----- Original Message ----- From: Herb Chong Subject: Re: Dumb Newbie Q - What Color ARE Color Negatives?
> Message text written by INTERNET:[EMAIL PROTECTED] > >> you have some influence, but not a lot. you take a picture of something > > remotely resembling normal contrast and color across your whole roll and > > they are likelty to get it right. take 24 pictures of a red barn door and > > you could get anything back. > > That is what is so strange to me. I mean logically I get how the machine > needs a color range (wheel) to set a color standard, but it still seems > weird -- counterintuitive that one roll of one color would create > problems.< Thats cause the picture with predominantly red is a colour failure negative. It falls way outside what the machine is expecting to see. The printer is doing the same thing with your negative that your light meter does with it, except it also has to look at the colour. Whatever it looks at, it presumes is 18% gray. Any attempt by the machine to second guess the negative with relation to massive colour casts is a potential disaster, though the new machines are doing it, and fairly well. The problem is, what colours do you tell it to ignore, and what colour casts do you tell it to correct? Ignore huge amounts of blue? Sure, it could be a scenic with lots of sky, so we program the machine to undercorrect that negative. Play with it till we get it right, and another neg will come along soon enough that will cause a subject failure anyway. Red is a hard colour for printers. We can't tell the machine what to ignore with red. It may be a barn door, it could also be a tungsten illuminated room. Both have ots of red in the colour mix, but one wants to be under corrected, the other over corrected. This is why printer operators are employed. It's to make the decisions that the machine can't make on its own. > > the computer that does this isn't very smart. older color processors would > be set a few times a day, whenever the operator noticed something off about > "normal" prints that are coming through, ones where the operator can make a > judgement as to whether the prints they are packing away look right or not. > with digital minilabs, the computer inside can adjust for every roll or > every frame, if so programmed. if the operator is lazy or doesn't know how > to, the computer inside can be told to assume that everything coming > through is meant to be processed as "average" and to "make the colors > right". when you take your first course on scanning and digital > manipulation of slides and negatives, all will become clear. I don't know how you define older. The oldest machine I worked on was a Kodak 5S-5, which still used vacuum tubes in the photomultiplier section. It required some operator input every negative. not several times per day. The Gretag 3140 High Speed Printer used a 16 bit computer that had to have it's boot up done with a series of binary switches on the front, and used a teletype terminal to communicate 4 letter commands to the computer. I happened to have an epiphany of sorts when I was on a training course for this machine, and got really good at programming it. At that, I never got more than 88% acceptable first run production out of it. The machines that I use now are still optical printers. If they are running optimized, they will, in theory, hit about 96% acceptable production. I find myself adjusting about 2/3 of the negatives I print in some fashion, usually density corrections. I don't have a lot of experience with digital minilabs yet. The only one I have operated can be run on auto, and has a pretty good hit ratio. As far as I know, there is no such thing as a minilab printer that can be told to adjust automatically for every roll. They all adjust for every frame. I have seen no correlation yet between what I do at home with digital imaging, and what I do at work, either with the digital to photographic paper printer or the optical printer. Other than shared colour theory, theyr doesn't seem to be anything in common. William Robb
