I would think the notion of a macro t/s is more appealing to people who don't fully understand the practical implications of the tilt concept rather than actual macro photographers. Changing the block of focus to a wedge of focus that decreases to a point could well result in less dof at macro distances. It would be great for almost two dinensional subjects, but move to three dimensions and that ever narrowing cone of focus is going to hurt, and you would need a lot more than 8° of tilt too, 20° would be a minimum even for a 45mm, take that to 90mm lens and you'd need even more to get the movement you'd need.
First of all, Canon has obviously thought about this if these lenses in fact do exist. A multibillion dollar corporation is not going to dismiss these most relevant of aspects. You're not the first person to think about this.
Secondly, do you have actual calculations to show what you claim, or are you just guessing? Would you be able to support your claims? I clearly don't have enough knowledge, and that's why I can't say whether you're right or wrong. Thanks in advance for your response.
Why would you assume I didn't think Canon would think of this? I am fully cognisant of the fact that I am not the first to think of anything, and am quite content with that realization. My point wasn't that Canon hadn't thought it through, it was posters clamouring for a macro T/S probably haven't.
Second, with regards calculations I can give you THE BOOK. http://www.trenholm.org/hmmerk/index.html#FVC
I can also point you to Harold M. Merklinger's two gif's that demonstrate the maths on Keith Coopers own page (I can't believe I am talking tilt shift with Keith Cooper :-) ) here http://www.northlight-images.co.uk/article_pages/using_tilt.html
To clarify my earlier answer. Macro T/S would have some limited benefit, particularly for small product work, such as ladies watches and jewelry, circuit boards and microchips, something with a flat, or close to it, face, something you might well use the current 90 TS/E and a tube for. But it wouldn't help the typical macro shot of a fly's head type scenario at all.
Macro and tilt can indeed be useful. I use a TS-E90 quite often with bellows or extension tubes.
As has been noted though, just tilting doesn't give more DOF, just 're-arranges' it to some extent ;-) Macro and camera movements are part of why I built a DSLR back for an old 5x4 camera some time ago. Interesting to play about with, but not something I've had a lot of practical use for.
None of the TS rumours I've seen have any real detail (no surprise there) but I have had several comments relating to encoders in lenses, putting T/S info into EXIF data (there was a Canon patent some time ago potentially related to this).
Still looking forward to what might turn up... ;-)
I have written here before about the possibility of tilt and shift data being recorded to auto correct vignetting and CA though using the expanded canvass method works well at the moment and any auto settings would almost certainly necessitate a dreaded trip into DPP, but something like the Hasselblad HTS 1.5 Tilt Shift adapter does but built into the lens makes sense. Plus the recent patent for the display of the plane of focus as a perspective grid on the LCD or computer, a much finer geared tilt would be welcome too, as well as the now compulsory seamless tilt and shift orientation rotation.
Make the IQ as good as the 24 TS-E MkII and reduce the body/rear tilt needed for parallel lines in retrofocus designed T/S lenses when using tilt/swing, which are both much easier the longer the focal length, and Canon will have two or three winners.