The future of colour analog photography is....

donbga

Member
Joined
Nov 7, 2003
Messages
3,053
Format
Large Format Pan
Please post your question in a new thread. You have major topic drift in your post.

Thanks,

Don Bryant
 

Rudeofus

Member
Joined
Aug 13, 2009
Messages
5,081
Location
EU
Format
Medium Format
Like so many here I had big scanning troubles when I used negative film which disappeared as soon as I switched to slide film. The neg scans looks incredibly grainy and noisy despite the claims by professional scanner guys that neg film is so great for scanning.

Here are my thoughts:
  • Most scanners, especially non professional ones, are not nearly as good as their data sheets claim. pellicle himself posted measurements where a V700 barely yielded 8 or 9 bits of real information per color channel.
  • If you look at the data sheet of negative film, you see that the density range of each layer is quite narrow while the density ranges start at different points. In order to scan e.g. Fuji Pro 160S, you'd have to set the sensitivity of the scanner so that D 0.2 gives full range. The lowest density of the blue layer is, according to the data sheet, about D 1.0, which yields 10^(-1) = 1/10 of the light.
Result: you scan with a Dmin of D 0.2 and a resolution of 8 or 9 bits. The blue layer, which already starts at D 1.0, will be scanned with less than 5 or 6 bits, the rest is noise! If you add any type of post scan color correction to this mess, you reduce the number of relevant bits even more.

This also explains why pros don't share the problems we have with negative film: they have good scanners which yield more than 9 bits of actual information per channel, thereby completely avoiding the noisy mess I described above.
 

pellicle

Member
Joined
May 25, 2006
Messages
1,175
Location
Finland
Format
4x5 Format
pellicle himself posted measurements where a V700 barely yielded 8 or 9 bits of real information per color channel.

I did? I only have Epson 3200 (now sold), 4870 and 4990. Any information I give on V700 is usually obtained from my friends who have given me scans but I did not drive the process.

Definately I've said that the blue channel is weakest on scans and significantly challenges the Dmax

Its true that there are advantages in scanning slide, but that does not mean there are not advantages in scanning neg. I personally would like a scanner which like a drum scanner allowed me to tune the optimally density of the media to the input scope of the sensor without needing to buy a drum scanner, but I don't think such creature exists.

I can't recall off the top of my head if I've compared blue channel between the Epson and the Nikon for the same neg... sounds like a good thing to do
 

donbga

Member
Joined
Nov 7, 2003
Messages
3,053
Format
Large Format Pan
This also explains why pros don't share the problems we have with negative film:

They shoot digital color which eliminates scanning altogether.

Don Bryant
 

Rudeofus

Member
Joined
Aug 13, 2009
Messages
5,081
Location
EU
Format
Medium Format
I did? I only have Epson 3200 (now sold), 4870 and 4990. Any information I give on V700 is usually obtained from my friends who have given me scans but I did not drive the process.
Sorry, I got that wrong. You did post results from the 4990 here in this thread, and stated it's close to the V700.
Definately I've said that the blue channel is weakest on scans and significantly challenges the Dmax
I don't think it's a Dmax problem, especially with negative film which doesn't even get all that dense (see the data sheets! ). The problem is rather the small Dmax-Dmin combined with awful bit resolution. Not surprizingly, the blue channel starts at the highest Dmin, so it suffers most from this. Don't forget that D is a logarithmic measure while all digital sensors record in a linear fashion.

Let's take this Fuji film once more. Dmin for the red channel is 0.2, if we assume 8 bits per color channel we assign the value 255 to D 0.2 (before channel inversion). Dmin for blue is 1.0, which would be assigned the value 255/(10^(1.0-0.2)) = 255/(10^0.8) = 255/6.3 = 40. This means we have only 41 distinct blue channel values for the whole range from bright white to pitch black. The green channel does slightly better: 255/(10^(0.75-0.2)) = 255/(10^0.55) = 255/3.5 = 72, almost twice as many discrete steps as the blue channel. Only the red channel yields the full 256 steps.

Note that the 8 bit resolution per color channel does not come from the file format used to store the data. It comes from the poor signal to noise ratio of the scanner line sensor!
Its true that there are advantages in scanning slide, but that does not mean there are not advantages in scanning neg.
There may be great advantages in scanning negs on a really good scanner. The crap the electronics industry is throwing at us seems to perform better with slide material. I failed miserably when I scanned neg film with a Nikon Coolscan V ED and gave up on color negative film as a result. Which is a shame as I really did like the look of Portra and the new Ektar.
They shoot digital color which eliminates scanning altogether.
I meant professional film scanner folks, not professional photographers.

Do you have a link where I can see some of your scans?
I can post some samples here if you want. Tell me what you want to see.
 

Alan Klein

Member
Joined
Dec 12, 2010
Messages
1,067
Location
New Jersey .
Format
Multi Format
Rudeofus: How about posting some trans and negative scans indicating the scanner type and film type and size (Good and bad)? Maybe we can help you get better scans of negatives.

My own results don't seem to indicate too much different between these two. The biggest difference is between 35mm and medium format. 35mm blocks up more in the shadows. Alan
 

Rudeofus

Member
Joined
Aug 13, 2009
Messages
5,081
Location
EU
Format
Medium Format

Dear Alan,

thanks for your offer! Yes, I did scan 35mm film strips, using the Nikon Coolscan V ED, when I gave up on neg film. I have in the mean time moved to an Epson V700, mostly because I added an RZ67 to my equipment.

My biggest beef with the Coolscan/neg film scans was the unbearable noise in dark regions. Think of a picture with black background which was literally peppered with red and green pixels.
 

donbga

Member
Joined
Nov 7, 2003
Messages
3,053
Format
Large Format Pan
They do? Maybe you should let us all know....I know I missed that....as have many other pros I associate with.
Save the sarcasm fpr P-Net or DPReview. The majority of pros don't shoot film any longer.

Don
 

cupcake_ham

Member
Joined
Feb 10, 2007
Messages
56
Format
Plastic Cameras
Save the sarcasm fpr P-Net or DPReview. The majority of pros don't shoot film any longer.

Don

My apologies....I thought this forum was for film users. Most of us film users don't care what the digital shooters are using....and don't need the reminder.
 

Alan Klein

Member
Joined
Dec 12, 2010
Messages
1,067
Location
New Jersey .
Format
Multi Format

You ought to try negative scans on the V700. I've scanned RB67 negative and trans with my V600 which is not even as good as the V700 and have had succes with it. You can see my gallery. I also had problems with red blotches when I set backlight on the scan. Then I stopped selecting anything except ICE and did all my processing in Elements 8 afterwards. Seems to work better. Try scanning "flat" with the supplied Epson software and adjust afterwards in your post processing software and see what happens. Alan.
 

nolanr66

Member
Joined
Feb 5, 2006
Messages
283
Format
35mm


There are galleries over at APUG.
 
Cookies are required to use this site. You must accept them to continue using the site. Learn more…