Yes, it does make sense to me. I think conclusions drawn from this thought experiment are not entirely correct. Under your experiment when sending you spatially create a beam - there is this beam as wide as parabolic dish and no light anywhere else. With zero interference (using the word quite freely) this beam "will be just as strong no matter how far away the receiver is". With roles reversed, you do not necessarily get directional transmitter. For the sake of argument consider omnidirectional light bulb (the same bulb without parabolic dish) as sender. Classical intro/ELI5 description of transmit power is fixed number of spatially evenly distributed omnidirectional "rays". The larger the distance, the lower number of "rays" you receive per unit of area (receive power scales with surface area of transmit profile, usually r^2). By adding a parabolic dish you effectively increase receive area to size of dish instead of sensor size. When sending, the area over which power is dissipated is constant instead of scaling with distance.

> You're claiming it won't matter how far away the sender is from the focus, which makes no sense.

You are correct in that it makes no sense, but the claim was "every improvement in sending is completely equivalent in receiving for a system that uses the same antennas for both directions", key word improvement. While this is not entirely correct, this approximation (that antenna improvement works both ways equally) is pretty close to reality in practical far field applications.

> With roles reversed, you do not necessarily get directional transmitter.

I don't get why you put "necessarily". You do not, period.

> [...snip...] but the claim was "every improvement in sending is completely equivalent in receiving for a system that uses the same antennas for both directions", key word improvement.

Yes, the claim is wrong:

1. With the sender at the focus and the receiver farther away, the mirror reduces the loss (due to distance, since everything else is assumed ideal as you already acknowledged) to ZERO. You literally cannot do better than reducing your loss to zero.

2. With the receiver at the focus and the sender farther away, the mirror DOES NOT even remotely reduce the loss due to the distance to zero. Loss is still ~1/r^2. Because the mirror only gets so much of the omnidirectional light. This should be very obvious.

I don't care how you slice and dice this, and I'm tired of arguing something this obvious, so I won't keep commenting. 1 is not equal to 1/r^2.

> I'm tired of arguing something this obvious

You're arguing a scenario that is impossible both practically and theoretically and then calling others pedants for pointing out the fundamental flaws in your thought experiment. You're literally arguing against the textbook and then asserting that your thought experiment is close enough to make your point. It might be time for you to stop assuming that everyone else is an idiot and entertain the notion that you might actually be wrong.

The trick here (in this Wi-Fi example) is that the end you control is the end with parabolic antenna - if you can make it to collect enough energy for RX from an omnidirectional source, you can also make it send enough energy towards that source for its antenna to pick it up.

But you're right that "every improvement in sending is completely equivalent in receiving for a system that uses the same antennas for both directions of communication" is wrong; there are different ways of improving TX and RX, even though some overlap a bit, to some extent.

No, as long as antennas and signal power is concerned, I stand by what I said: every improvement in transmission power has a equivalent improvement in receiving power. If you have a more specific doubt, please, feel free to ask.

> If you have a more specific doubt, please, feel free to ask.

Unless by "feel free to ask" you mean "I don't plan to answer", this doesn't really seem like a sincere suggestion. I already did this and you didn't answer.

I'm still on the move, but I'll answer eventually. Your previous post did not contain any questions.

Your thought experiment is not in line with what OP is claiming. You would need to vary the dish size and see how that affects tx/rx.

Have you studied physics and antenna theory?

> Light will come out parallel

No it won't, it's diffraction-limited. The best you can theoretically do is an https://en.wikipedia.org/wiki/Airy_disk

You cannot construct, with any arrangement of mirrors or lasers, a beam that does not diverge.

I'm not going to get into to the "gain wars" (yes, text bok, real-world you can have just one "good" (high gain) antenna, and it will work just fine for send and receive) - but your answer prompts the obvious question: What about lasers? Turns out, before optical lasers there were masers - microwave lasers, bit they were (are?) very limited power:

https://www.scientificamerican.com/article/first-practical-m...

I was assuming geometric optics... because it was simple enough to get the point across. Which of course you knew, but took the liberty to 'correct' anyway. It's quite impressive how you can always count on HN pedants to deliberately go out of their way to technically-correct you while making sure to completely miss the actual point you're trying to make.

You're the one that started in by trying to wrongly "correct" the statement "every improvement in sending is completely equivalent in receiving for a system that uses the same antennas for both directions of communication" (which is actually correct).

The fact that you can't have perfectly collimated beams and must make do with antenna pattern "lobes" is actually important here.

Think of a parabolic dish on a microphone at a sporting event. The dish blocks noise and interference from the directions you don't want to listen to, and less noise and interference is as useful as if the sender had more power. Same result of better s/n either way.

Or a TV satellite dish. If you were correct, you could rip off the dish and point the antenna forwards instead of backwards, and still watch TV. Not gonna happen.

> The dish blocks noise and interference from the directions you don't want to listen to

You shouldn't need to talk about noise or interference for the argument to make sense. Just imagine there isn't any. The argument yields absurd results either way: the power certainly does depend on how far the sender is from the focus.

You are assuming the mirror is parabolic.