Nehalem Overclocking @ XS

[revogamer]

4.8 on air!!! damn me wanty

[Sampsa]

Oh boy, looking forward :)~

[Movieman]

Yes, I am flying to Fugger's Light city this week, and I am taking some serious Kung Fu with me, from the crazy parts, to the super Kung Fu unlocking reserved for Manufacturers ...
The point I was to make is not the max frequency, it is more about fact that a Manufacturer demo is kind of pointless, out of the 100 000 Core i7 we already shipped the day of launch, it is obvious that some of them are just awesome crazy chips.
A manufacturer demo of extreme Overclocking is rarely representative to what the CPUs are on the market.
Since Conroe, I am very carefull not to use Picked processors (When I start controling the perf demos) ... I am going to show you what happen when you actually use picked parts.
The process technology people are actually able to understand what limit a CPU, so, if you look for it, you find a part that will be out of the distribution, but will still quality to sell. Those usually end up into the extreme edition and finish running at high frequency aircool into an OEM build machine, or in a Boxed Extreme Edition. Now, we of course sort them, and we got some extreme extreme parts.
Manufacturer of processors should stick to demonstrate what they are selling, not something that they cannot quaranty to its customer.
when I showed Skulltrail at 4.0Ghz, I knew that thermal cooling was the limits, and the QX9775 were all able to run at 4.0Ghz, if properly installed. (we did not want to quaranty it, for obvious reasons, but it was possible, the only impact was to decrease the life of the CPU)
If a processor does not always archive a frequency, I refuse to show it publically, it avoid to set up expectation that you can not follow up with a real product. Those are my guide lines, and I think everybody is fairly happy of it. I ll break the rule this week , but I want all of you guys to understand that it is not what a CPU maker should be showing. It is about setting expectation.

This baby is running at 4.8Ghz aircooled, without pushing voltage much:

I know that this baby will fly, because mathematically speaking, it shows the right caracteristic.

I am waiting 5 mores like this Monday, directly from the men who sort them.

(Notice that I go super honnest on the disclosure side)

4800 on air? Ok, you have my attention..

[dinos22]

oh i see
that's fair enough francois
4.8GHz on air
that is some CPU.....
i wonder what will happen to those CPUs after you are done with them....if a couple of benchers have these and the rest of the ppl dont it's all over benching wise as we'll never be able to find such freak of nature CPUs

[Micutzu]

Hello Francois and thanks for the insight.

We all understand that there are chips and CHIPS, some clock better and some clock worse and the luck of the pick is very important. But i was more interested in the problem that comes up around 1.5V wich seems to be present on all Core i7 chips, can you confirm and maybe share some info about it ?

Thanks.

[Sampsa]

oh i see
that's fair enough francois
4.8GHz on air
that is some CPU.....
i wonder what will happen to those CPUs after you are done with them....if a couple of benchers have these and the rest of the ppl dont it's all over benching wise as we'll never be able to find such freak of nature CPUs

You think it hasn't happened already ...

[Movieman]

You think it hasn't happened already ...

My guess is that you just got off the phone with Francois!

[massman]

4.8G? What kind of stability are we talking about?

[Fr3ak]

A manufacturer demo of extreme Overclocking is rarely representative to what the CPUs are on the market.

Due to all respect, what makes you think AMD used some cherry picked CPUs? Probably they did, probably they did not. They say they didn't. For the next official Intel demo, they will pretend the same with only a few people knowing whether that is true or not.

First of all, as a skeptic enthusiast I don't trust any manufacturer. I want solid proof from people I know and respect knowing where their stuff comes from. What you are trying to proof with your "experience" is beyond my horizon. Looks to me AMD hit a nerve and Intel has to react, even if your plan is entirely your private ambition, still do I get that picture as an outsider. That aspect aside, your experiment will have an impact on the benching scene.

On the one hand, it is exciting to see how far the very cherry, cherry picked CPUs can go, on the other hand, I won't even bother benching Nehalem knowing my CPU is crap and that I would have to buy/get 10.000 or more CPUs to find one that is way better than average and still does not watch what people with inside sources can get.

Now bring it on, show us 7 GHz and discourage AMD and the rest of us ordinary overclockers! =)

[Movieman]

Due to all respect, what makes you think AMD used some cherry picked CPUs? Probably they did, probably they did not. They say they didn't. For the next official Intel demo, they will pretend the same with only a few people knowing whether that is true or not.

First of all, as a skeptic enthusiast I don't trust any manufacturer. I want solid proof from people I know and respect knowing where their stuff comes from. What you are trying to proof with your "experience" is beyond my horizon. Looks to me AMD hit a nerve and Intel has to react, even if your plan is entirely your private ambition, still do I get that picture as an outsider. That aspect aside, your experiment will have an impact on the benching scene.

On the one hand, it is exciting to see how far the very cherry, cherry picked CPUs can go, on the other hand, I won't even bother benching Nehalem knowing my CPU is crap and that I would have to buy/get 10.000 or more CPUs to find one that is way better than average and still does not watch what people with inside sources can get.
Now bring it on, show us 7 GHz and discourage AMD and the rest of us ordinary overclockers! =)

But hasn't that always been the case?
Someone shows up with a cpu either bought or gotten from some inside source that does speeds that no one else can match.
It doesn't mean that your chip or mine are crap, just that a better one exists.
That my friend is life and to ignore that just because something better may or does exist is a very negative way to think.
I got to 4226 on air but does seeing this thread with the claims of 4800 on air bother me? No, it does not because I knew I did my best with what I have.
Look at the glass as 1/2 full vs 1/2 empty and you will enjoy life so much more.

[boblemagnifique]

it's always the problem of "pretesting" which is well known among manufacturers of memory for exemple...

I think we must be honest (we overclockers & press) if we have doubts about a piece pretested , because this isn't the representative of all the other pieces on the market , It's not correct for readers and future buyers

I would like to see more and more Overclockers say how many cpus & cp selected for to found the Golden Pc (or if someone has done) in the threads of overclocking

[Fr3ak]

Dave, it's not a negative way to think, it is a realistic view on things after taking the costs and probability of finding a good CPU into account. In case you want to play in the top 10, you would need such a 4.8 GHz CPU, though I would love to see some details about that system. The 965 I played around with, can't even run at 3.8 GHz WCG stable for more than 16 hours. That was with 1.35V. More voltage and either the CPU or the heatsink starts to melt So, I would say, "my CPU is crap" =)

From the results XS members posted, it seems like there are a lot of CPUs that don't OC too well, especially on LN2, though you don't even need LN2 at -50 :P:

This experiment will proof nothing if you ask me. It is interesting for sure, but as a response to the AMD demo, it has some negative side effects to me. I had the pleasure to meet Francois at Intel Germany in Munich this year and what he told me back then does not seem to match the attitude he is showing right now =/

[Movieman]

Dave, it's not a negative way to think, it is a realistic view on things after taking the costs and probability of finding a good CPU into account. In case you want to play in the top 10, you would need such a 4.8 GHz CPU, though I would love to see some details about that system. The 965 I played around with, can't even run at 3.8 GHz WCG stable for more than 16 hours. That was with 1.35V. More voltage and either the CPU or the heatsink starts to melt So, I would say, "my CPU is crap" =)

From the results XS members posted, it seems like there are a lot of CPUs that don't OC too well, especially on LN2, though you don't even need LN2 at -50 :P:

This experiment will proof nothing if you ask me. It is interesting for sure, but as a response to the AMD demo, it has some negative side effects to me. I had the pleasure to meet Francois at Intel Germany in Munich this year and what he told me back then does not seem to match the attitude he is showing right now =/

The chip I got using a TRUE and 87CFM 120x38mm fan would do Sp1M at 4226, Cinebench at 4005 and for WCG I use 3733 with 1.35v
I'm sure we'll agree that I don't have contacts inside Intel to get a cherry picked chip but maybe the one I got is on the better side of the lot.
I can't tell as it's the only one I've worked with.
Now just because you or I don't get one of the top 10 chips available doesn't mean we should be upset with those that do.
My point remains the same, there will always be someone that has better than what we do and I imagine that 2 weeks from now Francois might even find a better chip than the ones he has now. just life my friend and because you or anyone else doesn't make it to 4800 on air is no statement as to your or their abilities, it's all in what you were working with.
As to Francois here, I think he was very forthcoming and hid nothing.
He's just making the point that there are chips that will do more when chosen for certain characteristics.
Had he showed up with 5 chips and said " I grabbed these off the line as I was leaving Friday" and then showed 4800 on air then yes,people would have a valid complaint, but he didn't do that.

[Fr3ak]

The excuse about using chips from the line is that they could be the worst CPUs ever or the very best ever produced

I have no doubts about Francois being dishonest, I completely trust him and what he says. But coming up with this new Intel demo to show what cherry picked CPUs can do implies that AMD was using cherry picked CPUs, thus making them look bad, regardless if it was true or not, we can't tell.

I only hope by disabling parts of the CPU, Francois will manage to find a cure fore the Nehalem coldbug from which everybody can benefit.

[Drwho?]

well, now that I got your attention ... the 4.8Ghz on air can be ritch if you turn off the diode that protect the CPU ... it is just running at 120 celcius ... If you can't turn it off, well, your thermal Diode will "save" the CPU and stop you from doing it, and make the processor throttle.
What I am trying to get everybody to understand, a full disclosure is required when you do a Manufacturer demo, because so many tricks can be used, it is too easy to play with.
Just by changing 1 Byte, i got 800Mhz ... hummmm
I think every body got the point now ... it is easy to claim Mhz ...

[dinos22]

You think it hasn't happened already ...

well if it does i'll quit overclocking for points and just mess around with stuff that i find interesting from now on like RAM/board tweaking or even bench AMD unless they do the same thing ...lol not that i was trying hard but there was motivation there and incentive as it was fairly level playing field

current CPUs charles has are also monsters but there was at least one user based bin which has similar clocks so that is still fine but topping current highs with some 6GHz chips for select few would be absolutely unacceptable

Dave there was NEVER a case that a retail user binned cherries could not match or beat cherry ES CPUs until now (i guess we will find this out on the weekend when they bench it)

[dinos22]

well, now that I got your attention ... the 4.8Ghz on air can be ritch if you turn off the diode that protect the CPU ... it is just running at 120 celcius ... If you can't turn it off, well, your thermal Diode will "save" the CPU and stop you from doing it, and make the processor throttle.Mhz ...

oh i see so you are just taking screens then
try priming that CPU and see what its max is

[Drwho?]

oh i see so you are just taking screens then
try priming that CPU and see what its max is

all the pics with be posted before Thanks giving week end.

[dinos22]

looks up the date for Thanksgiving week on google hehehe
[edit]27 Nov for non-americans

[Fr3ak]

I don't think anyone has doubts about the ability of manufacturers to manipulate products to get higher speeds, but usually it turns out whether the statements are true or not when the product hits retail. AMD presented a 3GHz Phenom at either Games Convention or Cebit (can't remember), that never hit retail. Even most early CPUs could not be overclocked to 3 GHz. Why you are in such a rush now to show that AMD might have used dirty tricks is still something I don't get. In 6 weeks we will know the result when first CPUs are out and overclocked.

I only hope you will make available what gives you such a performance boost. Changing 8 bits sounds rather complicated. Would be nice to have some solid proof. You said that you could do 6 GHz or that you have done 6 GHz on Nehalem. Interesting if true, but without proof it is just a claim. Now that you made clear, that the 4.8 GHz were done using "dirty tricks" as you might call it makes it sound more reasonable. The CPUs I tried so far, were reaching 85C loaded on 4 cores with about 3.8 GHz and 1.35 V.

[boblemagnifique]

looks up the date for Thanksgiving week on google hehehe
[edit]27 Nov for non-americans

3 days yet ^^

PS for Dino : i have see the match of Rugby yesterday in the stade, congratz for the team Austalian , impressive power & mastery !!!

[Brian y.]

What i understand from Fugger's post is that there's a Vcore limit after wich the current just goes insane

So the CPU at some point of increasing the Vcore will reach a threshold where it starts to pull more volts then what you have set it to due to the leakage?

[Drwho?]

I don't think anyone has doubts about the ability of manufacturers to manipulate products to get higher speeds, but usually it turns out whether the statements are true or not when the product hits retail. AMD presented a 3GHz Phenom at either Games Convention or Cebit (can't remember), that never hit retail. Even most early CPUs could not be overclocked to 3 GHz. Why you are in such a rush now to show that AMD might have used dirty tricks is still something I don't get. In 6 weeks we will know the result when first CPUs are out and overclocked.

I only hope you will make available what gives you such a performance boost. Changing 8 bits sounds rather complicated. Would be nice to have some solid proof. You said that you could do 6 GHz or that you have done 6 GHz on Nehalem. Interesting if true, but without proof it is just a claim. Now that you made clear, that the 4.8 GHz were done using "dirty tricks" as you might call it makes it sound more reasonable. The CPUs I tried so far, were reaching 85C loaded on 4 cores with about 3.8 GHz and 1.35 V.

for 6Ghz++ , we are going to try and see! The phylosophy of Nehalem design was all the opposite, best performance in 130Watts ... so, it is already nice to see all those power saving technology running at 5Ghz++. Remember, Nehalem has power gating, the influance of low temperature on those new kind of transistor has to be understood.
The next week experiment will help us to understand more.

[T_Flight]

I don't think anyone has doubts about the ability of manufacturers to manipulate products to get higher speeds, but usually it turns out whether the statements are true or not when the product hits retail. AMD presented a 3GHz Phenom at either Games Convention or Cebit (can't remember), that never hit retail. Even most early CPUs could not be overclocked to 3 GHz. Why you are in such a rush now to show that AMD might have used dirty tricks is still something I don't get. In 6 weeks we will know the result when first CPUs are out and overclocked.

I only hope you will make available what gives you such a performance boost. Changing 8 bits sounds rather complicated. Would be nice to have some solid proof. You said that you could do 6 GHz or that you have done 6 GHz on Nehalem. Interesting if true, but without proof it is just a claim. Now that you made clear, that the 4.8 GHz were done using "dirty tricks" as you might call it makes it sound more reasonable. The CPUs I tried so far, were reaching 85C loaded on 4 cores with about 3.8 GHz and 1.35 V.

Freak what you are forgetting is DrWho works with this stuff for a living. He can do practically anything with these and has the knowledge to do it. The guy is a professional. He knows exactly how these chips work and could make them do practially anything that is physically possible. I wouldn't question him for a minute. he is a highly educated, highly experienced professional.

Now saying that, he can have fun too, and if he wants to clock a chip to 6GHz more power to him. he is under no obligation to share how he did it no more than anyone else here. If he wants to he can, or if he doesn't that is fine too. If he posts the results that would be really cool. He doesn't have to even post a verification for me. His word is good enough for me.

[Movieman]

The excuse about using chips from the line is that they could be the worst CPUs ever or the very best ever produced

I have no doubts about Francois being dishonest, I completely trust him and what he says. But coming up with this new Intel demo to show what cherry picked CPUs can do implies that AMD was using cherry picked CPUs, thus making them look bad, regardless if it was true or not, we can't tell.

I only hope by disabling parts of the CPU, Francois will manage to find a cure fore the Nehalem coldbug from which everybody can benefit.

Does it? or is it simply a demonstration of what can be done and have nothing to do with AMD or anyone else?
Again, it's all in how you look at things and is the glass 1/2 full or 1/2 empty.
I look for the positive and try and ignore the negative.