I am currently back onto an E5200 Pentium while I work this out, but an interesting thing happened yesterday.
I did the following:
- Flash the BIOS chip to 1.3.0 externally
- Insert new LGA 771 platform microcodes for 676, 677 and 67A (04 40 0F, 44 0B, 10 0A) and update existing LGA775 platform microcodes from 0016760C 0106760C to 0F , from 01067705 to 0B and from 01167A07 to 0A.
- Programmed all blocks directly with Win Flash (ticked all options and double checked)
- Shut down the computer, replaced the procesor and thermal grease, Clearned CMOS settings using jumper function and powered it on
- After that edited the BIOS settings (like tag, time, boot sequence and such, CPU Xeon was detected normally), saved to CMOS and restarted
After that windows started normally (Very quickly) but it still had E5200 dual core in its settings as the installed processor. During that time everything was blazing fast. Knowing this can happen if somebody replaces the processor with out doing an full restart I after that did an full system restart/shutdown (Shut down - Restart). Once that completed windows became much slower and has again started experiencing prolonged boot times and almost unbearable response times.
This has lead me to believe that there was something wrong with the system it self so I did a few tests. First that I tried was to boot my windows 8.1 installation disc, boot time was so long I literally went home, had lunch, got back and it was not done by then (eventually it did boot the installation menu but it was unusably slow), After that I tried an Mini Windows Xp boot disc (Hirens boot disc) and to my surprise, and expectation at this point, the mini windows XP system worked flawlessly, no lag no stutter.
This has lead me to believe that there is either something that is preventing usage of an Xeon processor in an modern system with this computer or I simply failed to install proper microcodes again.
Here are images of before and after my modification:
Maybe, even though an E5450 SLANU is an Code ID 676, I should have inserted 1067A Platform type 44 as well? I don't know.