Ive been working with Ektar for a while now and was continuously getting terrible colour shifts. Initially from shooting at EI. 50 - learnt that mistake pretty quickly :) However, I still found it difficult to post process without the colour shift problems. I've finally fixed everything...
I scan with a Nikon scanner and use Vuescan (its great software). I lock the exposure and base then choose an inbuilt film type that gives me the "least" colour cast in preview. In Photoshop Levels, choose Options then test combinations to achieve the most neutral image. The next step is my break through; make all tonal adjustments on the "L" channel in L*a*b mode. Within an RBG colour model, luminosity changes in Ektar has very strong impacts on chrominosity. By separating luminosity out from chrominosity you can make very significant tonal adjustments without those funky colour shifts. Just convert back to RGB when your done. Finally, where ever I see local colour casts with the image I will make a global adjustment manually then just introduce that via the history brush.
I even used to add some desaturation, especially to skin tones, prior to using L*a*b just to get rid of that strong red shift in Ektar - now no more need :D
Ektar does not like overexposure for sure but it loves to be optically printed onto Fuji CA or Kodak Supra paper which is much simpler sounding than your above hybrid workflow!
IMHO VueScan is terrible, not an APUG issue, but it can easily lead to lots of colour problems and clipping, which you could mistake for an analogue issue.
Optical printing isn't simple.. one has to get an enlarger, chemicals and paper etc. Labs print with laserjets onto crystal archive typically as opposed to directly off the neg. (ie from a scan).
I've been able to scan my Ektar negs with no changes to my software settings using my Epson scanner. It scans the same as my Portra.
I use curves and levels on the L channel of the L*a*b model as a standard procedure. Curves used in RGB always produce a colour shift, this can be more or less pronounced, but it's always there to my eyes.
I also find ColorPerfect by CFSystems a very effective piece of software, even though with the most contorted user interface and user manual I have ever come across.
One day I'll have my proper darkroom and I will be able to judge results by optical printing. At the moment I cannot judge colours in negatives on their own, it's only the scan I see. Sometimes life is much easier with slides.
Konica Impresa 50/ 120 had a wrong balance - too much red and too much green .
This is a wrong balance of Hue . Can be adjusted only in CMYK
In my hands, there were two actions which remarkably improved the colors of Ektar 100:
1. I tried to avoid underexposure - compare test attached (look for color, neglect differences in brightness)
2. I used Negfix8 - google for Negfix8 - I cannot post a link here.
Unfortunately, you all scan these to different densities with respect to the given speeds. You should lighten or darken the "off" exposures to match the density of the reference at ISO 100 and then compare them. Then it becomes a valid comparison.
Sorry, for the sloppy testing. I tried to better match the densities in the red-marked parts of the pictures. Nevertheless, there are still differences in the densities in the other parts of the pictures, probably in part due to the color shift.
Best regards, Frank