- Joined
- Dec 16, 2007
- Location
- Chicago, IL
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
More than a Bugatti!
And sorry, 15 KW is more than a 7 Ghz 2600k farm!
And that's likely more watts than a computer in the 1960s and 1970s!
It's better to just get a VapoChill'ed 2600k at 8 Ghz and a 2,000 W PSU should be fine.
It's not a joke and it costs 10,000,000$ US.
From what I could understand about the technology is that it only does one specific type of computation, reaaaaalllllyyy fast. (Discrete Optimization) - I don't really know what that is.
Thid - the server is one processor, just one The giant box is to cool it to almost absolute zero and keep any 'noise' from knocking the processor out of entanglement.
More than a Bugatti!
And sorry, 15 KW is more than a 7 Ghz 2600k farm!
And that's likely more watts than a computer in the 1960s and 1970s!
It's better to just get a VapoChill'ed 2600k at 8 Ghz and a 2,000 W PSU should be fine.
Well, it's 1st generation stuff, too. 50 years, or much less as fast as things seem to be progressing, we will have a descendant of that in our house.
It's better to just get a VapoChill'ed 2600k at 8 Ghz and a 2,000 W PSU should be fine.
That's what I'm thinking - just write software to utilize existing processors to run the optimization, and you may even get increased performance with more raw power for a fraction the cost. This sounds like snake oil. There's no way this thing can outperform a small farm made on a $100,000 budget. Even rich and (self proclaimed) "important" people get scammed.
Google said:It turns out that solving the hardest of such problems requires server farms so large that they can never be built.
...Let’s take unstructured search as an example. Assume I hide a ball in a cabinet with a million drawers. How many drawers do you have to open to find the ball? Sometimes you may get lucky and find the ball in the first few drawers but at other times you have to inspect almost all of them. So on average it will take you 500,000 peeks to find the ball. Now a quantum computer can perform such a search looking only into 1000 drawers.
That's really interesting. Not really a CPU, but a processor that optimizes things.... I can see why people would pay for that.
Seems like apples and oranges to me; Quantum computing uses 1, 0 and both, rather than 1 and 0.
From the blog:
How exactly does the thing optimize things? And what kind of things does it optimize? Anything?
How does THAT work? You can search a million drawers in 1000 drawers...? Huh? Quantum physics confuse me... and i have a pretty dang high IQ
Albert Einstein said:On quantum theory I use up more brain grease (rough translation of German idiom) than on relativity.
How exactly does the thing optimize things? And what kind of things does it optimize? Anything?
How does THAT work? You can search a million drawers in 1000 drawers...? Huh? Quantum physics confuse me... and i have a pretty dang high IQ
How exactly does the thing optimize things? And what kind of things does it optimize? Anything?
How does THAT work? You can search a million drawers in 1000 drawers...? Huh? Quantum physics confuse me... and i have a pretty dang high IQ
I'm thinking in terms of raw processing power, you could build a comparably 'fast' machine for significantly less money, and have it handle the 0, 1 instructions. Then build a second machine (again for significantly less) and have it handle the BOTH (0 and 1) instructions, and just connect the two using a simple interface that balances the work loads. I understand why having the ability to do three things at once per cycle is better, but for that cost it's a highly inefficient and laughable way of accomplishing the work. It's a good solution but doesn't need THIS to be its implementation.
I'm thinking in terms of raw processing power, you could build a comparably 'fast' machine for significantly less money, and have it handle the 0, 1 instructions. Then build a second machine (again for significantly less) and have it handle the BOTH (0 and 1) instructions, and just connect the two using a simple interface that balances the work loads. I understand why having the ability to do three things at once per cycle is better, but for that cost it's a highly inefficient and laughable way of accomplishing the work. It's a good solution but doesn't need THIS to be its implementation.