11 Replies Latest reply on Oct 20, 2010 1:14 PM by Doc_SilverCreek

    Authoritative answer on safe voltages

    max_Q

      Hi Guys,

       

      I'm hoping for a definitive statement from one of you to help guide my actions. I posted this question at the OCZ forums as well, but thought I should go right to the horse's mouth. I bought 8 gigs of OCZ Gold ram (OCZ3G1600LV4GK) and installed it in a new P55-Core i7-860 comp I built (Gigabyte GA-P55A-UD4P motherboard). I have it running at 1333 @ 8,8,8,18 1T, 1.5v with no issues. I knew when I bought the ram that it required 1.65v to run at it's rated speed of 1600, but I noted that it was "designed specifically for the Intel® P55 Chipset and subsequent Intel® Core™ i7, i5, and i3 (Socket 1156) processors" and I assumed that setting the memory voltage to 1.65v was no big deal.

       

      Well, after educating myself a little more I'm not so sure that elevated voltages are "no big deal".

       

      According to Intel's "Intel® Core™ i7-800 and i5-700 Processor Series Datasheet - Volume 1", Tables 7.4 and 7.6, 1.65v is the "Absolute Maximum Rating" for VDDQ (ram, right?). The normal operational limits for VDDQ however, are between 1.425v and 1.575v, with a typical voltage of 1.5v

       

      Now here's what Intel says about running devices outside the normal operational limits (in this case memory at 1.65v): "At conditions outside functional operation condition limits, but within absolute maximum and minimum ratings, neither functionality nor long-term reliability can be expected. If a device is returned to conditions within functional operation limits after having been subjected to conditions outside these limits (but within the absolute maximum and minimum ratings) the device may be functional, but with its lifetime degraded depending on exposure to conditions exceeding the functional operation condition limits." LIfetime degraded? Well, that's bad. The"functional operation limits" of VDDQ is a max of 1.575v, so 1.65v is clearly above "functional operation limits".

       

      In addition, the absolute maximum VTT is 1.21v. Here is what happens if you exceed it: "At conditions exceeding absolute maximum and minimum ratings, neither functionality nor long-term reliability can be expected. Moreover, if a device is subjected to these conditions for any length of time it will either not function or its reliability will be severely degraded when returned to conditions within the functional operating condition limits." Not function? Severly degraded? Well, that's even worse.

       

      Now, given that everyone and their brother is selling ram for the P55 chipset that is designed to run at 1.65v, and OCZ regularly recommends a VTT of 1.35v in their forums to get their ram stable at rated speeds, as in this thread, and they even sell 1.65v ram with the Intel Core logo on it - what is going on here?

       

      Given this info, with words like "neither functionality nor long-term reliability can be expected" and "it will either not function or its reliability will be severely degraded" I've been understandably reluctant to bump up my voltages to get my ram running at 1600. On the other hand, I don't want to leave the extra performance that I paid for on the table if there is no danger to the performance level and lifespan of my processor and chipset.

       

      I'm thinking that the recommendations of a VTT of 1.35v and DRAM of 1.65v can't just be some irresponsible attempt on the part of the memory manufacturers to get their ram running at it's rated values. It must be based on inside information. They wouldn't be selling ram at 1.65v for the P55 if there was the slightest issue with that, right?

       

      So how about someone from Intel going on the record. If the memory manufacturers are exceeding Intel's published voltage specs (which they are) and it's a bad thing, shouldn't Intel be coming down on them? After all, presumably there's going to be a whole rash of burnt up processors and chipsets and customers who won't be very discriminating in who they blame. Don't forget, these elevated voltages a often set by the ram using Intel's very own XMP profiles.

       

      Or are Intel's published voltages just very conservative? You don't have to say that, but maybe Intel could put out a statement saying that the voltage recommendations of the major memory manufacturers for their ram designed for the P55 chipset and Lynnfield processor are long term safe.

       

      The murkiness surrounding this issue is silly, and I'm tired of scouring the internet for reassurance that I can use one thing I bought in the way it was intended without breaking another thing I bought that it's supposed to work with.

       

      Thoughts?