past few days i've been flashing bios and updating cpu microcode, trying to get the later 10676/10677, 775 cpus working on a couple of boards, both are MSI but had medion bios. ones an MSI P4m900m-2 ver2.2, i picked this up about 4yrs ago for a win98 build as its a pci-e equivelent of my current 98 build (MSI PM8PM-V), i tried to flash a stock msi bios with awdflash and afudos by hotswapping the chip to another board but it wouldn't do it, didnt think i'd be able to without an external flasher so i shelved it.
then recently i tried uniflash and found it will allow you to hotswap chips between different boards re-detect them and flash any bios! absolutely brilliant tool!
the other board i have came from a medion pc i was given, when i opened it i was surprised to see a it had a PM8PM with medion bios, tho it didnt say MSI on it, it clearly was, so i flashed it with the bios from my MSI PM8PM-V, and i bricked it! so thats been shelved for 4yrs. 🤣. i just heard its compatible with the MSI PM8PM-IL bios, so all good now.
after sorting the bios out, and discovering uniflash, i can mess around with bios microcode without risk as i can hotswap and use uniflash to flash back the stock bios if anything goes wrong. ive had partial success on the P4M900M-2, after adding microcode for later core2duos/quads, my Q9450 didnt work at all but i did get it to post with a couple of core 2 duos that use the same cpuid/microcode as the e8600 so that cpu may be a possibility. i flashed 2 chips, one didnt display the cpu's model on post and winxp blue screened on boot, most likely as it couldnt detect the cpu model, the other bios did detect the cpu's model on post, and xp booted fine, problem is both bios i flashed had the cpu multiplier set at X6 instead of X13, so my cpus were running at 1200mhz instead of 2700mhz, and the vcore is 1.0v when it should be 1.3v, this is apparently set in microcode, so may not be a hardware issue as i thought. it may be i need a certain version of microcode or two are conflicing and its using the wrong one.
i then tried out my PM8PM-V, the only cpu's i have that work on this board are single core celeron D's, i wiped all the microcode and just added the 10676/10677 code for later core2duos/quads, it didnt work, it still worked with my celeron D even tho i removed the code for it, and also the multiplier was half of what it should be, it seems the microcode which is stored elsewhere in the bios partially supported the celeron thats why the multiplier was halved and i'm guessing the code for this cpu must be supplementary to that, in order to get it running at the correct ratio/voltage. knowing this i'm gonna have another go with the P4M900, i think im finally getting somewhere!