The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Looks like tilt is coming to XT natively

DaveRosenthal

New member
I feel like there are two big questions (along with the obvious "what is the focal length?"):
  1. Just tilt around the horizontal axis, or some other mechanics to enable tilt/swing on any axis?
  2. Is tilt encoded in the metadata, and, if, so, is there any point beyond just recording the number for posterity? Unlike shift, which is useful for LCC correction, it's not obvious what corrections it would enable.
 

ThdeDude

Well-known member
Just tilt around the horizontal axis, or some other mechanics to enable tilt/swing on any axis?
Good question. And only one axis?

(Although the new generation of tilt/shift lenses from Canon and Nikon allow tilt on any axis, they still allow tilt only on one axis.)
 

ThdeDude

Well-known member
Is tilt encoded in the metadata, and, if, so, is there any point beyond just recording the number for posterity? Unlike shift, which is useful for LCC correction, it's not obvious what corrections it would enable.

Good question! I don't think Nikon or Canon record and encode any tilt or shift information.
 

TechTalk

Well-known member
Is tilt encoded in the metadata,...
That would depend on the system. As an example, a Hasselblad using the HTS 1.5 Tilt Shift Adapter records tilt, shift, rotation (swing), focal length, aperture, distance, and any accessories or combination of accessories attached to the lens mount (e.g. extension tubes or converters) into the image metadata.
...and, if, so, is there any point beyond just recording the number for posterity? Unlike shift, which is useful for LCC correction, it's not obvious what corrections it would enable.
The data is used for automatic correction of vignetting, distortion, and chromatic aberration. Due to the number of variables which affect the projection of light rays onto the imaging surface, for the HTS 1.5 adapter their Phocus software has over 200,000 correction files in its database to apply automatic corrections using any of the six lenses along with any accessories it is designed to accommodate.
 

tcdeveau

Well-known member
Cats out of the bag:


Also 138mm with x-shutter for XT:


looking forward to hearing more from CI
 

Paul Spinnler

Well-known member
Cats out of the bag:


Also 138mm with x-shutter for XT:


looking forward to hearing more from CI
Well also the 138 HR is apparently now able to accommodate the X-shutter - so Rodenstock finished updating the construction ... nice!

It is a shame though to use it on an XT almost as the image circle easily covers 35mm left and right as per my own tests with the Alpa Pano ... will get that one for Alpa most probably given the increased flexibility. Quite a lot untapped potential on this lens with just 12mm shift left and right ...

Another observation - tilt is limited to 3 degrees while Alpa can go to 5 ... interesting.
 
Last edited:

tcdeveau

Well-known member
Well also the 138 HR is apparently now able to accommodate the X-shutter - so Rodenstock finished updating the construction ... nice!

It is a shame though to use it on an XT almost as the image circle easily covers 35mm left and right as per my own tests with the Alpa Pano ... will get that one for Alpa most probably given the increased flexibility. Quite a lot untapped potential on this lens with just 12mm shift left and right ...

Another observation - tilt is limited to 3 degrees while Alpa can go to 5 ... interesting.
Re: 3 degrees vs. 5 interesting indeed...maybe that's the tradeoff to get electronics in there? Regardless of the degrees, I'm still a fan of Alpa's T/S approach by putting the mechanism on an adapter (so one doesn't have to eat the cost of the T/S mechanism on every lens, for one).

Glad to see the updated 138mm. If I ever start traveling more and grow my piggy bank I'd love to add an Alpa 12 mount one to my kit.
 

Paul Spinnler

Well-known member
Re: 3 degrees vs. 5 interesting indeed...maybe that's the tradeoff to get electronics in there? Regardless of the degrees, I'm still a fan of Alpa's T/S approach by putting the mechanism on an adapter (so one doesn't have to eat the cost of the T/S mechanism on every lens, for one).

Glad to see the updated 138mm. If I ever start traveling more and grow my piggy bank I'd love to add an Alpa 12 mount one to my kit.
Yes Alpa is more modular and if one bought the 50 and 70 in XT there's no way of retrofitting tilt except if Phase One explicitly announces that this is possible which in the end would require a return to the cambo factory and probably also a significant upgrade cost ...

Another thing to note on the CI blog post is that the 32 HR also has shift and allows similarly to the 40 HR to create 4 stitch panos, although with a wider resulting FoV. So both have their uses, the 32 HR just yields more dramatic perspectives ...

It for sure looks super sleek in the new design though!
 
Last edited:

dchew

Well-known member
I’ve been discussing this off-line with some folks. Isn’t it true the XT rotates the whole camera around the tripod mount? If so, does this mean it will only tilt in landscape and swing in portrait?
What am I missing?

Dave
 

Paul Spinnler

Well-known member
I’ve been discussing this off-line with some folks. Isn’t it true the XT rotates the whole camera around the tripod mount? If so, does this mean it will only tilt in landscape and swing in portrait?
What am I missing?

Dave
I think this assessment is correct, but you can stitch then the portrait mode version to get a different aspect ratio ...
 

Ray Harrison

Well-known member
I have both the Rodenstock 40 & 70 in t/s + X-shutter that I'm using on the WRS-1600. Both are great to hike with. I don't find the cable too onerous and I get all of the necessary metadata except of course the shift (and now the tilt). For me, that's an OK tradeoff. I'd still like the ability to record it in the back at the time of capture manually and will continue to poke Phase on that, but I like having tilt & swing regardless of the orientation of the back, with 5 degrees each way. The Cambo is a bit heavier and larger but I really like the flexibility, the larger tilts/swings and the greater shift range, and I still get to rotate the back. I'm more interested honestly in what's coming for Cascable as well as any potential firmware update. I'm not sure what's coming with Capture One for iPad's tethering solution but hopefully in the coming months they'll start the beta testing on the first release. They're very careful to say that it won't all come in one go, so what roadmap Phase has with Cascable and how that plays with Capture One's grand tethering solution is hard to say.

What I really like to see though is Phase still doing stuff. I didn't think they weren't, but a bit of evidence is always good :D.
 

buildbot

Well-known member
I probably missed it, but does it day anywhere tilt is recorded? I actually could not find a reference to that.
 

ThdeDude

Well-known member
it will only tilt in landscape and swing in portrait?
My understanding too. Only tilt in one axis and that axis is fixed in relation to the sensor as you described it. The latter is a bit unfortunate in that landscape in portrait mode often shows the foreground and may need tilt the most.

Even my Linhof Techno is limited in that the rear standard has neither tilt nor swing.
 

Paul Spinnler

Well-known member
Now that things are happening, I wonder whether we might see a IQ5 in '22?
I heard it is still far off. There's also a component shortage. First there would need to be a better sensor available with higher megapixel count (eg 200, 250) and then they need time to develop a new back. I think we are talking 24/25 here subject to a better sensor coming out ...
 

Paul Spinnler

Well-known member
I have both the Rodenstock 40 & 70 in t/s + X-shutter that I'm using on the WRS-1600. Both are great to hike with. I don't find the cable too onerous and I get all of the necessary metadata except of course the shift (and now the tilt). For me, that's an OK tradeoff. I'd still like the ability to record it in the back at the time of capture manually and will continue to poke Phase on that, but I like having tilt & swing regardless of the orientation of the back, with 5 degrees each way. The Cambo is a bit heavier and larger but I really like the flexibility, the larger tilts/swings and the greater shift range, and I still get to rotate the back. I'm more interested honestly in what's coming for Cascable as well as any potential firmware update. I'm not sure what's coming with Capture One for iPad's tethering solution but hopefully in the coming months they'll start the beta testing on the first release. They're very careful to say that it won't all come in one go, so what roadmap Phase has with Cascable and how that plays with Capture One's grand tethering solution is hard to say.

What I really like to see though is Phase still doing stuff. I didn't think they weren't, but a bit of evidence is always good :D.
I still think the P1 marketing was a bit misleading in the sense that I thought the shift metadata will be used for auto LCC at one point. Maybe I read this in one of the dealer's blog posts at the time or so.

C1 was still part of P1 at the time. I thought they'd find a way to not only automatically correct vignetting depending on shift, but also the color cast. The problem seemingly is that LCC is also dependent on the f-stop so you need to create a huge library of LCC files to correct all scenarios if you don't find a smart way of doing it (eg interpolating between a few select LCC files for intermediate positions).

It seems now that it is still uncertain whether this will ever come to market which is kind of a shame as this would really be the most natural use case for shift metadata and a key differentiator then for the XT system as a fully integrated solution with added convenience factors.

Still haven't given up on this, but C1 has recently directed a lot more effort into making itself viable for various other brands and there really hasn't been any further investment in the area of further Phase One integration. In the old days C1 and P1 were more closely linked and there was a nice feature here and there for digital back users.

This shift of R&D focus is probably a result of them being now focused on just increasing revenues and broadening the customer base as a wholly independent company from Phase One with the only tie remaining being the service contract for camera integration - which they now also have with Leica and I guess Hassy soon. In the end the private equity company owning C1 wants to maximize subscribers as you get the highest exit valuation for the business if you have a good recurring revenue base as lifetime customer value of SaaS businesses is markedly higher than the CLV of customers in a license based software business model ... meaning best R&D money spent from a RoE point is anything which "steals" lightroom users rather than making the niche XT community which is captive to C1 anyhow a tiny bit more happy ...

Fingers crossed that auto LCC is still coming for XT users, but at this stage really can't hold my breath for it anymore as being an XT user you are effecrively a niche (within universe of P1) within a niche (having an IQ4 150 from P1) and it is understandable that the management of C1 would rather direct dev. resources to C1 mobile, Hassy integration, features for 35mm land Canon, Nikon, etc. before anyone is assigned to working out XT's auto LCC.

This means shift for now is relegated to auto vignetting correction and maybe as a nice pointer for you as a photographer how you shot it and this then could be used to select one's pre-made library of LCC files. But honestly - then you can rather take an Alpa camera with fixed detents and then always follow the same approach (ie shoot at f8, stitch in 3 shots left to right at the fixed detents) which negates a bit the use of the shift metadata.
 
Last edited:

Paul Spinnler

Well-known member
I suspect, they will wait until there is global shutter for the 40 x 53.4 mm sensor. This would give a IQ5 a new feature over the IQ4.
On this one I heard that the global shutter sensors aren't still (and won't be in the near future) up to par for high end photographic purposes especially with regards to dynamic range. There already is a 120 megapixel GS sensor


but it is not used because you cannot get the same IQ out of it than what can be done with the IQ4 150 - I also would not hold my breath for a GS sensor anytime soon. This one is primarily used for inspections in the industry where dynamic range is not the most important aspect, but rather being able to capture un-skewed imagery from a fast moving production lane.

Also developing a spec sensor requires a lot of capital and what is possible if you have a strong market position and strong customer base can be seen with Arri who developed the next-gen cinema sensor in the Alexa 35 which exceeds dynamic range of color negative film. They are able to do this because they have a 10y horizon and the whole cinema industry interested in their products ... and the IQ4 150 is already a fantastic sensor, so will require quite a bit of effort to exceed its specs, ie there needs to be a business case for it. Given the competition in the industry, who is willing to front double-digit millions for a next-gen 200 megapixel full frame sensor with high DR? ... so not clear at all when we will see an evolution of the IQ4 150 ... it is quite a mature technology at this stage and exceeding specs, namely DR, will require quite a bit of capital to achieve.

In terms of potential P1 pipeline it is therefore unlikely that we see:

XF 50 MKII
XT 50 / 70 TS
Next gen XF (ie multi point AF, or EVF)
Auto LCC
GS sensors
IQ5

anytime soon in fact due to RoE constraints ... what's more likely is another XT lens this or next year either below 32 HR or above 138 HR. Or, in the best case, more XT body options because here P1 can just modify a cambo design or pick up what comes out from Rodenstock. Let's say Rodie brings out a new 180mm lens - they have experience points how many XT lens they sell, they make an order and ship it to cambo for mounting and have a clear cut margin from that ... easy. The reason why we probably won't see a 50 HR T/S is because they need to order a certain batch size for that to make sense and given the XT 50 HRs have been sold already the incremental demand will be not so high after the initial hardcore customer base bought in. Only scenario in my view here is if they make a 4k upgrade offer courtesy of cambo to remount the lenses in the factory in Holland. This makes only sense if the unit upgrade economics are profitable for everyone which I think could be done, but not sure.

They could bring out an IQ5 if they replace all other internals in case there is no new sensor yet - like do something like they did when they released the trichromatic. Meaning some tweaks to the processing, screen, battery life or sensor stack. And look at the XF for example. Great camera. But would you invest millions in a re-design, fancy EVF, AF, etc. in this market environment for a mirror based system?

P1 itself is more focused on driving cultural heritage and industrial sales ... which requires focus on the sales org from a management perspective (ie you make more money selling 10 repro stations to the vatican than selling a few more XT lenses) and not hard capital, ie has a lot higher RoE if the revenues can't guaranteed for upfront investments.

In terms of new sensor, P1 is more or less at the mercy of Sony coming out with something new on their own ... and Hassy coming to the party with an 8k 100 megapixel mirrorless camera produced efficiently in China won't help make the business case for the ultra high end segment ... unfortunately. Which leaves P1 management with two options:

a) Go the Hassy route and make a cheap system priced at 10k (unlikely as they'd lose a lot of profit and the impact is uncertain in a saturated market)
b) They become even more high-end in the sense they will continue to sell 40-50k backs and 10k lenses or even slightly increase prices over time ... and will just milk it given the old shareholders of P1 probably eye C1 for the big equity upside anyhow ... and in the meantime try to sustain P1's revenues / slightly grow them via sales in the industrial / repro segments -> which I think is the strategy adopted for now
 
Last edited:

dchew

Well-known member
The problem seemingly is that LCC is also dependent on the f-stop so you need to create a huge library of LCC files to correct all scenarios if you don't find a smart way of doing it (eg interpolating between a few select LCC files for intermediate positions).
Movements, f-stop AND focus distance. My suspicion is that focus distance matters more than tilt. Tilt changes where the image sits within the image circle, but it doesn’t really change the angle of incidence on the sensor. Movements and focus distance do.
Dave
 

Paul Spinnler

Well-known member
Movements, f-stop AND focus distance. My suspicion is that focus distance matters more than tilt. Tilt changes where the image sits within the image circle, but it doesn’t really change the angle of incidence on the sensor. Movements and focus distance do.
Dave
Yes, you are probably right, making the problem even more complex ...
 
Top