Hi Andrew, I did some tests a few years ago by printing charts patches to exhibit small deltaE variations. My conclusion at the time was that the limit of deltaE of ≤ 1 was fine for colours with values of a or b > ±10 but as the colour approached the neutral (L) axis in the range 40 ≤ L ≤ 70 smaller values of deltaE were clearly visible. Most of my test subjects could detect differences of about 0.3 when the patches were compared side-by-side, and (almost a neutral grey). My conclusion from this was not surprising - for all my test subjects the Lab space was not perfectly perceptually uniform, and detecting differences in deltaE was significantly easier for values of -3 ≤ (a or b) ≤ 3 when comparing patches side-by-side. Simon simon@richstitch.com