[Solved] P55 UD4P Build - Page 2

One interesting thing I discovered when I started investigating these: the GB driver sites for these boards do not seem to contain 'F6 pre-load' drivers for the new Marvell SE9128 SATA3 controller, nor do drivers seem to have appeared on Marvell's site, yet...

This, unfortunately, seems to be due to 'designed in' limitations of the P55, or, more exactly, the fact that the PCIe controller is integrated on-die for the 1156 platform CPUs - leads me to believe there will not be a fix coming for this... True to form, Intel seems to be sticking to the X designation for 'chainsaw' ripper chipsets, and P for the more delicately programmable, but more intrinsically limited pieces...

on top of what the OP said about the new mobo P55A-UD4P it also has the lotes socket for you die-hard OCers

I was afraid to ask before, but curiosity got the better of me. What's FM?

Just ran. It fails. When I run it, as administrator or otherwise, it opens a pop-up window:
Title: Error
Message: NOT FOR THIS CHIPSET!

It also leaves a file behind it created called MS.ini. Don't know if that's useful or not.

Just sent it to you.

F*&king Magic!

Actually, a combination of:

A lot of practical experience, dating back to the days when we were soldering on a motherboard for which we had an actual, exact, real schematic, whose signals were slow enough to be seen on an actual real oscilloscope...

Having read enough Intel and JEDEC specifications to make my eyeballs bleed...

The knowledge that, at their heart, most of the memspecs are really physical time periods, say, to 'charge up' the impedance in the column address select line to the point where the DIMM will be readable, which makes it mostly a matter of conservatively calculating integer conversions, from clock counts (at some specified speed) to time periods, and back to memory clock counts (at a different clock speed)...

A few rules of thumb, as some sub-timings are not proportional to time periods, but are summations of others, sometimes +/- 1, sometimes upped to the next odd #...

Some educated guesswork - getting a stable tRFC is partially dependent on voltage (which you only want to use as a last resort, and sparingly then), and partially on the number of DIMMs involved...

Some dumb luck...

Glad I found this thread. To the OP:

niklas_13:

How did that Xiggy CPU cooler install? I have the original HDT-S1283 and I'm considering upgrading to the i5-750/GA-P55-UD4P setup myself. Wanted to buy the 1156 bracket for my S1283. Wanted to make sure my Xiggy was going to fit the Gigabyte board first.

Really don't wanna give up my Xiggy.

Thanks I assumed something with tall cooling fins (Corsair Dominator) wouldn't fit based on pictures of the board. So I was planning on doing the more standard sticks with the low profile heatsinks.

Appreciate the tip though.

My problems went away when i updated my bios to f4t and i also did a driver sweep of all ATI drivers and used cat 9.10

I did install cat 9.11 yesterday, so far so good