The whole sum on one line
Power out, plus both antenna gains, minus both cable losses, minus path loss, minus anything else you know about. With the defaults: 20 dBm out, 2.1 dBi at each end, half a decibel of cable at each end, and 106.2 dB of free space loss at 2 km on 2440 MHz. That lands at minus 82.7 dBm at the receiver. Against a sensitivity of minus 95 dBm there is 12.3 dB of margin.
Path loss is the term that dominates everything. It is 106 dB against a transmitter contributing 20. Nothing else in the chain is within an order of magnitude of it, which is why the interesting question is always distance rather than power.
Six decibels, and where it comes from
Free space loss goes as the square of distance, so doubling the range always costs 6.02 dB and no more. That number is the key to the whole page. Six decibels found anywhere — more transmit power, more antenna gain at either end, less cable loss, a more sensitive receiver — doubles the reach identically.
Where they are not identical is what they cost. Quadrupling transmit power for that 6 dB means four times the current at the transmitter and, in most places, a limit somewhere that is not yours to set. Six dB of antenna gain on the ground station costs a bracket and a bit of aiming discipline, and it is counted once on the way out and, if the same antenna receives, again on the way back. That is why link engineering is mostly antenna work.
Margin is not spare, it is the budget for reality
The default figures put zero margin at 8.19 km, which sounds like the range. It is not. That is the distance at which a perfect free space path with both antennas ideally oriented delivers exactly the weakest signal the receiver can decode, with nothing left for a rolling airframe, a hand in front of the transmitter, a reflection off a wet field or a tree line. Hold 15 dB back and the same sum gives 1.46 km.
Both numbers come from the same arithmetic. The difference between them is entirely the question of how much reality you have budgeted for, which is a judgement rather than a calculation, and the page takes it as an input for that reason.
What this cannot tell you
Whether the far end is above the horizon at all, whether the path clears the terrain, what the ground reflection does, how the antenna pattern actually looks once it is strapped to a fuselage, what else is transmitting on the band that afternoon, and how the receiver behaves near its threshold rather than at it. It also says nothing about what power you are permitted to radiate, on what band, under what licence — that is set by the radio authority where you are standing and it goes into the form as a figure you supply.
Questions people ask
What is free space path loss and why is it so large?
It is the spreading of a signal over the surface of an expanding sphere, and nothing to do with anything absorbing it. At 2440 MHz over 2 km it is 106.2 dB, which means the received power is about one forty-billionth of what was radiated. It grows 6.02 dB every time the distance doubles and 6.02 dB every time the frequency doubles, which is why the same radio reaches further on a lower band.
How much link margin should I hold?
That is a judgement this page will not make for you, which is why it is a field. What the margin has to absorb is a rolling airframe swinging the antenna pattern about, your own body between the transmitter and the model, ground reflections, foliage, and the difference between a manufacturer sensitivity figure and the same receiver in a noisy field. Whatever number you settle on, the page shows the distance at which you still have it.
Is more transmit power or a better antenna the better upgrade?
Decibel for decibel they are identical to the sum, so compare what each costs. Six dB of transmit power is a quadrupling of power and current, at one end only. Six dB of antenna gain is a bracket, weighs nothing in most ground stations, and counts on both the transmit and receive paths if the same antenna does both. The trade is that gain comes from narrowing the pattern, so a directional antenna has to be pointed.
Does this tell me my actual range?
No, and it should not be read that way. It gives the distance at which a clean free space path delivers exactly your stated sensitivity, which is the best case that physics allows and nothing like a field result. Real paths lose more, often much more. The number to plan around is the one where you still hold your margin, and even that is arithmetic rather than a promise.
What transmit power am I allowed to use?
That is set by the radio authority of the country you are in, for that band, for that kind of use, and this page holds no figure of its own. If you enter one, the page subtracts the radiated power from it and prints the difference — that is a subtraction between two numbers you supplied, not a compliance check, and it is not a verdict on anything.