Can offset presses actually do 1% increments of C or M or Y?

Question

When trying to achieve very bright colors, with no black, is it valid to move the other components in 1% increments: for example, 100*5*1*0? Or is that expecting a level of precision that is more than the press can actually deliver.
Years ago I worked with a printer who advised always rounding each of the 4 components to the closest multiple of 5. Obviously, controls are much more sophisticated these days. And, obviously, we spec out colors with 1% increments quite frequently. But can a press, in this case, one using the GRACol2006 profile, actually deliver? them?
Please note I'm not asking if 100*5*1*0 would be visibly different from 100*5*0*0. What I want to know is if I ask for 1% in one of the CMY channels, am I actually going to get it?

Answer 1

I found an interesting doc by Heidelberg:
http://www.heidelberg.com/corp/www/binaries/bin/files/dotcom/en/prinect/prinect_standardiz ation.pdf
Page 21 shows some tolerances for standardized offset processes:
Dot Gain: +-3% at the ends, +-4% in the middle
Inks C,M,Y,K: CIELab dE +-5%, Density +-8%
This sheds some light on the achievable absolute accuracy of levels and colors.
Whether ink inkrements of 1% can be reproduced measurably (density) is a matter
of process noise and parameter drifting, as John says..
But in fact this was the last question:
What I want to know is if I ask for 1% in one of the CMY channels, am I actually going to get it?
In my opinion it's impossible to achieve a safe reproduction of 1% ink.
Best regards --Gernot Hoffmann