Hi loved your article on scanning with the Nikon 800e.As it happens my old scanner has died and I dont want to use a epson flatbed scanner.Question – you only scanned the film with one frame ie you didnt take two or three frames and stitch them together ? This would result in a huge file which might even exceed the drum scanners results. How did you find the tonal range of the images taken with the 800e compared with the scanners.My gut feeling tell it should be better but one never knows.Any advise would be appreciated.There are no decent scanners coming through.
Thanks! glad it was helpful. I only made one shot scans without stitching. I’m not sure there would be that much benefit to a multi-shot approach, but maybe with very slow films you could extract a bit more resolution. It’s also possible that a set of blended images could reduce noise, but I doubt I’d be able to see it amongst the film grain.
With negative film you have a relatively limited density range that you are trying to capture. Assuming a density range of 1.2, which is typical, that’s only 4 stops. If you open the raw images in an app like RawDigger you can see just how much of the dynamic range the camera is capable of recording is used, and as expected it’s 4 stops. This makes it very important to expose to the right so you can get maximal tonal data captured. Even four stops exposed to the right with a 14 bit camera gives you a max of 15,360 shades of gray (2^14 – 2^10 or 16,384 – 1024). That’s the theoretical best and assumes perfect exposure. Reviewing my scans I get about half of that, which I feel is still good.
Note that this is also a limitation with CCD scanners, and some drum scanners. And with the scanners you have less control over the exposure so it’s harder to make a well exposed scan. Plus I think most scanners are 12 bit at best. The one exception is the Aztek/Howtek drum scanners with the capability to adjust the Log Amps to ideally set the brightness range sampled which can extract the full bit depth their analog to digital converts are capable of.
If we look at my Epson 4870 which claims a DR of 3.8. That equates to 12 bits per channel. With ideal exposure of a negative you are at 3840 shades of gray (2^12 – 2^8 or 4094 – 256). But with the Epson software I have a lot of trouble getting the scans close to ideal so the results are frequently not even half that good.
Also remember that since we are talking negatives any posterization will first show up in the highlights (you must invert the scan).
That’s a long way of saying with a 14bit capable DSLR and careful exposure you can get very good results. And with sloppy results you might get total crap, but that’s true of anything.
Hi loved your article on scanning with the Nikon 800e.As it happens my old scanner has died and I dont want to use a epson flatbed scanner.Question – you only scanned the film with one frame ie you didnt take two or three frames and stitch them together ? This would result in a huge file which might even exceed the drum scanners results. How did you find the tonal range of the images taken with the 800e compared with the scanners.My gut feeling tell it should be better but one never knows.Any advise would be appreciated.There are no decent scanners coming through.
Thanks! glad it was helpful. I only made one shot scans without stitching. I’m not sure there would be that much benefit to a multi-shot approach, but maybe with very slow films you could extract a bit more resolution. It’s also possible that a set of blended images could reduce noise, but I doubt I’d be able to see it amongst the film grain.
With negative film you have a relatively limited density range that you are trying to capture. Assuming a density range of 1.2, which is typical, that’s only 4 stops. If you open the raw images in an app like RawDigger you can see just how much of the dynamic range the camera is capable of recording is used, and as expected it’s 4 stops. This makes it very important to expose to the right so you can get maximal tonal data captured. Even four stops exposed to the right with a 14 bit camera gives you a max of 15,360 shades of gray (2^14 – 2^10 or 16,384 – 1024). That’s the theoretical best and assumes perfect exposure. Reviewing my scans I get about half of that, which I feel is still good.
Note that this is also a limitation with CCD scanners, and some drum scanners. And with the scanners you have less control over the exposure so it’s harder to make a well exposed scan. Plus I think most scanners are 12 bit at best. The one exception is the Aztek/Howtek drum scanners with the capability to adjust the Log Amps to ideally set the brightness range sampled which can extract the full bit depth their analog to digital converts are capable of.
If we look at my Epson 4870 which claims a DR of 3.8. That equates to 12 bits per channel. With ideal exposure of a negative you are at 3840 shades of gray (2^12 – 2^8 or 4094 – 256). But with the Epson software I have a lot of trouble getting the scans close to ideal so the results are frequently not even half that good.
Also remember that since we are talking negatives any posterization will first show up in the highlights (you must invert the scan).
That’s a long way of saying with a 14bit capable DSLR and careful exposure you can get very good results. And with sloppy results you might get total crap, but that’s true of anything.