MF colour reversal - scanning - learning curve

tikkathreebarrel

Suspended / Banned
Messages
2,150
Edit My Images
Yes
Currently scanning some 645 colour reversal film I shot at a Porsche Club Great Britain event at Castle Hedingham.

This is my first experience with MF colour reversal (not more than a couple of rolls of 35mm under my belt either!).

I love the deep contrast and the vibrancy of the colours in the film itself but I'm finding scanning with an Epson V500 to be a bit of a challenge.

On frames where I've got a light/bright/colourful object standing out against a dark background - so a predominantly dark frame - it's as if the scanner can't identify the edges of the frame and then it scans a significantly poorer image.

I know I should have held off posting until I've edited and uploaded some results but I'm not going to be able to show you a direct comparison of film and scan anyway...

Any thoughts?
 
Where'd I put it - in the For Sale F&C? Shows my age doesn't it! Sory guys, and thanks!
 
Are you using advanced options in the scan software (I assume there are curves / levels adjustments) - or just using easy / auto?
 
Also relevant - what software are you using? I can't help with Epson Scan or Silverfast but with VueScan it would be easy (if you're just having problems with the contrast) to make a couple of scans with different exposure levels and merge later. It might be that you've hit the limit with your scanner's DMax though.
 
Well I'm not much help :( but I've come across films that I call "unscanable" (h'mm is that in the dictionary).....well by me with a home scanner, in that no matter how I tried to adjust\correct the colours, in Photoshop, for a part of the shot it would ruin the colours for other bits etc.
Well maybe you have such a problem film?
 
Well I'm not much help :( but I've come across films that I call "unscanable" (h'mm is that in the dictionary).....well by me with a home scanner, in that no matter how I tried to adjust\correct the colours, in Photoshop, for a part of the shot it would ruin the colours for other bits etc.
Well maybe you have such a problem film?

I think it's going to be "problem underexposure" possibly.:cuckoo:

Also relevant - what software are you using? I can't help with Epson Scan or Silverfast but with VueScan it would be easy (if you're just having problems with the contrast) to make a couple of scans with different exposure levels and merge later. It might be that you've hit the limit with your scanner's DMax though.

I'm using Epson scan which gives me access to advanced trickery once I've got a preview scan (see below) but maybe I should try different software. I do have this issue with a few frames where most of the picture is so dark that it's as if the scanner cannot detect the boundary of the slide.

Are you using advanced options in the scan software (I assume there are curves / levels adjustments) - or just using easy / auto?

I am now thank you. :bang::bang: Some improvement to report so many thanks for the heads up. I'm not sure that I'm going to be able to recover every last frame without wrecking the overall colour distribution.
 
Is it Velvia 50 by any chance?
 
If the original looks visually satisfactory (nothing blown or blocked up) then in the scan software I would manually adjust highlight & shadow limits against a levels histogram, then use curves adjustments to tweak the tonal range of the scan between those points.

Colour balance is another matter ...
 
Last edited:
If the original looks visually satisfactory (nothing blown or blocked up) then in the scan software I would manually adjust highlight & shadow limits against a levels histogram, then use curves adjustments to tweak the tonal range of the scan between those points.

Colour balance is another matter ...

Rog, you da man! Your judgement is spot on.
 
Both the 50 and 100 flavours of Velvia are notoriously difficult to scan due to their density, flatbeds struggle to pull detail out of the shadows so if it's an underexposed shot then it's even harder.
 
Just save yourself the hassle with the colour balance and get an IT8 calibration target for £20 from Wolf Faust, you don't even have to have a program that does targetting as you can use several freeware programs such as SCARSE to generate a profile from a scanned target, that you then assign to a scanned image afterwards.
 
Both the 50 and 100 flavours of Velvia are notoriously difficult to scan due to their density, flatbeds struggle to pull detail out of the shadows so if it's an underexposed shot then it's even harder.


This ^


What would project just fine on a wall, can't always be extracted by a scan.
They can be soo so dense, the ones I've had problems scanning in the past are where the camera has metered for the sky and the subject is not too well lit, a grad would have done the trick, or meter for the subject and blow the sky, or a more balanced light level across the whole frame.
My bin is full of cocked up slide, I'm a cocked up slide 10th Dan genius..:)
 
This ^


What would project just fine on a wall, can't always be extracted by a scan.
They can be soo so dense, the ones I've had problems scanning in the past are where the camera has metered for the sky and the subject is not too well lit, a grad would have done the trick, or meter for the subject and blow the sky, or a more balanced light level across the whole frame.
My bin is full of cocked up slide, I'm a cocked up slide 10th Dan genius..:)


Should I?

be frustrated:gag:
pay the shop to scan 'em for me:help:
use slide sparingly :bat:
learn to use filters with slide
or just get over it and overexpose by a few stops:love:
 
Try to keep the exposure range within 4-5 stops using grads if required and see what a drum scanner can do for you if you want to shoot Velvia.

or just get over it and overexpose by a few stops

If you do this you blow the highlights and there is no realistic prospect of getting them back.
 
I'm gettin some deja-vu with this thread, posted it before somewhere.

A scan of cocked up RVP100F

wtp9u.jpg




recovery in post.....Fail


10zctxc.jpg




crop


2mfenwh.jpg





I dunno how I could be so stupid, there quite clearly no light on the train, it was never gonna work with slide with that sky, needed a proper rock HARD grad, I guess I was just too giddy just happening across this monster in the middle of nowhere.
Maybe it can be recovered with a few passes, or better hardware, I dunno, the fact remains that its a recovery job that should never have been.


This was shot in to the sun no grad, but it works because the rest of the scene is lit.......RVP100F


amvazm.jpg
 
I'm gettin some deja-vu with this thread, posted it before somewhere.

A scan of cocked up RVP100F

wtp9u.jpg




recovery in post.....Fail


10zctxc.jpg




crop


2mfenwh.jpg





I dunno how I could be so stupid, there quite clearly no light on the train, it was never gonna work with slide with that sky, needed a proper rock HARD grad, I guess I was just too giddy just happening across this monster in the middle of nowhere.
Maybe it can be recovered with a few passes, or better hardware, I dunno, the fact remains that its a recovery job that should never have been.


This was shot in to the sun no grad, but it works because the rest of the scene is lit.......RVP100F


amvazm.jpg

Nice summation of my problem reversed in that most of my "problem" frames have a nice bright car afront a backdrop of deep tree shade. Now I need to root through my box of filters to size for my Mamiya 645 lenses. Still learning - great! Just think, if I'd taken taken digital I'd have chimped on the spot:bonk:
 
BTW thanks John (joxby) for the explanation of why Velvia etc are so hard to scan. I sort of explains why I had so much of a higher hit rate on Kodachrome in Australia and NZ: plenty of light to go round!
 
Back
Top