[2013-11-28] FORBES Global Bitcoin Computing Power Now 256 Times Faster Than...

[simplydt]

http://www.forbes.com/sites/reuvencohen/2013/11/28/global-bitcoin-computing-power-now-256-times-faster-than-top-500-supercomputers-combined/

[Rannasha]

I hate these kinds of comparisons. Supercomputer processing power is measured in Floating Point Operations per Second (FLOPS), while SHA256 hashing is pure integer arithmetic. Furthermore, (super)computers can be programmed to all kinds of tasks, ASICs, which now make up almost all of the network, can do only one thing.

It's like saying a toaster is 256 times more effective than a lightbulb at making toast.

[PenAndPaper]

I hate these kinds of comparisons. Supercomputer processing power is measured in Floating Point Operations per Second (FLOPS), while SHA256 hashing is pure integer arithmetic.

On top of that they sound way to geeky and they dont provideany significant pr buzz. People want to hear stories about other people getting rich overnight with bitcoins 

[GenTarkin]

But the important thing to remember is, since they specialize in FPU(therefore probably being horribly weak at INT math)   ... given the pooled effort of the Supercomputers.... they would come nowhere close to being able to 51% attack Bitcoin.

[bassclef]

I hate these kinds of comparisons. Supercomputer processing power is measured in Floating Point Operations per Second (FLOPS), while SHA256 hashing is pure integer arithmetic. Furthermore, (super)computers can be programmed to all kinds of tasks, ASICs, which now make up almost all of the network, can do only one thing.

It's like saying a toaster is 256 times more effective than a lightbulb at making toast.

It sounds good to the average reader. We're not the average reader.

[DeathAndTaxes]

There is some use to comparisons like this.  There are two ways to brute force attack Bitcoin, the first would be to build out a massive dedicated farm, and the second would be for a government to temporarily use a dual purpose technology like a super computer.  That door has now closed and it is a good thing because the cost of such an attack is simply the operating cost of a existing supercomputer(s).

[bitbouillion]

The architecture of most supercomputers is not a secret. E.g. the No 1, Titan, consists of CPUs and GPUs of known type and speed (just look it up on Wikipedia). Then look up the hashing power benchmarks for these CPUs/GPUs and - voila - you have the sha-256 hashing power of Titan. I get 12 Thash/s, which is far away from 51% of the network's hashing power (as of now ~7000 Thash/s). I guess the top 10 supercomputer combined get 95% of the computing power of all supercomputers, so the other 490 supercomputers don't make a notable a difference here.

[mrb]

But the important thing to remember is, since they specialize in FPU(therefore probably being horribly weak at INT math)

It is absolutely false. Supercomputers are built from CPUs and GPUs that can do both integer and floating point operations just as fast as each other.

As a matter of fact: a modern Intel or AMD CPU core can execute 8 single precision floating point operations per cycle, or 8 32-bit integer operations per cycle. That's the exact same capacity.
 

[LiteCoinGuy]

let us agree to this: the bitcoin-network is pretty bad-ass  

[PenAndPaper]

But the important thing to remember is, since they specialize in FPU(therefore probably being horribly weak at INT math)   ... given the pooled effort of the Supercomputers.... they would come nowhere close to being able to 51% attack Bitcoin.

Becuase there was a chance for the 500 biggest supercomputers to turn bitcoin miners anyway