C1 hit 4.3 ghz

Sick stuff, I saw the other booted into windows P4 @ 4.3Ghz on <A HREF="http://www.vr-zone.com" target="_new">www.vr-zone.com</A>

My lil 2.66 is a screamer, will be benching this weekend.

You are limited to what your mind can perceive.

<A HREF="http://www.muropaketti.com/artikkelit/cpu/nw2800/index3.phtml" target="_new"><font color=green>New improved clickable!</font color=green></A>

2 Things I noticed about that article,

1) Why are they using PC800 RDRAM on a 533 FSB board?

2) Dealing with Liquid Nitrogen with no gloves to protect their hands is a big no no =)


Cooj

<i> I am ignorant no more! </i>

Yeah, I never thought LN2 cooling was this huge!

Still, I was surprised that the XP1800 could still kill it in raw FPU!
Though you can forget SSE2, it's hopeless for the AXP.

--
When buying an AthlonXP, please make sure the bus is at 133MHZ, or you will get a lower speed!

I wonder when they're coming out with a cooler setup for my Geforce card? I'd like to overclock that to 2 Ghz as well LOL


Save heating costs on your home, overclock your PC!!!

I was reading up about this particular system, and it seems that it ran stable at 3.9Ghz.

- - -
All good things must come to an end … so they can be replaced by better things!

People believe the extreme overclock is a sign of where the core can scale too eventually.....In other words, you may see an air-cooled P4 running at 3.9ghz someday.

This sig runs too hot.

Next stage is to build your own clock generator for the FSB..
1Ghz FSB anyone?

And yes I am serious (not about the 1Ghz FSB) but I am working on a 300Mhz clock generator to give me a 300Mhz FSB.

I will post results if I succeed.

What does a clock generator look like and how do you make one?

--
When buying an AthlonXP, please make sure the bus is at 133MHZ, or you will get a lower speed!

It is not too difficult to produce, basically they consist of:
A couple of capacitors
A Few resistors
A Crystal to set the frequency preciscely (300Mhz crystals are not too expensive).
The some sort of buffer circuit to make sure the clock continues to oscilate when connected to an external circuit.

I am used to building clocks but I have not ever attempted to build one thins fast, so I will post some info once I do it. Otherwise I will ask for help from my tutor in the electronics dept at my uni.

I can give you book references to build simple clocks.

whats so funny, I just need enough cash first incase I desroy my motherboard and Chip in the process.

(Donations accepted)

Sorry, don't shout......confused......

Just his post is on the 1st page.....long way away....

<A HREF="http://holicho.lib.net/top/020901/020901.htm" target="_new">2.5A broke 4.4Ghz</A> and worlds fastest <A HREF="http://holicho.lib.net/top/020830/020830.htm" target="_new">SuperPi</A>

Posted by macci elsewhere



You are limited to what your mind can perceive.

Dark now you're being over excstatic and too much "Into Intel".

Continuing to say that is rather useless, we know it has life, I mean if someone is not smart enough to ut the obvious, that the double-length pipeline WILL allow extreme scaling, should not even talk. I think you know that by now, so you really are acting more excstatic as if you don't know the obvious!

I stated before, if the 0.13m is reaching 4.5GHZ, the Tbred Bs CAN scale to 2.25GHZ, and since they can also reach 2.5GHZ, the 0.13m P4s with future refined steppings WILL reach 5GHZ, if they could reach 4.4GHZ now.

I think with water cooling or further extreme cooling, you can get a Tbird 1GHZ AXIA to 1.76GHZ which is also a 76%, nothing unachievable. I'll check overclockers.com to see.

--
When buying an AthlonXP, please make sure the bus is at 133MHZ, or you will get a lower speed!

You really want me to be picky, don't you? I've read somewhere about a few guys who OC'ed a 300A Celeron to 700mhz. That's an overclock of about 116%. They used LN2 for that.

Anyways, what I meant was the p4 can reach 10ghz, like Intel stated, even though several members of this forum doubt that.

- - -
<font color=green>All good things must come to an end … so they can be replaced by better things! </font color=green>
<P ID="edit"><FONT SIZE=-1><EM>Edited by Dark_Archonis on 09/02/02 04:23 PM.</EM></FONT></P>

Ahh I love this!
I will be even more picky, a 600MHZ Duron OCed to 1.7GHZ, a 183% overclock, my good man!
There is even one who got 2.1GHZ, but I will not include it because I find it a bit exaggerated and I don't beleive it too well.

Also they got the Duron to 1.3GHZ, a 116% OC, using also LN2. Both cores seem to be similar in clocking, because of their 10 stage architecture.

--
When buying an AthlonXP, please make sure the bus is at 133MHZ, or you will get a lower speed!

He wasnt being picky(good research btw eden) he was counteracting your bs statement amd had never had a better oc than that.

Heatsinks, if you dont overclock, use the <b>STOCK!</b>

There is NO LIMIT to what the clockspped an Architecture can achiver - as there is no limit to the speed any circutry can achive as long as you build it out of faster smaller cooler processes.
ofcourse P7 will hit 10Ghz - on the 90/65nm process.

anyway the heat the core produces will stop Pentium 4 much before its theortical limits. it will be intresting to see how far Intel can take P4 under 75w. right now the figure seems to be 3.2Ghz.

This post is best viewed with common sense enabled

It was working fine earlier.

Heatsinks, if you dont overclock, use the <b>STOCK!</b>

Did you change the default cpu, scroll down a bit.



Heatsinks, if you dont overclock, use the <b>STOCK!</b>

remember you can only change 1, amd cpus and intel cpus are in a different category.

Heatsinks, if you dont overclock, use the <b>STOCK!</b>

Yes but under their architecture, they are scaling pretty much similarly, clock-wise. If we had a 20 stage Celeron at 300MHZ, no doubt it could reach 1GHZ much easier than a 10 stage one, and since either is really limited, but rather one has problems at higher clocking, then yes I can say that the 2 10 stage CPUs are clocking fairly similarly. I can assume a copper implemented Celeron at 600MHZ, at 0.18m SHOULD go up to 1.6GHZ albeit extreme cooling.

But yea your point is right, in the end I shouldn't have said that, when I think about it.

Also why do you claim that the P4 can reach 10GHZ with -included- 0.09m? I would assume 0.065, since 0.09m will start at around 4GHZ, therefore end around 6-7GHZ. It will be hard to get 10GHZ but remember, Intel has new silicon technologies to add, and I heard BBUL packaging will allow lower wattage, not to mention Intel will also use SOI but much later on. Add that to process shrinks and stepping improvements, and yes, 10GHZ is not too shabby!

--
When buying an AthlonXP, please make sure the bus is at 133MHZ, or you will get a lower speed!

I didn't mean you take that as "Including 90 nm" I meant it might or might not achive that speed on 90nm. I dont know what intel is doing with P7 and the 90nm process but they already demoed 4.2Ghz pentium4 on 130nm and 10Ghz ALUs on 130nm using 1.75v - prescott will obviously never see such high voltages as it will the output 100w + but even that kind of heat can be dispanced Itanium 2 outputs 130w. even under lab conditions 10Ghz might be a litle far fetched for the 90nm process but who knows. it is very probable it will get close to 10Ghz (8Ghz-9Ghz) under lab conditions.

This post is best viewed with common sense enabled

AMD beating a dead horse?
Wow man, you need a serious CPU 101 class!
I may not have taken one, but ever since my big thread with imgod2u, I have learned tons of new things, and for one, the K7 IS FAR from being a dead horse. You're still one of those FUD-spreaders who like to claim that, which is false as hell. Raystonn, one of our old members who's an Intel employee, claimed K7 would die at 2GHZ or around it, my oh my was he wrong.

The K7 CAN go up to 5GHZ. If the P7 core is a double-wide pipeline, and the stages are all divided ones, to make it easier to compute, that means it should scale about 2, 2.5 times higher than a 10 stage design. Not to mention improving IPC will increase heat and hinder some scaling if it becomes too high. Your 15GHZ theory is in fact very far-off, because that would mean saying that the K7 will reach 7GHZ or around it. I am not just speaking with no knowledge, I am saying this with all of what I have learned. Intel can only push so far their 20 stage design before those tasks also will need to be divided in order to compute faster and finish faster.

It will however reach 10GHZ, albeit will require powerful technology behind it, like SOI and 0.065m.

--
When buying an AthlonXP, please make sure the bus is at 133MHZ, or you will get a lower speed!<P ID="edit"><FONT SIZE=-1><EM>Edited by Eden on 09/07/02 04:00 PM.</EM></FONT></P>

What, the K7?

We had a discussion before on why, and it seemed that it needed more TLBs, enhanced cache design. Most of us agree this is why. The Athlon has needed since day one a better cache design, like the P4's or P3's. Higher bitpaths would allow efficient bandwidth use.

--
When buying an AthlonXP, please make sure the bus is at 133MHZ, or you will get a lower speed!