[raylad]

So they simulated a chemical process, and the simulation ran 100B times slower than the actual chemical process?

Kind of like how simulating anything in detail tends to be a lot slower than the actual thing you're simulating?

Is the difference here that it's basically an analog rather than a digital simulation?

Not following if there was any breakthrough here or not.

[fnordpiglet]

> “Our experiment wasn’t a digital approximation of the process – this was a direct analogue observation of the quantum dynamics unfolding at a speed we could observe,” he said.

No, they didn’t simulate it in the way we typically simulate. They created a physical process that was an analogue of the actual process but 100b times slower so they could directly observe it.

[raylad]

That sounds awfully like an analogue simulation.

[chaosite]

Not an expert, but what I'm understanding is this:

With a regular simulation, whether analogue or digital, you create a mathematical model, and evaluate that model in some way to get a result, which hopefully matches the actual phenomena in whatever it is you were trying to simulate.

Here, it's more like instead of measuring the phenomena, they instead measured something that is physically related to that phenomena. There seems to be no modelling involved.

[quickthrower2]

Like a scale model aircraft in a wind tunnel?

[rbanffy]

Like a scale model aircraft in a tunnel, with a fluid that behaves like air, but just enough it can flow slower. Before FEA and CFD, people did a lot of work with miniatures. I remember simulating a long metal rod for an oil rig that was a plastic pipe filled with liquid mercury so that it'd flex the same way the kilometer-long metal rod would.

[colordrops]

Sounds like a semantics debate about the meaning of the word "simulation".

[fsloth]

No. An analogue system is not a simulation. A simulation is some approximation - a model - of a physical system that you compute with some finite precision.

An analogue is a physical entity with the same physics as some other thing.

A straightforward engineering example would be scale model of an airplane in a wind-tunnel.

[eesmith]

MONIAC was an analogue system which implemented a model - an approximation - of the UK economy. From https://en.wikipedia.org/wiki/MONIAC

> The MONIAC (Monetary National Income Analogue Computer), also known as the Phillips Hydraulic Computer and the Financephalograph, was created in 1949 by the New Zealand economist Bill Phillips to model the national economic processes of the United Kingdom, while Phillips was a student at the London School of Economics (LSE). The MONIAC was an analogue computer which used fluidic logic to model the workings of an economy. The MONIAC name may have been suggested by an association of money and ENIAC, an early electronic digital computer.

[nightpool]

You're getting confused between the words "Analog" and "Analogue".

Analogue: something that is similar or comparable to something else either in general or in some specific detail : something that is [analogous] to something else

Analog: of, relating to, or being a mechanism or device in which information is represented by continuously variable physical quantities

GP's comment has nothing to do with whether the simulation is "analog" (as opposed to digital), their point is that instead of being a "model" of a way a physical system might behave, the mechanism in this case has the exact same physical properties represented in a different form. An example might be doing a calculation in decimal vs doing it in binary—the exact steps you take will look different, and the answer might look strange, but there's absolutely no doubt about the fact that you'll get the same answer however you happen to compute it. Another synonym for the word "analogue" in the sense that the GP is using it might be "isomorphism"—you can prove that some transformation holds, and then you can do whatever you want to the transformed version and know that the results you get can be "transformed back" to the original form and reinterpreted.

[georgeburdell]

My (non-expert) brain maps it to this. Your output is a curve of some kind with the same mathematics as the problem you actually care about. The term “Quantum computing” can really refer to a number of quantum mechanical systems with fundamentally different properties (e.g., boson vs fermion) that have the same mathematics as some classes of problems

[Out_of_Characte]

I'm no expert but my understanding is that the word 'simulation' is a misnomer in this context. A better way to describe what they're trying to do is a proxy system. Effects observed in this proxy system would be used to predict interactions in reality. Whereas simulations are better at describing how real world interactions diverge from our models.

[magicalhippo]

> Kind of like how simulating anything in detail tends to be a lot slower than the actual thing you're simulating?

From the article[1]:

Our approach avoids the limitations of direct experiments on molecular systems, where only few observables such as spectra and scattering cross sections can be measured. [...] A further advantage comes from the ratio (r) of the ion’s natural timescale (ms) and the measurement speed (ns), leading to an increase in the observable timing resolution of r ∼ 10^6. This improves the achievable resolution of chemical dynamics measurements relative to ultrafast observations.

So seems this would be more like running a computer simulation with a super-short timestep, allowing you to extract more details of a process. It's only related to the wall-clock in that they're using a physical analog system, rather than a computer.

> Not following if there was any breakthrough here or not.

Again from the paper:

Our approach to quantum simulation using an MQB trapped-ion system makes chemical dynamics that are otherwise unmeasurable directly accessible in the laboratory. This is a key demonstration of the utility of small-scale quantum computational devices to offer practical insights into chemical dynamics and resolve intractable problems in chemical physics.

Seems the measurement itself was a showcase for the techniques developed.

[1]: https://arxiv.org/abs/2211.07320

[marginalia_nu]

Was expecting they'd be using the quantum Zeno effect[1], which really does slow down the time-evolution of what you're measuring.

[1] https://en.wikipedia.org/wiki/Quantum_Zeno_effect