alan@wisdom.BITNET (alan jaris) (09/14/86)
Hello
Has anyone out there have or has work with a Model 3000 Color Graphic
Recorder by Matrix Instruments? Can you please tell me how to calibrated
for a color negative film? or to polaroid film 559? I will apreciate your
help
alan
e-mail address alan@wisdom.bitnet or mcjaris@weizmann.bitnetjon@msunix.UUCP (09/19/86)
In article <8609140925.AA16260@ucbvax.Berkeley.EDU>, alan@wisdom.BITNET (alan jaris) writes: > Hello > > Has anyone out there have or has work with a Model 3000 Color Graphic > Recorder by Matrix Instruments? Can you please tell me how to calibrated > for a color negative film? or to polaroid film 559? I will apreciate your > help I am posting this in addition to mailing a response because I would like to know if anyone else has a better way of doing this. First of all, the QCR has the LUTs for Polaroid 559 in it, just send a 35 to the odd address and a 2 (for 2K) or a 5 (for 4K). I believe it goes: 0 - linear ramps, 2K 1 - Kodak Ektachrome 100, 2K 2 - Polaroid 559, 2K 3 - linear ramps, 4K 4 - Kodak Ektachrome 100, 4K 5 - Polaroid 559, 4K The numbers may go from 1-6 instead of 0-5. I don't know how accurate these LUTs are, but they are there to use. To calibrate for let's say, Kodacolor VR-G, use the linear ramps. The QCR tube face brightness drops off toward the edges like a monitor, so when you figure out the pattern you want to send, just use the center of the tube. The pattern we use is something like 18 squares by 12 squares, with thin black lines separating the squares. In the squares in the middle, about 9 by 6, we send varying values from 0 to 255. Each square is of course the same value throughout. In the outer squares, we usually put the same value, just so we can see how the brightness drops off at the edges. This is exposed through all three filters for a few frames, and then just through a single filter a few more times for each color filter. This is done for quite a few rolls of film, which are then sent to a bunch of different labs. The prints are then run through a Transmission-Reflection Densitometer, which is supposed to measure color density of photographs or transparencies. You need to calibrate this instrument before you use it with the colored transparencies provided. We usually use a MacBeth TR-924, which is far from perfect, but useful. After taking a statistical average of the color densities of all the grey and single color squares, you can figure out what sort of curve you need in your LUTs. Once you have the 256 values for each color, you send these to the QCR. I believe you send a 0 to the odd address and 768 bytes to the even address. I think you can calibrate your QCR with this simple procedure better than you can calibrate a monitor with a Minolta Color TV analyzer. Has anyone ever used one of these? It puts out UVS directly I believe. Most monitors aren't even close when measured with one of these things. BTW, 12-bit LUTs are nice in a film recorder, most of the good ones (Dunn) have 12-bit LUTs. I don't do this in my job, we have someone else here who is an expert in this area, and I in no way claim this is the best way. I can read a manual though, and I believe this is how it is done though. Undoubtedly, with better calibration equipment, you can get better results. "If we did it like everyone else, Jonathan Hue what would distinguish us from Via Visuals Inc. every other company in Silicon Valley?" sun!sunncal\ >!leadsv!msunix!jon "A profit?" amdcad!cae780/