Chasing accurate color from negative scans

froyd

froyd

Veteran
Local time
8:37 PM
Joined
Aug 14, 2006
Messages
2,319
As a new scanner owner, I'm overwhelmed by the myriad of approaches to get the best color from a negative. Some films, like Portra and Ektar have very distinctive looks, and I'm interested in preserving it as much as possible. After all, part of the appeal of using film, is to benefit from the individual looks of various emulsions, so I don't want to give up that perk in favor of a generic looking post processed image from LR.

From what I've seen on the web, ColorPerfect delivers results that are as close to the colors I remember seeing on the prints I used to order from photo labs. Not working on PS (just PSE) I don't want to go down that path quite yet. ...unless there's a plugin that works in LR.

I have tried the recommended Vuescan workflow of repeated previews of clear negative base to lock exposure and hen film base. I then edited the resulting image --to which no adjustments were made-- into LR and with a few adjustments, I managed to make it look better than what I received back from the Costco minilab, however, I feel that the colors, while good on their own merit, are not true to the film.

Interestingly, I found a random bit of advice online that might be specific for coolscan scanners, and which produced very good results right with minimum fussing: select the Color Fading filter and set the Color balance to Neutral --two big No No from all the other advice I read...and yet the resulting files only need the contrast restored to look great.

I don't know that I have much of a point here other than wondering if there's a difference between creating a great image from a neutral scan, and creating an image that faithfully preserves the film's personality.
 
I've used Colorperfect quite a lot, but recently started using a Pakon scanner. The supplied kodak software is as good and often much better in nailing the white balance. After that it's much easier to touch up in Lightroom.
 
Like Ranchu, I don't know the answer. I use Color Perfect and do like it but some times I meet a negative that needs a different approach. Personally, I think the scanner is providing unknown to me some interpretation that CP has difficulty with.

This fellow is a RFF member but he posted this on the Pentax forum. It has worked for me, one draw back is it is for a V500:

http://www.pentaxforums.com/forums/...darkroom/273685-scanning-ektar-my-method.html
 
I have the answer.
Take it easy on digital PP side. Scan it and check it in LR.
Most of the times if film is exposed, developed and scanned properly you don't have to do a lot of tweaks.
IMO, color film is more simple vs bw film to get it looks "naturally".
Cheers, Ko.
 
Setting the black/white point to an appropriate reference (film base, and then bright white thing in image) usually gets me 95% of the way there. Color curve layer for the rest. The only images that need more work than that are expired films and/or highly overexposed images.
 
Ranchu said:
I don't know the answer, but thanks for the coolscan tip...I'll try it next time I use it.
Click to expand...

Well, I hope it works for you, but take it with a grain of salt. I only tested it with one type of film (Kodeck Gold 400) on two frames (from the same roll) and with only one scanner (Coolscan IV).

Yesterday I tried with a different roll and the results in terms of color accuracy were 85% there but not as good as what I was able to accomplish by matching the R, G, and B curves manually, which I believe in not possible with all scanners.
 
Expose a frame or four with MacBeth Color checker.

Expose and color balance the white to 250, 250, 250. The films saturation and colors will match.. A modest contrast curve should be added as a scan is linear.

Stay off auto. Color balance manually and record settings.
 
An important part of the Color Perfect theory is to have a linear scan, and scan the negative as a negative. Then software honcho wants you to bring as untouched a negative scan to the CP software as possible, you then pick the film from a very long list of films. The film list is a profile made especially for each film. you then manually set Black and White Points, and remove color cast. It then always you to use editing tools (brightness, saturation, curves, etc). This all sounds good, but I still believe the first step, the linear scan, has some scanner software interpretation that foils the remaining steps somewhat.

Color Perfect also comes with a feature called Color Raw. Before starting anything you convert your raw file to a TIFF (called Color RAW Make TIFF) which has zero post software interference. It really looks ugly but when put into CP is surprising. You then can finish and send to PS so you can make your friends hair green, over-saturate and use HDR.
 
John- does Color Raw handle 64 bit RGBi files? If not, do I understand correctly that we lose the benefits of ICE when scanning 48 bits RAW files for use with Color Perfect?
 
Color Perfect has 64 bit capability. Perfect RAW is only for digital RAW files (I think). There are some methods of scanning that do eliminate ICE. This is really a problem. I don't have a work around for this. And sometimes I have to compromise my scans linearity if the negative has dust.

Also, you quoted my using Color RAW, it is not called Color RAW (my mistake). It is called Perfect RAW. I have never tried it with a negative scan as my scanner doesn't produce RAW. But I think the designers meant it for digital (camera) RAW.

They have a plug-in trial on their website, but the program really has a long learning curve so you may not like it unless you go back to school on the subject. I've been using it for 3 years (?) and I still have things to learn.
 
Jack, I tried your plug-in, but upon opening it in photoshop it gives me an error: "Failed to create histogram. Check your plug-in dirctory entitles."
I put your plug-in in: Adobe/Photoshop CS6/Plug-ins/Filters.
 
Ahh bummer. I don't have any experience with Filter Meister. Really liked the results you got in the other thread, Jack.
 
You must log in or register to reply here.
Back
Top Bottom