Core i7 Vdimm/VTT Graveyard :D

[Mungri]

Lets wait a few months and see if any of your Core I7's go boom =D.

I really hope they dont, but it will give us all a better picture after you have been using them for a long time.

[vservera]

nice topic there very impresive

[hiwa]

Run 920 @ chiller 10°C for over 100 hours

Vtt/Qpi V: 1.36/8
Vdimm : 2.04

[Benji Tshi]

According to Asus FR, their i7 run 7hrs/day for a long time @2vdimm (or more, i don't remember anymore)...I'll try to get more details about it.

[fragmasterMax]

I have a feeling this is going to materialise into another "hey i've been running my ram at 2volts, for 6 months now, and hasn't hiccuped yet" kinda thread.

[fragmasterMax]

If going over ~2 volts on the memory (or within .5v of vcore) really was such a risk, none of the boards would have that option enabled, as they (meaning intel +board maunfacturers)
would have 100x rma's on there hands. IT's not like they can tell if a chip has been oc'd or was just a bad apple.

[saaya]

Run 920 @ chiller 10°C for over 100 hours

Vtt/Qpi V: 1.36/8
Vdimm : 2.04

thanks

According to Asus FR, their i7 run 7hrs/day for a long time @2vdimm (or more, i don't remember anymore)...I'll try to get more details about it.

yeah, i heard about that too

I have a feeling this is going to materialise into another "hey i've been running my ram at 2volts, for 6 months now, and hasn't hiccuped yet" kinda thread.

yeah... the fact that NOT A SINGLE i7 has died so far is pretty good news, and i dont think we will see dead cpus anytime soon, maybe one or two out of a freak accident here and there...

and actually you dont need high vdimm, we managed to reach DDR3 2200 with 1.7v vdimm on our BloodRage board last week

IT's not like they can tell if a chip has been oc'd or was just a bad apple.

actually intel and amd CAN tell if you overclocked and overvolted your cpu afaik... whether they check and refuse rmas is another question though
rmas for cpus tend to be ridiculously low, so checking might not be worth the effort, its usually cheaper to just replace the cpu than do any checks and tests plus especially in the US people might sue them if they refuse an RMA which would cause a lot of issues... so just replacing a chip that costs them 50$ or so to make is easier

[Benji Tshi]

1.4 min, 1.55 max for QPI/DRAM Vcore
1.9 min, 2.2 max for vDimm
2.0 min, 2.2 max for vPLL

During all the week end, testing P6T Dlx, and no problems. I'm under Swiftech GTZ.
I'll go on next week with SS/DI and i'll tell you what's happening.

[dinos22]

i've started testing this weekend and my board likes 1.5-1.55v uncore voltage so i just use that hehehe

lets see how long this i920 lasts...i've gone up to 1.7vdimm as i didnt need more to run 2GHz 8-8-8-24-1T on these Corsairs yet ...just changed over my mounting gear on SS to LGA1366 so will do some cold runs

hopefully the CPU likes cold...down to -60C heh

once i start with SS i will run up to 2.3vdimm to see how these sticks like voltage

[GAR]

Dinos, 1.5v realy safe for regular usage? what board you using anyway?

[systemviper]

i've started testing this weekend and my board likes 1.5-1.55v uncore voltage so i just use that hehehe

lets see how long this i920 lasts...i've gone up to 1.7vdimm as i didnt need more to run 2GHz 8-8-8-24-1T on these Corsairs yet ...just changed over my mounting gear on SS to LGA1366 so will do some cold runs

hopefully the CPU likes cold...down to -60C heh

once i start with SS i will run up to 2.3vdimm to see how these sticks like voltage

also what corsairs do you have, i am looking to get my ram this holiday, waiting for BF to see if I can find some deals.

I was looking at the HyperX that do 2Ghz bu they are 3x1G for 250.00, but i love to get the 2Gx3 sticks of dominators instead, but I haven't heard much on performance.

I have my I7 and my GIG X58 Extreme will be here friday...

thanks

[dinos22]

i'm using Gigabyte X58-UD5 board with i920 and Corsair 1866 CAS8 3x1GB kit

[systemviper]

Seems we will be running almost the same same...

So what is the overclock you get for 1.4x vCore,
Is that regular DDR3 or the new improved versions

thanks for the info.


also i will be going almost straight to water
got my blocks and mountings already...

[3oh6]

but i love to get the 2Gx3 sticks of dominators instead, but I haven't heard much on performance.

word on the street is my kit should be in this week of their TR3X6G1600C8D. i will be running them on the EVGA X58 board. i can hit you up with some results as soon as i have them coming in. fire me a PM in that is the kit you are interested in Steve.

[T_Flight]

I've gopt those Same CAS8 1600 Dominators and might be able to get them booted in an hour or so. I'm working on it. I'm gonna run it stock for a few days, but might be able to get a Vanatge run in maybe tomorrow after I Run MemTest on it. It's gonna be a long night tonight...maybe all night.

I'm still concerned about this 1.65v stuff. Mine meet the spec, but I'd love top be able to up the voltages and tighten them up if possible. Not at the expense of a 1000 dollar CPU though. I'm watching this thread, and check it two to three times a day. I got a feeling it'll be months before we really know for sure what will happen long term.

[dinos22]

word on the street is my kit should be in this week of their TR3X6G1600C8D. i will be running them on the EVGA X58 board. i can hit you up with some results as soon as i have them coming in. fire me a PM in that is the kit you are interested in Steve.

looking forward to your results Jody
man 32M is FAST on these CPUs eek
i did a quick run at 4.2GHz only LSC and got 8m38s 32M Spi

imagine tweaking it all
far out

[T_Flight]

i've started testing this weekend and my board likes 1.5-1.55v uncore voltage so i just use that hehehe

lets see how long this i920 lasts...i've gone up to 1.7vdimm as i didnt need more to run 2GHz 8-8-8-24-1T on these Corsairs yet ...just changed over my mounting gear on SS to LGA1366 so will do some cold runs

hopefully the CPU likes cold...down to -60C heh

once i start with SS i will run up to 2.3vdimm to see how these sticks like voltage

Dinos,
Fugger had a 965 down to -95 if memory serves. It was in one of the first two i7 threads he posted about. Don't know if that will apply to the 920's too. Just thought I'd mention it in case you didn't see it. I'm pretty sure that was the number, but I'm pretty tired right now. Kinda taking a break. I'm trying to get mine booted for the first time. The butterflies are churning too.

[T_Flight]

Guys,
I got a questiuon on this. Anyone know where this 1.65v figure came from and how it was arrived at? Why 1.65v? I'm just curious what the reason was for it, and why. I know that the JEDEC was 1.5, but they were running higher than that before, but with this chip it has thouse warnings and they put stickers on some boards covering up the RAM slots. It worries me because these things seem to be doing fine, but that number is still there making me wonder.

[fragmasterMax]

i'm using Gigabyte X58-UD5 board with i920 and Corsair 1866 CAS8 3x1GB kit

Has anyone hear had problems running high speed/volt ddr3 on their i7 mobos?
Is the initial startup voltage 1.65, or the default for the ram? Did you have to run some slower ram first to set volts, and then pop in the high speed ram?

[miahallen]

AFAIK, Intel actually released "safe" voltages...1.65V was the highest reccommended for a stable system running 24/7.

I didn't have any trouble with mine. I'm running the Eclipse with missmatched Kingston DIMMs (2x 2000MHz 9-9-9-24 & 1x 1833MHz 8-8-8-24). They worked first time on auto...and I'm currently benching with them at 1600MHz 7-7-7-18 1T with 1.75V.

These i7's are really nice with memory it seems. They don't seem to need near as much voltage as on the older x48/P45 chipsets.

[saaya]

1.4 min, 1.55 max for QPI/DRAM Vcore
1.9 min, 2.2 max for vDimm
2.0 min, 2.2 max for vPLL

During all the week end, testing P6T Dlx, and no problems. I'm under Swiftech GTZ.
I'll go on next week with SS/DI and i'll tell you what's happening.

cool, thanks for the feedback!

Dinos, 1.5v realy safe for regular usage? what board you using anyway?

weve been running 1.55 uncore aka vtt for a while without any issues so far...

Dinos,
Fugger had a 965 down to -95 if memory serves. It was in one of the first two i7 threads he posted about. Don't know if that will apply to the 920's too. Just thought I'd mention it in case you didn't see it. I'm pretty sure that was the number, but I'm pretty tired right now. Kinda taking a break. I'm trying to get mine booted for the first time. The butterflies are churning too.

theres barely any scaling going that cold tho... you can bench at -15 and get almost the same clocks as -75 :/
tried the intel, giga and asus boards, all the same :/

Guys,
I got a questiuon on this. Anyone know where this 1.65v figure came from and how it was arrived at? Why 1.65v? I'm just curious what the reason was for it, and why. I know that the JEDEC was 1.5, but they were running higher than that before, but with this chip it has thouse warnings and they put stickers on some boards covering up the RAM slots. It worries me because these things seem to be doing fine, but that number is still there making me wonder.

ask intel
thats the figure they came up with... its 10% of 1.5v... 1.5 plus .15 = 1.65v...

Has anyone hear had problems running high speed/volt ddr3 on their i7 mobos?
Is the initial startup voltage 1.65, or the default for the ram? Did you have to run some slower ram first to set volts, and then pop in the high speed ram?

havent heard a single thing so far, all sticks seem to boot up just fine

AFAIK, Intel actually released "safe" voltages...1.65V was the highest reccommended for a stable system running 24/7.

I didn't have any trouble with mine. I'm running the Eclipse with missmatched Kingston DIMMs (2x 2000MHz 9-9-9-24 & 1x 1833MHz 8-8-8-24). They worked first time on auto...and I'm currently benching with them at 1600MHz 7-7-7-18 1T with 1.75V.

These i7's are really nice with memory it seems. They don't seem to need near as much voltage as on the older x48/P45 chipsets.

yeah, we hit 2200 32m stable with only 1.7v vdimm on BloodRage last week, i dont know how well it clocks on BlackOps and other x48 boards, but DEFINATELY not to 2200 with 1.7v ^^
and its just a regular kit, not handpicked, and its not even samsung but micron d9gtr i think

[dinos22]

Has anyone hear had problems running high speed/volt ddr3 on their i7 mobos?
Is the initial startup voltage 1.65, or the default for the ram? Did you have to run some slower ram first to set volts, and then pop in the high speed ram?

no problems here man
it just starts up with SDP settings it has
i loaded the XMP profile
went to windows and did a lot of testing
superstable even with F2 bios (super super early bios)

[saaya]

SOME boards jack up vdimm vcore and vtt automatically when you load an xmp profile or set a high memory speed and leave the voltages on auto/default!
the voltages they load are supposedly safe, but you might want to set them a little lower and manually just to be safe.

bad news, i heard of one cpu that died with 2.25v vdimm...
might have been a ES cpu or ES board, im waiting for more details and will keep you updated.

[Benny Lodewijk]

SOME boards jack up vdimm vcore and vtt automatically when you load an xmp profile or set a high memory speed and leave the voltages on auto/default!
the voltages they load are supposedly safe, but you might want to set them a little lower and manually just to be safe.

bad news, i heard of one cpu that died with 2.25v vdimm...
might have been a ES cpu or ES board, im waiting for more details and will keep you updated.

YES.. you are right saaya, i have this experience with Patriot Viper DDR3-1600 6GB Kit. When XMP enable, suddenly vdimm set to 1.72 volt.

[fragmasterMax]

AFAIK, Intel actually released "safe" voltages...1.65V was the highest reccommended for a stable system running 24/7.

I didn't have any trouble with mine. I'm running the Eclipse with missmatched Kingston DIMMs (2x 2000MHz 9-9-9-24 & 1x 1833MHz 8-8-8-24). They worked first time on auto...and I'm currently benching with them at 1600MHz 7-7-7-18 1T with 1.75V.

These i7's are really nice with memory it seems. They don't seem to need near as much voltage as on the older x48/P45 chipsets.

cool, thanks for the feedback!

weve been running 1.55 uncore aka vtt for a while without any issues so far...

theres barely any scaling going that cold tho... you can bench at -15 and get almost the same clocks as -75 :/
tried the intel, giga and asus boards, all the same :/

ask intel
thats the figure they came up with... its 10% of 1.5v... 1.5 plus .15 = 1.65v...

havent heard a single thing so far, all sticks seem to boot up just fine

yeah, we hit 2200 32m stable with only 1.7v vdimm on BloodRage last week, i dont know how well it clocks on BlackOps and other x48 boards, but DEFINATELY not to 2200 with 1.7v ^^
and its just a regular kit, not handpicked, and its not even samsung but micron d9gtr i think

no problems here man
it just starts up with SDP settings it has
i loaded the XMP profile
went to windows and did a lot of testing
superstable even with F2 bios (super super early bios)

SOME boards jack up vdimm vcore and vtt automatically when you load an xmp profile or set a high memory speed and leave the voltages on auto/default!
the voltages they load are supposedly safe, but you might want to set them a little lower and manually just to be safe.

bad news, i heard of one cpu that died with 2.25v vdimm...
might have been a ES cpu or ES board, im waiting for more details and will keep you updated.

Thanks guys thats good to know