Ivy Bridge's IMC v limit and 1.65 RAM. Any sign of degradation over time?

[El Mano]

I'm thinking about upgrading my RAM and searching for the new sticks I learnt that Ivy's IMC voltage is 1.5 rather than previous 1.65V.
Has anybody experienced any kind of degradation due to 24/7 RAM OC using more than 1.5V?

I'd like to keep my CPU for a few more years so if there's a real risk of damaging the chip I'll have to buy a slower kit.

Thank you!

[msimax]

I just had to replace a 3770k for bad mem controller after about a month running ram @1.72v

[hersounds]

From june 2012 to now i'm still in use Avexir core series 2800 Mhz xmp 1.3 default without troubles & degradation ---this kits use by default 1.65 V

[El Mano]

I just had to replace a 3770k for bad mem controller after about a month running ram @1.72v

I feared this. I've searched the internet and found a number of people who suffered the same issue: CPU dead after RAM OC @1.65v and higher.

On the other hand, a number of people claim to be able to use OC speeds at 1.5V. I'll try this and see what happens.
Thank you very much guys.

[dengyong]

I've used 1.65v memory with my 3770k since they came out with no problems. I'm running this 16GB 2400mhz C10 kit now. http://www.newegg.com/Product/Produc...82E16820231589

[st0ned]

I think vccIO and VccSA will be more prone to kill your CPU then vRam

[Ket]

I haven't experienced a CPU going bad due to RAM voltage, or I should say I haven't experienced it using RAM voltage up to 1.65v and a VCCSA voltage of 1v. Most good memory these days tends to only need about 1.6v to reach its max speeds anyway.

[sin0822]

You should realize that boards like the M5E and the Z77 Oc Formula both increase VCCIO and VCCSA without telling you when you set higher dividers. MSi and gigabyte and others don't do this as far as I know, but those other two do as to stabilize CPUs with bad IMCs. So DRAM voltage prob isn't killing the CPU, even though it is possible it could. However you need to ask youself, if they are OCing their memory to 1.7v+ 24/7 then how far are they OCing their CPU? lol most people OC CPU further than memory 24/7. But hey if the CPU dies then just get the warranty from intel.

[BlindFreddie]

Running high Vdimm without high VCCSA or VCCIO will kill or at least severely damage the IMC on Sandy Bridge & I assume Ivy is the same. And it may damage stuff in the CPU other than the IMC.

I have a very sick 2600K running in an Asrock Z68 Pro3-M:
When the CPU and M/B were brand new, I put in 2 x 1GB sticks of DDR3 which has JEDEC SPD specs at 1.5V.
Booted with default BIOS settings. Didn't enter BIOS at all in fact. Went ahead updating Windows drivers from the Q9650/P45 ones that were on the HDD from the previous machine.
After an hour or so, the machine started to crash with increasing frequency. Went into BIOS and discovered that the idiots at Asrock had selected the XMP profile of the RAM by default: 1600MHz C7 1.9V. All that UEFI cr*p and they didn't do important things like put tables of CPU max voltages that would trigger interactive user warnings before these thresholds could be exceeded.
I haven't played with BIOSes in any other Z68 boards, but I've done so with 3 Z77 boards (Asrock, Asus, GB) and none default to the XMP.

Anyway the IMC is severely damaged. Can only run RAM in 2 of the 4 slots, at max speed of 1033MHz, which is also the minimum. Vdimm above 1.54V causes further degradation. Running CPU at 4.1GHz, 1.23V (loaded), crunching WCG 24/7. Attempts at running Vcore more than 1.23V result in further degradation. About 1 year ago, the modem hung while I was away, the machine ran out of work & sat idle for about 5 days. The Vcore soared all the way to 1.24V (from 1.23) and the machine degraded some more. It was running at 4.2GHz before this event, now 4.1 is the max.

Bottom line: Tread carefully with the Vdimm control. Makes you wish for LGA775 again!

[Ket]

Running high Vdimm without high VCCSA or VCCIO will kill or at least severely damage the IMC on Sandy Bridge & I assume Ivy is the same. And it may damage stuff in the CPU other than the IMC.

I have a very sick 2600K running in an Asrock Z68 Pro3-M:
When the CPU and M/B were brand new, I put in 2 x 1GB sticks of DDR3 which has JEDEC SPD specs at 1.5V.
Booted with default BIOS settings. Didn't enter BIOS at all in fact. Went ahead updating Windows drivers from the Q9650/P45 ones that were on the HDD from the previous machine.
After an hour or so, the machine started to crash with increasing frequency. Went into BIOS and discovered that the idiots at Asrock had selected the XMP profile of the RAM by default: 1600MHz C7 1.9V. All that UEFI cr*p and they didn't do important things like put tables of CPU max voltages that would trigger interactive user warnings before these thresholds could be exceeded.
I haven't played with BIOSes in any other Z68 boards, but I've done so with 3 Z77 boards (Asrock, Asus, GB) and none default to the XMP.

Anyway the IMC is severely damaged. Can only run RAM in 2 of the 4 slots, at max speed of 1033MHz, which is also the minimum. Vdimm above 1.54V causes further degradation. Running CPU at 4.1GHz, 1.23V (loaded), crunching WCG 24/7. Attempts at running Vcore more than 1.23V result in further degradation. About 1 year ago, the modem hung while I was away, the machine ran out of work & sat idle for about 5 days. The Vcore soared all the way to 1.24V (from 1.23) and the machine degraded some more. It was running at 4.2GHz before this event, now 4.1 is the max.

Bottom line: Tread carefully with the Vdimm control. Makes you wish for LGA775 again!

1. Any settings defaulted to for memory in the UEFI are directly read from the speed data stored on your memory modules. If 1.9v was set, that is the fault of the memory manufacturer incorrectly setting parameters in the modules speed data, not asrocks.

2. "Soaring" voltage to 1.24v from 1.23v? That is a very, very minimal amount of fluctuation you can expect to see on any board.

3. High VCCSA and VCCIO is a bad idea... stick to default voltages generally for these. If you run high voltages here you will hurt the CPU.

On another note, my 2500k has recently died, it never had more than 1.4v going through it at 4.5GHz (about 1.36v actual under load), VCCSA was never higher than 1v, and memory voltage never higher than 1.6v. The CPU has died after roughly about 2.5yrs of use constantly running BOINC with load temps around 90c. Not sure why load temps were so high they always were nomatter what I did, anyway, a 3570k will be on its way soon.

[Planet]

1. Any settings defaulted to for memory in the UEFI are directly read from the speed data stored on your memory modules. If 1.9v was set, that is the fault of the memory manufacturer incorrectly setting parameters in the modules speed data, not asrocks.

That isn't necessarily true. Alot of boards these days do automatically adjust alot of the memory settings. Also if his XMP profile was rated at 1.9v that means that it was a fairly old kit that doesn't have a compatible profile with newer Ivy Bridge boards.

[Ket]

Settings are automatically adjusted in some cases if attempting to OC yes, or if somebody is a lazy OCer and just loads predefined included OC profiles, but the way the post read everything was default settings and the board "just decided" to change some settings, which would never happen.

[sin0822]

is there any XMP that sets over 1.65v?

[Splave]

Hmmm, I've had the same chip running 32m about 5 hours a day 5 days a week. Anywhere from 1.92 to 2.1v on my voltage hungry samsungs. Still as strong as ever...

As far as board auto raising voltage attached to memory divider....I'd hope your setting your own voltages, if your caring enough to run faster than standard mem.

[Ket]

is there any XMP that sets over 1.65v?

Not that I've ever seen and I even have a real old kit of Ripjaws dating back to the 1156 days yet even that kit only sets 1.6v via XMP.

[BlindFreddie]

The memory modules concerned are G-Skill "F3-12800CL7D-2GBHZ - DDR3-1600 CL7-7-7-18 1.9V", from my Asus P5K3 Delux board.
I also have 2 x 1GB Kingmax "FLGD45F-B8MF7 MAEH DDR3-1600" that run CL7. It's not printed on the labels but I think they need 1.9V for CL7 too. Labels date them as 2008 and 2009.

Ket: "1. Any settings defaulted to for memory in the UEFI are directly read from the speed data stored on your memory modules. If 1.9v was set, that is the fault of the memory manufacturer incorrectly setting parameters in the modules speed data, not asrocks."
I haven't recorded all of the timings in the SPDs, but as I said, there are (much slower) JEDEC timings with 1.5V set. It's only the XMP profile that sets 1.9V.
Every board that I've set up other than the Z68 Pro3-M has defaulted to JEDEC timings, and required the user to explicitly select the XMP.
If Asrock had done that, the damage would not have occurred.
They should also realise that there are such DDR3 modules out there and put a guard against applying more than 1.65V (say) without explicit user permission.
Too much "form over function"!

I haven't taken the risk of setting timings & 1.5V Vdimm manually with the current OCZ 1.65V sticks, swapping in the G-Skills and booting the modules again, as I don't know whether the manual settings would be preserved across memory module changes.
Even if it worked it would be too risky to leave that way in case the CMOS got cleared somehow.

Ket: "2. "Soaring" voltage to 1.24v from 1.23v? That is a very, very minimal amount of fluctuation you can expect to see on any board."
Yep! (Do you detect just a little hint of sarcasm maybe?) 1.23V is also a very low max voltage to avoid fast degradation of an SB CPU. The CPU core seems to have been damaged as well as the IMC. And Vcore has never been north of 1.3V, and that was only applied for a couple of min during initial setup experiments.

Agree re. VCCSA and VCCIO, tho higher VCCSA might get you a higher mem divider.
"... my 2500k has recently died ..."
I'm not overly surprised, and expect it was a combination of high Vcore, high temps, and time.
You might have got away with 1.36/1.4V, but 90 deg C with that would have been OTT. Remember that unlike metals, where electrical resistance increases with temperature, in semiconductors resistance decreases with temp, thus increasing the leakage currents and power, raising the temps even higher! In early transistor circuits, such thermal runaway was a major problem.
1.36V seems a lot for SB at only 4.5GHz. Maybe you had a slow leaky chip.
Back in the Q9650 days leeghoofd posted that he was backing off Vcore on all his machines to 1.25V, which I thought was rather conservative. But he was right, and it's still not a bad limit for current machines that are 100% loaded 24/7.
Your cooling must have been suboptimal too. Pushing a good HS like a TRUE with decent fan to 90C would require well over 250W I think.

Splave: "Hmmm, I've had the same chip running 32m about 5 hours a day 5 days a week. Anywhere from 1.92 to 2.1v on my voltage hungry samsungs. Still as strong as ever... "
Well, my experience was as I described, and I assume it was caused by Vdimm = 1.9V.
I'd advise getting some less hungry memory than your samsungs. DDR3 is cheap now. Have you found much performance increase from running your memory faster anyway?

"I'd hope your setting your own voltages, if your caring enough to run faster than standard mem."
I already had the mem sitting idle, and I didn't get time to set my own voltages before the damage occurred.

I haven't tried to RMA the 2600K as it's not Intel's fault, and putting 1.9V on the IMC would have voided the warranty, so it would be unethical too.

[Ket]

The memory modules concerned are G-Skill "F3-12800CL7D-2GBHZ - DDR3-1600 CL7-7-7-18 1.9V", from my Asus P5K3 Delux board.
I also have 2 x 1GB Kingmax "FLGD45F-B8MF7 MAEH DDR3-1600" that run CL7. It's not printed on the labels but I think they need 1.9V for CL7 too. Labels date them as 2008 and 2009.

Ket: "1. Any settings defaulted to for memory in the UEFI are directly read from the speed data stored on your memory modules. If 1.9v was set, that is the fault of the memory manufacturer incorrectly setting parameters in the modules speed data, not asrocks."
I haven't recorded all of the timings in the SPDs, but as I said, there are (much slower) JEDEC timings with 1.5V set. It's only the XMP profile that sets 1.9V.
Every board that I've set up other than the Z68 Pro3-M has defaulted to JEDEC timings, and required the user to explicitly select the XMP.
If Asrock had done that, the damage would not have occurred.
They should also realise that there are such DDR3 modules out there and put a guard against applying more than 1.65V (say) without explicit user permission.
Too much "form over function"!

I haven't taken the risk of setting timings & 1.5V Vdimm manually with the current OCZ 1.65V sticks, swapping in the G-Skills and booting the modules again, as I don't know whether the manual settings would be preserved across memory module changes.
Even if it worked it would be too risky to leave that way in case the CMOS got cleared somehow.

Ket: "2. "Soaring" voltage to 1.24v from 1.23v? That is a very, very minimal amount of fluctuation you can expect to see on any board."
Yep! (Do you detect just a little hint of sarcasm maybe?) 1.23V is also a very low max voltage to avoid fast degradation of an SB CPU. The CPU core seems to have been damaged as well as the IMC. And Vcore has never been north of 1.3V, and that was only applied for a couple of min during initial setup experiments.

Agree re. VCCSA and VCCIO, tho higher VCCSA might get you a higher mem divider.
"... my 2500k has recently died ..."
I'm not overly surprised, and expect it was a combination of high Vcore, high temps, and time.
You might have got away with 1.36/1.4V, but 90 deg C with that would have been OTT. Remember that unlike metals, where electrical resistance increases with temperature, in semiconductors resistance decreases with temp, thus increasing the leakage currents and power, raising the temps even higher! In early transistor circuits, such thermal runaway was a major problem.
1.36V seems a lot for SB at only 4.5GHz. Maybe you had a slow leaky chip.
Back in the Q9650 days leeghoofd posted that he was backing off Vcore on all his machines to 1.25V, which I thought was rather conservative. But he was right, and it's still not a bad limit for current machines that are 100% loaded 24/7.
Your cooling must have been suboptimal too. Pushing a good HS like a TRUE with decent fan to 90C would require well over 250W I think.

Splave: "Hmmm, I've had the same chip running 32m about 5 hours a day 5 days a week. Anywhere from 1.92 to 2.1v on my voltage hungry samsungs. Still as strong as ever... "
Well, my experience was as I described, and I assume it was caused by Vdimm = 1.9V.
I'd advise getting some less hungry memory than your samsungs. DDR3 is cheap now. Have you found much performance increase from running your memory faster anyway?

"I'd hope your setting your own voltages, if your caring enough to run faster than standard mem."
I already had the mem sitting idle, and I didn't get time to set my own voltages before the damage occurred.

I haven't tried to RMA the 2600K as it's not Intel's fault, and putting 1.9V on the IMC would have voided the warranty, so it would be unethical too.

- Your clue is in their title They are a kit that require 1.9v so the mainboard reads that from the speed data on the modules, you needed to adjust timings, voltage and speed manually. No mainboard has pre-programmed JEDEC standards in its BIOS / UEFI, every mainboard simply reads the speed data from the memory modules and loads pre-programmed failsafe values from the modules speed data. It was also a bit daft not to research your new system before porting over old hardware but hey I'm sure we have all made a oversight before. If you had checked though, you would of found out very quickly that its recommended not use use higher than 1.6v on the memory for 24/7 usage due to the IMC being sensitive to voltage.

- My 2500k died still for reasons unknown really, while 90c is on the higher side of things for 1.36v @ 4.5GHz its still a good 15-20c away from the Tjmax so the CPU really should of lasted about another year at those temps. I always suspected the solder used on the 2500k between the core and IHS wasn't applied properly when being manufactured and thats why core temps hit 90c. Now the CPU is half dead I might just try and knock off the IHS to find out for certain.

- I wouldn't say its unethical to RMA your CPU as in all honesty you made a genuine naive mistake when it come to the memory, thats not a crime and I would think even intel wouldn't mind RMAing a CPU for such cases.

[sin0822]

unethial to RMA a CPU? Intel just writes off losses into their taxes, since they are an american company some years i bet they wish more people would RMA.

[Ket]

True, if intel wanted to they could probably write off every RMA as a business expense.

[Alex-Ro]

I ran 1.9V for so many hours on different ivy chips that i can't count,no problem.For daily-use,i can't pronounce a verdict as i never exceeded 1.6 ,minimal gains from 2400 upwards,but i can say one thing for sure.Ivy is a far more robust animal compared to sandy-bridge,sandy-bridge was like an immature girl,i've seen a lot of chips that died for absolutely no reason,even at stock vcore,far more than other generation.

"I haven't tried to RMA the 2600K as it's not Intel's fault, and putting 1.9V on the IMC would have voided the warranty, so it would be unethical too."

Oh man,this is just poetry at it's finest,keep dreaming...