But lets say for the sake of argument (and to rephrase my original question) that you plan to export to only one media type, say Fujiflex. Forget about flexibility for the moment. Do you or do you not agree that starting soft proofing in ACR (instead of waiting until the end just before printing to soft proof) will yield superior prints? Let me be specific by what I mean "starting soft proofing in ACR
Anything is possible but I don't see how this could be the case considering this is just the process of producing parametric instructions which are non destructive and can take place at any time. In Lightroom, thanks to virtual or proof copies, it makes even less sense. You can immediately spin off a proof copy after doing all the heavy lifting in terms of image adjustments if and when necessary and apply those output specific edits without polluting a master.
Step 1: Inside ACR before you do anything, change the color space to your Fujiflex printer ICC profile.
Step 2: Do all your needed capture adjustments, such as capture sharpening and color correction, et. al.
Step 3: Whilst still in ACR, change the color space back to Adobe RGB (or Prophoto).
Step 4: Open image in PS and immediately go to View>Proof Setup>[select your Fujiflex ICC profile].
Step 5: Complete all edits, resizing and output sharpening.
Step 1: That has zero effect on the data,
it's a soft proof.
Step 2: Ditto, capture sharpening is output agnostics. That's why Bruce designed this multiple step sharpening workflow. Now here's where we could go down a divided trail in processing of color corrections, et. al. Do you edit
based on the output (such as setting end points)? You
could but don't have to, the edits again are simply instructions you can alter at any time and ideally, if you're editing for a specific target, you do this on an iteration such as a
proof copy (advantage LR but you can use
Snapshots in ACR: Check with Jeff about this workflow!).
Step 3: There's no changing of color space; all processing in ACR and LR occurs using a variant of ProPhoto RGB. What you're changing is the soft proof. That could affect your decisions yes, I'd suggest doing so on a master image is both unnecessary and limiting.
Step 4: Fine, that's an iteration of the master raw data. Any edits you make now can be both output specific or not, just do it on layers and label them as such.
Step 5: Fine.
I'm expecting you to say yes.
Sorry, the workflow implications are too simple for a yes or no answer.