**Hi mods - not sure if this best belongs in CPUs or overclocking, so posted in both. Feel free to delete one.**

Was it just a typo all along?

Just noticed there is a June 2010 revision to the i5 700 and i7 800 datasheet. Not very exciting until I got to table 7-4. The castrated Vtt absolute maximum of 1.21 from the previous version of the document that has been the bane of Lynnfield overclockers (apart from ones like me who always thought something fishy was going on and ignored it) has gone up to 1.4 where it jolly well should have been all along. Also note Vcc has come down to 1.4, and ram volts has gone up to 1.8!

http://download.intel.com/design/processor/datashts/322164.pdf

I wont even begin to speculate as to what was going on at intel when they published the first doc - probably a work experience kid having a laugh. My i5 breathes a small sigh of relief knowing that whilst it's life is probably only gonna get harder now, at least the 1.3V Vtt it's running wont kill it . So unleash your i5 750s, folks *

Liam

*Liampope takes no responsibility for fires, loss of property or life......

It's all a ploy to get us to slowly kill off our i5 cpus so we will have no choice but to take up their new sockets next year /ott conspiracy mode over

My i5 has only even needed 1.2vtt for 4Ghz stable and I've never really took into consideration if that was considered high or low. However, having previously come from a DDR2 + X38 based motherbard, all voltages seemed "low" in comparison

YAY for VTT!

Is this for all new chips or existing units? ie will this affect my 750 or only newer ones?

White paper is for 750 and 8xx series.

you read the 96 pages??!!

Temi_D said:

you read the 96 pages??!!

Click to expand...

Dedication

Temi_D said:

you read the 96 pages??!!

Click to expand...

LOL not exactly. In fact it was only really these specs I was interested in when I noticed the revision, as I was aware that the datasheet for socket 1156 Xeons already had 1.4 for Vtt so something seemed amiss. Now they've brought it all in line. Who knows whether they've done further testing, or whether the reasons for the lower voltages previously were political or even a mistake? Guess it dont really matter. And yes this revision would retrospectively refer to all i5 700 and i7 800 series chips.

Time to crank up the clocks

if this is true, then id imagine RAM makers are gonna have a field day with no more 1.65v limitations

With a max of 1.4v most (all?) cant reach 4.4Ghz any more. But I was previously one of those who harped about the 1.21 Vtt limit (You would think that Intel of all people would know their processors)

thingemajib said:

if this is true, then id imagine RAM makers are gonna have a field day with no more 1.65v limitations

Click to expand...

That's assuming they pay much attention to Intel's specs in the first place. When Lynnfield Vtt was supposedly 1.21 absolute max, plenty of mobo XMPs were auto-setting Vtts to anything from 1.3 up to 1.4. Were they taking big liberties with people's kit, or had Intel already told them those values were safe? Hmmmmmm...

Lleot said:

With a max of 1.4v most (all?) cant reach 4.4Ghz any more. But I was previously one of those who harped about the 1.21 Vtt limit (You would think that Intel of all people would know their processors)

Click to expand...

Lol - I don't think people pushing 4.4 on these chips really care about longevity or Intel specs anyway

That's great news! Might actually be able to push 3.8 - 4 ghz without worrying about hitting 1.21 All I need are some Vipers, then I'll see how far VTT's been holding mine back. Cheers Liampope!

Is the VTT the volt setting for the IMC?

and giving ram 1.7 won't kill an i5?

My mobo repeatedly states in the manual that putting ram volts above 1.65v will irreversibly damage your cpu. Intel really screwed ram makers.

Has anyone actually seen an i5/7 die from excessive IMC voltage?

The actual intel spec is the differential between the VTT and the RAM voltage - this should not be more that 0.5.

With the orginal max VTT of 1.21 RAM voltage should not be more than 1.71. Now with 1.4 as the max, RAM volts could go up to 1.9 - if the stix can handle it I guess?