Measuring noise distortions helps us appreciate the noise tolerance of the scanner’s electronics, so this section can be found in any scanner-related review on our site. We have two test steps here.
The graphs below show the dependence of the amount of random noise on the reflecting capability of the sections of the Kodak IT8.7/2 Q-60E1 target (it is a transparent 4x5” target). Bigger SNR values are better, indicating the scanner’s higher tolerance to noise.
Correlated noise comes as an inherent defect of the CCD technology (electrical crosstalk of the light-sensitive array) as well as due to instability of the intensity of the scanner’s lamp. The amount of correlated noise (and of random noise, too) is somewhat lower with slide scanners than with flatbed ones. It is explained by the use of smaller lamps whose intensity is more stable. This minimizes noise deviations in the scanned image.
I put down only one correlated noise value for the F-3200 – 9.82 – although I usually produce two for flatbed scanners. The fact is I couldn’t place the target into the holder vertically. That’s also why I couldn’t scan line density patterns – the target didn’t fit into any of the holders.