[Beltiras]

There are a couple of those that might be relevant to work I am doing. Just wondering what the horizon should be for someone to dive into QC. The engineering doesn't seem to be anywhere near where a mid-sized company can buy or rent access to a reasonably sized machine any time soon (thinking about timescale next 10 years or so). Super interesting thou.

[marcosdumay]

There are no reasonably sized machines.

QC is currently basic research. It's not clear when machines large enough to be useful will exist, and I don't understand the reasons behind the current development speed. But keep in mind that we have computers with ~20qubits, and people have been consistently adding 3 more qubits every 2 years, for a rather long time.

[Beltiras]

Isn't this the same as development of classical computers? We are still in the phase that classical computer were in in the 20's to 40's. Once you crack the lid and make something mass-producible, exponential advancement will take over for as long as physics allows it (which in QC is an extremely deep well).

[marcosdumay]

I can't imagine how to trace a parallel. Even before that, classical computers stayed a very long time "on the lab", and by that time the expansion in functionality was way more visible than any expansion of size. Only after people kind of agreed on what a "computer" can do (at the 40's) that we got any clear size expansion, that was exponential.

As a comparison, we have a very nice idea on the requisites of a quantum computer. Nobody is exploring this space. Also, we have that very clear linear growth on their size, while we didn't have even a clear size measure for classical computers.

It looks like something very different to me. Really, that linear growth is unsettling in many ways.

[dangirsh]

Recent announcements from IBM and Google indicate they will jump from ~20 to ~50 qubits. My guess is that near-term growth will not remain linear.