Search results

I don't know which software tools are to be used, since I've never used a Nehalem CPU. However, I think you should better ask in a true overclocking forum. ;)

The Xeon X5690 doesn't seem to have an unlocked multiplier, but you can overclock locked Nehalem-based CPUs by changing their reference base clock.

I don't think so, at least I haven't found anyone mentioning that their overclockable Xeons suddenly stopped being overclockable after installing the latest Windows patches or modding their BIOSes to integrate the new microcodes.

Thank you, I updated the list. :)

What's the source of your information?

I assume there are quite a few people who want to get one of these CPUs, however I haven't seen them in the real world. I'm not sure if they really exist.

Correct, that is exactly what the numbers mean. :)

According to Wikipedia, the E5-2699C v4 has very shitty turbo steps compared to the regular E5-2699 v4: List of Intel Xeon microprocessors - Wikipedia

I remember reading somewhere that CPU-Z shows the processor's current VID instead of the actual voltage if it can't access or don't support the voltage sensor. This would at least explain the voltage displays of the CPU-Z validations.

I've edited the first post to reflect this discussion.

So you are saying that this CPU-Z validation is fake? Intel Xeon W5580 @ 5612.23 MHz - CPU-Z VALIDATOR It shows as D0 stepping (which is the production stepping of this CPU) and an multiplier of 42.

Thank you _haru, I edited the list to reflect that.

I also had an Xeon E5-1680 v2 in the past and I could overclock it pretty well. 4,5 GHz (on all cores) on stock voltage was my usual setting.

Summit Ridge and Pinnacle Ridge do support unbuffered ECC RAM. If the mainboard manufacturer implemented it to its BIOS, it also works fine.

Correct, basically it isn't worth it trying to overclock the E5-26xx series. Either go E5-16xx v1/v2/v3, or use the consumer CPUs.

The mounting holes of EATX and SSI EEB formfactor boards are definitely not the same. Only three of them are identical, the other ones are in different positions.

It should work with only one CPU in the first CPU slot.

I'm asking specifically for this configuration because it is rumored that this would be the memory configuration for Threadripper 2. If this is true (and not one channel per die), I'm interested in the performance impact. :)

I really have hoped for a few more benchmarks. Also a configuration like two channels populated on die 1 and 3, and no RAM on dies 2 and 4 would be interesting. :)

This doesn't explain how he is able to change the CPU's multiplier to 40 and above.