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!

Rodenstock 28HR and IQ150, IQ250 or Leaf 50... anyone ?

archivue

Active member
I still dreaming of a nice live view for my arca RM3D... my lenses are

28HR, 35XL, 45ASD, 55ASD, 72L, 90ASD, 135ASD, 180ASD.

it looks like the 35 isn't working properly with this sensor.

For the others, it should be ok ? maybe the 45 will cause problems ?

But what about the 28HR ?

( same question for the IQ3100... but i won't be able to buy it... ).

Thanks
 

torger

Active member
As usual no straight answers can be given to these questions. Will the shots be crosstalk free? No. The retrofocus Rodenstocks are designed to be crosstalk free with Dalsa 6um, and that's the limit. Every sensor after that get crosstalk issues to some extent.

Will it actually be visible? It depends. Will you actually care? Maybe, maybe not. Depends on how important you think color separation / tonality is and how you good you are at detecting degradations of it.

The 28HR works better than the 35XL. For center shots you'll probably be fine, and since the 28HR only have 70mm image circle you can't shift much anyway.

The 45 and 55 Sironar-Digital is analog designs and thus not as retrofocus as the Digarons. They've not been tested much with these sensors AFAIK, but the 55 is probably similarly demanding as the SK60XL, fine with center shots, problems with large shifts. The 45 is probably similarly demanding as the SK43XL, if so it will be fine for center shots, but a bit problematic with shifting, although I guess you can do 10mm without noticing issues in most situations.

My guess is that your first impression if you do some casual tests will be that "this works", but the risk is that you will after a longer period of time either get used to less-than-top-notch tonality or discover that you do get issues in an image here and there.

If you get to test the back and you want to make a quick and easy worst case testing, then make an LCC shot with a red filter (you could tape a red gel on your LCC card), and then a normal one, and apply the normal LCC shot to the red shot, test for the shift amounts you like to use, in both landscape and portrait orientations. If it forms a perfectly flat red surface then you will have no issues, if you have a residual cast then you have some crosstalk that can't be canceled out and may pop out now and then in real images.

In some cases you can rely on manual color correction techniques as described here:
The Image Quality Professor's Blog » Blog Archive How to fix the sky with the Skin Tone tool
if you think that's okay it could be the sensor for you.

As you probably now I'm a bit biased in this as I certainly wouldn't make these type of sacrifices to get live view. One aspect that I don't want to jeopardize tonality and the other that I think "enough is enough", it's been moving in this direction for years and I don't think it's fair by the manufacturers to put incompatible systems in the hands of the customers and hope they won't notice. The Sony sensor is designed for strong retrofocus lenses and should be used with such.
 

Paul2660

Well-known member
From my use, I found the following to be true of the 28HR when used on the IQ150, results should also apply to the 250, and 350.

1. The 70mm IC and 1.3 crop sensor work better together than the full frame sensors, in that you can get between 9 and 8mm of shift and 10mm of rise fall. After this, the IC indicator of the lens starts to show and you get the hard dark vignette common with all Rodenstocks.

2. The sweet spot for shifts I found to be 5mm to 7mm

3. The LCC of shifts past 4mm will show a lot of red cast as will the actual image captured, the red cast will be visible almost as a line on the images and will seem possibly excessive.

4. The LCC when processed will eliminate 100% of the red cast (at least for me and my work), In fact the LCC from Phase One does an amazing job to both correct the red cast and return the shifted part of the image to a normal color cast, even with a solid blue sky. You may need to tweak a solid blue towards the edges. (see post in this forum that talks about blog post by the image quality professor).

5. You will see on a solid color, like sky, very faint banding on shifts. This banding is not the same as the rigid microlens ripple of the CCD shifts, but instead a much more wavy and faint series of banding. It only really shows up in solid colors. Phase is aware of this as I pointed it out during the time I worked with the IQ150, but so far as I am aware, they have not worked up a solution to remove it or have even acknowledged it is there. My dealer agree with my findings. I found that using the Topaz Denoise software removed most of it, as it has a de-banding setting. This same issue is also visible on IQ100 images, and I have seen it in most of the Alpa test shots (where they have a building shot against a solid blue sky). For my work, THIS issue was more note worthy than the cross talk problems and I am a bit disappointed that Phase One has not found a solution for it.

Overall, the 28HR works very well. Sure the colors on shifts will be effected by some cross talk issues but as hard as I looked, I never really found it in my shots as the LCC works very impressively on correcting the magenta cast.

If you get a chance, try to setup a demo with your glass, as the only real way to determine if the 50MP chip will work, is to try it on your shooting environment.

Paul C
 

archivue

Active member
Thanks !

So it looks like it's worth trying one of these back !

i wonder if using stitching we are able to shoot the same image (in term of angle of view ) as an IQ300, or if there is more limitation.

For my personal artwork, i don't use really wide lens, but for architecture sometimes i need to go wide but stitching can be used !
 

torger

Active member
I have also noticed the wavy ripple but I haven't worried too much about it. Just like the Dalsa-specific microlens ripple this Sony-specific ripple should be possible to clean up with a purpose-made algorithm, although it may actually be more difficult due to the lower frequency. I see the crosstalk as a much more severe problem as it's much more difficult or probably even impossible to clean up with a post-processing algorithm.

However in the current state when the ripple still haven't been cleaned up in Capture One, and indeed crosstalk issues quite rarely become evident, the ripple may very well be the more severe issue.

It's now been a couple of years and if they had wanted they would have fixed it a long time ago, so maybe they consider the problem to be small enough to ignore?
 

dougpeterson

Workshop Member
Hundreds of files with various tech camera lenses and various digital backs:
https://digitaltransitions.com/another-20gb-of-100mp-phase-one-raws/
https://digitaltransitions.com/phase-one-iq250-tech-cam-testing/

Generally most of our clients that have looked into a CMOS back for use on a tech camera have moved forward with the purchase. There are drawbacks (well described on this forum and even within this thread) such as the reduction (ranging from small to significant depending on the lens and back in question) of the usable image circle due to various factors (loss of color fidelity, artifacts, light fall off etc). But the appeal of live view, stellar long exposure, and high ISO is strong enough that it has offset those drawbacks for most of our clients that have done that evaluation.

I suggest you get your hands on one and do your own testing of course, but generally speaking, I think you'll be very happy.
 

torger

Active member
Does Phase One have a plan for the future, or is this "the last chance" to use the Rodies with the latest backs?

I'm thinking that the next sensor generation is likely to have even higher pixel count, and thus smaller pixels and thus even narrower angular response, and thus further reducing performance with Rodie wide angles. The patience among tech cam users seems to be almost infinite though, so perhaps they will accept an additional round down in compatibility.

What I think the plan for the future is that the resolution of future sensors will be so high that you can get away with using an XF, place it level and crop the part of the frame you need for your architecture shot, and keystone correct if there is any residual need. A new ultra-wide lens would be desired for that workflow though. You could fall back to stitching heads instead, which compared to the tech lens artifact cleanup workflows doesn't seem too bad.

In any case it seems to me that if you buy a tech wide angle system today you are very likely buying into a dead end -- which may be okay considering for how long digital backs hold up, people are still shooting backs made more than ten years ago. If you want to stay in the upgrade loop however, it's high risk and I haven't seen any sign from the manufacturers to ease the worry. The backs are designed for the SLRs and then the tech users get what they get.
 

archivue

Active member
Does Phase One have a plan for the future, or is this "the last chance" to use the Rodies with the latest backs?.
same applies for nikon shooter and PC lenses... and still no 35mm PC lens in any brand... if i was zeiss or sigma...


we are a small market...
 

vjbelle

Well-known member
Thanks !

So it looks like it's worth trying one of these back !

i wonder if using stitching we are able to shoot the same image (in term of angle of view ) as an IQ300, or if there is more limitation.

For my personal artwork, i don't use really wide lens, but for architecture sometimes i need to go wide but stitching can be used !
I used to shift a lot with my Credo 50 as I wanted either a larger FOV or wanted a particular aspect ratio. If you stitch two Portrait images 10mm left and right you will end up with an image that will have just about the exact pixel dimensions and FOV as a FF back (with slight cropping). That was a way I got around the crop factor if I needed the extra FOV. I did this a lot with my lenses between 60mm and 100mm. So.... if you shoot with a 60mm instead of the crop factor turning that lens into a 78mm you end up with a 60mm FOV image.

I can't comment on the 28 as I don't own one but I do have the 35XL and was able to use rise easily to 10mm. I have also been able to get very satisfactory results with 10mm rise using my 100MP back. I haven't put it through the same torture test as Paul with clear blue sky but will see what happens when I get the chance..... although I am sure his findings are accurate.

Victor
 

torger

Active member
same applies for nikon shooter and PC lenses... and still no 35mm PC lens in any brand... if i was zeiss or sigma...


we are a small market...
Yeah, I've been waiting for Canon TS-E upgrades of the longer focal lengths for years too. The wides seems to do great and is very popular, but the 45 and 90 stays old. And indeed no 35 which is my favorite field of view, represented by the SK47XL in my Linhof system. The closest there is seems to be the Contax 35mm PC lens which can be used via adapters, but it lacks tilt, not too important for architecture but I use tilt quite often in landscape.

We're indeed a small market, but with some will from the manufacturers I think there's growth potential. The market seems traditionally to be thought as consisting of rational thinking professional architecture photographers, but if you choose to expand into landscape and enthusiasts there's solid interest to find. Phase One's A system indeed seems to be aimed at that, but I think it's halfheartedly carried through, it's just a minimal effort bundle with no real hardware adaptations to back it up. Milking the last money out of the systems until they fade away due to lacking support from the digital backs.

I think it's sad, because I don't think it have to be this way. All it takes is a sensor that truly is made for these lenses (no microlenses, introduce lightshields and we're all set), and then some marketing romancing classic large format photography with movements, just like Leica is romancing range finder photography that sells their M system.
 

jlm

Workshop Member
how do people do product photography without a longer lens and tilt, using 35mm? focus stack only?
 

torger

Active member
how do people do product photography without a longer lens and tilt, using 35mm? focus stack only?
Focus stacking seems popular, but there is the 90mm TS-E for Canon and 85mm PC-E for Nikon, those aren't that bad.

And while wide angle tech does seem a bit endangered from a technology aspect, at the same time combinations like A7r + Universalis + longer tech lens seem to do really well in the product photography genre.
 

ErikKaffehr

Well-known member
Hi,

My solution on the A7rII is a HCam Master TS II combined with a Hasselblad Macro Planar 120/4 CFi, that combo offers ample shift and tilt.

For next travel, I will carry the HCam Master TS II combined with a Contax 28-85/3.3-4 zoom, for tilts.

The image below was shot on the Hcam Master TS II with the Contax 35-135/3.3-4.5 at around 60mm:


I also have a Hasselblad Flexbody with a P45+ back. I might use that under studio conditions, but it didn't work for me in the field.

I find that live view is a game changer in using tilts, especially when used with peaking.

Best regards
Erik


Focus stacking seems popular, but there is the 90mm TS-E for Canon and 85mm PC-E for Nikon, those aren't that bad.

And while wide angle tech does seem a bit endangered from a technology aspect, at the same time combinations like A7r + Universalis + longer tech lens seem to do really well in the product photography genre.
 
Top