I have recently did a play on Jean Carles method just using different math, and the presumption that I can tell when it is right. The idea being that at some point it is too much of one, and eventually it will not seem like too much. This certainly will not replace the full on experimentation, because it doesn't take into account aging, nor side by side comparison, it is good for something.

Simply start with your two ingredients by putting 1 drop each into a container and using strips or skin if you are in dilution sniff. If it is too strong for one of them, then add 1 drop of the other.

What this does is goes from 1 drop each = 5:5 ratio that that Carles had. Now, if you put in another drop in one so that you have 1:2 that is similar to 3:7. Now, if you went too far you can add another drop of each one giving you 2:3 which is exactly 4:6. If it wasn't too strong, simply adding 2 drops to one side gives you 1:4 which gets you 2:8 and so on like that...

1:1 = 5:5 -- if it smells right 5x each side
2:3 = 4:6 -- if it smells right double both sides
2:5 ~ 3:7 -- if it smells right you can simply add 1 and 2 to actually make it 3:7

4:16 = 2:8 -- will give you more than 10 units
4:36 = 1:9 -- will give you more than 10 units

1:4 = 2:8 -- if it smells right double both sides
1:9 = 1:9 -- if it smells right you are already at 10 units

The idea is that you can go until it is how you want it to smell, or if it goes too far, you can pull back by simply adding equal single increments to both sides. If you are walking the the numbers like this, then there is no numerical difference except the 2:5 but it is all forward progression. And with some calculations you can actually walk backwards. I just showed the ones that come out even, but you can see there are more. It is simple enough calculation. A really nice table can be done in excel. But the real benefit is that you are not carrying 5 samples but only 1. But you sacrifice the side by side comparison ability, unless you simply make strips along the way and compare them later.