Molecular level 3D construction....?

[Ein_Wannabe]

Hey guys, just thought I'd bounce a random thought off of you.

Since we can now "3D Print" on a molecular level, would it be possible to construct a body (crystalline maybe) using say Carbon G60 and other compounds to create multi-level data storage?
I know there have already been advances in storing data in crystals, but from what I've read so far they are only on small discs.

Maybe during the "construction" of such a thing different electrical pathways could be created by using different carbon molecules.

Also while I have your attention, any thoughts behind using extremely dense carbon as a form of extreme heat-shielding?

[John Cuthber]

We can. Slowly. Very slowly.

[EdEarl]

I am no expert.

 

Currently microchips are characterized as 2D. Although, there are several layers of silicon to make a transistor, these transistors are not stacked one on top of another. However, I have read several reports that labs are working on 3D microchips having multiple layers of transistors. It will allow them to increase gate (transistor) densities which allows faster computers and faster memories.

 

As for carbon heat shielding, the space shuttle used carbon-carbon heat shields, because it has a high melting temperature. However, carbon-carbon, diamond, and graphite are good heat conductors.

[Enthalpy]

3D chips have been a subject of discussion for over 30 years. Up to now, they're only a buzzword to describe but more than 2D chips.

 

- Static Ram has been made with P-Mos in polysilicon stacked over bulk monocrystalline silicon N-Mos.

- Intel calls 3D its Mos with vertical channel. Everything else in the process is flat. No stacking, especially.

- Some assembling manufacturer stacked complete 2D chips, especially Dram, to increase the packing density and reduce connection length.

 

Not really thrilling, is it? One would have expected as many transistors stacked as there are side-by-side, to integrate quintillions of them - or to reduce the on-chip wiring length and eventually break the propagation delays that prevent for a decade chips from improving.

 

Building chips atom by atom would take a prohibitive time up to now, even for the few atomic layers that are active on a cm² chip. More and more people think at it, so a breakthrough may come in a not so near future.

 

-----

 

At ablative re-entry heat shields, carbon is interesting because upon deterioration, it makes an opaque gas or dust that protects the craft from heat radiated by the shock wave. It's also nice as it takes much heat to evaporate.

 

Both advantages require the ablative material to evolve as a gas and to transmit little heat, which favours a carbon-rich organic material over carbon, and rather a fluffy over a compact one - the latter being not mandatory if a deeper layer insulates the craft. This makes phenolic resin a good candidate for instance, as it's sounder and easier to use than tar.

 

Re-usable protection is a different approach, taken only to re-enter from low-Earth-orbit up to now, for which carbon-carbon is one good material.

[EdEarl]

3D chips have been a subject of discussion for over 30 years. Up to now, they're only a buzzword to describe but more than 2D chips.

 

Until recently chip manufacturers have been able to shrink transistors and make faster integrated circuits with higher transistor densities and more capability. Thus, they have not needed a new technology to continue shrinking computers. There is a limit, which they are near or perhaps have already encountered. When they cannot make smaller transistors, they will need a new technology. There are at least two candidates, quantum computing and 3D chips.

 

Labs have made both quantum computers and 3D chips, and the race is on to transfer that technology into production. I do not believe computer improvements will halt, and the exponential rate of technology improvement tends to say once a thing is demonstrated in the lab, it can be mass produced faster than most people estimate.

 

One of the computers I used in university classes was a CDC6600 (a Seymour Cray supercomputer), which had a megabyte of hand-wired cores (magnetic donuts) that cost about a dollar a bit. If you had told me I would someday carry a portable telephone in my shirt pocket with more computer power than that $10M supercomputer, I would not have believed it. Conversely, it is probably hard for students today to understand the size and lack of power in the CDC6600.

[Enthalpy]

 

Until recently chip manufacturers have been able to shrink transistors and make faster integrated circuits [...]

 

Labs have made both quantum computers and 3D chips [...]

CPU have gained no serious speed since the Code 2. Then 3.3GHz, now 3.6GHz with 20% more cycle efficiency, in so many years. The AVX hardware will perform in one cycle instead of two at the coming Core generation, which will improve the software that uses the AVX instructions.

 

No lab has made a quantum computer, nor even has a research path towards a demonstrator.

[EdEarl]

CPU have gained no serious speed since the Code 2. Then 3.3GHz, now 3.6GHz with 20% more cycle efficiency, in so many years. The AVX hardware will perform in one cycle instead of two at the coming Core generation, which will improve the software that uses the AVX instructions.

 

No lab has made a quantum computer, nor even has a research path towards a demonstrator.

True that the cycle times have not increased in a few years, but the number of processors per microchip has been growing, which increases power per chip. I am aware of the laws of diminishing returns on number of processors, but additional processors allow my laptop to do run more than one process at a time, which helps.

 

I'll give you the lack of a quantum computer. i saw something once in which someone claimed limited success, but cannot find it now.

[John Cuthber]

If you have a 'phone that can tell which way up it is then you already have a 3 d chip.

http://en.wikipedia.org/wiki/Accelerometer