Nehalem Overclocking @ XS

[TheKarmakazi]

I think that saying this is a response to the new AMD hitting 5ghz is a little unfair. Weve seen retail nehalems at 5+Ghz already, no need to prove that. I think DrWho is just trying to shed some light on these new chips for us overclockers! We should be thanking him for his valuable insights and intricate knowledge of this architecture. Not trying to put the Red vs. Blue spin on it (god knows there are enough threads about that s(h)ift already just look in the news section).

[T_M]

I think that saying this is a response to the new AMD hitting 5ghz is a little unfair.

And my outsider's opinion is that is exactly the case, or why else 'advertise' it?
Intel doesnt need to fly to Fugger's house to test cherry modded CPU's to simply get a better understanding of what happens at low temperatures
Flying to Fugger's house on the other hand does get a lot of publicity

I really hope these modded/super cherry parts never make their way onto hwbot btw.

[Drwho?]

I think that saying this is a response to the new AMD hitting 5ghz is a little unfair. Weve seen retail nehalems at 5+Ghz already, no need to prove that. I think DrWho is just trying to shed some light on these new chips for us overclockers! We should be thanking him for his valuable insights and intricate knowledge of this architecture. Not trying to put the Red vs. Blue spin on it (god knows there are enough threads about that s(h)ift already just look in the news section).

Dudesss, I am like anybody else, I like competition, this is fun when AMD start showing the teeth

Before I compete, I like setting up the rules, and explain the themes ... I just did that. Some people think that I was "shooting" at AMD, but in fact. I was just spelling the rules of the game (Since they did dissable the thermal diode, allowing them to do the trick I did to get to 4.8Ghz)

Guys, I really want to be fair and honnest, do not give me the bad guy shirt here, I went open 100%.

Francois

[T_M]

I dont think anyone is saying you are a bad guy Francois
Just people questioning your motives, which you have just explained.
If this is being done in direct retaliation to AMD's recent event, then Intel would not be the first company to do so (wrt a competitor). Seen it many times!

So if you dissable the diode, can you also run Intel processor at -196C pot temperature?

[TheKarmakazi]

Absolutely I think its great that AMD has an ace up their sleeve. Competition makes all aspects better for the consumer (price, performance, choices). Would love to OC some AMDs again, and have them neck in neck in terms of performance against an intel. Thats why I havent purchased an i7 yet, I'm waiting too see how AMD will perform.

Im just very interested in any info on the coldbugs, and potential things happening at extreme OC's (which I believe was this threads intent). Whether or not AMD was using cherry picked cpus or not is another matter completely. I like the transparency being displayed here also, because that way it heads off any speculation.

[Charles Wirth]

Its an honor to be given this opportunity as well as the knowledge to be gained is invaluable.

My current chip is a good chip, the chip that Francois is showing is a cherry chip. I see that it scares Oliver, heh

I have seen the Amps fly on many boards and many chips, the voltage bug Peter is talking about may be the same, we need a better description.

[eva2000]

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 ...

woah 120C.. interesting to see how long the cpu lasts at those kinds of temps too

[miahallen]

Dudesss, I am like anybody else, I like competition, this is fun when AMD start showing the teeth

Before I compete, I like setting up the rules, and explain the themes ... I just did that. Some people think that I was "shooting" at AMD, but in fact. I was just spelling the rules of the game (Since they did dissable the thermal diode, allowing them to do the trick I did to get to 4.8Ghz)

Guys, I really want to be fair and honnest, do not give me the bad guy shirt here, I went open 100%.

Francois

What are the chances we will also be able to disable the thermal diode?

[Drwho?]

What are the chances we will also be able to disable the thermal diode?

None, this is too dangerous, one booting without the right voltage and cooling is fatal.
Trust me, it would become a dissaster, it is like removing the self preservation of the human kind ... watch the movie "The happening".

[miahallen]

None, this is too dangerous, one booting without the right voltage and cooling is fatal.
Trust me, it would become a dissaster, it is like removing the self preservation of the human kind ... watch the movie "The happening".

That's what I figured

[boblemagnifique]

None, this is too dangerous, one booting without the right voltage and cooling is fatal.
Trust me, it would become a dissaster, it is like removing the self preservation of the human kind ... watch the movie "The happening".

there must be related to the Cold boot / Cold bug, it seems that Hipro has managed to circumvent the problem with a prototype he used to contest MSI Europe in Amsterdam i think

many uses LN2 here, but in some countries the price is very expensive, the most economical a long term is the cascade into 2 or 3 stages but it's difficult to adjust the temp when the cpu cold boot quickly

it looks like 920/940 for the best cooling seems to be the WaterChill / dod for the base clock is limited by cold. I have test a few the 920 / 940 step B0 and it's better than my 940 C0 ES (not lucky )

For the 965XE, it seems the "petit bonheur la chance" ....

Nehalem of overclocking is fairly easy to find the limit cpu but very difficult to find the golden chips in Cold

[dinos22]

What are the chances we will also be able to disable the thermal diode?

None, this is too dangerous, one booting without the right voltage and cooling is fatal.
Trust me, it would become a dissaster, it is like removing the self preservation of the human kind ... watch the movie "The happening".

boooooo

party pooper

[massman]

None, this is too dangerous, one booting without the right voltage and cooling is fatal.
Trust me, it would become a dissaster, it is like removing the self preservation of the human kind ... watch the movie "The happening".

I think you missed the title of these forums: XTREMEsystems.

If we don't kill our cpu's by disabling the thermal diode (via PECI?), we certainly will by using excessive voltage to get the cpu to run higher frequencies

[T_M]

None, this is too dangerous, one booting without the right voltage and cooling is fatal.
Trust me, it would become a dissaster, it is like removing the self preservation of the human kind ... watch the movie "The happening".

For many of us here, that is the case every time we boot up. So why not give us the option?

[Charles Wirth]

Right now there is no option to disable it in bios, it might be the same as cutting the THERMTRIP trace.

[Harshal]

Intersting discussion(s) in last few pages.

Looking forward to the results

[Sampsa]

I really hope these modded/super cherry parts never make their way onto hwbot btw.

Imho live overclocking events will become more and more valuable where everyone have exactly same parts / time restrictions.

[massman]

Right now there is no option to disable it in bios, it might be the same as cutting the THERMTRIP trace.

That has been tried before on the Wolfdale and failed to remove the coldbug. Afaik, the Thermtrip trace is only the consequence of thermal issues, so even if you'd cut the trace, the issues would already have occured and therefor cutting it has no effect on the coldbug. Hipro already has made some sort of coldbug remover and he did say it hasn't got anything to do with the Thermtrip.

My bet would be on the Prochot trace. As it's possible to change this trace from uni-directional to bi-directional, you can change it from being solely an output to being both output and input. Now, the Prochot trace is connected to the TCC, Thermal Control Circuit, which is designed to protect the processor from any damage in case of thermal issues. If the processor has thermal issues, the TCC can either:

-Alter the Frequency or voltage of the processor in order to get the cpu within safe temperature range
-Turn on Clock Modulation, which apparently is rapidly turning the clocks off and on at a duty cycle which should reduce power dissipation by about 50%.

Normally, the when the TCC is enabled, the Prochot signal is turned on, which indicates that the TCC limits have been crossed. Normally, this is an output, however, if you change it to an input signal, you make it possible to influence the TCC, telling it there's no temperature problem. If you then force Prochot to be disabled at all times, the TCC won't kick in as it's informed by Prochot.

This is not a fact, people, just a theory. I might be off by miles!

[Frontl1ne]

Some people think that I was "shooting" at AMD, but in fact. I was just spelling the rules of the game (Since they did dissable the thermal diode, allowing them to do the trick I did to get to 4.8Ghz)

DrWho, not having a stab at you or anything, but do you have solid evidence that AMD did disable the thermal diode and/or use cherry picked chips when they did the demo?

Edit: Am thoroughly interested to see how well your cherry picked chips oc as well. Sounds like a monster so far.

[Fr3ak]

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.

Don't get me wrong. I know exactly who DrWho? is. He is a genius in my eyes and I enjoyed every minute of talking to him. I don't question his motives or him being dishonest, but the way he introduced his OC demo in the AMD thread was a bit unlucky, to say the least.

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.

Propagating this upcoming demo in a thread that is about AMD's OC demo gives it a negative look and the impression that it has something to do with AMD, you can't argue with that I am talking about this thread btw: http://www.xtremesystems.org/forums/...d.php?t=208369

Its an honor to be given this opportunity as well as the knowledge to be gained is invaluable.

My current chip is a good chip, the chip that Francois is showing is a cherry chip. I see that it scares Oliver, heh

I have seen the Amps fly on many boards and many chips, the voltage bug Peter is talking about may be the same, we need a better description.

I don't doubt that this gathering will bring us more insight into this phenomenum. Probably I haven't been clear enough: I have 2 converns:

1) Advertising this demo in the AMD OC demo thread to show what can be done with manipulated CPUs implies AMD has cheated. Francois could have come up with this idea in a more elegant way

2) My fear is that those CPUs will find their ways into Hwbot, thus making it impossible for everyone else to catch up by finding a equally good CPU. Getting "ordinary" hardware for free, but getting cherry picked stuff in a way that no one else can cherry pick in the same way, is something way different.

None, this is too dangerous, one booting without the right voltage and cooling is fatal.
Trust me, it would become a dissaster, it is like removing the self preservation of the human kind ... watch the movie "The happening".

You seem to forget where you are =) Do you think anyone here using LN2 and 2 Volts on C2D/C2Q cares about the loss of a CPU? If it can be done in a way that it is obvious the CPU has been manipulated thus losing all warranty, why not give extreme overclockers that option?

[jasonelmore]

I know this question isn't going with the current conversation but what are your guys current NB temps on a 920 or 940 when taking your blck to around 200 or more? Particularly on Asus boards like the Rampage II or P6D.. Is water cooling needed to do this or would be buying a NB block be a waste of money for this chip-set? I'm not talking about anything too extreme here. around 4 to 4.3 ghz on water.

[[EvQ]RheA]

4.3 on water shouldn't be too hard to get to; I was running some quick runs at 1.61x and was stable enough for some benching...

[Fr3ak]

Depends on what board you are using. On some boards, I noticed that the NB temp skyrocked after a little increase in voltage. Having a simple fan in the NB area helped a lot. But lower temperatures will not help you with your overclock. 222 MHz BCLK is far as you can go, depending on your CPU, but from what I can tell, NB temperature has next to no influence in max BCLK. I reached the highest speed with the Biostar and the MSI board without additinal cooling and NB temp was in the high 70's due to bad contact.

[Sampsa]

offtopic:

Here is pic of Finnish singer Jorma Kääriäinen:



And here is Francois Piednoel from Intel:



Wtf.. lost twin brother?

[woffen]

Hahaha. They actually look like eachother quite a lot.

Hyvä löytö