Originally Posted by wyverexauctor
It's actually painfully obvious what the maximum safe voltage for a Sandy Bridge CPU is. It's the default voltage setting at stock speeds, anything beyond that is going past Intels recommended specs. With that said if you want to overclock you're going to need higher than defaults when it comes to Vcore. As a guideline I would say no more than 1.4V but again, this sort of voltage is going to have an impact on the CPU's overall lifetime.
Not true. Intel's job is to get your chip stable at 3.3GHz, or stock speed. They find the lowest voltage it is stable and that becomes your chip's VID. But just because your chip is stable at that speed at 1.0v does not mean that that is the max safe voltage. Intel hasn't tested that yet, and so it is unknown.
While it is true that anything above a chip's VID will "degrade" it. Its all about the time constraints that determine what is actually safe. If your chip's VID is 1.0v then perhaps 1.1v will kill your chip in 100 years, 1.2v in 80 years, 1.3v in 50 years, 1.4v in 20 years, and 1.5v in 5-10 years. But at any point does that really matter? According to Moore's law, most of your computer components will only last you 2 years, so even doubling that to 4 years, provides you a reasonable time constraint. If a certain voltage gives your chip a 6-10 year life, then its most likely a maximum safe voltage.
Of course the above times/voltages are made up, the point is that stock voltage is not your max, its your normal voltage that is required to run stock speeds only.
Originally Posted by Sin0822
those are SVIDs operating ranges, DVID has the same type of table.
VID equates to the range of voltages which the processor might be given at stock, so your VID would be probably like 1.25v, SVID increases your VID automatically, but it tops off around 1.44v as in I have never seen VID exceed 1.44v on its own through the mechanism of SVID. .
Page 72 on your document:
The processor uses three signals for the serial voltage identification interface to support
automatic selection of voltages. Table 7- 1 specifies the voltage level corresponding to
the eight bit VID value transmitted over serial VID. A ‘1’ in this table refers to a high
voltage level and a ‘0’ refers to a low voltage level. If the voltage regulation circuit
cannot supply the voltage that is requested, the voltage regulator must disable itself.
VID signals are CMOS push/pull drivers. Refer to Ta b le 7- 9 for the DC specifications for
VID range is only up to 1.52v because SVID has the capability and that is over and beyond what Intel felt they could program allowing enough room for OC.
Intel never specified a maximum vAbsolute or max VCC which would be considered max vcore. VID isn't Vcore.
For SBe Intel does the same except defines everything. THey actually said max is 1.4v for vabsolute for SBe in their datasheet. This past week at CES i talked to their OC engineer again, he told me 1.4v sounded a bit low to him as Vabsolute, he was going to look into it.
I have had MANY CPUs go above 1.5v all of them showed signs of degradation, when you always hoot for the max multiplier and you use over 1.5v you feel that you need more and more volts to boot into windows, i would never recommend to someone that they do so for 24/7.
Your argument holds up. So I'll agree to stop saying 1.52v is maximum, but still in my personal opinion and experience 1.5v is my recommended maximum.Edited by kennyparker1337 - 1/17/12 at 7:14am