MP 1,1-5,1 Core 2 Extreme QX9775 in MP3,1

[lnx64]

I'm someone who loves to tinker, even if it's useless and for no reason. I know that the Xeon X5482 and Core 2 Extreme QX9775 are nearly identical. In fact, fairly identical if there are any differences. Something I have wondered though as both are also C0 steppings and LGA771, could the QX9775 be used in a MP3,1 at all? Has anyone actually tried?

 

[lnx64]

Ok I was able to obtain 2 QX9775's from my girlfriend (she used to have a Skulltrail build), and can say: It works at least on MY system.

https://imgur.com/dFk2CY1

 

[startergo]

Very similar to X5482.

 

[LightBulbFun]

Ok I was able to obtain 2 QX9775's from my girlfriend (she used to have a Skulltrail build), and can say: It works at least on MY system.

https://imgur.com/dFk2CY1

very cool to see they work! I had a feeling they would as they would as they are C0 stepping but still good to have confirmation, I wonder if you can use throttle stop to exploit the unlocked VID and multiplier and overclock them?

do you have a link to a CPU-Z validation btw? that would be neat to see

(I wonder why windows says virtualization disabled...)

 

[lnx64]

very cool to see they work! I had a feeling they would as they would as they are C0 stepping but still good to have confirmation, I wonder if you can use throttle stop to exploit the unlocked VID and multiplier and overclock them?

do you have a link to a CPU-Z validation btw? that would be neat to see

(I wonder why windows says virtualization disabled...)

So for virtualization, that's the firmware's legacy boot, Mac Pro's for some reason disable VT-x when you legacy boot, including this machine, which is more of a debugging kit than a production machine.

Now for CPU-Z, here's where things get very very confusing:

https://imgur.com/pFjdrXE

CPU-Z still see's the X5482 I had in the system last, while AIDA64 as you saw, reported it correctly. I can't run a proper validation if CPU-Z still thinks the last CPU is installed. I thought the CPUID was obtained directly from the CPU? Or is this more of a problem with the ROM simply not properly supporting a QX9775?

This particular debugging machine cannot do a PRAM reset (holding the keys does nothing, it's completely ignored).

 

[lnx64]

UPDATE: It no longer boots. After a few reboots I was unable to get back. This didn't last long at all. I think what may have happened was it was forcing CPUID platform ID 40 on it, which is what CPU-Z may have been seeing? After a few reboots later platform ID 10 couldn't be loaded as this boot ROM simply doesn't have it..

I have no idea what was going on, but in the long run, it is *not* compatible. You may get one or two boots out of it before something updates. (This kind of reminds me when I went from an E5462 to X5482, the X5482's were NOT running correctly at first.)

Damn, oh well.

EDIT: It was Windows Fast Boot. Shut down apparently never truly shuts down Windows (some form of hybrid hibernate). That appears to be why this worked until I actually rebooted.

 

[pc297]

Anyone tried to slap a couple of those or QX9770/X33XX (Yorkfield LGA771) in a MP2,1 to see if we have more luck than with Harpertown Xeons?