Suppose China used this for generating bitcoins...

[jimbobway]

"Tianhe-1A was designed by the National University of Defense Technology (NUDT) in China, and it is already fully operational. To achieve the new performance record, Tianhe-1A uses 7,168 Nvidia Tesla M2050 GPUs and 14,336 Intel Xeon CPUs. It cost $88 million; its 103 cabinets weigh 155 tons, and the entire system consumes 4.04 megawatts of electricity."

http://voices.washingtonpost.com/blog-post/2010/10/supercomputer_in_china_super_f.html


Ummm, what do you think??? Guaranteed monopoly?

[1715143086]

1715143086

[grondilu]

"Tianhe-1A was designed by the National University of Defense Technology (NUDT) in China, and it is already fully operational. To achieve the new performance record, Tianhe-1A uses 7,168 Nvidia Tesla M2050 GPUs and 14,336 Intel Xeon CPUs. It cost $88 million; its 103 cabinets weigh 155 tons, and the entire system consumes 4.04 megawatts of electricity."

http://voices.washingtonpost.com/blog-post/2010/10/supercomputer_in_china_super_f.html

Ummm, what do you think??? Guaranteed monopoly?

Even if there were the only one who generated bitcoins, it wouldn't be of much use if they don't spend it.

That the point of fixed amount money system :  any corner is only temporary, otherwise it's useless.

[LZ]

Yeah, we will try!

[MoonShadow]

Ummm, what do you think??? Guaranteed monopoly?

No, not guaranteed.  Not even likely.  The difficulty would adjust in short order, and they would, at best, be left with collecting the new coins for as long as they were willing to commit the resources to the endeavor.  I seriously doubt that they would consider it worth it for long.

However, if they were willing to commit the entire thing to generation of bitcoins for a short time immediately after a difficulty adjustment, ram through the generation as fast as they can, and then stop as the difficulty adjusted so high that it took an hour or more to find a single block solution; then they could dick with the network for a very long time.  But this could also be done by any group of people willing to act in collusion to the same end.  Still, this would only frustrate the network for as long as they were willing to keep it up, but never break it.

A more dangerous attack would be to secretly use such a supercomputer to build a false blockchain before the whole of the network can develop a total network power in excess of what something like this could achieve.  Which is why anyone who has a computer that idles most of the time should donate clocktime to generation, regardless of their personal odds of grabing a block.

[grondilu]

Ummm, what do you think??? Guaranteed monopoly?

No, not guaranteed.  Not even likely.  The difficulty would adjust in short order, and they would, at best, be left with collecting the new coins for as long as they were willing to commit the resources to the endeavor.  I seriously doubt that they would consider it worth it for long.

However, if they were willing to commit the entire thing to generation of bitcoins for a short time immediately after a difficulty adjustment, ram through the generation as fast as they can, and then stop as the difficulty adjusted so high that it took an hour or more to find a single block solution; then they could dick with the network for a very long time.  But this could also be done by any group of people willing to act in collusion to the same end.  Still, this would only frustrate the network for as long as they were willing to keep it up, but never break it.

Also, I doubt they've built this machine in order to generate bitcoins ! 

[ShadowOfHarbringer]

Also, I doubt they've built this machine in order to generate bitcoins !  

Nah, at this point nobody is willing to spend millions of dollars just to kill some "crap money transfer network" (this is how politicians will understand it probably).
I seriously doubt any government will take interest in bitcoin, until it becomes mainstream, or at least half-mainstream.

There are many targets which are also disruptive to current false fiat money currencies and much easier to take down (because they are centralised) - such as Liberty Reserve, Pecunix, Webmoney and so on.

[BioMike]

Ah... Chinese people and their prestige projects (they really like them).

IMHO they do a lot of things that have a high "look at us" factor, but poorly executed to be really useful.

Reminds me of a conference in Vienna on radiation induced mutagenesis in plants.
Chinese big boss steps up for a presentation, tells some story about sending seeds into space to
have them irradiated by cosmic rays (all fine so far, from a scientific view, but the real added value might be little).
Then someone in the audience asked how much radiation the seeds received. He couldn't answer and I think they
never measured. Ouch.. epic, expensive fail.

I've got a few more examples of these type of things. Do note that there is also a lot of good research done in China,
but when a lot of money is involved it is often spend on the easy stuff (a.k.a. build a super computer) but the
efficiency to use it can be a lot less.

[SmokeTooMuch]

http://top500.org/lists

I wonder how it will rank in the soon coming november 2010 list.

[TTBit]

Until China can produce fake bitcoins, they wont want the real ones.

[ShadowOfHarbringer]

Until China can produce fake bitcoins, they wont want the real ones.

This is interesting. Can You elaborate ?

[kiba]

Until China can produce fake bitcoins, they wont want the real ones.

This is interesting. Can You elaborate ?

Chinese like to inflate their money supply so they can increase their export.

[MoonShadow]

Until China can produce fake bitcoins, they wont want the real ones.

This is interesting. Can You elaborate ?

Chinese like to inflate their money supply so they can increase their export.

Yeah, they are always so proud of their official GPD growth numbers going up at about 8% per year, but recently it's come out that they have been quietly inflating their currency base at about 4% per year since 2003.  Considering GDP is more accurately a measure of Gross Consumer Spending, such a revealation not only cuts the knees off of the idea of a decoupled China boom, it also means that their adjusted GDP is barely beyond their middle class population growth.  That is a doubling rate about every 16 years, BTW.  Even the target rate of inflation used by the European Central Bank of 2% has a doubling rate of about every 33 years.  But I'm not really sure which is worse, because China still has a growing middle class population, which would actually mitigate some of the effects of inflation by reason of a growing domestic demand.  Europe, on the other hand, has a static or falling middle class population rate; so even a 2% inflation rate under such predictitable demographics is a destructive number.