I agree. Also, I'd love to hear Keith Cooper's take on those lenses.
Thanks ;-)
There was an interesting Canon patent last year which looked at autofocus and tilt, which at first thought seems a bit of a weird idea.
USPTO
However, one of the hardest aspects (I've found) of explaining/teaching the use of such lenses is using tilt - not to randomly change focus/tilt settings for the so called 'tilt-shift' effect, but actually placing the plane of focus exactly where you want it, especially at relatively short distances. The trick can be in appreciating that when a lens is tilted, those numbers on the focus ring no longer mean much.
The patent includes the ability to define two points on a surface and have the camera change the tilt/focus to place a plane of focus through them. Now, it ignores setting the tilt axis, but I can see how defining two points on a plane, the AF system could supply the necessary info in conjunction with the lens focus setting and tilt, to arrive at the correct settings - or tell you that it couldn't be done.
How much of this patent is just covering potential features and how much reflects potential design is another matter though ;-)
Back to the two suggested lenses... and what we have now.
The 14 is quite extreme, I'm not sure how much shift would be available. The TS-E17 already matches the FOV of a ~10mm lens if you shift and stitch. This starts showing the quite extreme distortion from a (correct) rectilinear projection. That said, when using a 14mm I've often thought that a few mm of shift would be nice ;-)
The TS-E24 mk2 is pretty good, but obviously could be improved, but for what? It's fine with my 5Ds (and the R5 I tried recently), so I'd want a hefty MP R5's' to go with one... I'd note though that I tried the 24mm with the S1r and pixel shift mode, where an up/down stitch gave getting on for 300MP and some nice pics...
The current 17 is the (slightly) weaker of the two - I recently had a go with the Nikon 19mm and liked it, but I'm still happy to use the 17 for my day to day work.
My recent testing of the R5 was specifically to see how it was with T/S lenses - great, oh and the polarising filter adapter was a nice thing for occasional use too.
The EVF and level actually made hand held use of the 17 a lot easier with shift, especially for shifting (heresy I know, but i like using both hand held)
So, for the putative RF versions...
I don't mind fly-by wire focus, but please give me more focus throw (or the option for it) than I get with the new 50/90/135. I had a TS-E50 here with the R5, and had forgotten the reduced throw at medium-far distances. Electronic focusing, focus peaking and a good EVF would make my own use of the lenses easier. It could be that AF just works with shift, which might be of some use?
One thing that's been missing for me from existing lenses is EXIF data with movement/distance info, allowing for lens correction - even when using tilt. The Hasselblad HTS1.5x adpter has encoders for movements, so it can be done (at a price...)
For the two new lenses, I do wonder whether I'd prefer a manual version that just has great optics, with less complexity?
So, yes to new T/S lenses, but I do wonder about the complexity (and price)
One other thought that occurred... let's say we do get AF and motorised movements. How are they going to explain what you do with it ;-) If nothing else it could help with sales of my tilt/shift book
(coming in November I'm now told - he casually notes ;-)