Have you tried using archive.org? If you're able to come up with any links to the old support pages or a URL to the BIOS file itself, then archive.org might have an archived copy. I've sometimes been able to retrieve old files that way.

Unfortunately I think it might be difficult to find anybody else who has the same board to dump a BIOS from - Slot-2 and 440GX boards in general aren't very common.

I did try that, but ultimately i did get the BIOS files 😀 .. Turns out you should just do a search for the Gateway board number, and "BIOS". Then it returns the correct site, and they actually do have the files available for download . The board is a Gateway 4000427, now it's got a force flashed copy of Gateway's P17 BIOS on it, it fixes the cosmetic issues, but still the fan headers don't work.

The headers might have been overloaded by somebody. There's probably some tiny 3pin transistors near each header, they might be burned out. It can happen when somebody has a bad fan and ends up trying it in both headers.

I'm reviving this thread to see if OP or anyone else has a copy of this BIOS? It's for the Gateway OEM MS440GX motherboard. BIOS string is 4M4SG0X0.15A.0020.P10. The P17 version (4M4SG0X0.15A.0029.P17), or any version newer than P10 would be helpful. I tried searching with the bios string, gateway spearfish, MS440GX, the motherboard number (4000427), etc, but haven't had any luck. 😢

I found this, but it's the intel BIOS for the retail MS440GX. https://www.infania.net/misc/moboarchive/Prem … 40gx/index.html

Getting closer...

https://web.archive.org/web/20011209114145/ht … /4m4sg0x0.shtml

Some of the old support material is still on the Gateway site (at least, last time I looked), but can't be accessed via Gateway's own search engine. Try using Google and search on the gateway site only.

Have you looked at the file on DriverGuide? - 7505904.exe. Here's a scanned local copy renamed with the zip extension

https://www.driverguide.com/driver/detail.php … driverid=884407 - file contents are at the bottom of the page

Wow! Thank you. That worked perfectly and now my board supports Pentium III Xeon (Tanner) CPUs!

I also thought it would remove the 1MB L2 cache limit, but it looks like that’s still in place. 512K and 1M Xeons work fine, but 2M Xeons result in a cache disabled error. 😵

Katmai500 wrote:

That worked perfectly

where is this GIF from?

at the 0:27 of the video.
Vinamilk ad, don't know if it original or they rip it off somewhere
https://youtu.be/1LfM0fgxio4?t=27

Found the animated gif thru Google image search - did a quick crop / resize and hey presto! - never even noticed it was a product ad (actually thought it was from an animated series called Shaun the Sheep), though having watch the Youtube link, I like their other ads.

Katmai500 wrote on 2019-07-05, 22:22:

Wow! Thank you. That worked perfectly and now my board supports Pentium III Xeon (Tanner) CPUs!

I also thought it would remove the 1MB L2 cache limit, but it looks like that’s still in place. 512K and 1M Xeons work fine, but 2M Xeons result in a cache disabled error. 😵

Interesting, I was pondering getting a pair of 2MB Tanners for my Intel MS440GX, but am now having second thoughts. I have the 4M4SG0X0.86E.0050.P20 07/02/1999 BIOS and wonder if this will support 2MB.

I have just fixed SIV to report the fan speeds on my MS440GX motherboard, but the ADM9240 Fan 1 input always returns 0 RPM 🙁. I suspect there may be a hardware issue with the board and would like to know if other MS440GX motherboards report a none zero speed for Fan 1, does your motherboard report a speed for Fan 1 please?

The MS440GX has a fan multiplexor that seems to switch the Fan 1 input between the three case fans with Fan 2 being four CPU fans, but I could not find any software that supported the fan multiplexor. As I develop SIV this was no real issue and as of the SIV 5.49 release SIV does as you can see below.

I was able to resolve the 2MB issue by installing a VRM 8.3 module in the empty socket. After that, 2MB CPUs and dual CPUs worked fine! The module I bought is intel P/N 697701-602.

I’ll check on the fan speed and report back when I get a chance.

Katmai500 wrote on 2020-06-18, 03:12:

I was able to resolve the 2MB issue by installing a VRM 8.3 module in the empty socket. After that, 2MB CPUs and dual CPUs worked fine! The module I bought is intel P/N 697701-602.

I’ll check on the fan speed and report back when I get a chance.

Thank you and that's good news, I will check which VRM I have later. Before you got the VRM 8.3 module did you have one and if so which one was installed please? Do you know if it's possible to get 700 or 900 MHz Xeon (Cascades) CPUs to work?

There is now SIV V5.50 Beta-00 in which I tidied up the MS4400GX fan multiplexing.

The VRM socket was empty before I bought the VRM 8.3 module. So, without the VRM socket populated it seems that you can use only one CPU with up to 1MB of L2 cache. The one I bought can be easily found for purchase by googling the part number I provided. If you already have the VRM socket populated it’s most likely an 8.3 module in there already.

My board is Gateway OEM running the Gateway P17 BIOS, so it only supports up to 550 MHz Tanner CPUs. I’ve read that the intel retail version of the board works with Cascade CPUs, but don’t have any personal experience with it.

The one I have is which looks slightly different, but thus far I can't find if it's 8.3. I only managed the find one Intel 697701-602 for sale in the UK, so I ordered it.

I tried a pair of 700MHz + 1MB L2 SL49P and the system would not post with the post card reporting 38 (Memory high address line failure at 01000-0A000 (non-fatal)) 2F (Enable cache before system BIOS shadow), see http://www.bioscentral.com/postcodes/phoenixbios.htm.

BTW As it a Gateway rather than an Intel SIV may fail to report the fans, if this is the case and you post the two Menu->File->Save Local files it should be trivial for me to fix this.

Anonymous Freak wrote on 2016-10-03, 11:59:

I know I'm replying to the original post, but there was a special internal BIOS build for the MS440GX that supported Coppermine-based (also called "Cascades") Xeons. I had that board running stably with a pair of 2 MB Pentium III Xeon 700 MHz CPUs for quite a while. (Until well in to 2004 when I finally replaced it with a "modern" computer as my daily-use PC.)

I just found the above post and guess I need the "special internal BIOS", I wonder if anyone has it.

Katmai500 wrote on 2020-06-18, 11:54:

The VRM socket was empty before I bought the VRM 8.3 module.

I just got the same module, it works fine with my 550MHz 1MB Xeons, but as I expected the 700MHz 1MB Xeons still fail with the 38 2F post failure.

Katmai500 wrote on 2019-07-05, 22:22:

Wow! Thank you. That worked perfectly and now my board supports Pentium III Xeon (Tanner) CPUs!

I also thought it would remove the 1MB L2 cache limit, but it looks like that’s still in place. 512K and 1M Xeons work fine, but 2M Xeons result in a cache disabled error. 😵

An interesting development with BIOS P17. It seems that ONLY Pentium III Xeons work now.

• No POST with a single Pentium II Xeon (450/512 or 450/2M)
• No POST with dual Pentium II Xeons (450/512)
• Pentium III Xeon 550/2M CPUs run at 500 MHz

Not being able to use PII Xeons is a disappointment. I tried flashing it again, but that didn't fix the problem. Now I need to track down an older BIOS to try to flash it back if I want to return this system to its original configuration.

Does anyone have a link to or copy of any older Gateway BIOS version?

• 7505904.exe 437 KB 6/15/99 Pentium II BIOS update 4M4SG0X0.15A.0029.P17 <- What I have now. Weird that it's even called "Pentium II BIOS Update" when it doesn't work with Pentium II Xeons
• 7505527.exe 411 KB 4/15/99 Pentium II BIOS update 4M4SG0X0.15A.0026.P15
• 7504834.exe 344 KB 12/15/98 Pentium II BIOS update 4M4SG0X0.15A.0020.P10