elkcub
Silicon Valley, California
The ability to discriminate between levels of brightness (luminance) obviously depends on how bright it is to start with and the size of the areas compared. For instance using big bold paterns at high light levels on a professional 10-bit computer monitor it's apparently possible to detect differences as small as 0.1%. Of course simultaneous comparisons within a field of view are not what we are talking about here. We are asking whether it is possible to spot a 3% difference in transmission between binoculars.
I've mentioned before the little experiments I've done on myself and others masking objectives to reduce transmission. When conducted in a way to represent the way most people compare binoculars, no one I've tried could spot a 10% difference. Here's an even simpler test for those with a light meter and a dimmable room light. Get someone else to edge the brightness up or down while you look away (as you would while comparing binoculars), checks the meter reading, and note any difference you think you might see. Some might find the results surprising.
David,
I've done the same and concluded that I must be an absolutely terrible judge of brightness. :-C
Ed