This isnt that complicated. Digital sensors have dynamic range just like film does. At some point they clip at higher light levels and at some point they just produce noise at lower light levels. THAT is the dynamic range of the sensor itself and it doesn't matter what the bit depth of the A/D is after the sensor, you cannot get more Dynamic Range in the output than the sensor itself has by increasing the bit depth.
That said, if the bit depth is too little compared to the dynamic range of the sensor there will be problems because without enough output shades of gray (bit depth) there would be obvious visable banding as the sensor's dynamic range increases due to technical improvements. But you have to remember that increasing the bit depth of the output isnt increasing the dynamic range of the sensor, it is only making whatever dynamic range the sensor has fully usable. I do not claim to be an expert on this but my understanding on this is that the sensors are limiting dynamic range at this point, not the bit depth of the a/d conversions so just increasing bit depths of todays sensors will not increase the recorded dynamic range, just more more invisibly finer shades of gray possible out of the same limited recorded dynamic range. JCO -----Original Message----- From: Shel Belinkoff [mailto:[EMAIL PROTECTED] Sent: Friday, November 05, 2004 1:47 AM To: [EMAIL PROTECTED] Subject: RE: USAF target and resolution tests Greater bit depth provides greater dynamic range. That was discussed here a week or so past, and that's what I understood from the likes of John Francis and others, whose opinions and technical expertise I have come to trust Anyway, all the technical talk gives me a headache. Amplitude shmaplitude (to paraphrase another thread), I'm only reporting what I've seen and what I've come to understand from those, both on and off this list, who are true experts when it comes to working with digital files. Like I said, I'm mostly ignorant about these things, and maybe my terminology is sometimes incorrect, but I stand by my statement, qualifiers and all. So, if you want to argue your point on technical grounds and theory, I'm outta here, because I just don't know enough of the terminology and will get lost very easily. I just know what I've seen and what the experts have shown and told me. Shel > [Original Message] > From: J. C. O'Connell <[EMAIL PROTECTED]> > To: <[EMAIL PROTECTED]> > Date: 11/4/2004 10:29:16 PM > Subject: RE: USAF target and resolution tests > > I think you might have misunderstanding of what higher bit > depth means. > > Bit depth is a amplitude resolution parameter, not > a dynamic range parameter. Dynamic range of a digital sensor is > independent of the bit depth of the output. More bits does not mean > more dynamic range, it just means more gray shades. > > Bit depth is the number of grey shades from **output** pure black > to **output** pure white, dynamic range on the other hand is the number > of **input** fstops between the > sensor's pure white (clipping)output and the sensor's dark noise(pure > black) output . Two different digital > sensors can have same bit depth but different dynamic range > or vice versa.... > > What I was referring to about specialized films is that super low > contrast films could have a greater DYNAMIC RANGE than digital for > extremely contrasty scenes and super high contrast films could have a > better amplitude resolution (bit depth) for extremely low contrast > scenes than digital. > > > JCO > > -----Original Message----- > From: Shel Belinkoff [mailto:[EMAIL PROTECTED] > Sent: Friday, November 05, 2004 12:58 AM > To: [EMAIL PROTECTED] > Subject: RE: USAF target and resolution tests > > > I'm saying that, from what I've seen of high bit RAW files, yes, I > believe they can. Again, take my comments with a grain of salt (and > note the > qualifiers) as I'm still just learning this stuff, and have just > started to work with digi RAW files. Remember, digital can be very > well matched with the scene, and there's control for manipulation > throughout the workflow. > > Shel > > > > [Original Message] > > From: J. C. O'Connell <[EMAIL PROTECTED]> > > To: <[EMAIL PROTECTED]> > > Date: 11/4/2004 9:50:03 PM > > Subject: RE: USAF target and resolution tests > > > > Are you saying that digital sensors can capture as wide > > a scene contrast range as the widest range (low contrast) color neg > > films can? > > > > Are you saying that digital sensors can capture as narrow > > a scene contrast range as accurately range as the highest contrast > > color slide films? > > > > JCO > > > > -----Original Message----- > > From: Shel Belinkoff [mailto:[EMAIL PROTECTED] > > Sent: Friday, November 05, 2004 12:12 AM > > To: [EMAIL PROTECTED] > > Subject: RE: USAF target and resolution tests > > > > > > I'm not sure your assessment of digital (especially 12-bit or > > greater > > RAW > > files) is correct. Maybe with the 8-bit digicams that are so much in > > use, but not with a higher end DSLR with 12-bit or 14-bit capture. > > > > Shel > > > > > > > [Original Message] > > > From: J. C. O'Connell <[EMAIL PROTECTED]> > > > > > > > Film has major advantage over digital in that > > > the film type selection can be matched to the requirements. > > > Digital > > > is > > > > > more of a general purpose capture which I do not think would do as > > > well as film on very low or very high contrast scenes shot with > > > correct types of film. > > >
