More advanced scanner for 4x5 than Epson flatbed?

[bob carnie]

Bob, a master like you makes quality prints with anything, what counts is the printer, this was true with the enlargers and it remains with digital/hybrid. You may have problems when somebody else "edited" the image before.

there is a old lab rat saying... garbage in garbage out.

[Pere Casals]

there is a old lab rat saying... garbage in garbage out.

it does not fail...

[Jim Andrada]

Yes, one would assume that the high end scanners used good sharpening algorithms - perhaps the best available when they were built. Which was a long time ago.

I've looked for sharpening artifacts in the scanned negs and haven't found any that I could see

[Pere Casals]

Yes, one would assume that the high end scanners used good sharpening algorithms - perhaps the best available when they were built. Which was a long time ago.

I've looked for sharpening artifacts in the scanned negs and haven't found any that I could see

Anyway it's difficult that embedded sharpening generates visible artifacts, my guess is that this is pixel level sharpening, which should have a radius of only say 2 or 3 pixels.

Another thing was the sharpening a Frontier/Noritsu had in the user level, I'd say that image enhancing at that level could sharpen way more. The embedded sharpening level in the firmware should fight the scanning shortcomings, while in the user sharpening level we would fight the defects in the taking (shake, bad focus), but not what the photographer placed in the OOF. Those are different missions.

[interneg]

Though I have used the Epson, Fuji Frontier, Flextight , Creo and tested the drum scanners it is interesting to see some of the behind the scene stuff... my frontier was good up to about 16 x 20 and then the files fell apart.

I recall the Jorge wars well...

Feels like we've been round the same path on trying scanners - the scanner that I'd be really interested to try is the Durst Sigma 45 for minilab-like raw speed & 4x5 handling. I'd agree about Frontier scans - it's also interesting to see how it fairly ruthlessly trims parts of the gamut & highlight/ shadows to fit the paper LUT.

Going forwards, I think sorting out inversion methods that actually mirrors the behaviour of RA4 paper response automatically on the raw output of current CMOS camera sensors with the speed of a Frontier will be where things go next - speeding up the basics before the creative dodge & burn would be immensely handy.

[Pere Casals]

inversion methods that actually mirrors the behaviour of RA4 paper response automatically

This has an straight solution.

The base is a converting 3D LUT that has been automatically crafted from an IT8 calibrated scan of a representative negative, and from the resulting IT8 scan of the RA-4 print. Those two images deliver a conversion 3D LUT, done.

Form that base we adjust the 3D LUT to suit our taste, that's all. Of course we may make several 3D LUTs from several RA-4 interpretations we like.

This is not Science Fiction, it's what several Top Notch film wedding photographers have done these years, those who take care about serious color management.

In fact many film emulation software base its color conversions in that, to convert from DSLR color to film colors.

Still conversion from film scan to RA-4 will always be better, because it has the right spectral footprint in the capture.

3D LUT Creator... even the cheap $99 version does it. I gets integrated in Ps, for total convenience.

[bob carnie]

Here is a question for both interneg and Pere... what I find with any of my scanners is the old film do not have presets or if they do lets say VPS Kodak one of the most wonderful colour negative films I have used. the fact that the film from this era has sometimes deteriorated I have a hard time finding a good preset or profile onboard to use.. any advice on how to balance old film that may have deteriorated a bit.
i get film from all possible scenerios... current cheap store bought film processed in Honduras mini labs to old film from 60's 70's and 80's with suspect process control. I am always find it hard to navigate this situation as the presets are for optimal situations but I get non optimal film.

[Tin Can]

https://nofilmschool.com/tags/color-grading

[Corran]

Nice link! I have been working with some 3rd-party footage this weekend and need to brush up on more advanced color grading, as I don't normally have to grade others' footage with a more cinematic approach.

[interneg]

Here is a question for both interneg and Pere... what I find with any of my scanners is the old film do not have presets or if they do lets say VPS Kodak one of the most wonderful colour negative films I have used. the fact that the film from this era has sometimes deteriorated I have a hard time finding a good preset or profile onboard to use.. any advice on how to balance old film that may have deteriorated a bit.
i get film from all possible scenerios... current cheap store bought film processed in Honduras mini labs to old film from 60's 70's and 80's with suspect process control. I am always find it hard to navigate this situation as the presets are for optimal situations but I get non optimal film.

Sounds like some of the stuff that I get landed with - the main trick is not doing the inversion in the scanner software - Flexcolor can be particularly bad for some odd colour casts and sharpness issues. I tend to scan the film as a negative in 16-bit with some of the frame rebate, possibly with a slight curve applied to open the denser areas that will become highlights, then take it to Photoshop to sample & divide out the base colour as a layer, then an inversion layer, then curve layer where, using the clipping warnings, I individually clip each of RGB black points until just starting to clip in image area & then white points until just before clipping (ignore specular highlights & dust!). I then add two more curve layers, one set to colour blend mode for finer colour correction (for example, current Portra 400 needs a little warmth at this step to look correct) & one for tone curve work. After that, it's on to the rest of the masked curve layers etc. At this point you usually have a good colour starting point for most films - though it depends on how damaged/ faded the films are.

If it's useful, I can write this up as a clearer set of instructions & pm them.