checking very large number for primality.

[a1call]

Originally Posted by Nick

It is time for you to start thinking a little deeper before posting here.

At what fraction of the speed of light do electrical signals travel in wires?

For a task which would take many thousands of years, would a switch to optical signalling make any practical difference?

Once you have worked that out, you may understand the incredulous reactions.

Electronic logic gate (Building blocks of any typical computer) switching times are much slower than and are irrelevant to speed of electrons (or electric signal more accurately) in an electric wire.
https://en.wikipedia.org/wiki/Propag...ay#Electronics

An optical gate (say the AND gate) based on interference of two input beams will switch at the speed of light (essentially instantly).
Yet this is not what I was referring to as an optical computer. There are other approaches to computation that are task specific and are not logic gate based.
For the record, I stand by my assertion.

[a1call]

A general estimate from Scientific American:

A computer based on beams of light rather than electric currents might be capable of a trillion operations per second. The crucial component, an optical analogue of the transistor, has been built

https://www.scientificamerican.com/a...ical-computer/

Again, there are potentially much more instant approaches to task specific computation than Logic-Gate based computers. What are they?
Well they are mostly proprietary at the moment.

[kladner]

Originally Posted by a1call

Here is the Wikipedia article for those interested:
https://en.m.wikipedia.org/wiki/Optical_computing

Interesting article. There do seem to be many hurdles to surmount.

(Aside: I must get my Irony Detector detuned a bit.)

[retina]

Originally Posted by a1call

... the speed of light (essentially instantly).

Not even close.

[a1call]

Originally Posted by retina

Not even close.

In actuality, I spent a number of years researching optical gates (more than a decade ago). I came up with a concept and discussed it with a leading Scientist in the field. The title of this post is pretty much what I got back from it.

I was under the false assumption that different beams of light readily interfered with each other. After all red and green pixel's light emitted from a monitor (there were only tube monitors available then) would interfere to produce yellow light. If things were only so simple it would be very simple to design an AND and an Inverter gates, which would theoretically entail the possibility of designing any complex logic gates or electronic/computer circuits. In such a circuit the red light would represent a binary 0 while the green light would represent binary 1.
Alas it turns out a computer monitor fools our eyes by simultaneous simulation of red and green receptors in our eyes (a characteristic of actual yellow light) into thinking we are seeing yellow light emission. In fact the yellow light frequency is quite different than the interference frequency of the green and red light (green-light-frequency - red-light-frequency).

But fortunately things have progressed since.
you might want to look up Non-Linear Media in this context.

ETA. Another prerequisite for 2 beams of light interfering with each other is that they must be coherent (This too took a number of years to sink in).

[retina]

Originally Posted by a1call

In actuality, I spent a number of years researching optical gates (more than a decade ago). I came up with a concept and discussed it with a leading Scientist in the field. The title of this post is pretty much what I got back from it.

I was under the false assumption that different beams of light readily interfered with each other. After all red and green pixel's light emitted from a monitor (there were only tube monitors available then) would interfere to produce yellow light. If things were only so simple it would be very simple to design an AND and an Inverter gates, which would theoretically entail the possibility of designing any complex logic gates or electronic/computer circuits. In such a circuit the red light would represent a binary 0 while the green light would represent binary 1.
Alas it turns out a computer monitor fools our eyes by simultaneous simulation of red and green receptors in our eyes (a characteristic of actual yellow light) into thinking we are seeing yellow light emission. In fact the yellow light frequency is quite different than interference frequency of the green and red light (green-light-frequency - red-light-frequency).

But fortunately things have progressed since.
you might want to look up Non-Linear Media in this context.

ETA. Another prerequisite for 2 beams of light interfering with each other is that they must be coherent (This too took a number of years to sink in).

None of that addresses your assertion in any way about the speed of light being instant. And none of it has anything to do with the topic here about the time and resources required to test a number with 10^1523 digits. What are you talking about?

[a1call]

Originally Posted by retina

None of that addresses your assertion in any way about the speed of light being instant. And none of it has anything to do with the topic here about the time and resources required to test a number with 10^1523 digits. What are you talking about?

used for saying that something is mostly true, but not completely trueThe list is essentially complete.

http://www.macmillandictionary.com/d...sh/essentially

Also instantly means quickly, not instantaneously. I was not implying 0 delay, but extremely quickly.

I think my post is on topic, since the OP asks about a fast alternative to computing the primality of a large integer.

[a1call]

Here is the closest Wikipedia match I can find in regards to Non-Linear Media.

https://en.wikipedia.org/wiki/Nonlinear_optics

[VBCurtis]

Let's start with the number of particles in the universe, 10^85 give or take a magnitude or ten.

Exactly how do you propose to do a computation on a number that contains 10^1523 digits with a computer that fits in this universe? Note that merely expressing this number without using exponential shorthand would require each particle in the universe to hold 10^1400 digits.

But don't let total ignorance of the size of the number (let alone the size of the task of primality testing) get in the way of telling us how close you are to a magical computer using interference of light to make yellow out of red and green photons.

Perhaps you could do the thread a favor and estimate the additional magnitude of computational effort is required per digit added to the exponent. For instance, how much harder is prp-testing a number with 10 million digits vs 1 million digits? That's one more digit in the power. We can presently test numbers with 10 million digits. How hard is 100-million digits? 1 billion? 10 billion? (Hint: you won't see a test performed on a 100 billion digit number in your lifetime)

100 billion digits is 4 orders of magnitude bigger than present tests.
OP's number is One thousand five hundred orders of magnitude bigger.

You're going to need a bigger universe.

[a1call]

How would that argument not apply to the following comment by yourself earlier( not to mention with conventional computing)? Hmmm?

Originally Posted by VBCurtis

Finding a factor to show compositeness is OP's only hope.

[Batalov]

...is the highest number. Fuggeddaboudit!