Ad
Ad
Ad
Pages: « 1 [2]   Bottom of Page
Print
Author Topic: LUT for transforming Negative film to positive  (Read 4366 times)
Tim Lookingbill
Sr. Member
****
Offline Offline

Posts: 1153



WWW
« Reply #20 on: March 07, 2014, 07:15:25 PM »
ReplyReply

Quote
Links?  I haven't seen these videos.

Here's three I found...

https://www.youtube.com/watch?v=RejaetB-g-o

https://www.youtube.com/watch?v=B2gLT0SWzpQ

https://www.youtube.com/watch?v=iMO50AlGyrw

Several more can be had entering specific search terms but I believe those three offer different software edit strategies to arrive at their final results as well as advanced to quite simple hardware capture setups.

What I meant by source space is the actual RGB numbers that show up in ACR which is linear RIMM which uses ProPhotoRGB colorants. The main point is using a DSLR's internal A/D converter to write the source Raw for ACR to formulate the preview under the hood starts out far more accurate especially coming off a Canon which are known for their accurate looking captures with regard to hue/saturation.

Economical flatbed scanners (my Epson 4870) for me have always been strapped with not so desirable color conversion software that delivers somewhat dull results unless you get into elaborate Vuescan/Silverfast linear processes in Photoshop as you're doing now. Also you can't build an ICC profile for color negative film so accuracy of how the software interprets color is anyone's guess. But at least working in ACR on DSLR Raw files gives the option to use DNG profiles which you might look into creating to see how close it gets you after inverting to positive as a better starting point.

It would be interesting and informative if you posted results using your processes on several different frames and film brands to get a gauge on consistency and complexity. When I did the "capture as positive" on Kodak UC 400 negs I'ld get over blown saturation after inverting especially flowers but OK results on other scene types.
« Last Edit: March 07, 2014, 07:34:50 PM by Tim Lookingbill » Logged
WayneLarmon
Jr. Member
**
Offline Offline

Posts: 50


« Reply #21 on: March 07, 2014, 09:24:29 PM »
ReplyReply


Quote
Here's three I found...

They do seem to make it seem easy, don't they.  I use somewhat more exacting capture methods and I always work from RAW and work in a large color space at 16 bits until I'm done.  I do manual focus using 10x liveview, focusing on the grain.  Starting with an 18 megapixel 60D, I can usually get 8-10 megapixels worth of detail out of negatives.

I'm in the process of writing up the methods I use.  I haven't written up how I capture the images yet, but I use a Canon 60D + Canon 60mm macro lens and a PhotoSolve Extend-a-Slide to hold the negatives.  I use a custom version of the  PhotoSolve negative carrier.   

Right now I use four CRI 93 daylight balanced CFLs in a softbox, with two additional stops of Rosco #2026 diffusion filter suspended about 1/2 inch from the front of the film.  This diffusion reduces the effect of scratches.

Quote
Also you can't build an ICC profile for color negative film so accuracy of how the software interprets color is anyone's guess.

Canon cameras very well may convert color better.  A photographer on a different forum said that he spent years trying to photograph his wife's artwork (she is a professional) with a Nikon camera.  He said that he didn't get acceptable color until he switched to a Canon 50D.

Quote
But at least working in ACR on DSLR Raw files gives the option to use DNG profiles which you might look into creating to see how close it gets you after inverting to positive as a better starting point.

As I said, until I read what Bart said about the importance of doing the initial white balance adjustments on a linear RAW conversion (earlier in this thread), I had been opening up the RAW files of the camera scanned negatives in ACR, adjusting the R, G, and B white and black points in ACR's point curve tool, exporting as ProPhoto 16 bits to PS and using PS's Curve tool to finish the conversion.  Some conversions are easy.  Others are hard.

After reading what Bart said about not using ACR because it doesn't expose a linear RAW conversion, I tried to find the best way of doing a linear RAW conversion.  (Covered in previous posts in this thread...)   The DPP approach works best.  I went back and revisited some of my past negative conversions and I think that the DPP approach converges to decent looking color faster than the ACR approach.  I think, but I need to do more testing.

I still haven't figured out Raw Therapee.  As I noted in an earlier post, it looks like Raw Therapee developers are working on incorporating dealing with negative conversions in RT.  Apparently RT (like ACR) does some "helping" under the hood that enhances normal images but works against you when you work with negative images.  So they are trying to nip this problem in the bud by doing the negative conversion earlier in the pipeline.  (But this isn't in the release version yet.  I don't think.)

I have experimented with tweaking DNG profiles with the Adobe DNG Editor.  This is sort of a last resort because it adds time to the workflow.  But a tweaked DNG profile will work for other images that were taken in the same light.  And film stock.

As I said, I am still writing.  Here is what I have thus far:
http://www.frogymandias.org/imagery/

Most of this was written while this thread has been progressing.   Comments are welcome!

Quote
It would be interesting and informative if you posted results using your processes on several different frames and film brands to get a gauge on consistency and complexity. When I did the "capture as positive" on Kodak UC 400 negs I'ld get over blown saturation after inverting especially flowers but OK results on other scene types.

Negatives capture a lot of dynamic range.  But they usually don't exhaust the density of the film.   Even with the low dynamic range (compared to Nikon) Canon sensors, I've never run out of dynamic range at the camera level.  When shooting negatives.  (I'm not going to make this statement for slides.  I did my slides several years ago and I didn't examine them critically.  Slides require more camera dynamic range.) 

 I'm planning on adding to my web pages.  The pages thus far cover two negative conversions.  I supply the RAW files so you can try to duplicate my conversion.  Or improve on them.

Logged
Pages: « 1 [2]   Top of Page
Print
Jump to:  

Ad
Ad
Ad