[Decivox]

Hey guys,

So I based my overclock off of this guys template. I have a Q9550 C1 running at a stable 4GHz, (471FSB 8.5multi). I am using a Gigabyte EP45-UD3P. My voltages are as follows.

***********Motherboard Voltage Control*******

Load-Line Calibration [Enabled]
CPU Vcore [1.39] (1.36 Windows idle and load)
CPU Termination [1.34]
CPU PLL [1.59]
CPU Reference [0,83]

MCH Core [1.4]
MCH Reference [0,88]
MCH/DRAM Refernce [0,93]
ICH I/O [1.57]
ICH Core [1.1]

My core temperatures differ between 65-70 C on load (keeping in mind maximum rated is ~72 C). I am just concerned that I may be giving my processor too much juice, as I hear they dont take alot that well.

Oh yea I am on air.

 

[mattlyall06]

cpu pll should be 1.57 that is the MAX

all else looks good. 1.39vcore is NOTHING

 

[dralb]

Quote:

Originally Posted by mattlyall06 cpu pll should be 1.57 that is the MAX all else looks good. 1.39vcore is NOTHING

I had thought there was no benefit for a PLL higher than 1.5v. Also, do these not suffer from failure casued by running higher than 1.3625? Granted, if LLC is off and the droop gets it below that (not that you can ever trust a software voltage reading 100%)

 

[Decivox]

Quote:

Originally Posted by dralb I had thought there was no benefit for a PLL higher than 1.5v. Also, do these not suffer from failure casued by running higher than 1.3625? Granted, if LLC is off and the droop gets it below that (not that you can ever trust a software voltage reading 100%)

Hmm interesting, I will lower the PLL to 1.5v again and see if its still stable. I have to leave to work now, so I will run OCCT and post my results when I get home in like 5 hours.

 

[mattlyall06]

Quote:

Originally Posted by dralb Also, do these not suffer from failure casued by running higher than 1.3625?

that has been debunked time and time again. 1.3625 is the max VID, not the max VCORE. there are people who have been running 1.4v and higher for months on end without a hitch (with a 45nm cpu)

edit: yeah, you might even get away with 1.5pll. i'd try that first then if that doesnt work go with 1.57

 

[dralb]

Quote:

Originally Posted by mattlyall06 that has been debunked time and time again. 1.3625 is the max VID, not the max VCORE. there are people who have been running 1.4v and higher for months/years on end without a hitch (with a 45nm cpu) edit: yeah, you might even get away with 1.5pll. i'd try that first then if that doesnt work go with 1.57

Years? Also, I was under the impression that VID is a rating and not adjustable. (as in, why would they release a VID limitation when it is static for each CPU?)

I am just curious as I know it was a problem right out of the gate, but I haven't heard of anyone having this issue lately and I know many have run higher. Might be moot to the OP, though, as I am not sure how much more he wants to run due to temps.

I need to brush up. Will be getting a 45nm eventually, lol.

 

[DuckieHo]

Quote:

Originally Posted by mattlyall06 that has been debunked time and time again. 1.3625 is the max VID, not the max VCORE. there are people who have been running 1.4v and higher for months/years on end without a hitch (with a 45nm cpu)

Really?

 

[mattlyall06]

ok ok, maybe not YEARS. but definitely for an extended period of time.

 

[10acjed]

http://download.intel.com/design/pro...hts/318726.pdf

Page 19 lists absolute maximum voltages for the Q8xx Q9xx as 1.45

Page 24 explains Vcc overshoot of .050v, meaning that there will be a load to idle voltage spike of .050v. IE: if your running 1.45v than at load to idle transition your cpu will recieve a voltage spike of 1.5v. If you run at 1.40v than the highest spike will be 1.45v and still within the max voltage of 1.45v.

These are all taken from intel datasheets and usually listed around the same place for each cpu.

VID (1.3625) is different than max voltage range (1.45)

Quote:

Table 2-2 specifies absolute maximum and minimum ratings only and lie outside the functional limits of the processor. Within functional operation limits, functionality and long-term reliability can be expected. At conditions outside functional operation condition limits, but within absolute maximum and minimum ratings, neither functionality nor long-term reliability can be expected. If a device is returned to conditions within functional operation limits after having been subjected to conditions outside these limits, but within the absolute maximum and minimum ratings, the device may be functional, but with its lifetime degraded depending on exposure to conditions exceeding the functional operation condition limits. At conditions exceeding absolute maximum and minimum ratings, neither functionality nor long-term reliability can be expected. Moreover, if a device is subjected to these conditions for any length of time then, when returned to conditions within the functional operating condition limits, it will either not function, or its reliability will be severely degraded.

 

[Decivox]

Quote:

Originally Posted by Decivox Hmm interesting, I will lower the PLL to 1.5v again and see if its still stable. I have to leave to work now, so I will run OCCT and post my results when I get home in like 5 hours.

So changing my PLL from 1.59 to 1.5 made my overclock unstable, so I will raise it to 1.57. So the PLL does make a difference.

Quote:

Originally Posted by 10acjed http://download.intel.com/design/pro...hts/318726.pdf Page 19 lists absolute maximum voltages for the Q8xx Q9xx as 1.45 Page 24 explains Vcc overshoot of .050v, meaning that there will be a load to idle voltage spike of .050v. IE: if your running 1.45v than at load to idle transition your cpu will recieve a voltage spike of 1.5v. If you run at 1.40v than the highest spike will be 1.45v and still within the max voltage of 1.45v. These are all taken from intel datasheets and usually listed around the same place for each cpu. VID (1.3625) is different than max voltage range (1.45)

So what you are saying, is that a Q9550 can run at 1.40 volts fine and not have their lifespan quickly degraded?

 

[10acjed]

Quote:

Originally Posted by Decivox So changing my PLL from 1.59 to 1.5 made my overclock unstable, so I will raise it to 1.57. So the PLL does make a difference. So what you are saying, is that a Q9550 can run at 1.40 volts fine and not have their lifespan quickly degraded?

Actually that is what intel is saying..

And a big thanks to Pizzaman for pointing this out to me a few weeks back on the Q6600 being 1.55v abslute max voltage.

Quote:

Within functional operation limits, functionality and long-term reliability can be expected.

Like I said, page 19 list functional operation limit up to 1.45v, when reading the datasheet you must also take into the account of the Vcc off shoot (spike) of .050 which leaves you at 1.40v..

It is probably the reason people are seeing the 45nm cpu running up near 1.4v with no issues...

This is also the same with FSB VTT voltages as well, I know everest reads that voltage.

My CPU FSB is set to 1.4 in bios for this screen.

 

[Decivox]

Alright, thanks a ton 10acjed.

P.S, the 1.57 PLL made my system stable again, and dropped the temps abit.

 

[EricM9104]

Quote:

Originally Posted by mattlyall06 cpu pll should be 1.57 that is the MAX all else looks good. 1.39vcore is NOTHING

on a 45nm Intel, yes.
1.3625v is the maximum reccomended vcore.

 

[10acjed]

Quote:

Originally Posted by EricM9104 on a 45nm Intel, yes. 1.3625v is the maximum reccomended vcore.

that is the VID range of the chip. That link to the .pdf data sheet from intel is a great read explaining all of it..

It explains Vcc overshoot (voltage spike) and is usually also the reason you see a difference between what voltage you set in bios and what voltage is actually applied to the chip at idle. It also explains why there is vdroop to account for the Vcc overshoot.

Essentially if you have LLC on or a vdroop mod it also raises the overshoot from load to idle (.050v).

It basically states that within a voltage range of -0.3 to 1.45v is considered the "functional operation limits" of the cpu.

The only time it suggests 1.3625v is when it is talking about the VID range of the chip, the VID rang is the area of voltage needed to operate. Meaning a 45nm cpu will never need more than 1.3625 to operate at default settings.

Quote:

Individual processor VID values may be calibrated during manufacturing such that two devices at the same core speed may have different default VID settings. This is reflected by the VID Range values provided in Table 2-3. Refer to the Intel® Core™2 Extreme Processor QX9000 Series and Intel® Core™2 Quad Processor Q9000, Q9000S, Q8000, and Q8000S Series Specification Update for further details on specific valid core frequency and VID values of the processor. Note that this differs from the VID employed by the processor during a power management event (Thermal Monitor 2, Enhanced Intel SpeedStep® technology, or Extended HALT State).

Again this is not me making things up as I go, this is directly from the intel data sheets made readily available at the same site we get our 1.3625 info from
http://processorfinder.intel.com/det...px?sSpec=SLB8V

 

[10acjed]

PS:

You must also understand that this is also true of the FSB VTT voltage, and if your pushing 1.45 - 1.5v to it your causing damage to your cpu. This was a pretty frequent read a while back over at xs.org that the FSB VTT voltage were causing chips to burn up when people were thinkin it was core voltage.

FSB VTT voltage can be read with everest, and as long as you are below 1.4v on that you will be safe, if above 1.4v you may be causing silent damage to the cpu and wont realize it until it is too late....

Heat is also a bad thing, running your cpu close to the voltage limit and heat limit is not recommended by me, I suggest staying a good 15 - 20c below max recommended if you are going to push the voltage limits.. 60c or lower is ideal...

The heat part is just opinion on my part. So take it or leave it

 

[Chris Ihao]

Personally I like to play it safe with costly components like this. Like it says in my sig, I have 1.352v going to my q9550 at idle, and 1.304 volts at full load @ 3.6 ghz (running p95). Max temp in a hot room is 67c, but I would never keep it at these temps under "normal" use.

I did not even dare to raise the vcore bios setting above 1.36250 before I had xflashed my p5e with a Rampage Formula bios, which had voltage monitoring in the bios.

Also, in my honest opinion it isnt wise to "gamble" that the specific piece of cpu chip can withstand a so and so high voltage setting above the listed recommended, like many people seem to do. Especially if you arent very rich and have financial capability to replace it without much difficulty.

I feel I can sleep well on my settings, and thats more worth than breaking 3.8ghz, so to speak

 

[Prong]

1.57 Cpu PLL is the max for 45nm Quads ? Who says that ? I thought it was 1.60 max.