# Another step forward for diamond-based quantum computers

Building simple quantum gates is common, but creating something that could be built on transistor-like scale is a huge challenge. Now, boffins from the Technical University of Vienna, Japan's National Institute of Informatics, and NTT's Basic Research Labs are offering an architecture they reckon can be scaled up. What the …

1. #### eh?

So, why are 4.5 billion qubits needed for Shor-2048?

There's this bit in the paper:

"For a logical error rate pL ≤ 10^-18, d ≥ 32 is required [73]. Therefore, a logical cell requires V = (5d/4)^3 = 40^3 cluster cells. To perform a logical CNOT gate requires a cluster volume 2 × 2 in cross section and two logical cells in temporal depth."

40 ^ 3 = 64,000

64,000 x 2048 = 131,072,000

131,072,000 x 8 = 1,048,576,000

Maybe "d" actually equals 52? (((5*52)/4)^3)*2048*8 = 4,499,456,000

...and that's where I stop making stuff up.

1. #### Re: eh?

That looks like a nice bit of reverse engineering, though I have no idea whether it's right or not.

It does emphasise the problem with quantum computing - all the stuff about simultaneously occupying all possible states ignores the error rate, which means that to get acceptable accuracy the calculation may have to be performed many times in either parallel or series, which negates some of the advantages. It's going to be interesting to see where the crossover points between conventional and quantum computing appear.

1. #### Re: eh?

>which means that to get acceptable accuracy the calculation may have to be performed many times in either parallel or series, which negates some of the advantages.

At least it's easy to check the results of the factorisation with a classical computer.

1. #### Re: eh?

> which negates some of the advantages.

Not really

> At least it's easy to check the results of the factorisation with a classical computer.

Because this.

We also have "quantum error correction" which actually makes sure that the calculation stays on track and the problem of information loss is considered solved. No, I haven't though about this at all.

2. #### Fibre to the chip

I assume that block and translucent strip going up from it is to do with coupling photons into where they are needed..

Has any work been done on laying fibre optic tracks on PCBs or within chips?

1. #### Re: Fibre to the chip

If it works like my ISP, then it's fibre to the chip, then a converter and modem hookup once it's inside.

3. #### FOR GODS SAKE MAN !!!

It's Monday morning !!!

How on earth am i supposed to understand any of that till at least 1:00 PM !!

4. You'll understand that by 1 PM ?

Lucky guy.

5. #### You learn something every day

> UC Santa Barbara researchers demonstrated controlling a nitrogen vacancy centre in diamond using photons.

I had no idea nitrogen needed a job. Lucky for it, having clerks that work at the speed of light. :(

6. #### Proper Science

All proper science should include diamonds, nitrogen and some manner of light related injector/gun/cannon, or something that melts faces off. Otherwise there's simply no way to have any good fiction about it.

1. #### Re: Proper Science

"All proper science should include diamonds, nitrogen and some manner of light related injector/gun/cannon"

Quite so. Now Mr Bond, let me explain at length how all this works before I feed you to the sharks.

## POST COMMENT House rules

Not a member of The Register? Create a new account here.