Sandy bridge Owners Please post max multi and default vid

[chew*]

Collecting some data, may or may not be usefull.

Time will tell.

Model of cpu.

Default vid of cpu. Turbo , C states, EIST must be disabled and stock multi must be used.

max post multi, etc the multi you can get past blinking cursor or 4E post code, or whatever the multi limit is of your chip. Cstates 1E 1 3 6 disabled, EIST disabled.

2500K

1.156 58X
1.140 55X
1.175 55X
1.195 55X
1.220 50X?

2600K

1.230 vid 57X
1.230 vid 56X
1.240 vid 55X
1.215 vid 55X
1.210 vid 54X
1.220 vid 54X
1.220 vid 53X
1.245 vid 52X
1.240 cid 52X
1.240 vid 52X
1.245 vid 52X
1.250 vid 52X
1.245 vid 51X
1.250 vid 51X
1.250 vid did not test high multi... maybe 51x?
1.265 vid 50X

[thebanik]

hmmm...damn nice idea, since I guess no1 has been able to find a way of selecting the best chips. Will pass on my info in a while from a Dud CPU which does only 5.2Ghz

[some_one]

Default vid of cpu.

Care to expand a little on this. VID changes not only with multi, up to a point, but can also change several steps depending on load on the same multi.

EDIT: I've add some values for 2500k. Note how the VID drops off at 37x, I would imagine this would happen at 38x for 2600k.

[chew*]

Care to expand a little on this. VID changes not only with multi, up to a point, but can also change several steps depending on load on the same multi.

EDIT: I've add some values for 2500k. Note how the VID drops off at 37x, I would imagine this would happen at 38x for 2600k.

Default VId for all chips regardless of model is different.

This is the voltage your chip has with all the crap turned off and no turbo, c1 eist, c3, c6. It's the voltage intel gives the chip for whatever they consider is stable for that piece of silicon while staying in its TDP rating.

Its why some people are benching 5.5 at 1.48v.......and some need 1.65V............ its also why some people can prime at 4.5 with 1.20 or less volts and some need 1.35v for the same speeds......

The question is.......is intel locking low leakage chips, or are they locking potentially good ocing chips as well, etc high leakage.

Your table indicates 1.220 for 33 multi which is stock, I talked with someone that has a vid at 1.050, 1.14, and 1.156 for default.....

What I am seeing is trends with batches, if not by max multi, but with voltage required for XXX.

A chip with unlocked 55 multi but needs 1.65 to do 55X probably won't last long.....also a chip like that is probably not desireable for 24/7 users.

I have seen multis all over the place in batch L050A933.........but one thing has remained consistent..............every chip needs retarded volts for like 5.4 or higher........in the 1.65 range......All chips are also reporting a rather high default vid..

[Hoss331]

The vid on my good 2600K is 1.23, boots up to 57x now but probably 58 when its cold. The chip posts and starts loading windows at 56 x 104 but doesnt make it into windows. The other chips I tested were 1.245vid - 52xmax, 1.245 - 51x, and 1.25 - 51x.

[chew*]

All same batch hoss?

[Hoss331]

The 57x chip and 1.25 vid 51x chip are of the same batch L045A912 but the other 2 with the 1.245 vid were 2 separate batchs.

[some_one]

Default VId for all chips regardless of model is different.

This is the voltage your chip has with all the crap turned off and no turbo, c1 eist, c3, c6. It's the voltage intel gives the chip for whatever they consider is stable for that piece of silicon while staying in its TDP rating.

So just to clarify, you just want just the HFM at idle without power saving? Note that C1 AFAIK isn't usually an option in the BIOS and needs to be disabled by the OS. BIOS option is usually enhanced C1 (C1E).

its also why some people can prime at 4.5 with 1.20 or less volts and some need 1.35v for the same speeds......

Hey that sounds like my chip, the 1.35V one

I talked with someone that has a vid at 1.050, 1.14, and 1.156 for default.....

That's quite a bit of difference. Could it be the PLL that's holding the lower VID chips back?

[chew*]

So just to clarify, you just want just the HFM at idle without power saving? Note that C1 AFAIK isn't usually an option in the BIOS and needs to be disabled by the OS. BIOS option is usually enhanced C1 (C1E).

Hey that sounds like my chip, the 1.35V one

That's quite a bit of difference. Could it be the PLL that's holding the lower VID chips back?

Yes to line 1.

Not sure, I think its intel holding the chips back but curious if they are holding back potentially good chips etc high leakage.

[babalouj]

So my chip does x59 multi and starts to load windows before it freezes (no clue about stability at higher volts, was just testing for blinking cursor). It sucks though since all I want is a chip that does 5ghz with low volts..... I have a 2600k L050A933 SN: A1948 default VID of 1.24V. It takes me 1.38V for 4.8ghz on water though.

[chew*]

So my chip does x59 multi and starts to load windows before it freezes (no clue about stability at higher volts, was just testing for blinking cursor). It sucks though since all I want is a chip that does 5ghz with low volts..... I have a 2600k L050A933 SN: A1948 default VID of 1.24V. It takes me 1.38V for 4.8ghz on water though.

Heh I will gladly trade you a chip that does low volts at 5 gig if that indeed posts at 5.9.

Seems very odd as I have been all through that batch and max is 54X so far however the voltages your reporting are very normal for that batch.

No doubt your getting them from Cambridge like myself.

I would go conservative with volts, that batch degrades very quickly.

Will check through our chips this week, I think we have one that might suit your needs, only needs 1.48 for 5.5 benchable.

[poke349]

According to TurboV, my default vcore is 1.165.
If I prime it with everything at stock settings, the load vcore is 1.168.

Is that the number you're looking for?


btw, I haven't tested my max multi - nor do I plan to test it. I don't have the balls or the cooling to put more than 1.5v into my chip...

EDIT: It is benchable at 50x. But I don't plan on going any higher.

[babalouj]

Here is a video of my 59x multiplier. http://www.youtube.com/watch?v=BI3__FvF0VU

This is what you are talking about right?

[Dumo]

[QUOTE=chew*;4754334]Heh I will gladly trade you a chip that does low volts at 5 gig if that indeed posts at 5.9.

QUOTE]QFT

@Babalouj...Can you try it with H2O? Most likely if you can boot up to that point then it will go 57X with better cooling than air

@chew*....Micro Center in Cambridge? Probly it used to be tweeter or compusa long ago...talking about my first Intel 486DX

[babalouj]

I am on watercooling. Check the youtube link of the video I just made. Oh and chew, I did buy this at the Cambridge MC

[chew*]

Here is a video of my 59x multiplier. http://www.youtube.com/watch?v=BI3__FvF0VU

This is what you are talking about right?

Yes sir that is the ticket right there, all i can say is wtf

I have to return a fallen soldier tommorow as well, preety sure i know where your golden low volt chip is batch wise.

Whether I can persuade them to give me one is a diff story.

From trend I see it should be L042B143 for 2600K.

Batches from 2500K to 2600K are 1 offs.

Its equivalent is L042B144 for 2500K and they do this. My chip 55X 1.175

[Sam_oslo]

Default VID=0.990 here. At least this is what I get with all C-states=Auto in BIOS, and Power Option=Balanced in W7.



It has been doing this, so far, didn't try higher OC yet, at least not seriously.

2600k Batch# L040B165, @4.5GHz 1.27v 1-hour Prime Small FFT

[chew*]

C states, eist, turbo need to be disabled sam for correct vid also default multi.

.990 seems to low for 2600K

[Sam_oslo]

This is with turbo and c1, c3, c6=disabled in BIOS, and Power Option=Balanced in W7.



Still the same. Just booted up and there was some activity, couldn't wait, LOL, and got 0.005 more.

[chew*]

This is with turbo and c1, c3, c6=disabled in BIOS, and Power Option=Balanced in W7.



Still the same. Just booted up and there was some activity, couldn't wait, LOL, and got 0.005 more.

yah windows 7 balanced + power saving is screwing with you then,thats the voltage its using when it drops speed to 1600.

Whats bios report the voltage at when auto idling in bios, should be 1.1+ more than likely 1.2 something on 2600K.......

[Sam_oslo]

OK, I'll try to look in BIOS. I'm running out now, and will post it in a few hours or so.

[dinos22]

Here is a video of my 59x multiplier. http://www.youtube.com/watch?v=BI3__FvF0VU

This is what you are talking about right?

does it boot at 58x?

[chew*]

does it boot at 58x?

It may be he has EIST, C1 C3 and C6 on dino's gonna double check with him tommorow.

If i'm right thats not a 59X chip actually.

If i'm wrong then this data will be preety useless but possibly not, batch may skew things a bit, IIRC in that batch I saw low 51X/52X multi chips all the way at 1.265vid.

There should be a trend of lower vid with higher multi and higher vid lower multi.

There should also be some vid's and multis washing over each other but not by much like the 1.245vid at 51x and 52x likely due to batch variation.

What it all means and how it can help us remains to be determined.

I have a couple ideas but I need large scale data.

[chew*]

According to TurboV, my default vcore is 1.165.
If I prime it with everything at stock settings, the load vcore is 1.168.

Is that the number you're looking for?


btw, I haven't tested my max multi - nor do I plan to test it. I don't have the balls or the cooling to put more than 1.5v into my chip...

EDIT: It is benchable at 50x. But I don't plan on going any higher.

Bios will report your default vid, auto vcore, auto multi, all power saving features plus turbo off, reboot check volts. That will be your hard coded vid intel gave chip.

You don't have to bench your chip or boot to windows, set turbo on C1 on everything else disabled increase turbo multis till you get a blinking cursor hang when posting, you don't even need unsafe voltage.

I really need accurate data for this to make any sense.

[Sam_oslo]

If I set turbo and c1, c3, c6=disabled, and vCore=Auto in BIOS, then I get vCore=1.224v in BIOS .