Mike,
Thanks much for the detail on this - useful to appreciate. I've just looked around in my 500 or so scans of the history pictures so far. I can't compare I think to the focus of attention you're taking, especially without the film scanner for comparison. You're right by the way - although I had read Koren and friends debating 3200 vs. the 2450, the information that most swung choice was the photo-i.co.uk article where the Epson 3200 compared quite favorably to a 2700 lpi Nikon film scanner. Page 13 of it, I think.
What I seem to see myself in 3200 scans is that grain is very prominent and much a resolution dominating factor in Leica Tri-X shots - which look that they'd make good 10x15 prints - but quite possibly not in the caliber you're shooting for, especially when I read about these contests. I appreciate that resolving the grain itself can be part of the aesthetics in ways of looking at certain pictures, have made some myself that way upon a time though again no doubt not at the level you're tracking, and can't know if what scans here can achieve is what you want.
On the other hand, to what you said about possible softness, perhaps I more than you am in favour with the idea that the sensor should allow final decision on sharpening to the algorithms in the Photoshop pass. Given of course that dynamic range is properly accounted as below, and assured that this would be better if I used full 16-bit Photoshop. I haven't experimented with the USM control on the Epson software to see if it can do as well, but it is always adjustable.
On finer-grained color films one can also resolve grain, but again I'm not sure vs. what you want, and resolution limit is evidently elsewhere - camera lens, overall focusing, shake, oldness of slides, etc. - and/or scanner.
I feel free to think unless could test otherwise that with more modern films and setup, one could go farther than this scanner can - why people are ready to pay on most fit technology for work there.
I'm not sure how to get at what you seem to have felt about exposure controls. The Epson Scan Twain software has three levels of control mode - full automated 'home', 'office', and 'professional'. In the professional mode, it will still automate prescan including multiple image identification, and then turn over to let one adjust it seems everything that's useful - and do that on a per-image basis within a multiply identified scan. This auto-selecting is easily fine-adjustable afterwards of course, and works on most kinds of material given inter-item contrast - it is particularly good at auto-preparing from the included holders that handle for ex&le 2 six-image 35mm strips at a time, or 4 slides, or one or two images as appropriate from large format films.
There are settable grey correction (white balance), histograms, and tone curves on color channels, with presets and named saved preference lists on all. There is arbitrary segment-by-segment constructability if you like for curves. There are fully automated histogram and tone correction buttons also if you want the machine to try its shot, as well as usual contrast/brightness/saturation/hue etc. for simpler adjustments. Resolution or better, delivered document line resolution with automated resolution setting is there too.
I know I've used all of these at various levels of need to assure that the scan itself will be good from some pretty challenged materials, including with mixed items on the scan bed, and of course it can make a huge difference. Constructed curves really saved one particularly valued family memory with a difficult situation. S&ling briefly through my results, including some of the odd ones, the scans all appear to have nicely filled out histograms.
What I'm wondering is if it might have been possible to miss this richness - the documentation is pretty minimal, if I find the on-disk doc here on my laptop lists all the features if approached with persistence
. I think their idea was to layer things very thoroughly, usual reasons.
I remember just exploring, finding checkbox choices etc. to turn on the available dialogs, so some is also hidden until dug into this way. It seems to me that discovering the ability to adjust individual scans so completely from multiple items identified on the bed was also something you had to realize was there, working forward from the box-selected previews.
And then VueScan may not hook up to what the Epson software is able to manipulate internally, as I am sure your results are conclusive there.
Anyway, for the level of detail and interest I know can exist with persons in this forum, the best is surely as you've done, to try one out for purpose before committing, if this kind of scanner looks interesting.
I liked that I could get results that seem at least to the degree of own needs for information which seems contained on these slides, along with handling all the odd sizes of sheet and roll film that more than a century can deliver - plus all the reflective materials, photos and other things, that I have.
That's personal, as it should be - and great that we can find items to fit so well for varying needs.
The rest of the money vs. having a film scanner in tandem, I place away toward camera/s
Thanks again, for your insights on what else might be available and needed past this scanner. It could be too that I still misunderstand what you were getting at with the control issues, so be sure to let know if anything there.
Regards, Clive