Sabine Hossenfelder has thrown cold water on the Simulation Argument.
http://backreaction.blogspot.com/201...-computer.html
Among other things she says that you cannot get quantum effects from classical bits.
She also says that you would be able to see the clashes between discretization and general relativity.
Now bear in mind that a simulation would not have to simulate an entire universe. It need not simulate anything that is not at least indirectly observed by conscious observers in the simulation. And it does not need to simulate everything to a particle level, since we very rarely observe individual particles. The simulation need only model particles if there is some process that actually observes the action of individual particles.
So I suggest a thought experiment to test her claim that we would be able to detect these discrepancies. 100 scientists are locked in a room and have no communication with the outside world except via a group of scientists on the outside.
The scientists inside can devise experiments and send the instructions to the scientists outside to carry out.
But suppose the scientists inside become suspicious that the scientists outside are not really carrying out the experiments but rather calculating the expected results, adding a bit of fake randomness for instrumentation and measurement imprecision, and then sending the results back inside.
We assume that the outside scientists are just as smart as the inside scientists.
Is there an experiment that the inside scientists could devise where they would be able to detect that the experiment had not really been carried out?
Only science that has been achieved so far is allowed, no full quantum computing.
My guess is no.
(NB thisis under science rather than philosophy since Hossenfelder's is a scientific take on the simulation argument.
http://backreaction.blogspot.com/201...-computer.html
Among other things she says that you cannot get quantum effects from classical bits.
She also says that you would be able to see the clashes between discretization and general relativity.
Now bear in mind that a simulation would not have to simulate an entire universe. It need not simulate anything that is not at least indirectly observed by conscious observers in the simulation. And it does not need to simulate everything to a particle level, since we very rarely observe individual particles. The simulation need only model particles if there is some process that actually observes the action of individual particles.
So I suggest a thought experiment to test her claim that we would be able to detect these discrepancies. 100 scientists are locked in a room and have no communication with the outside world except via a group of scientists on the outside.
The scientists inside can devise experiments and send the instructions to the scientists outside to carry out.
But suppose the scientists inside become suspicious that the scientists outside are not really carrying out the experiments but rather calculating the expected results, adding a bit of fake randomness for instrumentation and measurement imprecision, and then sending the results back inside.
We assume that the outside scientists are just as smart as the inside scientists.
Is there an experiment that the inside scientists could devise where they would be able to detect that the experiment had not really been carried out?
Only science that has been achieved so far is allowed, no full quantum computing.
My guess is no.
(NB thisis under science rather than philosophy since Hossenfelder's is a scientific take on the simulation argument.
via International Skeptics Forum http://bit.ly/2SUnTt2
Aucun commentaire:
Enregistrer un commentaire