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!
IQ180 vs Up-Res A7R2
- Thread starter rhern213
- Start date
I mainly got the A7RII to accompany my FPS-IQ180 set-up either for evening or night architecture shots where the higher base ISO and the cleaner files at longer exposure times of the Sony is a clear advantage to the PhaseOne back and for longer focal lengths.
I did not have much time yet for direct comparison of the two systems but what I see is, that it's incredible which quality can be achieved by a 135 format sensor camera and a good lens today. But the camera needs sharp lenses;-) 4.5 um pixel size is really tiny and too much sharpening in post can add some "pseudo-random" structure to the plain areas of the picture (maybe due to the not-lossless compression of the Sony raw).
Of course the size of the Sony + Metabones set-up is so nicely tiny, I was not longer used to carry around such a little camera.
(Ok if I add my full shift lens line TSE17, TSE24, Contax 35, HB 50FLE, HB 100, HB 180 w/ Mirex and a sturdy tripod with geared head ... the weight and volume advantage is rather little.)
The real lifeview of the Sony CMOS is great, no question, but I am used to work with the IQ to frame by estimation, focus by estimation, exposure by estimation, do a test shot at open lens aperture, re-frame, re-shoot, zoom in to 100%, re-focus and re-shoot, then the view will stay at 100% and immediately show me the corrected sharpness, zoom out, set new aperture and exposure time, which can be set together/invers with the FPS, a cool feature and do the final shot, so no big problem wiht that as long as one is not under pressure to get a particular situation or light.
Btw, the Sony for view at 100% in play mode gives a not perfectly sharp image with a lot of artifacts, so bit difficult to determine the sharpness, I think the jpg engine of the Sony is just not great (display already set to better quality).
I modified a Benro L-Plate a bit to screw it directly with two screws to the Metabones. This works very well with the TSE lenses, the combination camera/ Metabones/ Sony sits better in the center of gravity compared to a mount at the camera.
What I noticed is that I got used to the 1.33 : 1 format of the IQ and I really like it (in landscape and in portrait orientation) for architecture photography, using the 1.5 : 1 format of the Sony I always think ... hmm here not wide/high enough. You see with the sample picture from Guangzhou that with the IQ and same lens that there is the option for a wider crop with the IQ which is pretty cool.
I think the colors can be tweaked in post to match quite well in both systems and the Sony has about 1 stop better shadow recovery.
I did not have much time yet for direct comparison of the two systems but what I see is, that it's incredible which quality can be achieved by a 135 format sensor camera and a good lens today. But the camera needs sharp lenses;-) 4.5 um pixel size is really tiny and too much sharpening in post can add some "pseudo-random" structure to the plain areas of the picture (maybe due to the not-lossless compression of the Sony raw).
Of course the size of the Sony + Metabones set-up is so nicely tiny, I was not longer used to carry around such a little camera.
(Ok if I add my full shift lens line TSE17, TSE24, Contax 35, HB 50FLE, HB 100, HB 180 w/ Mirex and a sturdy tripod with geared head ... the weight and volume advantage is rather little.)
The real lifeview of the Sony CMOS is great, no question, but I am used to work with the IQ to frame by estimation, focus by estimation, exposure by estimation, do a test shot at open lens aperture, re-frame, re-shoot, zoom in to 100%, re-focus and re-shoot, then the view will stay at 100% and immediately show me the corrected sharpness, zoom out, set new aperture and exposure time, which can be set together/invers with the FPS, a cool feature and do the final shot, so no big problem wiht that as long as one is not under pressure to get a particular situation or light.
Btw, the Sony for view at 100% in play mode gives a not perfectly sharp image with a lot of artifacts, so bit difficult to determine the sharpness, I think the jpg engine of the Sony is just not great (display already set to better quality).
I modified a Benro L-Plate a bit to screw it directly with two screws to the Metabones. This works very well with the TSE lenses, the combination camera/ Metabones/ Sony sits better in the center of gravity compared to a mount at the camera.
What I noticed is that I got used to the 1.33 : 1 format of the IQ and I really like it (in landscape and in portrait orientation) for architecture photography, using the 1.5 : 1 format of the Sony I always think ... hmm here not wide/high enough. You see with the sample picture from Guangzhou that with the IQ and same lens that there is the option for a wider crop with the IQ which is pretty cool.
I think the colors can be tweaked in post to match quite well in both systems and the Sony has about 1 stop better shadow recovery.
ErikKaffehr
Well-known member
Hi,
Thanks for the images! I have downloaded them on my MacBook and did have a quick look. I would say that comparing Canon 17 TSE on the Sony and the IQ-180 favours the Sony, as it has smaller pixels. But, if you are shooting the TSE shifted it may be the perfectly valid comparison.
I have seen that you are using the Alpa FPS with the IQ-180. How do you focus that thing? Sorry for asking, I have tried to use my P45+ on a Hasselblad Flexbody and it has been a real hassle.
Best regards
Erik
Thanks for the images! I have downloaded them on my MacBook and did have a quick look. I would say that comparing Canon 17 TSE on the Sony and the IQ-180 favours the Sony, as it has smaller pixels. But, if you are shooting the TSE shifted it may be the perfectly valid comparison.
I have seen that you are using the Alpa FPS with the IQ-180. How do you focus that thing? Sorry for asking, I have tried to use my P45+ on a Hasselblad Flexbody and it has been a real hassle.
Best regards
Erik
GrahamWelland
Subscriber & Workshop Member
I'm not Chris but I also use the 17/24 TSE lenses with both the Sony A7r and Alpa FPS, although in my case with an IQ150.
How do you focus the 17 TSE with the FPS? It's really pretty simple with live view, even with a CCD IQ back. The FPS will open up the shutter for focusing. However, it's almost a moot point with the 17TSE in particular because it has such enormous hyperfocal range anyway. You can pretty much just zone focus it.
voidshatter
Member
Hi, many thanks for sharing the RAW files! I can see that for your shutter speed the A7R-II does have some advantage in terms of shadow recovery. By stitching the A7R-II surely yields more details than the IQ180. Why do you get a black circle in the IQ180 file? The 17mm TS-E should be able to cover the whole sensor.
However, for long exposure shots, the A7R-II may not be a good choice.
For long exposure shots:
A7R-II performs even worse than Canon 5DSR
A7R-II heats up quickly during long exposure
A7R-II darkframe comparison (RAW files) against 5DSR and other Sony sensors
A Chinese review of A7R-II doing poorly in long exposure
Base ISO or high ISO, the A7R-II is not suitable for long exposure. You might even get better image quality with your IQ180 in the 10s-120s territory.
Based on the comparison between A7 and A7-II, i.e. no degradation due to 5-axis IS, it is probably true that BSI sensors are not suitable for long exposure. If it's true then it's such a shame that the Rodenstock and Schneider wide lenses cannot be resurrected by a perfect CMOS digital back based on BSI...
ChrisLivsey
New member
Indeed many posts on the weight/size point overlook that, for many situations where it is the ultimate best quality that is required, the Sony advantage is minimal at best.
If you're in a situation where you can stitch with the Sony, then you can also stitch with the IQ.
It's simply not a valid argument.
The image circle of even the 17mm TS-E is not as large as the Phase Sensor..... therefore vignetting occurs. I don't understand the allure of using 35mm lenses with the FPS as there is, IMHO, nothing to be gained over current cameras. The resulting file size remains the same, or smaller, because of the constriction of the image circle. Usually, if not almost always, the dedicated camera body for the given lens (Nikon, Canon, etc) has more than surpassed the electronics in the FPS and would be easier to use. I do think that the examples show just how far and fast Sensor technology is moving on.
Victor
Last edited:
Doesn't it depend on the lens being used - assuming we're talking about getting good results from shifting, rather than just being able to shift ?
The question is: does the new type of sensor used in the Sony behave as badly as the one in the IQ180 with large shifts where the angle of light hitting the sensor at the edge of the IC is acute ?
Since I've yet to hear anything about how it performs in such situations (and the BSI sensor design is predicted as behaving well with light striking it at an acute angle), I wouldn't reach that conclusion just yet.
The allure of using the 17mm TS-E on full frame MF with the FPS is that it gives you a field of view that is not possible with any other lens.
It's not a matter of subjective humble opinion, but one of objective fact.
And until the release of the 11-24, it gave you a field of view that was not possible on any other lens/sensor combination that has ever existed.
Shashin
Well-known member
No. It is an irrelevant condition as you should be assuming equal angle of view. In both cases, you would go to a longer focal length lens for the shifted/stitched image which would perform better than a shorter one.
There are two possible scenarios here.
If you're talking about shift-stitching to create an image with a wider field of view than is usually available from a given lens:sensor combination, then assuming you're utilising a good lens, the MF solution is going to be advantageous simply because you will always be able to capture in a single shot what the Sony solution would require multiple exposures with lens shifts to accomplish.
And I'm sorry, but the IQ180 does NOT behave "badly" when paired with good lenses such as the Rodenstocks. Does the A7R2 perform well with these lenses? Bit of a moot point if you ask me - if you have to shift the sensor multiple times and take multiple exposures to get the same composition as a FF MF sensor is going to deliver with one exposure, just what are the circumstances in which it would be advantageous to do that?
The second scenario (and actually, the one that is probably more relevant here, because the point raised was "By stitching the A7R-II surely yields more details than the IQ180."), is that what was being inferred was that by nodal stitching with the A7R2 you will get more detail than you would with the IQ180. Yes - of course that's an option. But if you're going to nodal stitch, then you can do so with the IQ180 as well. When it comes to nodal stitching, for the same number of exposures to deliver a set field of view, an 80MP MF back will always deliver more resolution than the A7R2.
Kind regards,
Gerald.
Long exposures
I mean and need only "long" exposures, not loooooong exposures like Void;-)
Example: At a late sunset, just before the sun light completely disappears, I typically shoot with the IQ at ISO 35 f11 8s. That's about the maximum exposure time acc. my experience (using ACR) without deterioration of the file quality.
At 20-60 seconds, the plain areas become to a certain extend grainy, the sharpness on pixel level is decreasing, more and more hot pixels show up, even with the dark exposure applied. (C1 handles this better than ACR)
With the Sony A7RII I tested 30s at ISO 100 (with NR = dark frame ON), still can lift the exposure or the shadows enormously and get a very good result. 30s ISO 100 is 4 stops difference EV to 8s ISO 35, so that's a usable advantage for my application.
Sample: Photography in near full darkness (crop), ISO 50, 30s, lifted 3 steps in ACR
TSE 17 vignetting with IQ180
W/o shift the TSE 17 image circle is "just" sufficient for the 54x40mm sensor, but my sample picture was with full 12mm shift on the long side, so you see what you get, still impressive IMO.
TSE 17 12mm shift left + 12mm shift right + stitch on IQ180
This gives a damn wide pano image with below 10mm 135 field of view and up to 12000 pixels width.
I mean and need only "long" exposures, not loooooong exposures like Void;-)
Example: At a late sunset, just before the sun light completely disappears, I typically shoot with the IQ at ISO 35 f11 8s. That's about the maximum exposure time acc. my experience (using ACR) without deterioration of the file quality.
At 20-60 seconds, the plain areas become to a certain extend grainy, the sharpness on pixel level is decreasing, more and more hot pixels show up, even with the dark exposure applied. (C1 handles this better than ACR)
With the Sony A7RII I tested 30s at ISO 100 (with NR = dark frame ON), still can lift the exposure or the shadows enormously and get a very good result. 30s ISO 100 is 4 stops difference EV to 8s ISO 35, so that's a usable advantage for my application.
Sample: Photography in near full darkness (crop), ISO 50, 30s, lifted 3 steps in ACR
TSE 17 vignetting with IQ180
W/o shift the TSE 17 image circle is "just" sufficient for the 54x40mm sensor, but my sample picture was with full 12mm shift on the long side, so you see what you get, still impressive IMO.
TSE 17 12mm shift left + 12mm shift right + stitch on IQ180
This gives a damn wide pano image with below 10mm 135 field of view and up to 12000 pixels width.
Strange - I have exactly this outfit too! FPS, IQ180, A7R, metabones, etc.
Am not a super wide user though, so no TSE for me - Mainly Otus on my A7R, and the Sony 55mm too. for walk around the 40mm Canon is super light & compact on EF adapter. (and ridiculously cheap!)
I use a 50/60/90 on my FPS mostly. the 60 is very light and great angle of view, and its f4. Plus it nicely = the 40mm on A7R
Still waiting for Phase/Sony next generation MF back to arrive with EVF outlet point, so I can use a viewfinder with live view focus/ peaking, then we have will truly have Mirrorless MF with the FPS. Can't wait.
Unless I am missing something the FOV on the IQ180 - after croping out the vignetting - will be approximately the same as the FOV on the Sony A7rll and the pixel dimensions are approximately the same. If the FPS is the 'only' system available then I see your point..... but if it were me I would mount that lens on an A7rll.
Victor
Exactly. The IQ180 combined with Rodenstock HR-W glass can produce a file that is just way above any 35mm or smaller format file. I have yet to see samples of the results when the A7RII is used with the HR-W's though.
The Rodenstock HR-W lenses are just a cut above any other lens I have used, it is no contest. (Well, the Cine Leica Summilux-C lenses are another story, but I have yet to see one used for still photography. They look AMAZING on a RED or Alexxa camera though
).I get stunning results with the IQ160 and 40mm HR-W even when shifted a lot. I really do not wish for anything better in most landscape situations. The color gradations are rendered superbly.
The new Sony looks very nice but the file compression seems an issue when pushing the files a lot, something that is a non issue with the PhaseOne backs and C1pro. The Nikon D810 at base iso (64) is the camera that mostly resembles medium format digital in quality (color and file pushability) not the A7R/A7RII.
If Sony fixes via firmware the file compression then things might change but until then, there are better choices. I am sure it can be done and be offered as an option even if it sacrifices a lot of speed, something the landscapers care little about anyway.
This may be stretching the limits of that lens to the extreme. It doesn't matter how many pixels are in the final stitched image but rather the quality of those pixels. Even with my longest lens on my STC I limit shifting to 10mm. If I need more than than then I resort to panning which many times will give better results (Architectural images being the exception). I have found that shifting beyond certain limits results in very disappointing images. YMMV.
Victor
voidshatter
Member
Gerald insists that CCD is good enough for that area (post here). On the other hand, my experience with CCD agrees with your statement here. Would you be able to share RAW files in this area (i.e. long exposure but not too long) to see the shadow recoverability, for a comparison between the 80MP Dalsa and the A7R-II?
Originally I speculated that in the 30 second territory the A7R-II could offer more details in the shadow than the IQ180, but now it appears that the A7R-II performs rather disappointing at 30 seconds when compared against its predecessor A7R.