This applies only to French Press, cupping and other immersion methods were a cap of ground coffee forms above the liquid (if you stir, this doesn't apply; and worse yet, you are heading for the fifth circle of coffee hell where the over-extracting gluttons reside ).

If you smell the cap, it starts out similar to the dry aroma, with mostly the fruit, flowers origin flavors. As the cup steeps, you add deeper chocolates, caramels and other roasty aromas. So the procedure is simple enough -- as soon as it starts smelling funky, push down, finish the extraction and pour.

There's a caveat. I have an aged Sumatra that starts out smelling like rotting wood and which takes about two minutes to smell better. In this case, it goes from good to the roasted funk smell after about four minutes. I'm guessing initial funkiness may be the case with fermented coffees as well.

In any case, I stumbled on this idea brewing this year's Nekisse. This is a coffee that was mistreated on its way from Ethiopia -- It is hard to roast and brew, and tends to taste faded. But even in this sorry state, a few flashes of its magic are left. I brewed this coffee with my regular timing, and it tasted like I was drinking it through an old burlap sack. After a few tries, I realized that the cap went from floral to chocolate in about 45 seconds, and from there to an old burlap sack in just under two minutes. Now I get a tasty cup by keeping it this under extracted, and raising the coffee to water ratio to compensate. That is, I brew it half the normal time, and at 1.5 times the normal coffee to water ratio.

Anybody else have experiences where they had to drastically change the brewing time to get a good cup?

When I do a vac pot, I don't rely on time. I just keep on smelling for the right aroma then cut the heat. This led to way more successes than timing the process. Been doing it for years and never had trouble even with decafs. I found that the "time method" tends to need adjustment for different coffees, but the "snif method" doesn't

I tried the idea with an Aeropress. I had a hard time distinguishing changes in aroma over time. I wonder if the small amount of water relative to the amount of coffee diminishes the effect. I'll be switching to a different blend soon and will try again.

I've used this technique for years with tea before I was ever interested in coffee, so when I started brewing coffee I adapted it pretty quickly, I think it's very useful for any method where you can easily smell the coffee while it's brewing

Would the sniff test work, though, for methods that require a drawdown (CCD is what jumps to mind). For those, wouldn't you want to begin the drawdown just before the smell is appropriate?

But then I'm ignoring my first question... how do I know the smell is right?

Pardon me, but I think my head is about to a'splode.