Journalist Question - Is bitcoin the most powerful computing network ever?

[bawish]

The network hashrate hit 85 petaflops the other day. From what I can tell, none of the other distributed computing projects (BOINC, SETI, Folding) come close.

Subsidiary questions:

How come the network hashrate is so variable? It's down to 19 petaflops today.

Is it fair to compare the bitcoin network to a supercomputer like the one in Tianjin (which reports say gets 2.5 petaflops)?

I'd very much appreciate a quick response!

[1714922758]

1714922758

[1714922758]

1714922758

[1714922758]

1714922758

[1714922758]

1714922758

[Basiley]

The network hashrate hit 85 petaflops the other day. From what I can tell, none of the other distributed computing projects (BOINC, SETI, Folding) come close.

Subsidiary questions:

How come the network hashrate is so variable? It's down to 19 petaflops today.

Is it fair to compare the bitcoin network to a supercomputer like the one in Tianjin (which reports say gets 2.5 petaflops)?

I'd very much appreciate a quick response!

1. this horserpower comes from voluntary contribution of members anywhere around Globe. and most of computers - usual home PC's. which can be or canot be used for this purposes from time to time.
2. its fair. BTC only lack more functional software to use this horsepower for other[than hash-crunching]. true for nearly ALL highly-paralell computation with DP-precision-hungry-requirements.
3. sure, you are. and welcom.

[Jack of Diamonds]

Yes. Tianhe's current computing power is made up of 7168 nvidia tesla 1070-series GPU's which would equal about 1 terahash/second, or only 1/7th of the network capacity.

So the bitcoin network block chain can't be attacked by any (known) supercomputers in existence and a 51% attack would require most of the supercomputers in the world.

Also, the bitcoin network is growing in computing power exponentially. Tianhe-1 will most likely stay at it's current limits until they build something better.

[caveden]

For doing the type of calculations it does, I think you can safely assume there is no network out there as powerful as bitcoin. But the type of hardware mostly used for hashing blocks might not be that effective when used for other types of calculations, so depending on what you need to do, I'm not sure the bitcoin network is the strongest one available.

Regarding variation, estimates of the total computing power are done using the number of blocks produced in a time window and the difficulty factor. If the time window is too short, like a few hours, you'll see wild variations. You should get estimates which use larger time windows, like one or two weeks.