The common f-number we use in photography embodies the assumption that the magnification is much less than 1. When you're taking a picture of a mountain 3 miles wide and that will be imaged 30 mm wide on the sensor, the magnification is so small that the common f-number approximation is quite valid.

However, this breaks down as magnification approaches 1, and fails dramatically above 1. The "real" amount of light you get is described by the f-number times 1 / (1 + M)2. M is the magnification, which is the linear dimension of the actual object divided by the linear dimension of its image on the image plane. As you can see, this correction factor is close enough to 1 as long as the magnification stays well below 1. At exactly 1:1 magnification, note that this factor is 1/4, which means you are two f-stops down.

So how close to macro is close enough so that you have to consider this effect? We can work the equation backwards and see what magnification results in the minimum factor we care about. Let's say anything less than a 1/4 f-stop is slop so that we don't care. 1/4 f-stop means the equation must result in 2-1/4 = .841, which results in M of .095, or a reduction factor of 11. So as long as you aren't closer than 11:1, you can pretty much ignore this effect. For a "35mm" frame, which is 36x24 mm in size, this means you are fine as long as the subject is 400x270mm, or about 16x11 inches.

Of course any thru the lens metering will see the light actually delivered thru the lens and automatically compensate for this effect. Since just about every camera has that nowadays, your pictures will still be properly exposed, but you might notice the increasingly worse tradeoff between f-stop and shutter speed as you get more into macro range.