The magic 'Six' (6GHz) has been broken

tck

http://www.xtremesystems.org/forums/showthread.php?t=42655&page=1&pp=25

using a Prescott Pentium 4 running at 3.60 Ghz,

i wouldn't even know where to get liquid nitrogen!

thats one hell of an overclock

grimloch

good god that is some XTREME overclock

Töpher

Holy dog poop!

grimloch

tck wrote:

i wouldn't even know where to get liquid nitrogen!

you can get it from companies like gasworld.com or something but you need permission of some sorts to buy it, its not easy to come buy at all

dubdvd

i better start saving now holy st*t i thought i was doing well getting to 4.2 ghz did you see his 3dmark03 score 1800 and odd ...now thats the way to go :eek: my kind of guy

KdjaCL

Didnt toms hardware do this last year sometime?

kdjac

RuggieBear

KdjaC wrote:

Didnt toms hardware do this last year sometime?

kdjac

yeah but i think it was 5ghz..

OfflerCrocGod

It's not stable though they have to bring it down to 5.4Ghz to get a benchable stable system.

dubdvd

OfflerCrocGod wrote:

It's not stable though they have to bring it down to 5.4Ghz to get a benchable stable system.

yeah but still it would be so sweet to see that 6ghz on your desktop even for a couple of minutes

unkel

The chip does 5.4G Superpi already at 1.55V and -90C

Hehe, kewl

koneko

liquid nitrogen is pointless, I don't even take any notice of benches done with it. you can't run a system cooled by liquid nitrogen, it'll run for a few minutes so you can bench and then what?
Benching is to give an indication of how the system will perform in real world tasks, like gaming. you can't game on a system that's cooled by liquid nitrogen, it's completely pointless.

MR DAZ

now thats the way to go my kind of guy

Dubdvd......Is something you want to share with us (only Joking)

RuggieBear

koneko wrote:

liquid nitrogen is pointless, I don't even take any notice of benches done with it. you can't run a system cooled by liquid nitrogen, it'll run for a few minutes so you can bench and then what?
Benching is to give an indication of how the system will perform in real world tasks, like gaming. you can't game on a system that's cooled by liquid nitrogen, it's completely pointless.

Now apologies if this is a really stupid question....but why not? Does the nitrogen warm up and dissapate...

elexes

i wonder what sort of stable oc he can get ...

koneko

It does disappate/disappear. You'd have to keep topping up the liquid nitrogen every few minutes.

Would you want to get your protective gear on and do this every few minutes, or have someone standby while you play your games?

http://www20.tomshardware.com/cpu/20031230/5ghz-13.html#conclusion_525_ghz_liquid_nitrogen_cools_the_p4_to_a_new_speed_record

RuggieBear

koneko wrote:

It does disappate/disappear. You'd have to keep topping up the liquid nitrogen every few minutes.

Would you want to get your protective gear on and do this every few minutes, or have someone standby while you play your games?

http://www20.tomshardware.com/cpu/20031230/5ghz-13.html#conclusion_525_ghz_liquid_nitrogen_cools_the_p4_to_a_new_speed_record

LOL...Cheers, Koneko. I had an inkling it was something like that.

don't think it's worth it. Somehow i can't see my mates risking their digits so that i can play games that bit quicker....

gline

koneko wrote:

liquid nitrogen is pointless, I don't even take any notice of benches done with it. you can't run a system cooled by liquid nitrogen, it'll run for a few minutes so you can bench and then what?
Benching is to give an indication of how the system will perform in real world tasks, like gaming. you can't game on a system that's cooled by liquid nitrogen, it's completely pointless.

Its not pointless, it is just to see how much hardware can be pushed and used at. This type of benchmark is not to " give an indication of how the system will perform in real world tasks, like gaming", it is to show how much u can get out of the hardware, to get a glimpse of what the future cpus will be capable. It is just to prove that these chips are capable of doing mmuch more, its obvious you wouldnt be using this on your desktop

Capt'n Midnight

6GHz <Yawn>

Back in 1998 they were talking about Room Temperature Supercomputing - 16 GHz SiGe HBT Fast-RISC - Computing at GHz Speeds DAAG 55-97-1-0316
http://www.ecse.rpi.edu/frisc/slides/HTMT98/

Zoom forward to August 6, 2001 and we have 100GHz hardware http://www.elecdesign.com/Articles/ArticleID/3770/3770.html

researchers at IBM Microelectronics have reformed the structure of the SiGe heterojunction-bipolar transistor (HBT) to catapult its cutoff frequency (fT) to 210 GHz.
...
The company believes this transistor will drive communications and networking chips to 100 GHz

Anyone know where they are today ???

edit - Back in 2000 these guys were thinking even faster
http://www.eet.com/story/OEG20000410S0057

"We're working with chemists now who are designing molecules that when attached will act like transistors that can switch at 100 trillion times a second," said Lyding.

cerebus

Originally posted by Capt'n Midnight

Anyone know where they are today ???

Another EE Times article from earlier this year -Vitesse (VTSS) had some InP frequency dividers running at 152GHz.

Some googling for fastest transistor shows a device running with an fmax of > 500GHz - this is from 2003, so I'm sure there must be something faster out there by now.

Superman

Ok i guess it is pretty good but really useless!
Who needs to crunch all those numbers and even if they do how stable can it be

koneko

Yeah thanks gline :rolleyes: I know that (I used to have a prometeia aswell)

That doesn't take away from my point, it's still useless. If you can't run your system like that, the benches (other than giving indications into future speeds etc etc) mean nothing in real world computing imho.

I mean, I see your point, but I see people on forums getting all hyped over it and saying omg look at those scores, when the scores themselves mean nothing. The actual speed they got it to, fair enough, but the 3dmark scores are pointless.

vibe666

Superman wrote:

Ok i guess it is pretty good but really useless!
Who needs to crunch all those numbers and even if they do how stable can it be

that's the same thing people have been saying for years about computing power. who needs it right now? probably nobody, but things do progress.

people write better software that can do more, so we need faster processors to take full advantage, and so on and so forth.

it's the same with everything, a gradual evolution.

whpo'd want to sit and type a short message on a fiddley phone keypad when you can just call someone instead and tell them what you wanted to say? that'll never take off will it?

I'm off on a tangent again aren't I?:rolleyes:

Oh well.

you get the idea anyway.