Page 1 of 1

Flash Bios from a Similar Motherboard?

Posted: Sat Nov 22, 2003 5:23 pm
by rojmiller
I would like to know whether it is safe to flash my bios from a similar motherboard from the same manufacturer. The situation is as follows:

My board is a Chaintech 7aja2E, which has the KT133E chipset. There does not seem to be much info on this chipset on the web, but what I've found indicates that it may just be a repackaged KT133A chipset, not "officially" certified for the 133MHz bus. Tests on boards with this chipset (e.g., Abit KT7E) show it runs stably at 133MHz, and can even run up to 143MHz.

Chaintech make a KT133A board (7aja2) that appears to be identical to mine (same manual), except for the one chip difference. There is a 100MHz/133MHz jumper on the board, and on my board it is soldered in the 100MHz position. However, in my bios I can change the setting so the bus can run at up to 130MHz (with a 4 divisor for the PCI bus). I presume that the other bios can set the bus at 133MHz.

According to Chaintech, my board can support up to an Athlon 2000+ (130X12). The 7aja2 can support up to a 2600+, which must mean that the other bios must be able to set the multiplier as well (both boards have jumpers for automatic multiplier, and setting it at up to 12X). The only thing in the manual that distinguishes the 2 boards is the chipset description and the instruction for my board to set the bus speed (soldered in, not jumpered as the manual says) to 100MHz.

So my question is, can I safely flash to the 7aja2 bios? Seems to me that the only problem that I am likely to get would be some possible instability, but that I am highly unlikely to get a dead motherboard out of it. Any agreement or disagreement with this assessment?

Just to let you know, I have absolutely no need to actually do this flash. I am running the board with a 950MHz Duron, with no immediate plans to upgrade. However, I have built 10-12 computers from scratch for others, and the first thing I do is flash to the latest bios - I am not afraid of flashing, but am aware of the risks involved. I just gotta know whether this board that I picked up for $20 can run a Athlon 2600+ (with some cooling on the chipset, I presume). Any one want to lend me an Athlon 2600+ for a while to test this out? :D

Posted: Sat Nov 22, 2003 5:35 pm
by Rainbow
If the jumper is soldered to 100MHz, there must be a reason why. Most probably because the chipset does not work at 133MHz.

Posted: Sat Nov 22, 2003 6:01 pm
by rojmiller
Nope, there are several reviews on the web for that chipset that says it runs fine up to 143MHz. The reviews say that the chipset preceeded the KT133A, and was less expensive, and was not "officially" certified at 133. They suggest that there was an agreement with VIA not to officialy run it at 133. And my board can set it in the bios at 130, and can run an Athlon 2000+ at 130X12!

The reviews suggest that the chipset is actually a KT133A, but with different packaging (plastic vs. metal). So I suspect that a little extra cooling would solve any problem of the increase in speed from 130 to 133 anyway!

Each manufacturer seems to have a different approach to support of the chipset. Some have not supported it beyond Athlon 1.3 GHz. Chaintech seems to have gone the farthest so far, with its support.

But the question really relates to the bios - would it really be very different for these two boards, other that for the options presented? I have seem suggestions where people have used the bios from other manufacturers and booted ok, with only somre stability problems. Seems to me that flashing this other bios should be low risk, as should trying out the 133MHz bus (paying attention to chipset temperature).

Posted: Sat Nov 22, 2003 8:35 pm
by Rainbow
The BIOS will not help you as the FSB is hardwired to 100MHz. You'll need to remove the wire from there and solder in a jumper. It might even work with your current BIOS.

Posted: Sat Nov 22, 2003 8:53 pm
by rojmiller
How can it be hardwired to 100MHz? As I mentioned, I can go into my current bios and change it to 130MHz (32.5MHz PCI bus) right now. No big deal to hardwire it to 133MHz, if the new bios supports it (which is still the question....). I'll just clip the wire near the 100MHz connection, flip it over, and solder it to the 133MHz connection.

But the question still is, is there any chance thatl the other bios will somehow create a bad flash, stopping me from booting? Seems to me if the other board has the actual jumper referred to in the book, it's bios will run fine at 100MHz as well - after all, the board was originally produced for the 100MHz Durons and Athlons. I suspect that the other bios is virtually the same, except allowing for higher FSB settings (overriding the jumper), and maybe even some new, (higher than 12X) multiplier settings (which my bios doesn't have).

Posted: Sat Nov 22, 2003 8:59 pm
by Rainbow
With the jumper set to 100, the FSB can be <133MHz. If it's set to 133, the FSB can be 133MHz and >133MHz.
The BIOS from different board might work or might not work - as always.

Posted: Sat Nov 22, 2003 11:21 pm
by edwin
Rainbow wrote:With the jumper set to 100, the FSB can be <133MHz. If it's set to 133, the FSB can be 133MHz and >133MHz.
The BIOS from different board might work or might not work - as always.
This would hold true for the current bios as well. Chances are that with the other bios and the jumper soldered in the 100MHz postion, you still cannot go beyond 130MHz in the bios.

Posted: Sun Nov 23, 2003 7:45 pm
by rojmiller
Maybe I can generalize my question a little more. In any case when someone is looking to use another bios for a motherboard, what are the things to look for to increase the chances of success. Is a bios from the same manufacturer critical? Is the chipset critical? What is the bios initializing? I think I remember reading somewhere that having the same keyboard controller chip was essential. Or is it all random, just flash, hope, and have a backup plan?

I also remember reading somewhere how to tell if you can flash your bios chip in another motherboard, e.g., whether the other chip has the same characteristics (voltage, pinout etc.) so that you can boot the other motherboard and re-flash your chip there in the case of a bad flash (backup plan). Can't seem to find it anywhere, so if anyone could post a link to a good reference it would be appreciated.

Posted: Sun Nov 23, 2003 7:51 pm
by edwin
Hotflashing:
http://www.pppr.sk/rainbow/hardware/hotflash.html

Requirements for a cross-flash: Same I/O chip, same keyboard controller, same chipset, same onboard audio/LAN/video (if any), same USB controller (if any), same PCI addressing scheme. The layout may differ. Your mileage may vary. This holds true for boards from the same manufacturer and between manufacturers.

Posted: Sun Jan 18, 2004 9:46 pm
by rsshnry
I have a 7AJA2E board also. I recently put in an XP2400 running the 12/05/2001 BIOS. When it booted up, it was identified as an unknown processor at 1500 mhz (15x 100). Since the processor was unlocked and the 2100's and up have remapped multipliers, I set the dipswitches on the board for 12.5x (12.5x= 20x). After booting, the XP was identified as an unknown processor at 2000 mhz.

In an attempt to get it identified correctly and hopefully have the option of running it at 133fsb, I started looking at using the 7AJA2 BIOS's. I was able to successfully flash the board with the 3/11/2002 BIOS (identified the processor as unknown as expected). I then tried the 11/18/2002 BIOS which should have indentified the processor correctly. However after the step where the Award flasher asks if you want to save the old BIOS, it gave the message "The program file's BIOS-lock string does not match with your system" and gave me a DOS prompt. After pressing CTRL-ALT-DEL the system would not boot. After turning off the power to the power supply and starting up again, the system booted successfully. I got the same result from the 10/31/2003 BIOS. The BIOS experts here can probably explain this better than I can, but it looks like Chaintech has added something to prevent later 7AJA2 BIOS's from being used in the 7AJA2E.

Since my 2400 runs fine on the 7AJA2E, I suspect the your 2600 will probably also run fine, just identified as unknown during start up. As far as running it at 133fsb-I suspect the board is capable of it, but I'm not sure how to make it happen.

Posted: Sun Jan 18, 2004 11:08 pm
by edwin
by unsoldering the piece of wire and putting a 3-pin header in place, then put the jumper in the 133MHz position. This will be done entirely at your own risk and again: your mileage may vary.