I as just wandering if this is a practice that people still carry out? I was always told that the very last step before printing you should set your output levels for printing. I use an Epson 2200 and I am unable to distiguish black past a certain value (often around a value of 10-15 depending on the paper). The same applies for white (normaly 250-253) is distiguishable.

Although I now use Lightroom with a custome ICC profile , I still round trip into PS to do my sharpnening and local contrast adjustment - then use a levels adjustment layer and set the output levels to 10 and 251. It does seem to flatten out the contrast though which means prints can come out a bit flat.

Do people bother with this practice? Any other recomendations?

Jon

Jon, if you have a good (custom) profile, then you don't need to do the levels adjustment. If you are seeing blocked up shadows or flat highlights, then the profile is suboptimal.

Jon
I do this in Curves rather than Levels. Place a point at the mid tone and one towards the highlights to lock them, then move the black point up until the darkest significant shadow reaches 15 (RGB), and the same for the brightest highlight. I normally set the values when editing, rather at the end.


Brian Price

Eric is right. The profile should maximize tonal separation at the ends of the scale. In practice this requires multiple readings of shadow patches in particular before building the profile. This precaution often separates excellent custom profiles from poor ones. Perhaps due to time expediency , the latter , unfortunately abound.
Production of a clipping curve negates the effectiveness of the profile and prevents the attainment of max. black.
In practice sometimes a small reduction of white is useful to prevent blank highlights, which show as surface effects on some papers in files that are themselves a little clipped. The eye/mind resets it's own white point.
In cases where shadows are clipped on the file I find it preferable to lower Color Density 10-20% in
Paper Configuration (Ep 9800, PS/Print) rather than adjust curves which lowers global contrast.
Cheers,
Brian,
www.pharoseditions.com.au

Brian, I am assuming this has to do with it being harder to get reliable readings of darker patches (due to more noise in the measurements)? Do you do multiple readings, then average?

Eric