Paul,
At the risk of adding to the confusion or covering things you already know, here is a bit of background about D and Dmax:
In the context of light measurements, density (sometimes called optical density or absorbance) is defined as:
D = log(I0/I)
where I is the intensity of light detected by the sensor, and I0 is the input intensity, or the intensity if there were perfect transmission or reflectance. For reflectance measurements (eg scanning a print), I0 is the intensity of light that hits the surface, and I is the intensity of light that is reflected and detected. For transmission measurements (e.g. scanning a negative), I0 is the intensity of light that reaches the material from the source, and I is the intensity that reaches the sensor.
The important thing about the equation above is that it is logarithmic. D=1 means that 10% of the light is transmitted or reflected; D=2 means that 1% is transmitted or reflected.
The term Dmax has two common meanings. The first (which is one usually discussed on this form) is the maximum value of D attainable for a particular material (e.g. the darkest black for a paper and ink combination or the maximum density attainable for a particular film). The second meaning is the maximum density that a scanner or densitometer can record. I'll just address the second meaning here.
As Tony has said, the Dmax for a scanner is not all that easy to specify, as it depends on the signal to noise ratio one is willing to accept. The basic problem is this: as D increases, less light reaches the sensor, and the fall of is very rapid. From D=2 to D=3, the intensity goes from 1% to 0.1% of the initial intensity. The noise level, from scattered light and other factors, stays roughly the same as the detected intensity decreases, so that the S/N ratio decreases.
The ADC that Tony mentioned is the analog to digital converter, which does just what the name implies: it converts the analog measurement of light intensity into digital number. In principle, the final limitation on Dmax is the number of bits in the output number. If 8-bits are used, then the smallest number above zero that can be recorded is 1 and the largest is 255. If I0=255 and I=1, then Dmax=2.4. With 16-bits, Dmax=4.8. This is almost always what manufacturers cite as Dmax. It doesn't say anything more than the number of bits used by the ADC, and it says nothing about the S/N at a given D. As Tony says, its bogus.
The good news is that negative films rarely if ever have Dmax themselves that seriously challenge even a modest scanner. Slide films can be more problematic.
Since what you are concerned about is detail in the shadows, there is no real S/N problem, since the signal is the highest for those regions. What is important is how you set the high and low points when you scan. These basically adjust the range of light intensities that will be recorded. If the low density setting cuts off the histogram, then you will lose information from the low density regions of the negative (the shadows in the print). Similarly, you can lose highlight information if the histogram is cut off at that end. Generally, it's best to adjust the sliders so there is just a bit of extra room at both ends of the histogram. That ensures that you get the most information that the ADC is capable of.
Some people also suggest that you scan negatives in positive mode. This makes the scan look like a negative, which you then invert in Photoshop or whatever editing program you use. I don't know of any obvious reason that this should be better, but at one point, I did convince myself that I got better results with Nikon scanners that way, and it is the way I do things. But I'm not sure that it makes that much difference. It's worth fooling around with using your scanner and software.
My general experience is that scanners can recover just about every bit of detail that is present in the thin parts of a negative (and the dense parts, for that matter). I've never tried to compare, but it's probably at least easier to capture the shaddow detail in a scan than in a wet print.
I hope this helps more than it confuses.
David
--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Paul" <paulmwhiting@...> wrote:
Show quoted textHide quoted text
>
> I know what DMax is on a b/w print, it's where the blacks are at their darkest. But when I see specs for scanners, I get confused. Scanner specs include DMax, but is that for scan of a positive print or negative film? If I want good DMax on my b/w print, isn't that the region of least density (with some detail as in a shadow) on the negative? But if the scanner DMax is for a negative, that's going to be the highlights (with some detail so the highlight is not blown out) on the print, right?
>
> So which do DMax specs on the scanner mean? Nor sure if I'm explaining this very well, but I hope I'm making some sense!
>
> TIA,
>
> Paul
>