So I just finished a really fun little experiment, and the results are certainly interesting.
I hooked up our scan tool, and looked at ambient temperature, coolant temp, and intake air temp. I ran two runs, one with the stock filter setup, and one with the snorkel removed. This simulates the same as if I had a big round filter sitting on top of the transmission.
I completed this all within about 20 minutes, so ambient didn't change, and was 10 degrees (all temps in Celsius). Engine was at full operating temp, 85.5 degrees at the time of measurement. Runs were completed cruising on the same stretch of road, in the same direction to account for wind, and at the same speed (100 km/h).
With the factory intake in place, I had an intake temp of 18.5 degrees.
With the snorkel removed, that temp jumps to 28.5 degrees, a full 10 degrees higher.
The rule of thumb for intake temp vs. Hp is every 10 degrees Fahrenheit increase will lower HP by 1%. So...
18C = 64F. 28C = 82F. 82 - 64 = 18F. So approximately a 2% HP loss would be expected due to the higher intake temps that the 'improved' open-filter will 'gain'.
Rather interesting... sure my methods are rather simple, but I used the same info the car uses to make the power everyone is so interested in scavenging. So really... my methods may be about as complete as possible. Short of a dyno (which then removes the effects of motion, which is a HUGE contributor to the complete experiment) to measure actual output... this really is as good as it gets.