A quite hard scene extremely underexposed. My idea was to get deep shadows, the midtones be close to the shadows and the highlights and reflections be the subject.
Not really getting there
+ theres this illumination problem @ the top and specially @ the bottom
CAT backlight values are hue=41.9º and chroma=37.1%
That’s because we need to increase the exposure so much that the contrast reveals these flat field and vignetage issue , because it increases its contrast too.
That’s something you could do with a normal exposure too. Unfortunately, your exposure level is too low to be able to do that correctly on a print.
How I process:
1 Disable the White balance module
2 Set Demosaic to VNG
2 set Exposure to have the top of the data almost touch the top of the field, in parade mode.
3 Set input and working profile to IdentityRGB-elle-V2-g10.icc (easy to find on internet)
4 In Color calibration, set CAT to none (bypass) and the channel to 0, 1, 0 in the Gray tab
the color one was indeed a light leak that i get, idk, 1 out of 200 photos, maybe even less. Have no idea how it happens. Even took a whole test roll with each photo pointing the lens into the light, away from the light, etc and holding it this way and that way… Writing down everything. Not a single photo came out with the light leak
In the BW it was either a dev prob or a scanning rig issue, and after
Then i guess that narrows it down to scanning issue
I think it will be a couple days until i can try to scan it again taking care of exposing better. Will report back when that
Then this process brings some other questions
is this particulary for BW scans?
Why? Is this for this image in particular or for analog scans in general?
Had a read on the DT docs and its not obvious here why we would choose it even tho it mentions
VNG4 is better suited for use on images with low-frequency content
why? Will have to read further on the official page but
I can’t think of any reason why you’d ever want to actually edit images in the IdentityRGB working space.
furthermore i guess i have some trouble with these iccs, coz: opened the image, saw the error “couldnt find the icc defaulting to…” and the image looked fine, like the export you posted, properly added the iccs and everything went massively out of whack; just black and white, no greys.
Yes, because we only grab the green channel, we don’t need to change anything in this module, risking to mix the channels.
Because it’s a simple demosaic algorithm that avoids cross-talk between channels.
See Ansel | Film scanning
This profile kind of “bypass” the profile conversion step. We need to avoid any channel cross-talk, just take the green channel.
The green channel is the one that contains the more details of your BW film.
Still, the image results. Are they better? i guess, yeah. But not that much better. I dont think.
I will keep scanning with a better exposure in the future, obv. But I guess that wasnt truly the problem of this scans. There are other problems to find in my editing that actually contribute much more.
But im grateful for all the information that ive recieved and all ive learned here
The issue was on the scene’s exposure on film or the development. The scanning is correct here, but inefficient exposure leads to no data recorded on your film in shadows, and inefficient development leads to poor contrast value. Always try to expose for the shadows, because on negative film, the clipping occurs in the shadows.
Exposure = data amount. On negative film, the more you expose, the more you get data on your film. Development = contrast. The more you develop, the more you have contrast and the fulcrum on neg film is in the blacks so it’s not risky. (I don’t know if you process your film yourself but it’s a good thing to remember too
