sandybridge take note

Quote:

---

Originally Posted by Movieman

Boy did you take the words right out of my mouth.
5GHz will be everyones goal but I can understand that.
They see benches from "talented"OC'ers showing 5.4.-5.5 and everyone builds the system and before ever booting to windows sets the bios at 5GHz..:rofl:
If I can offer anything it's that these chips will have variables.
We'll see the dogs that need 1.4v for 4200 and the gems that will do 5.5 at under 1.4v..
Learn the chip,see what it does and slowly bring it up.
Be smart here and be happy with what your chip does and forget what anyone else gets except for general knowledge for your own testing.

---

Wise words from Mr. Movieman :up:

I feel as though yes there can be processor degradation but from what i heard most of the dead CPUs were from too much BLCK, which I still don't quite understand. Now in the past Intel recommended 1.375v for bloomfield and then vabsolute was 1.55v. Now I don't understand why intel has to publish a datasheet and just say VID upto 1.52v. From what i take with that is if you keep voltage at auto vid will change upto 1.52v when processor is OCed to a very high frequency like 5-5.2ghz. i have actually seen my VID goto 1.475v at 5.2ghz. i did see voltage degradation from 1.53v and then the week after i needed 1.56v and then 1.58v, now that was at 5.2ghz on a D1 stepping chip. i did make note of it in my article, those are with LLC added. From what i saw was that the temperatures on air were very high at auto, and it is well known that heat and voltage will kill a chip. From talking to a lot of people I have gathered that they do not just die because of nothing, extreme overvolting and heat can result of the death of any chip. Companies like MSi that have published OCing guides have stated that high voltage is 1.55v. Now i understand that 1.5v is a lot to fathom for a 32nm chip, but heck so is 5+ghz.

it does seem like retail chips are much weaker than initially thought, their average OCes are not 5ghz, more like 4.5-4.8.

Quote:

---

Originally Posted by Movieman

If I can offer anything it's that these chips will have variables.
We'll see the dogs that need 1.4v for 4200 and the gems that will do 5.5 at under 1.4v..
Learn the chip,see what it does and slowly bring it up.
Be smart here and be happy with what your chip does and forget what anyone else gets except for general knowledge for your own testing.

---

I like all my chips to be nice clockers... a plain "ole" average chip just doesn't cut it... to be sold off at the feed lot.

Quote:

---

Originally Posted by sin0822

it does seem like retail chips are much weaker than initially thought, their average OCes are not 5ghz, more like 4.5-4.8.

---

I'm finding that with my chip as well. 4.7 is pretty close to stable, 4.8 isn't stable yet.

Makes me happy to see you say that too, I've been over your OC guide so many times now asking myself, what the hell am I doing wrong ?

Quote:

---

Originally Posted by PolRoger

I like all my chips to be nice clockers... a plain "ole" average chip just doesn't cut it... to be sold off at the feed lot.

---

That's ok but you can find out if they aqre gems or dogs without killing them.
That's really the point I was trying to make.
For the other guy just topped 5GHz in 3 mins after initial boot up.
Ran SP1m at 5088mhz.
Seems to be stable at 4588mhz on 1.35vcore.

yes just like with my 5.2ghz chip i can run spi, but i have to drop down to 5.1 for benches with .6v less, b/c of heat. The 4.8ghz chip will do 4.8ghz and bench 4.8ghz it has room with voltage, but I think no CPU PLL for D1 is killer. Maybe Intel is afraid that it would make their retail D2s look bad, which they are doing on their own. I have been looking around very busy places looking for SB OCes, maybe 1 over 5ghz, one at 5ghz, and a lot under 5ghz, more between 4.5-5. Here people are hitting over 5ghz no problem it seems tho, but this is one place that real OCers gather.