So currently my conclusion is that I will still keep using the scanner approach, but I am more aware about the fact that with certain materials (such as silver based film), the error marginal is approx. The de-facto highlight zone for calibration is VIII, and around it's densities there are almost 0.1D difference with my scanner and densitometer I have used.ĭifference with other densitometer? I don't know, but I assume that two different densitometers would also give different readings. The difference between my scanner approach and densitometer has createst impact on zone system calibration. Light source effect is similar like in enlarger. The light source of densitometer might be diffused or it might not be. The light source of scanner is usually quite soft, like diffuser head. Also the type of the light source alters the result. The spectrum of the light source is not only thing that causes differencies. Which one gives more accurate reading from film X developed in developer Y remains mystery. But not for all kind of silver based films. So for stouffer wedge, both were 'calibrated'. As did my scanner when applying correction function. However densitometer readed stouffer wedge accurately. The densier the negative, the more difference there was. The readings were bit off from my scanner-densitometer readings. I found this recently when I measured my negatives with densitometer. The color of the scanner (or densitometer) light have it's own co-effect. BUT I have to warn that scanner, like real densitometer, does not give exact readings with all materials. The Y-axis is vuescan measured density and X-axis is real density on the stouffer wedge. Here is my correction curve as for example. My approach was to measure densities that vuescan gives from stouffer wedge, then create correction funtion which maps vuescan's (or scanners) densities to real densities that wedge represents. Before that, I used spotmeter as densitometer. I have been using vuescan as an densitometer with my epson 3200 flatbed scanner for years. Press the control key and the log density values will be displayed where the RGB values are normally shown. Below is the VueScan tab.Vuescan will function as a densitometer (within limits). It's just that this seemed like an interesting idea and a valid way to measure densities by comparing to a known density. PS: Please don't ask me to just buy a real densitometer. So if any forum members have a calibrated transmission wedge, Vuescan and a flatbed scanner (preferably one of the Epsons), I'd be grateful if they could try out this little experiment to see if it works. But I am away from home right now and don't have access to the gear to try this out. Now, of course, since it's a calibrated wedge, you know the actual density of that step and that should also be the density of the point you read in the negative (or pretty darn close).

Now in the scan of the step wedge, find the step which has the closest greyscale value to that slot.

Now, open both in photoshop in 8bit grayscale (I'm only thinking BW at the moment) and with the dropper tool read the value at a point in the picture where you want to read the density. So, for example, you would lock the exposure at 5 and scan the transmission wedge and then a negative at that same exposure. So I got to thinking - If I scan a calibrated transmission step wedge - the ones from Stouffer or Kodak - at a locked exposure, and then scan a negative at the same exposure on a flatbed like the Epsons (3200 to v750) comparing the greyscale values in photoshop could give us the density quite accurately. Now, doing this means that we can control the exposure from frame to frame and not have the scanner tweak its autoexposure. VueScan has a very useful exposure lock feature under the input tab which lets one lock the exposure on the various channels of scanners that support them. I was using VueScan the other day and wondered about something that I hope some forum members may help me with.

Vuescan multi exposure