What lies beyond ASICs?

[enigmazr]

Without getting too far ahead of ourselves, I am curious what the next big move in mining hardware will be.  I had a few conversations with bitcoin enthusiasts at Bitcoin 2013, but no one had a clear answer.  My nascent thoughts lead me to believe that quantum computing may lead to faster mining.  Although the company D-Wave manufactures quantum computers, they are very expensive at the moment which will make it difficult to develop mining software for these uber machines.     

[1714871044]

1714871044

[1714871044]

1714871044

[1714871044]

1714871044

[1714871044]

1714871044

[1714871044]

1714871044

[1714871044]

1714871044

[enigmazr]

Ok, and that amounts to what? A board with hundreds of chips?

[cp1]

Faster smaller cheaper chips

[mprep]

Faster smaller cheaper chips

Much, much faster. It'll simply completely eliminate the purpose of gpu mining.

[J35st3r]

We did this already https://bitcointalk.org/index.php?topic=227285.0

But I'm new, so it probably comes up twice a week. Newbie forum is like Groundhog Day (the film), stuff just keeps happening, again and again (and again)^oo.

[enigmazr]

This is how I see the score.

	Quantum Computing	Classical Computing
Speed	+++	0
Size	+	0
Price	0	+++
Development	0	+++

Currently, classical computing wins; however, as we continue to develop quantum algorithms of the type that can be used in mining, and quantum hardware becomes more approachable, it will quickly succeed classical computing as the forerunner in mining.

This vision may be years or decades out, but it seems inevitable that as we began to realize the power of quantum computing, its applicability to Bitcoin mining will become all that more salient.

[enigmazr]

We did this already https://bitcointalk.org/index.php?topic=227285.0

Thanks for pointing this out.  I guess I could specify that I'm curious as to the nature of how quantum computers could be used to mine.  I speculate that they would operate at a higher hashrate.

[J35st3r]

This vision may be years or decades out, but it seems inevitable that as we began to realize the power of quantum computing, its applicability to Bitcoin mining will become all that more salient.

Applied pablovium Phlebotinum (TV Tropes)

Edit Oops mis-remembered, took a while to cross-check  (doesn't help that I'm on a RasPi with two background vanitygen's running). Still pablovium sounds pretty good too.

[Svennisen]

Yep, just faster, smaller chips.

[enigmazr]

This vision may be years or decades out, but it seems inevitable that as we began to realize the power of quantum computing, its applicability to Bitcoin mining will become all that more salient.

Applied pablovium Phlebotinum (TV Tropes)

Edit Oops mis-remembered, took a while to cross-check  (doesn't help that I'm on a RasPi with two background vanitygen's running). Still pablovium sounds pretty good too.

I totally agree that quantum computing is phlebotinum; however, quantum information processing does exist as a burgeoning science.  The scientific journal Quantum Information Processing is a testament to the field's credibility.  For those who haven't peeked into the machine, a quantum computer may very well serve as a black box that is fueled by pixie dust.  However, those of us who have a deep understanding about quantum computing see the potential it may serve to the bitcoin community.

[crazynoggin]

These faster and smaller chips will likely be even more expensive than ASIC's.

[spike420211]

full-custom is the next step up from ASIC.
You're talking budgets of Intel/AMD/NVIDIA to build these.
I imagine the budget for a USB full-custom, using way fewer transistors,
would be hella cheaper.
And on a smaller volume scale- think TI/Linear Technologies/Analog Devices/International Rectifier.

I don't see Bitcoin as the driving force for quantum computing,
but Butterfly Labs gets rendered obsolete.
[You'll still need GPU's for SolidWorks and 3D printing]

[spike420211]

These faster and smaller chips will likely be even more expensive than ASIC's.

not necessarily, see my post above this.

[socket]

full-custom is the next step up from ASIC.
You're talking budgets of Intel/AMD/NVIDIA to build these.
I imagine the budget for a USB full-custom, using way fewer transistors,
would be hella cheaper.
And on a smaller volume scale- think TI/Linear Technologies/Analog Devices/International Rectifier.

I don't see Bitcoin as the driving force for quantum computing,
but Butterfly Labs gets rendered obsolete.
[You'll still need GPU's for SolidWorks and 3D printing]

BFL is full custom ASIC design...

[spike420211]

any ASIC is fixed transistor location
fixed transistor sizing.
the routing connections done [aluminum/copper] are custom.
Full-custom allows re-sizing and specific placement of transistors,
optimizing speed, possibly minimizing heat/power.
The fact thet they can even do this in USB allows for a paradigm
with an order of magnitude improvement available,
maybe before full custom.

[socket]

any ASIC is fixed transistor location
fixed transistor sizing.
the routing connections done [aluminum/copper] are custom.
Full-custom allows re-sizing and specific placement of transistors,
optimizing speed, possibly minimizing heat/power.
The fact thet they can even do this in USB allows for a paradigm
with an order of magnitude improvement available,
maybe before full custom.

I think you just hit the jackpot on buzzword bingo.

[IBCBen]

Threads like this make me giddy with anticipation of the meme's complaining about the new 900000000% difficulty increase since the first asic was invented. We will look back and cherish these memories. lol

[spike420211]

I think you just hit the jackpot on buzzword bingo.

actually, that's how you describe the engineering involved.

[Dasneko]

This is based on the notion that bit mining is a lottery.

In the case where its blind luck that decides if your hash is the right one Quantum Computers suffer a major drawback. Since there is no definite "right answer" you cant "cheat" your way to the right hash using quantum functions and the actual calculation is so simple that you wont gain anything from using other properties of quantum mechanics to speed it up.

Its more cost efficient to get inexpensive ASIC's that hardly cost anything to run.

[r3wt]

This is based on the notion that bit mining is a lottery.

In the case where its blind luck that decides if your hash is the right one Quantum Computers suffer a major drawback. Since there is no definite "right answer" you cant "cheat" your way to the right hash using quantum functions and the actual calculation is so simple that you wont gain anything from using other properties of quantum mechanics to speed it up.

Its more cost efficient to get inexpensive ASIC's that hardly cost anything to run.

and someone with a brain has answered the question