Sandy bridge Owners Please post max multi and default vid

[OC Nub]

2600k-vid-1.250, 52x

2500k-vid 1.210, 54x boots, 55x no 4E code and freezes at windows load screen.

2500k-vid-1.220, 54x

I went by what the bios sets default with all power savings and turbo disabled, not cpuuz reading.

[Dumo]

L042B243 *A1174

1.25V - 57X

58X @99.8 max(F8x)

UD7 bios F7f

[poke349]

With EIST, C1E, C3, C6, speed step, turbo - all off, I'm getting:

BIOS/TurboV: 1.165
Idle: 1.192 - 1.200
Load: 1.152

Screenies later.

Almost bench stable at 52 x 100. Haven't checked if it's bootable though. I'll post again when I have more results.
But I won't be pushing my vcore too much. So whatever result I post later should be marked as "not maxed".

[Utnorris]

Well, this is where I am confused Chew. I have the Asus deluxe and I can set 68 in the bios, with my vcore at 1.54v and I can get to the Windows boot screen before it freezes. I can't go over 1.54v because I keep getting the "Overvolt protection" kicking in and I have no idea how to turn it off. But my point is, that I do not think just because I can get to the Windows logo that that is my max multi. By the way, the batch is L042B208, probably going to go swap it out tomorrow since I think it is not a good chip, 5Ghz at 1.46v seems to be stable.

[chew*]

Well, this is where I am confused Chew. I have the Asus deluxe and I can set 68 in the bios, with my vcore at 1.54v and I can get to the Windows boot screen before it freezes. I can't go over 1.54v because I keep getting the "Overvolt protection" kicking in and I have no idea how to turn it off. But my point is, that I do not think just because I can get to the Windows logo that that is my max multi. By the way, the batch is L042B208, probably going to go swap it out tomorrow since I think it is not a good chip, 5Ghz at 1.46v seems to be stable.

C3 C6 EIST off.

we have already mentioned this a couple times.

For higher vcore you need to go set voltage monitioring for vcore to ignored on ASUS, after that you can boot at 1.990 if your heart desires but i don't suggest it, instant death.

Those of you reporting less than 1.20v with 2600K are setting something wrong.

My Asus only reports that low with 2500K as it should.

[Utnorris]

ok, I will try it in a bit. I just downloaded a new bios so I need to flash to that first. I am getting random reboots and I am not sure if the chip is going or if it is a bios issue, so I am trying to solve that first.

[poke349]

A followup:

EIST, Turbo, C1E, C3, C6 - All disabled

BIOS: 1.165
Idle: 1.192
Load: 1.152

52x - 1.53v -> Benchable
53x - 1.53v -> Barely Benchable (no screenshot though)

I'm won't push vcore any higher. So put me down as "not maxed".
Note that I am running 4 x 4GB of ram and 5 hard drives in the machine. These might play a factor.


(click to enlarge)

[Utnorris]

ok, so it looks like for my chip, L042B208, x52 is my max multi. Best I can tell vid is 1.18v, that's the default vcore. It's a pretty dog of a chip, more reason for me to swap it out tomorrow.

[Falkentyne]

Chew,

My 2600k has 1.230v VID at 3.4 ghz and the max multiplier with PLL DISABLED is x49. Enabling PLL and the chip goes higher, but I haven't tried posting with higher than x52. Someone mentioned that the chips usually go +6 over the max multi with PLL disabled, after enabling PLL override.

You need to also clarify the original post:
Are those results with PLL override Enabled or Disabled?
Seeing how many x54-x57 results you have (I find it hard to believe there are cpu's that can do X57 with PLL override disabled), I'm guessing that is from "Enabled", right?

Can you distinguish the results from each setting?

[chew*]

Chew,

My 2600k has 1.230v VID at 3.4 ghz and the max multiplier with PLL DISABLED is x49. Enabling PLL and the chip goes higher, but I haven't tried posting with higher than x52. Someone mentioned that the chips usually go +6 over the max multi with PLL disabled, after enabling PLL override.

You need to also clarify the original post:
Are those results with PLL override Enabled or Disabled?
Seeing how many x54-x57 results you have (I find it hard to believe there are cpu's that can do X57 with PLL override disabled), I'm guessing that is from "Enabled", right?

Can you distinguish the results from each setting?

I'm pretty certain that everyone posting is using PLL enabled especially to test max multi

If not we would be seeing a very large amount of chips that couldn't post at even 50X.

We could also test that way, not really necessary though. 50X no pll = 55X with pll, and so on.

Your chip is a 54X if your doing 49 no pll.

[Falkentyne]

Ah, that's fine. I just don't want to jack up the vcore if that's needed to post at x55, which I'm guessing may be max (at least not right now, since I no longer have a fast backup system), and even though I have no problems setting BIOS settings with PLL disabled, not looking forward to repeated boot loops and/or having to clear CMOS if x56 starts looping forever :P

I'll definitely post the result though if I test that high.
Having PLL disabled and enabled results would see if there's a multiplier range that enabled helps over disabled or if its just random.

So far seems like my chip is "middle of the road." Runs hot at high volts though...

/rant on:
Can't complain, I guess. I am mad that I considered my QX9650 c0 to be "middle of the road", when in fact it was probably a golden chip; it needed 1.375v (the second QX) for 4 ghz. The person I sold it to could run it at 4 ghz at only 1.28v prime stable. Unhappy to know the old P5WDH and the 4-pin power was holding my chip back...

Last golden chip I had was a 600e that ran at 900 at default volts and 930 at 1.65v, 155 fsb. That was the highest I could take it due to Abit BH6 1.1 FSB limitation, 150fsb was stable with infineon ram only, and chipset fan cooled, and 155FSB was flaky stability even with max "WBxtune" tweaks as that was at the very limit. May have been able to reach 1 ghz on a newer board...
/rant

[Nostromo]

It's one off equivalent for 2500K reacted the same way.

L050932 One wprime run after it could never pass prime 95 with like up to 1.40v at 4.5, prior it was fine, kept dropping 2 cores, then eventually just stopped posting.

Other batches haven't seen that issue but a buddy had a C batch that he also lost, his cooling was only air though, he was also at 5.6+.

Se if i can get his info later.

Hmm, I have a chance to pick up an L050A i think 853 or 953 can't remember, should I pass? dont want a super degrade chip.

[Timoleros]

New tests today..!!
2500K batch#3044B310
Auto Vcore with turbo,EIST, C1 C3 and C6 disabled and default multi (x33)
Bios 1.195v... cpuz reads 1.176

Max multi with Air 56x but with lower bclk
Maybe i can go higher with better cooling..

[Utnorris]

I'm pretty certain that everyone posting is using PLL enabled especially to test max multi

If not we would be seeing a very large amount of chips that couldn't post at even 50X.

We could also test that way, not really necessary though. 50X no pll = 55X with pll, and so on.

Your chip is a 54X if your doing 49 no pll.

I believe I did mine with it it disabled. I will go back in and check.

[some_one]

I have always been wondering about this mysterious VID-thingy and power usage, so I'm trying to follow this thread for several reasons.

Just got the ThrottleStop, I'm not familiar with it yet. I hope the setting are right. This is still with turbo and c1, c3, c6=disabled in BIOS, and Power Option=Balanced in W7, but this is reporting 0.995 too.
O don't know why that Maximum Turbo Limit is reportorial 59 . I hope it's not confused

Well the VID-thingy is the voltage the CPU requests from the Voltage Regulator. Say for instance the CPU is operating at the 30x multiplier then Intel have programmed a Voltage for that particular multi which is sent to the Voltage Regulator telling it what Voltage to provide to the CPU. What the CPU ends up with though will depend on a few things such as loadline, offset voltage or fixed voltage and also there seems to be a change with load not attributed to loadline but the way the VID is programmed in Sandy Bridge.

Looking at your SS it can be seen that EIST, C1E, C3 and C6 are disabled but not C1. I think chew really means C1E when he says C1 so I think that's okay. You might find with turbo that the BIOS has disabled it by not reporting turbo ratios to the OS. If you want to be absolutely sure turbo is disabled tick the "disable turbo" box.

Your VID at 16x is reported as 0.995V and that is correct, it's just that IIRC your using fixed Vcore so the voltage regulator ignores what the CPU is requesting and gives whatever the BIOS wants instead.

The maximum turbo limit is what has been set by your BIOS, not what you are using but the maximum you can request from the CPU. IIRC this use to be 57 but I guess with a few people hitting 58x with the help of BCLK and the 57x multi then Asus decided to raise it a couple of notches. The maximum that could be programmed in theory is 255x. Also I believe there was a change with Sandy Bridge on how the turbo limits work so you might find they work in ThrottleStop only with Nehalem. Not too sure about that.

[crossg]

C3 C6 EIST off.

Those of you reporting less than 1.20v with 2600K are setting something wrong.

My Asus only reports that low with 2500K as it should.

Probably.
I went back and checked made sure Turbo, EIST, C1E, C3 and C6 were off then booted back into bios and got 1.096v in bios the same in windows fluctuates between 1.176 and 1.184v, then when I put a load on it it went to.1.144. LLC maybe doing something in the background?

[chew*]

very possible cross, will toss another 2600 in the pro this weekend and check what we had set.

[poke349]

very possible cross, will toss another 2600 in the pro this weekend and check what we had set.

Same, I'm getting 1.192 - 1.200 idle with all that stuff disabled and all other settings at stock.
I posted a couple of screenshots yesterday.

[splmann]

L042B166 one from 70

Max Multi 59

[zalbard]

Yeah, but your bclk is <100...
Don't most of them gain 1x or 2x multi with underclocked bclk?

[chew*]

Yeah, but your bclk is <100...
Don't most of them gain 1x or 2x multi with underclocked bclk?

Yep our 5743 vantage chip ( 1 from 18 on our own dime ) does 59 fine with underclocked bclk.....bclk need to be at 100 for this.....not that it matters.....

59X is boring, show me 5900 with acess to 70 chips........

[Utnorris]

Hmm, I have a chance to pick up an L050A i think 853 or 953 can't remember, should I pass? dont want a super degrade chip.

That's the one I picked up yesterday and I am at 5Ghz at 1.44v, but I think my MB is having issues with the latest bios, otherwise I think it could do it at 1.38v.

[Mad222]

@ Chew
I made a mistake on the reported VID I told u b4
That was 2 cores Vid...not accurate !
I'll update a full list when I finish the rest (2500k+2600K)

Here's one:-
2500K Default Vid 1.165v...x55 on some boards & 59 x 100 (with Eist/C1E/etc disabled) on others

[dejo]

My 038b214 will make it to windows and bench at 49x and bench with the pll overvolt off but will only make 51x with it on

[chew*]

My 038b214 will make it to windows and bench at 49x and bench with the pll overvolt off but will only make 51x with it on

Thats very interesting dejo.

What mobo are you using?