walwal]

Tried to hotflash with v1.1 and v1.1e bios codes downloaded from ECSUSA website. Nothing works. With a working chip the flash freezes and the old bios is still there.
Does anyone have a working bios or am I downloading it wrongly somehow? Help much appreciated

[NickS]

From Rainbow's site, Uniflash 1.28 known bugs include:
-Winbond W49F002U doesn't work - looks like write protected (sometimes???)
-256KB BIOSes don't work on some i430VX/HX boards (AWDFLASH works!)
-Some VIA and AMD chipsets don't work - maybe because non-standard flash write protection on some boards.

Explain a little more about how you are hot swap flashing - sounds like you are just using the one board ? you have a ROM that works and you use that to boot the board and then flash ?

Random thought: have you turned off all system BIOS shadowing/caching ?

[Rainbow]

Caching/shadowing should never affect BIOS flashing because Flash ROM is mapped at the top of 4GB which cannot be shadowed and is always uncached. The only exception might be ALi FINALi chipset but I found not problems there.

[NickS]

Caching/shadowing should never affect BIOS flashing because Flash ROM is mapped at the top of 4GB which cannot be shadowed and is always uncached. The only exception might be ALi FINALi chipset but I found not problems there.

Useful info, thanks, Rainbow.

[walwal]

Er, thanks Nick. Just realized the chip was a Winbond W49F002U
Where do I go from here?
Tried the recommended AWDflash flasher but that wont proceed because it says wrong chipset. Assume that means it doesnt work with the VIA chipset in the Soyo board being used for the hotflash.
Still got the pc chips m599lr board which AJZchips rescued with the correct bios upgrade. Would the AWDflash work in that for a hotflash perhaps - its sis 530 i think and the ecs board is sis also? The bios chip is now ASD.
Or if the Winbond is write protected perhaps the boot block is still there and the awdboot thing might work. Dont understand how to even try that though - dont speak DOS.
Now we know what went wrong any help in recovering much appreciated.

Wally in Walthamstow

[NickS]

I don't think you would lose much other than time by trying to hot-swap flash in the PC-Chips board. Rainbow knows the ins and outs of flash chips better than I. The "AWDboot thing" is pretty simple if it's working. It detects that the BIOS is corrupt (if it is) and tries to boot from diskette. The problem comes when you correctly flash the wrong BIOS and it doesn't try to boot from diskette.

When it tries to boot from diskette, what you have to do is create a boot diskette. In Windows 95/98 all you have to do is format the diskette and tick the "copy system files" box. I believe in ME you have to bring up a DOS box with start-run-"command" and then type "sys a:".
Then you create a text file called "autoexec.bat" on the "A" drive - (warning, you may need to turn off "hide extensions for known file types" or you will end up with a file called "autoexec.bat.txt")

Autoexec.bat must contain the necessary command line to carry out to flash - "awfl772 /sn yourbios.bin /py /cc /cd" for example. Those files must be on the diskette.

All that is useless if it isn't trying to boot from the diskette anyway.

[walwal]

Success
Used aminf330 from Rainbow's site (ami flasher with award bios ) and yes the m599lr mobo. Flash took forever but booted at once.
The problem? The boot screen now says flash type unknown which is odd because it means the bios doesnt recognise itself! I guess its something to do with that version of the Winbond bios chip. It was my fault in the first place for not reading the instructions and obeying them
Thanks for all the help everyone.
Nick: it was your post that made me read the uniflash page from top to bottom that solved the mystery.

Wally

[NickS]

@Rainbow - would I be right to guess that the bootblock is for the wrong chip if he's getting the "unknown flash type" message on boot ?

[Rainbow]

Unknown Flash Type = BIOS does not know the Flash ROM chip installed - either the BIOS simply does not support that chip (old BIOS version or replaced Flash ROM chip) or the write protection is enabled.