Change Page:
< 12 | Showing page 2 of 2, messages 31 to 34 of 34
Delirious
-
Total Posts
:
11461
- Joined: 11/15/2007
- Location: at my computer
-
Status: offline
- Ribbons: 54
- Folding: 104,454
-
|
Re:Approved memory
Thursday, June 14, 2012 11:59 AM
( permalink)
yeah, I did get mine to run at 2400 but the bios posting was buggy. I'm trying to get 2666 to work, then I can work down from there.
[ Site Rules ] "Be quick to listen, slow to speak and slow to become angry" Affiliate Code XZUMV9TJW5 if you think education is expensive wait until you get the bill for ignorance
|
|
|
|
feniks
-
Total Posts
:
7631
- Joined: 8/2/2008
- Location: NJ-NY, USA
-
Status: offline
- Ribbons: 37
- Folding: 13,279,128
-
|
Re:Approved memory
Friday, June 15, 2012 9:42 AM
( permalink)
it seems that voltage and current specs for gen3 (ivy bridge) i7 desktop line are mostly same as for gen2 (sandy bridge). http://www.intel.com/cont...p-vol-1-datasheet.html same max vcore (1.52v), same standard vccio 1.05 (+5% max at stock), same vccsa 0.971v max limit at stock, etc. I'd assume it's same safe to up vccio to 1.20v (considered max safe for overclocking) and re-test stability, also along with upped vccsa (which even gets bumped up automatically in Z77 FTW BIOS) for better stability of faster memory. just so you know, I recently had to up VCCIO to daily 1.160v because sometimes my 2133mhz memory (4 sticks) on Z68 FTW would crap out after a few days of running 24/7 and throw a BSOD 124 under light load even when using VCCIO at 1.15v daily (which was bench stable). note, this is not to be confused with a random BSOD 124 at idle which was related to vcore voltage dropping sometimes a tad too low at long idle (probably because of spread CPU core voltage spectrum locked in as enabled) - the solution was to leave C1E & EIST enabled in BIOS, so the multiplier drops to 16x at idle so it would make the manually dialed vcore still sufficient, instead of upping vcore to compensate for too low idle vcore (at a cost of higher vcore under load).
|
|
|
|
essenbe
-
Total Posts
:
24
- Joined: 5/10/2012
-
Status: offline
-
-
|
Re:Approved memory
Sunday, June 17, 2012 6:49 AM
( permalink)
I don't know if this will help, but intel specs on Sandy are max VCCIO of 1.2V and VCCIO must be within 0.5V of dram voltage. I have another brand Z68 board and am running dram Voltage at 1.65V and VCCIO at 1.16250. It works with Sandy Bridge and Ivy Bridge.
Cooler Master HAF X- Asus Maximus V Extreme - i7-3770K - 8Gb G Skill DDR3-2400- EVGA SC 670 4GB - XSPC Raystorm RX360 - Corsair AX850 - 256GB Samsung 830, 128GB Samsung 830- 2X 1TB Samsung Spinpoint. 
|
|
|
|
feniks
-
Total Posts
:
7631
- Joined: 8/2/2008
- Location: NJ-NY, USA
-
Status: offline
- Ribbons: 37
- Folding: 13,279,128
-
|
Re:Approved memory
Sunday, June 17, 2012 12:30 PM
( permalink)
essenbe
I don't know if this will help, but intel specs on Sandy are max VCCIO of 1.2V and VCCIO must be within 0.5V of dram voltage. I have another brand Z68 board and am running dram Voltage at 1.65V and VCCIO at 1.16250. It works with Sandy Bridge and Ivy Bridge.
no, not anymore since SB, or link the intel white paper where it states the VCCIO must be within 0.5v from vdimm. I can run my system perfectly stable with default 1.05v vccio running 4 mem sticks at 1600MHz using 1.65V only for speeds at or above 1867MHz I need to bump vccio (to 1.16v) so they stay stable - mind that is for 4 sticks - and even that is not true if I run 2 sticks instead of 4 all works fine at lower vccio ...
|
|
|
|