It looks like you're using an Ad Blocker.

Please white-list or disable in your ad-blocking tool.

Thank you.


Some features of ATS will be disabled while you continue to use an ad-blocker.


Japanese Supercomputer breaks Petaflop Barrier

page: 1

log in


posted on Aug, 31 2006 @ 11:59 PM
I did a search and didn't find a thread on this. I guess initial estimates thought it wouldn't be broken till 2010. That is ridiculously fast!

BTW, the Petaflop barrier is the ability of a computer to do one quadrillion floating point operations per second.

One quadrillion = one thousand trillion

A "floating-point number" is any real number including fractions, negative numbers, and numbers written in scientific notation. Floating-Point Operations is a common way to rate the speed of a microprocessor.

[edit on 9-1-2006 by worldwatcher]

posted on Sep, 1 2006 @ 05:08 PM
Damn that's fast...


posted on Sep, 1 2006 @ 05:46 PM
That's not fast. That's waaaaay faster than fast. That's so fast Deep Thought needs a towell.

posted on Sep, 2 2006 @ 05:23 AM
cripes. wonder how long it would take to render something on that?
erm... where do i plug my speakers in

posted on Sep, 2 2006 @ 07:09 PM
It better have an ATI Videocard.....

That is insane. What cools that thing?

posted on Sep, 2 2006 @ 07:22 PM
i wonder what the overclock on the puppy could be just imagine adding quad sli cards to

posted on Sep, 2 2006 @ 07:45 PM
Please stop the graphics card comments, such a system is USELESS for 3d games with raster graphics.

Also floating point operations is not a good way of mesuring overall perforamance.

posted on Sep, 2 2006 @ 07:46 PM
On the website you linked to, a fellow named Bryan takes serious issue with the story. I've lifted a couple sentences from his post:

MDGrape-3 is not a real supercomputer. It is a special purpose machine to run ONE specific program. MDGrape-3 is built specifically for Molecular Dynamics. MDGrape-3 can't even run the benchmarks needed to get on the supercomputing charts. Bryan

It is still durn fast!

[edit on 9/2/2006 by donwhite]

posted on Sep, 2 2006 @ 09:12 PM
well i bet it can fun molecular dynamics test with pretty cool results.
still i wonder if anyone will make a super computer just as powerfull but for virtual reality, that would be super-cool.


posted on Sep, 2 2006 @ 11:45 PM
These are all far too narrow a view. Computations this fast supercede the notion of a specific use... molecular dynamics... virtual reality... these don't take in the full breath of what it means to compute at such speeds.

I struggle to describe it. Lets just say... in a universe where everything can boil down to a single equation, being able to evaluate that equation quickly means being able to render anything. And I do mean anything.

A petaflop is incredably fast. So fast that "fast" really doesn't do it justice. But, it still pales in comparison to what is theoretically possible.

In an age where the doubling rate of all human knowledge is getting smaller with each passing day, soon a "petaflop" will be as laughable as "640K of memory is more than anyone will ever need."

posted on Sep, 3 2006 @ 01:49 AM
with that in mind it only realy took us, oh... lets say 60 years to develop from the "colosus" to this mamoth super computer and those similar to it. just think what will happen in the same amount of time for the future.
the next objective would be an exaflop, which would be 10 to the power of 18 =
1 000 000 000 000 000 000 = 1 quintillion

posted on Sep, 3 2006 @ 09:18 PM

Originally posted by kipman725
Please stop the graphics card comments, such a system is USELESS for 3d games with raster graphics.

We're just havin fun here and joking around. Anything wrong with that?

posted on Sep, 4 2006 @ 01:14 AM

USELESS for 3d games with raster graphics.

Yes very true but it wasnt designed for 3d games.

Besides. even if it were useful in that capacity, it would be pointless

posted on Sep, 4 2006 @ 01:19 AM
Ludicrous Speed... GO!

Sorry... couldn't resist a SpaceBalls reference here.

posted on Sep, 4 2006 @ 02:41 AM
ok i think this point has gotten through. it is designed for a spacific perpose, that is molecular dynamics. it wasnt designed spacifically to beat all other computers out there. mabey somewhere a bunch of very rich nerds are designing a Petaflop gaming system somewhere. i dont think that this will be released any time soon so dont get your hopes up.


posted on Sep, 4 2006 @ 10:56 AM
Like I said... this is way beyond specific purpose.

You couldn't play games on Google's server farm, either. But to my knowledge there is no other database system that rivals it.

The fact that this machine can evaluate equations at such speeds is where the breakthrough lies. That fundamental process can then be applied to any specific use desired.

new topics

top topics


log in