P55 Roundup: Gigabyte P55-UD6 / P55-UD3 and MSI P55-GD80

Motherboards/Intel S1156 by massman @ 2009-09-20

Almost two weeks after the official release of the Core i5 platform, we review three LGA1156 motherboards. We put the Gigabyte P55-UD3 and UD6 through the same batch of tests as the MSI P55-GD80 and check who comes out on top. Next to that we have some surprising results for those who want to use Turbo mode and also have a look at the extreme overclocking capabilities of both high-end motherboards. Last but not least we publish our findings on the maximum voltages for the Core i5 platform.

  • prev
  • next

Core i5 and Vdimm

Core i5, the platform where Vdimm is not deadly?

Everyone who has followed the launch of the Core i7 in the beginning of 2009 will know that one of the biggest concerns of both end-users, motherboard and memory manufacturers was the memory voltage limitation set by Intel. From the absolute beginning, Intel warned everyone that using a Vdimm exceeding 1.65V was playing with fire. Well, maybe not that dramatic, but most certainly not healthy for the CPU. The reason had to be searched in the field of electromigration, but to save you a lot of time I won't go deeper into physics part; more interestingly how end-users had to work with this limitation and, even more interestingly, how to work around this issue.

First of all, let's go back at the Core i7 (LGA1366) specifications and have a look at why the 1.65V Vdimm limitation was set in place.

Madshrimps (c)


Madshrimps (c)


The standard values for the VTT voltage, which basically is the voltage for the uncore die (not core die) on which the memory controller is located, vary between 1.045V and 1.22V. Given the critical value of 1.65V for the memory, the suggested difference between VTT and Vdimm of <0.5V is a safe limitation as Intel acknowledges that it's possible to have a maximum difference of (1.65V - 1.045V) 0.605V. When looking at the absolute maximum ratings, we see that VTT can be set at 1.35V maximum and Vdimm at 1.875; the difference is 0.525V. Strangely enough.

Ok, onto the Core i5 specifications then:

Madshrimps (c)


VTT voltage varies between 1.045V and 1.15V; given supposed critical value of 1.65V, difference varies between 0.5V and 0.605V

Madshrimps (c)


The absolute maximum ratings are pretty clear: the Vddq (=Vdimm) can not exceed 1.65V, and only match 1.65V if the VTT is running at 1.21V; difference here is 0.44V.

Madshrimps (c)


The question I want to bring forward here is as follows: Given the big fuss Intel made regarding the maximum memory voltage when releasing the first Core i7 series, why didn't they do the same with the new generation of Core i5 processors? If exceeding 1.65V on a Core i7 platform is, according to Intel, indeed not within safety ranges, how can they not tell anything about the new safety ranges on the Core i5 platform, which has a lowed stock VTT voltage to begin with and for which the Intel data sheets clearly indicate that 1.65V Vdimm is the absolute maximum?

Honestly, as a performance enthusiast I hardly ever stick to the 'safety' voltages as indicated by Intel or motherboard manufacturers, but that doesn't take away my concern for the mainstream users who, unknowingly, will use 1.65V rated memory.
  • prev
  • next
Comment from Massman @ 2009/09/21
Recent findings actually push me in the direction of the GD80 in terms of overclockability.

32M is both faster and higher clocked with same or a bit less voltage :-)

Comment from leeghoofd @ 2009/09/21
Great to see MSI reacting quickly ! any news for ya DFI arrival ?
Comment from Massman @ 2009/09/21
Via via I heard a board was sent. Let's hope for the best
Comment from leeghoofd @ 2009/09/21
bout time... hope the bios is already pretty mature !

 

reply