Luminous Landscape Forum
Equipment & Techniques => Medium Format / Film / Digital Backs – and Large Sensor Photography => Topic started by: BJL on February 07, 2020, 08:48:01 pm
-
A bit of news that I missed earlier from Fuifilm's "X Summit 2020" video presentation (I find video a painfully slow and inconvenient way to receive most sorts of information; a plain old web page with text and stills would vastly better): Fujifilm has announced plans to add a 400MP pixel shifting mode to the GFX 100, using the IBIS system:
https://www.dpreview.com/news/7648446596/fujifilm-says-new-400mp-pixel-shift-mode-is-coming-to-its-gfx-100-camera-system (https://www.dpreview.com/news/7648446596/fujifilm-says-new-400mp-pixel-shift-mode-is-coming-to-its-gfx-100-camera-system)
If the lenses (current or forthcoming) can indeed sustain this resolution, this seems like further pressure on the larger format 645 film SLR derived systems to adapt to an even more extreme, narrow, high-end niche or else risk fading away as current users and equipment retire.
(In situations where the multiple frames needed for this can be taken, DR can presumably be enhanced by combining even more frames, nullifying the DR advantage that larger sensors have so far for single-shot photography.)
-
Fuji first mentioned this long before when the GFX100 was still on the development board, then when the camera was close to release, decided to bring the camera to market without this feature.
Glad to see it's back on the board, but per the video you mention and some other websites, there is no mention of when it will be made available.
I have no need for 400MP output, but would more welcome a feature like Phase One has on the 4150 frame averaging. If the pixel shift mode enhances DR that would be a benefit. I am sure it will require a tripod and subject movement will be limited. It will also take a very superb raw conversion. LR/ACR's conversion for the Pentax K1 was terrible and never improved upon (Abode's typical one and done approach). Since Phase One is now formally supporting the GFX cameras I hope that Capture One will support the pixel shift images (C1 never supported the Pentax K1 pixel shift). Pentax's Pixel shift was different in that it did not increase resolution 4x, but instead gave increased details and much better signal noise/DR, vastly cleaner images.
Personally I had hoped for a improvement in firmware for the AF on the camera as I find it's lacking in low light/low contrast situations, similar to the 50s and other Fuji X series cameras I have used.
Paul C
-
Paul, yes the timeline is a definite “eventually”; which will we see first, this or the Canon R body with 8K video?!
My guess is that this is mainly for the very special niche of documenting flat art works, which is what Hasselblad markets its 400MP pixel shifting backs for. By my reckoning, once the aperture is large enough to avoiding overwhelming diffraction effects, the DOF when viewing the image large enough to see that fine detail would be less than with an f/1 image in 35mm format viewed “normally” (viewing distance comparable to image size) That’s why I said _flat_ art—no sculptures!
-
Paul, yes the timeline is a definite “eventually”; which will we see first, this or the Canon R body with 8K video?!
My guess is that this is mainly for the very special niche of documenting flat art works, which is what Hasselblad markets its 400MP pixel shifting backs for. By my reckoning, once the aperture is large enough to avoiding overwhelming diffraction effects, the DOF when viewing the image large enough to see that fine detail would be less than with an f/1 image in 35mm format viewed “normally” (viewing distance comparable to image size) That’s why I said _flat_ art—no sculptures!
With a 3.76um-pitch Bayer sensor and a diffraction-limited lens, you need to stop down to between f/11 and f/16 to avoid aliasing artifacts. 16-shot pixel shift would drop that to about f/5.6.
https://blog.kasson.com/the-last-word/balancing-real-and-fake-detail-part-2/
Jim
-
With a 3.76um-pitch Bayer sensor and a diffraction-limited lens, you need to stop down to between f/11 and f/16 to avoid aliasing artifacts. 16-shot pixel shift would drop that to about f/5.6.
https://blog.kasson.com/the-last-word/balancing-real-and-fake-detail-part-2/
Jim
That sounds about right, though I was talking about choosing apertures to avoid significant loss of detail to diffraction (about 2 to 3 times pixel pitch, so f/8 to f/11 for this example) whereas you are aiming at some deliberate diffraction softening to avoid aliasing, and so going about one stop smaller. But note that my DOF description used "equivalent f-stops", to allow fo the fact that to see the increase in resolution one will do something like print at the same PPI and view from the same distance (giving the same "angular resolution"), so with greater enlargement and thus stronger OOF effects and less of the image appearing sharp: decreased perceived DOF.
In relative terms, any doubling of linear resolution like 100->400MP reduces the diffraction-related aperture ratio threshold by a factor of two, so doubles CoC size on the sensor, which makes each CoC four times as large relative to the pixels. Then at equal PPI display, each CoC is four times as big, giving 1/4 the DOF, or "four stops less DOF" in my comparisons to what one sees at "normal viewing angle". So f/5.6 sounds sort of OK if you think about what one sees from a normal viewing angle, but that way of viewing is only good for seeing about 10-16MP worth of detail, so seeing what 400MP reveals requires at least 5X more enlargement and reduces DOF to what on would "normally" see at f/1.1.
(A former participant in these forums argued more stringently for a "true DOF" measure based on CoC no larger than pixel pitch, or twice pixel pitch, or something like that, since anything beyond that is visibly OOF if you look closely enough; I am a bit more flexible about viewing options!)
As indirect evidence, look at the various "giga-pixel" plus images; when done without focus stacking, the scenes all seem chosen to consist only of distant subjects so that hyperfocal focusing can keep it at all sharp.
But then there is focus stacking! There is an impressive result from that in this Hasselblad promotional site: https://www.hasselblad.com/stories/goran-liljeberg--exploring-hasselblad-400mp-multi-shot/
-
That sounds about right, though I was talking about choosing apertures to avoid significant loss of detail to diffraction (about 2 to 3 times pixel pitch, so f/8 to f/11 for this example) whereas you are aiming at some deliberate diffraction softening to avoid aliasing, and so going about one stop smaller. But note that my DOF description used "equivalent f-stops", to allow fo the fact that to see the increase in resolution one will do something like print at the same PPI and view from the same distance (giving the same "angular resolution"), so with greater enlargement and thus stronger OOF effects and less of the image appearing sharp: decreased perceived DOF.
In relative terms, any doubling of linear resolution like 100->400MP reduces the diffraction-related aperture ratio threshold by a factor of two, so doubles CoC size on the sensor, which makes each CoC four times as large relative to the pixels. Then at equal PPI display, each CoC is four times as big, giving 1/4 the DOF, or "four stops less DOF" in my comparisons to what one sees at "normal viewing angle". So f/5.6 sounds sort of OK if you think about what one sees from a normal viewing angle, but that way of viewing is only good for seeing about 10-16MP worth of detail, so seeing what 400MP reveals requires at least 5X more enlargement and reduces DOF to what on would "normally" see at f/1.1.
(A former participant in these forums argued more stringently for a "true DOF" measure based on CoC no larger than pixel pitch, or twice pixel pitch, or something like that, since anything beyond that is visibly OOF if you look closely enough; I am a bit more flexible about viewing options!)
In my mind, the purpose of pixel shifting is not to get sharper images (pixel shift does not reduce the size of the sampling aperture), but to reduce aliasing.
Jim
-
In my mind, the purpose of pixel shifting is not to get sharper images (pixel shift does not reduce the size of the sampling aperture), but to reduce aliasing.
Jim
Good point: even if some use pixel-shift multi-shot for higher resolution (Hasselblad advertises that as one benefit), another use is improving color accuracy and reducing aliasing, while still intending to print an view at the same size as with 100MP single shot files. Also, I have not seen any camera maker claiming that its multi-shot mode fully doubles the linear resolution, so my judging DOF by the standard of doubling linear print size is unfair. Going back to viewing no larger than for 100MP single-shot mode, diffraction-limited DOF gains back a factor of four to a distinctly more widely ueful "f/4 equivalent" or so (and akin to what one would see with about f/3 in 35mm format.)
-
As someone who almost exclusively shoots static, planar abstracts from a tripod, I welcome the multishot/"oversampling" approach that hopefully will come in a firmware update.
Color detail and aliasing are remarkably improved on the Hasselblad MS implementation. I assume it will be a similar leap on GFX100.
I should have put "planar" in quotation marks above, because in the real world stuff is almost never that flat. Focus stacking dramatically expands the shooting envelope for what I do, too.
I'm imagining how slow my computer will crawl when we can bring both of these tools to bear at the same time!
-
Also, I have not seen any camera maker claiming that its multi-shot mode fully doubles the linear resolution...
And that's for a good reason. The slanted edge MTF is determined by pixel aperture, not pixel pitch.
Jim
-
As someone who almost exclusively shoots static, planar abstracts from a tripod, I welcome the multishot/"oversampling" approach that hopefully will come in a firmware update.
Seriously?
-
Seriously?
I don't understand your point.
-
I don't understand your point.
Free? If Free, that would be amazing...
Peter
-
Up front bias disclosure: My company (Digital Transitions) chooses to sell cameras (Phase One) that do not do pixel shifting and chooses not to sell cameras (e.g. Sinar, Hasselblad, Fuji) that do. So I'm obviously biased. But I also have quite a lot of experience working with clients to evaluate against these options.
On Color Accuracy: Multishot does not improve color accuracy*
Resolution: Multishot can improve resolution but the subject and camera must be perfectly static, and the lens and aperture must be able to resolve the subject at the small aperture values required. The severity of those constraints increase with the resolution being discussed. For 50mp sensors it was already very challenging to use pixel shifting successfully. For 100mp sensors those constraints are very very considerable.
Aliasing: Multishot will almost always reduce aliasing (which is good), but can also introduce significant pixel-level artifacts that can be difficult to detect without reviewing every image at every part of that image at 100%.
If using multishot in a workflow environment (e.g. trying to get a certain amount of work done in a certain amount of time) it's also notable that multishot captures requires more capture time, more processing time, more pops of the flash (if using strobe), more shutter actuations (if using a mechanical shutter), reduces raw compatibility, and reduces the number of viable shooting environments. And while it does not explicitly prevent you from using other techniques like focus stacking, exposure bracketing, frame averaging, or stitching, it does significantly increase the number of frames required to accomplish them.
I'm biased, so I get that my opinion here is of limited value. But I would very strongly urge anyone considering a multishot solution (new or used, current or old, fuji hassy or sinar) to test it before pulling the trigger. It isn't coincidence that the vast majority of US museums, libraries, and archives have left multishot platforms for single-shot platforms over the last decade (before which they were a very significant presence). And when you make that comparison please make sure you're not just comparing the multishot mode of a camera to it's own single-shot mode; many of the multishot cameras came with raw processors with very mediocre single-shot raw processing algorithms. That, at least, will not be a problem for the GFX which ships with the very capable algorithms in Capture One for single shot. Though, I suspect you'll need a separate app to do multishot with the GFX, but I don't claim inside knowledge on that.
*It would, in theory, improve the pixel level color accuracy of subject matter that is 1) totally random 2) exactly one pixel in size and 3) perfectly sharply rendered and 4) falls exactly on the pixel grid, not between. That is not a scenario you'll find in real-world photography.
-
Free? If Free, that would be amazing...
Peter
Fuji has introduced other things to the GFX line that were extremely useful as free firmware updates. Focus bracketing is the example that comes to mind.
-
Up front bias disclosure: My company (Digital Transitions) chooses to sell cameras (Phase One) that do not do pixel shifting and chooses not to sell cameras (e.g. Sinar, Hasselblad, Fuji) that do. So I'm obviously biased. But I also have quite a lot of experience working with clients to evaluate against these options.
On Color Accuracy: Multishot does not improve color accuracy*
Resolution: Multishot can improve resolution but the subject and camera must be perfectly static, and the lens and aperture must be able to resolve the subject at the small aperture values required. The severity of those constraints increase with the resolution being discussed. For 50mp sensors it was already very challenging to use pixel shifting successfully. For 100mp sensors those constraints are very very considerable.
Aliasing: Multishot will almost always reduce aliasing (which is good), but can also introduce significant pixel-level artifacts that can be difficult to detect without reviewing every image at every part of that image at 100%.
If using multishot in a workflow environment (e.g. trying to get a certain amount of work done in a certain amount of time) it's also notable that multishot captures requires more capture time, more processing time, more pops of the flash (if using strobe), more shutter actuations (if using a mechanical shutter), reduces raw compatibility, and reduces the number of viable shooting environments. And while it does not explicitly prevent you from using other techniques like focus stacking, exposure bracketing, frame averaging, or stitching, it does significantly increase the number of frames required to accomplish them.
I'm biased, so I get that my opinion here is of limited value. But I would very strongly urge anyone considering a multishot solution (new or used, current or old, fuji hassy or sinar) to test it before pulling the trigger. It isn't coincidence that the vast majority of US museums, libraries, and archives have left multishot platforms for single-shot platforms over the last decade (before which they were a very significant presence). And when you make that comparison please make sure you're not just comparing the multishot mode of a camera to it's own single-shot mode; many of the multishot cameras came with raw processors with very mediocre single-shot raw processing algorithms. That, at least, will not be a problem for the GFX which ships with the very capable algorithms in Capture One for single shot. Though, I suspect you'll need a separate app to do multishot with the GFX, but I don't claim inside knowledge on that.
*It would, in theory, improve the pixel level color accuracy of subject matter that is 1) totally random 2) exactly one pixel in size and 3) perfectly sharply rendered and 4) falls exactly on the pixel grid, not between. That is not a scenario you'll find in real-world photography.
Doug, pixel shift does increase color accuracy of high-spatial frequency information by reducing (and in my experience eliminating) false color Bayer demosaicing artifacts. I agree that, because of the vibration and subject motion constraints, it has limited utility. It's great for static fabric shots.
-
Fuji has introduced other things to the GFX line that were extremely useful as free firmware updates. Focus bracketing is the example that comes to mind.
You have to admit that hi-res Is a lot more than an update. That's a major selling point. I would be most pleasantly surprised if Fuji gave that away.
Peter
-
You have to admit that hi-res Is a lot more than an update. That's a major selling point.
So was focus bracketing. And IMHO, that's even more useful.
-
So was focus bracketing. And IMHO, that's even more useful.
I hope that they do Hi-res with an update. Many will smile...
Peter
-
In the past, all such upgrades for features from Fuji have come with firmware, as Fuji mentioned the pixel shift solution from day on, it's safe to assume the feature will not be a added cost, as I can think of no feature addition in the past where Fuji charged.
Paul C
-
Up front bias disclosure: My company (Digital Transitions) chooses to sell cameras (Phase One) that do not do pixel shifting and chooses not to sell cameras (e.g. Sinar, Hasselblad, Fuji) that do. So I'm obviously biased. But I also have quite a lot of experience working with clients to evaluate against these options.
On Color Accuracy: Multishot does not improve color accuracy*
Resolution: Multishot can improve resolution but the subject and camera must be perfectly static, and the lens and aperture must be able to resolve the subject at the small aperture values required. The severity of those constraints increase with the resolution being discussed. For 50mp sensors it was already very challenging to use pixel shifting successfully. For 100mp sensors those constraints are very very considerable.
Aliasing: Multishot will almost always reduce aliasing (which is good), but can also introduce significant pixel-level artifacts that can be difficult to detect without reviewing every image at every part of that image at 100%.
If using multishot in a workflow environment (e.g. trying to get a certain amount of work done in a certain amount of time) it's also notable that multishot captures requires more capture time, more processing time, more pops of the flash (if using strobe), more shutter actuations (if using a mechanical shutter), reduces raw compatibility, and reduces the number of viable shooting environments. And while it does not explicitly prevent you from using other techniques like focus stacking, exposure bracketing, frame averaging, or stitching, it does significantly increase the number of frames required to accomplish them.
I'm biased, so I get that my opinion here is of limited value. But I would very strongly urge anyone considering a multishot solution (new or used, current or old, fuji hassy or sinar) to test it before pulling the trigger. It isn't coincidence that the vast majority of US museums, libraries, and archives have left multishot platforms for single-shot platforms over the last decade (before which they were a very significant presence). And when you make that comparison please make sure you're not just comparing the multishot mode of a camera to it's own single-shot mode; many of the multishot cameras came with raw processors with very mediocre single-shot raw processing algorithms. That, at least, will not be a problem for the GFX which ships with the very capable algorithms in Capture One for single shot. Though, I suspect you'll need a separate app to do multishot with the GFX, but I don't claim inside knowledge on that.
*It would, in theory, improve the pixel level color accuracy of subject matter that is 1) totally random 2) exactly one pixel in size and 3) perfectly sharply rendered and 4) falls exactly on the pixel grid, not between. That is not a scenario you'll find in real-world photography.
Doug, all good points.
Note, that FA on Phase one can't tolerate any movement either, and needs a totally static camera solution.
As for the Fuji and Pixel shift, your points are all well taken, and I agree that the key will be how it's implemented. If Fuji works out the solution in camera, and saves are a raw as the Olympus cameras do, I feel it could work very well, where as if Fuji takes the Son approach where you take the four shots and then have to open them in a Sony piece of software then save as a dog, I don't feel it will work as well. Also I would hope since currently Phase One/Capture One claim to have full support for the GFX camera, that when Fuji implements it, if it's a solution where the image is saved in camera, that C1 will eventually support it. Capture One has an excellent raw conversion for the existing GFX files, so I have faith that Capture One will work well on the pixel shift images.
As to your points on limitations, as a landscape shooter, I can think of many locations where wind would not be an issues (Monument Valley, Grand Canyon, Bryce Canyon, Any shot of a Colorado 14 foot peak) all come to mind. In my location, wind would be a factor for sure but I would still want to try it as I use FA in a similar work flow. FA can't tolerate movement, but there are many times I can combine parts of a shot in post.
Pixel shift on the K1 did not increase resolution, but did dramatically increase details, dynamic range and overall image quality. The problem with the K1 pixel shift was terrible raw support from both Adobe and Capture One. Capture One never supported it (what a shame) and the Adobe one and done was terrible. Where as many other software companies, with a lot less money and size were able to come up with excellent solutions that did allow for movement (iridient, Sliky Pix and Raw Therapee all do a great job).
I would hope that with the upcoming Fuji solution, the overall image quality will be improved, and dynamic range possibly improved as with Pentax. I have no issues with the need for a static camera as I tend to use a tripod most of the time.
Overall, for a 10k camera, if Fuji comes out with this at no additional change, it's still a great return on investment for any photographer using the camera. And in time Fuji hopefully might look at FA also as it's been proven by Phase One to make a considerable improvement to the overall image quality.
Paul C
-
Very happy to see this coming!
Cheers,
Bernard
-
I have read quite a few threads about that new pixel shift feature. I do see that there are use cases, but most of them come with strong limitations or are targeted to a very specific group of photographers. I am wondering if those software development resources would not have been better spent on other improvements that target a larger audiences. On the other side that feature might just be a side product of the pixel shift implementation of the new X-T4 (under the assumption that it will have a similar IBIS setup to the GFX 100 and that they are planning on offering that feature for other cameras as well).
Let's see how their implementation turns out. This is not a choice for customers after all.
-
I don’t expect pixel shift functionality with the upcoming X-T4. This would be very difficult to implement on a 6x6 X-Trans color filter array.
Regards,
Jaap.
-
I have read quite a few threads about that new pixel shift feature. I do see that there are use cases, but most of them come with strong limitations or are targeted to a very specific group of photographers. I am wondering if those software development resources would not have been better spent on other improvements that target a larger audiences. On the other side that feature might just be a side product of the pixel shift implementation of the new X-T4 (under the assumption that it will have a similar IBIS setup to the GFX 100 and that they are planning on offering that feature for other cameras as well).
Let's see how their implementation turns out. This is not a choice for customers after all.
At least this pretty much puts an end to the life of the Hasselblad H system... not that I am happy about that as a H6D-100c owner, but the only specificity that could still have convinced new photographers to buy into the system at this point in time is now gone and the revenue stream associated to it as well.
Some rich Chinese may still prefer the shades of grey of the H body over the dark black of the Phaseone XF... but that's pretty much it. ;)
Cheers,
Bernard
-
Fuji has avoided pixel shift on x-trans cameras. They did not pursue it with the X-H1 which has IBIS.
Odds are the XT4 will not have it either due to same issue.
GFX100 has traditionally designed Bayer pattern which allows the use of pixel shift design considerations more acceptable.
Paul C
-
Partly in response to this thread, I've done a theoretical study on the effect of pixel shift on resolution:
https://blog.kasson.com/the-last-word/does-pixel-shift-increase-resolution/
Jim
-
Thanks Jim!
I can vaguely understand why the MTF is unchanged with the “every pixel sampling” of your luminosity-only model. What is less clear is the effect with Bayer CFA. In the simplest case of four shifts by full pixels, one measures red, blue and green (twice!) at each location, so like Foveon X3 and a bit more (Dare I call it X4?). That seems good for something, changing or eliminating the spatial averaging needed in Bayer demosaicing. Reports are that the shift modes of Olympus and Pentax do add some visible benefit.
But with luminosity resolution more important for our eyes, it might not add much in practice.
-
Thanks Jim!
I can vaguely understand why the MTF is unchanged with the “every pixel sampling” of your luminosity-only model. What is less clear is the effect with Bayer CFA. In the simplest case of four shifts by full pixels, one measures red, blue and green (twice!) at each location, so like Foveon X3 and a bit more (Dare I call it X4?). That seems good for something, changing or eliminating the spatial averaging needed in Bayer demosaicing. Reports are that the shift modes of Olympus and Pentax do add some visible benefit.
But with luminosity resolution more important for our eyes, it might not add much in practice.
The 4-shot sampling doesn't increase sharpness, but, done right, completely eliminates the false-color artifacts associated with the Bayer CFA. In addition, it avoids any loss in sharpness caused by interpolation during Bayer demosaicing.
The 16-shot sampling does all that and in addition allows the elimination of some more aliasing.
There are undeniable advantages (and disadvantages) to pixel-shift shooting, but pure MTF is not one of them, because the sampling window is 4 times are area of what it should be for a sensor with half the pitch of the actual sensor.
Jim
-
At least this pretty much puts an end to the life of the Hasselblad H system ... the only specificity that could still have convinced new photographers to buy into the system at this point in time is now gone and the revenue stream associated to it as well.
It could be so, unless the H system finally gets a new "full frame" back exceeding the detail (pixel count) of the new 44x33 mirrorless systems. Because after the technical input in this thread, my prediction is only slightly changed: the only "specificity" that I see being strong enough to maintain a market for new gear in the far bulkier and more expensive 645 SLR-based cameras is larger sensors that give more detailed images. And even that is a thin advantage, with an only modestly larger format (47% area, 21% linear); the best that 645 SLRs can hope for is Sony to keep making 53x40mm sensors with the same unit cell as in 44x33mm sensors, giving a 21% higher (linear) resolution and so prints of 47% greater area at equal PPI.
Most other advantages typically claimed for larger formats seem mostly neutralized by things like improving multi-shot (DR stacking, etc.), leaving mainly "more detail" and the love of big, bright OVF images. The OVFs might provide enough reason for some people to keep using 645 SLR gear, but I doubt it will be enough to attract enough new customers and sales for the longer term.
-
Partly in response to this thread, I've done a theoretical study on the effect of pixel shift on resolution:
https://blog.kasson.com/the-last-word/does-pixel-shift-increase-resolution/
Jim
Interesting, and thanks!
My experiences using the multi-shot and micro step on the Imacon/Hasselblad and Sinar backs really made me a believer, but I've been a bit underwhelmed by the pixel shift on the newer cameras based on my expectations from the older backs. I've just got the Sony A7rIV and haven't yet tried their 16 shot mode. My first test with multishot, I taped up a bunch of stuff to a wall including a dollar bill and shot some images. The things that really impressed me the most was not the increase in hard line things like the print and the cloth fibers in the dollar but stuff like my finger print on the scotch tape used to fix the bill to the wall. This was subtle thing, very faint, and it wasn't there at all in single shot and appeared with remarkable clarity in the multi-shot. And the multishot images just had so much more palpable illusion of depth to them when you stepped back and looked at the whole image and not just 100%.
Each frame is taken with the same sensor regardless of single or multi - so is it right to assume a smaller pitch in your calculations? In a way with shift you're actually sampling 4x the area for each pixel and in my empirical experience this results in 1.5 - 2 stops more usable DR and being able to stop down further without noticing diffraction softening - exactly what you'd get from a larger sensel and opposite of what you'd get with a smaller sensel. And also why assume a monochromatic sensor? One of the biggest advantages with pixel shift/multi shot is that you're recording true color information for each site. I think there must be a better way to evaluate the loss of information between an interpolated RGB data set and a true measured RGB data set - it's not so much that pixel shift has more resolution as it has more accurate data. I think another question mark is how the frames are assembled since it's proprietary software and there are some differences in methodology.
-
Each frame is taken with the same sensor regardless of single or multi - so is it right to assume a smaller pitch in your calculations?
Yes. The pitch of the reconstructed image is finer than the pitch of the sensor.
In a way with shift you're actually sampling 4x the area for each pixel and in my empirical experience this results in 1.5 - 2 stops more usable DR...
Yes, pixel shift does increase DR, as would taking 16 images of the same subject without shifting.
...and being able to stop down further without noticing diffraction softening...
The measurable diffraction softening in the absence of aliasing is the same. However, you have the ability to sharpen the pixel-shift image more since it has fewer aliasing artifacts.
- exactly what you'd get from a larger sensel and opposite of what you'd get with a smaller sensel. And also why assume a monochromatic sensor?
To make the analysis tractable, and to get away from issues about the sharpness and aliasing changing with the demosaicing algorithm.
-One of the biggest advantages with pixel shift/multi shot is that you're recording true color information for each site.
Absolutely! There is a dramatic reduction in color aliasing.
I think there must be a better way to evaluate the loss of information between an interpolated RGB data set and a true measured RGB data set - it's not so much that pixel shift has more resolution as it has more accurate data.
That last sentence is precisely my point!
I think another question mark is how the frames are assembled since it's proprietary software and there are some differences in methodology.
Yes. A reason to assume a monochromatic sensor is to sidestep all that. It's complicated enough as it is.
Jim
-
So... after all the math is over and done with (thank you Jim and others), I'll just add: go have a look at the Hasselblad samples from five or six or however many years ago when they came out with their big campaign, if you haven't before. The difference is incredible. The single-shot images, which were fine a few seconds before, immediately look a bit like mud.
If you shoot static things, not on a timeline, as a way of making art, and your goal is to print large (and one day, "display large," of course), this is a huge step up, indeed.
I hope Phase can develop a stabilization-based body one day, not just so that they remain competitive, but also because then Doug will start telling us how awesome it is. :)
-
I hope that they do Hi-res with an update. Many will smile...
Seriously!
-
So... after all the math is over and done with (thank you Jim and others), I'll just add: go have a look at the Hasselblad samples from five or six or however many years ago when they came out with their big campaign, if you haven't before. The difference is incredible. The single-shot images, which were fine a few seconds before, immediately look a bit like mud.
A major driver in that case is that Hasselblad has very dated/mediocre single shot raw processing in their Phocus software.
If someone wants to make a comparison they should do so between a best-in-class single shot and best-in-class multishot system.
-
A major driver in that case is that Hasselblad has very dated/mediocre single shot raw processing in their Phocus software.
If someone wants to make a comparison they should do so between a best-in-class single shot and best-in-class multishot system.
Multi-shot seems to show significant IQ gains with the Olympus and Pentax implementations too, so it seems hard to put it mostly down to poor handling of regular one-shot conversions.
-
A major driver in that case is that Hasselblad has very dated/mediocre single shot raw processing in their Phocus software.
That may be so, I wouldn't know. I just downloaded the TIFFs from their site.
If someone wants to make a comparison they should do so between a best-in-class single shot and best-in-class multishot system.
That sounds like a great test. As soon as Fuji finishes shipping this feature, we'll all have a chance to compare GFX100 multishot to GFX100 singleshot.
In addition, I suppose we could also compare GFX multishot to Phase singleshot. :)
But, seriously: I'm not shelling out 3-5x the money for a Phase system with what, for me, is substantially inferior shooting ergo. So, the compelling comparison for me will be: does multishot improve the results with the camera I can afford. I am pretty sure I know what the answer will be already. If it also happens that GFX multishot is better in some ways than Phase 150MP singleshot, that's nice to know in an academic sense.
-
That sounds like a great test. As soon as Fuji finishes shipping this feature, we'll all have a chance to compare GFX100 multishot to GFX100 singleshot.
I see no reason why, at a pixel level, it won't look like the a7RIV pixel shift shots. Think of the a7rIV sensor as a crop of the GF 100 sensor.
Jim
-
A major driver in that case is that Hasselblad has very dated/mediocre single shot raw processing in their Phocus software.
If someone wants to make a comparison they should do so between a best-in-class single shot and best-in-class multishot system.
[/quote
I've done that, still impressed with the CF 528 which is Hasselblad - no problems with the software. Workflow is simpler with single shot, but I won't be saying it's best even with the 150mp tri color. TPO really missed an opportunity not going with the multi-shot. If you are doing art reproduction subtle stuff gets missed without having the non-interpolated color. Very curious to see how the Fuji 100mp camera does with pixel shift. Really seems to me that Phase missed an opportunity to be as you write 'best in class" (what a tired expression) by not offering a multi-shot option.
-
I see no reason why, at a pixel level, it won't look like the a7RIV pixel shift shots. Think of the a7rIV sensor as a crop of the GF 100 sensor.
Jim
Well, to Doug's point, if there are substantial differences in the magic software soup, then things could turn out differently. This isn't (necessarily) just about hardware equivalency.
If you made me guess, though: sure. It seems reasonable to suspect that it will be about the same.
-
A major driver in that case is that Hasselblad has very dated/mediocre single shot raw processing in their Phocus software.
If someone wants to make a comparison they should do so between a best-in-class single shot and best-in-class multishot system.
I've done that, still impressed with the CF 528 which is Hasselblad - no problems with the software. Workflow is simpler with single shot, but I won't be saying it's best even with the 150mp tri color. TPO really missed an opportunity not going with the multi-shot. If you are doing art reproduction subtle stuff gets missed without having the non-interpolated color.
That's not the evaluation of the overwhelming majority of high-end cultural heritage institutions (museums, libraries, archives). Very very few of them are still using multi-shot cameras, in contrast to 15-20 years ago when many (most?) of them did. Specifically, many (most?) of them are now our clients.
As I've said numerous times in the past: there are use cases where modern multishot can still be helpful, but they are rare, the difference is modest, and the constraints are significant. Most of what you see written about multishot comes from the golden era of multishot (when 16mp to 22mp sensors dominated and poor single-shot raw processing).
If Fuji adds a multishot mode via firmware the users should try it; free features added after launch are an unequivocal good thing – on that, Fuji and Phase One agree. But if those users think firmware will magically turn their system into the equivalent of a 400mp camera, as many will based on how multishot is advertised, there will be a lot of disappointment.
-
Doug, it doesn’t surprise me that the increase in the resolution available in single shot reduces the proportion of use cases that need the extra effort and cost of a multi shot back; where 50 MP single shot was not enough so a MS was worth the tens of thousands of dollars more; now 150MP (or even 100MP) single shot is as good or better, and good enough for a larger share of use cases.
But Fujifilm is proposing a 100MP+MS for far less than the cost of any 150 MP single shot solution.
P.S. The “400MP” description can be misleading, but I’m not sure that it’s deliberately deceptive; it is necessary to use that many numerical “pixel” values to encode the extra data. It’s the hazard of the acceptance of pixel counts as a proxy for resolution.
-
Well, to Doug's point, if there are substantial differences in the magic software soup, then things could turn out differently. This isn't (necessarily) just about hardware equivalency.
If you made me guess, though: sure. It seems reasonable to suspect that it will be about the same.
With Sony, we aren’t limited to the manufacturer’s development software, either for single or multiple shots. I think it will be the same with Fuji. In fact, I don’t see how Fujifilm could prevent that.
-
Doing a multi-shot specific back isn't really a huge upsell outside of very specific markets. Doing an IS setup for a chip with really small pixel sites does make sense. Let Fuji keep working on it, and when it's perfected at that level, it'll be ready for the next IQ, or should we call it the 'IS'? Since Sony isn't making a bigger chip, the only way to higher resolutions is smaller photo sites, which may need the IS plus the gyros in the XF body for non-tripod use. Part of what Phase has to balance is how can we keep the photographer working, not dealing with a camera issue.
You don't need in-body stabilization on the IQ4, but it sure would be nice. I'm going thru some shots from MLS Cup last year and it's not pixel perfect, even at 1/1250th. Blame me, my technique & the lack of having a bunch of static cameras on remotes.
(https://pbs.twimg.com/media/EJmqMdyUcAA7hdb?format=jpg&name=4096x4096)
-
With Sony, we aren’t limited to the manufacturer’s development software, either for single or multiple shots. I think it will be the same with Fuji. In fact, I don’t see how Fujifilm could prevent that.
That is true, but it does not invalidate my statement. Fuji could well ship a proprietary algorithm that outperforms any other available one. In the same light, Sony's algorithm may be the best one available for their files today. (I have no idea.)
-
That is true, but it does not invalidate my statement.
Any "magic" developed by third parties could be used with any pixel-shift cameras they chose to support. If they support both Sony and Fuji 3.76 um cameras, then the playing field will be level.
Fuji could well ship a proprietary algorithm that outperforms any other available one. In the same light, Sony's algorithm may be the best one available for their files today. (I have no idea.)
Sony's reconstruction algorithm is bested by at least one third party, and probably two.
Reconstructing 4-way and 16-way shift stacks is pretty simple, but compensating for camera and subject motion is not.
Does Fuji sell raw developers for the GFX cameras at present? I am unaware of their existence. If they don't, why should they start now?
Jim
-
You don't need in-body stabilization on the IQ4, but it sure would be nice. I'm going thru some shots from MLS Cup last year and it's not pixel perfect, even at 1/1250th. Blame me, my technique & the lack of having a bunch of static cameras on remotes.
The painful thing is that an imperfect 150mp image may contain as little as 20 or 40mp worth of image detail, depending on how much movement there was obviously.
Odds are that a GFX-100 or a Sony a7rIV with their image stabilization are able to deliver a lot more real world detail than an IQ4-150 shot imperfectly.
Cheers,
Bernard
-
The painful thing is that an imperfect 150mp image may contain as little as 20 or 40mp worth of image detail, depending on how much movement there was obviously.
Odds are that a GFX-100 or a Sony a7rIV with their image stabilization are able to deliver a lot more real world detail than an IQ4-150 shot imperfectly.
Coming from a background of wedding photography, mostly using available light (you're moving, subjects are moving, light is not always abundant) I really don't understand the obsession/overstatement of the shutter speeds required to handle various levels of detail.
IS is a nice feature; if you're using a camera that has it, you should use it where applicable. It does expand the range of subject illumination in which you can shoot hand held. But cameras without IS are still, absolutely, hand holdable. More here: https://phaseoneiq4.com/yes-you-can-hand-hold-150-megapixels/
Not to mention that not every shot requires/benefits from maxing out the resolution of every last pixel. There are many image quality reasons to shoot a given given camera other than resolution, such as dynamic range, color, tonality, lens look, sync speed, etc, do. So even if you're in a range where you're "only" getting 50mp worth of detail from a 150mp camera, that doesn't automatically preclude it's use to the creation of beautiful images (unless the specific shot you are doing is one where micro detail is a leading/major aesethic contributor).
-
Coming from a background of wedding photography, mostly using available light (you're moving, subjects are moving, light is not always abundant) I really don't understand the obsession/overstatement of the shutter speeds required to handle various levels of detail.
IS is a nice feature; if you're using a camera that has it, you should use it where applicable. It does expand the range of subject illumination in which you can shoot hand held. But cameras without IS are still, absolutely, hand holdable. More here: https://phaseoneiq4.com/yes-you-can-hand-hold-150-megapixels/
Not to mention that not every shot requires/benefits from maxing out the resolution of every last pixel. There are many image quality reasons to shoot a given given camera other than resolution, such as dynamic range, color, tonality, lens look, sync speed, etc, do. So even if you're in a range where you're "only" getting 50mp worth of detail from a 150mp camera, that doesn't automatically preclude it's use to the creation of beautiful images (unless the specific shot you are doing is one where micro detail is a leading/major aesethic contributor).
That sounds about right, except for one critical argument: you don't really need a 150 MP camera for beautiful images, good colours, good dynamic range and leaf shutter lenses. If you're buying a 150 MP camera, the primary value proposition is the 150 MP. Otherwise, you do have lower MP count options, offering the exact same "aesthetic contributors" for a fraction of the cost.
And for wedding photography, 150 MP is overkill anyway.
I do need 150 MP for my large gallery prints (I would need a single shot 400 MP if one existed). So squeezing every single pixel of every camera I had is a critical requirement, for me. While a non-stabilised camera restricted my freedom of movement and creativity, tying me to a tripod in at least 50% of the situations, the IBIS allows me to access more creative angles, more difficult locations, more rapid light changing situations, while still realising the main value proposition of the camera I bought: currently, the 100 MP of the GFX 100. Even if pricing was identical (which is far from being the case) I would still have to choose as my 1st option the 100 MP stabilised camera vs the non stabilised 150 MP. The 150 MP would be my second option / camera, and I would consider it primarily for those extra MPs (leaf shutter lenses and tech lenses are another reason, but for my shooting they are maximum 20% of the shooting scenarios).
Simply because the GFX 100 allows me to focus more on creating images, than fiddling with my tools. It gets out of the way faster, like any good tool should. I've been on the "slowing down will make your shots better" path, and that while it may be true for beginning photographers who need to get out of the spay-and-pray mode of photography, for those of us who know what they need from their tools and their images, the IBIS-mirrorless-modern AF combination is really unlocking new creative possibilities. I would rather slow down to explore more angles, more fleeting light or movement scenarios, than slow down just because I need to focus on operating 3 metal legs and multiple knobs with all the paraphernalia associated with that. And even in those situations where I need to commit to doing that, I will do it primarily because those extra MPs count, because in my 1.80 by 1.20 to 3m by 2m gallery prints, every single bit of properly captured and processed detail counts.
It's hard to consider seriously the argument that even if you're spending the money for 150MP it shouldn't really matter if you're only getting 50MP out of that. The 10x price difference (German car level of investment) is hard to swallow based on "other aesthetic contributors" - especially when those same "contributors" can be realised with different tools for a fraction of the cost, especially considering the final display medium. While that argument may work for rich dentists and snobs, I wouldn't trust a real image creator making that argument. Sorry if I come across as harsh, but I spent too much time, energy and money feeding the marketing machines of camera makers and sellers, who kept convincing me that I always need more and more and more diminishing-returns-"special"-"magic'-"unquantifiable aesthetics"-gear, while in reality I only needed ... 400 MP stabilised single shots, which nobody can deliver (small laughing break allowed) :))) And yes, while I'm in the target audience for that, I will really need 400 realised MPs not 50 MP with "added aesthetic benefits". So please, let's try a better sales pitch, next time :)
-
That sounds about right, except for one critical argument: you don't really need a 150 MP camera for beautiful images, good colours, good dynamic range and leaf shutter lenses. If you're buying a 150 MP camera, the primary value proposition is the 150 MP. Otherwise, you do have lower MP count options, offering the exact same "aesthetic contributors" for a fraction of the cost.
And for wedding photography, 150 MP is overkill anyway.
I do need 150 MP for my large gallery prints (I would need a single shot 400 MP if one existed). So squeezing every single pixel of every camera I had is a critical requirement, for me. While a non-stabilised camera restricted my freedom of movement and creativity, tying me to a tripod in at least 50% of the situations, the IBIS allows me to access more creative angles, more difficult locations, more rapid light changing situations, while still realising the main value proposition of the camera I bought: currently, the 100 MP of the GFX 100. Even if pricing was identical (which is far from being the case) I would still have to choose as my 1st option the 100 MP stabilised camera vs the non stabilised 150 MP. The 150 MP would be my second option / camera, and I would consider it primarily for those extra MPs (leaf shutter lenses and tech lenses are another reason, but for my shooting they are maximum 20% of the shooting scenarios).
Simply because the GFX 100 allows me to focus more on creating images, than fiddling with my tools. It gets out of the way faster, like any good tool should. I've been on the "slowing down will make your shots better" path, and that while it may be true for beginning photographers who need to get out of the spay-and-pray mode of photography, for those of us who know what they need from their tools and their images, the IBIS-mirrorless-modern AF combination is really unlocking new creative possibilities. I would rather slow down to explore more angles, more fleeting light or movement scenarios, than slow down just because I need to focus on operating 3 metal legs and multiple knobs with all the paraphernalia associated with that. And even in those situations where I need to commit to doing that, I will do it primarily because those extra MPs count, because in my 1.80 by 1.20 to 3m by 2m gallery prints, every single bit of properly captured and processed detail counts.
It's hard to consider seriously the argument that even if you're spending the money for 150MP it shouldn't really matter if you're only getting 50MP out of that. The 10x price difference (German car level of investment) is hard to swallow based on "other aesthetic contributors" - especially when those same "contributors" can be realised with different tools for a fraction of the cost, especially considering the final display medium. While that argument may work for rich dentists and snobs, I wouldn't trust a real image creator making that argument. Sorry if I come across as harsh, but I spent too much time, energy and money feeding the marketing machines of camera makers and sellers, who kept convincing me that I always need more and more and more diminishing-returns-"special"-"magic'-"unquantifiable aesthetics"-gear, while in reality I only needed ... 400 MP stabilised single shots, which nobody can deliver (small laughing break allowed) :))) And yes, while I'm in the target audience for that, I will really need 400 realised MPS, not 50 MP with "added aesthetic benefits". So please, let's try a better sales pitch, next time :)
Using the same arguments i choose a 35mm option with very fast AF & IBIS & 1.4 lenses ( and an abundant lens choice form 800mm to fisheye)
Also a lot cheaper if you drop them in the heat of the moment
-
Using the same arguments i choose a 35mm option with very fast AF & IBIS & 1.4 lenses ( and an abundant lens choice form 800mm to fisheye)
Also a lot cheaper if you drop them in the heat of the moment
And for 99.9% of current photographic needs / tasks / jobs, that sounds about right.
The only reason I chose medium format was because I needed the resolution for the large size gallery prints. And had the GFX 100 not been available, with those features, lenses and post-processing options and especially at that price point, I would still shoot 35mm. But I'm a serious minority. Most photographers don't print or don't need to print that big, and I also believe, as earlier stated that 99.9% of the contemporary creative requirements can be met with a solid 35mm kit.
Which makes it a lot harder to try and sell a 10x more expensive kit, with the argument "if 150 MP fails, you can still get 50 MP of pretty".
-
... cameras without IS are still, absolutely, hand holdable. More here: https://phaseoneiq4.com/yes-you-can-hand-hold-150-megapixels/
The linked example fits with my observation that, with careful hand-holding, I can do one or two stops better than 1/f in35mm format — for sharpness that is adequate for normal causal viewing; say about 3K (6MP). And I agree that the video world has a far better way of describing resolution, with linear measures like "4K" or "2000 lines per picture height", so 150MP is about 14K and only needs about three or four times the shutter speed of 35mm format film. (Bulky cameras have an inherent stability advantage too!)
Then again, that example is wide-angle in bright sunlight, about the easiest case for hand-holding. Where IBIS has most amazed me is dimly lit indoor scenes with no flash or tripod usable.
-
Any "magic" developed by third parties could be used with any pixel-shift cameras they chose to support. If they support both Sony and Fuji 3.76 um cameras, then the playing field will be level.
Sony's reconstruction algorithm is bested by at least one third party, and probably two.
Reconstructing 4-way and 16-way shift stacks is pretty simple, but compensating for camera and subject motion is not.
Does Fuji sell raw developers for the GFX cameras at present? I am unaware of their existence. If they don't, why should they start now?
Okay, let me try again, because it feels like we're talking past one another now.
Doug suggested that an algorithm that reconstructs MS from a given camera might look great relative to an inferior algorithm that reconstructs SS on the same camera, and that to make the best decision we should compare the best MS only to the best SS end-to-end implementation, inclusive of the software processing.
I suggested that we will soon be able to compare GFX MS to SS and of course Phase SS.
You suggested that the improvement will look roughly like the A7r4 MS, presumably since the GFX and A7r4sensors are approximately the same per unit area.
I responded in an attempt to play devil's advocate by saying that Doug's statement suggests that any improvement may not look quite the same, since the A7r4 MS reconstruction algorithm may be different than the GFX MS one. (I also said that I doubt that it will be the case that it is much different.)
You said a bunch more things, some of which add a lot more information (who has written better algorithms, what is available today, etc.) that is interesting, but I believe none of which invalidate what I wrote. You ask me to speculate about manufacturers' intentions, when all I was trying to do was leave open the theoretical possibility that the software could, indeed, make a meaningful difference.
Yes, I think it is unlikely that Fuji will have some special algorithm that is much better. No, I do not know who writes the GFX algorithms that are in LR, C1, Iridient, Phocus, dcraw, or whatever else. I would have assumed that Fuji would at least help one or more of these dev teams optimize one or more of these algorithms, so when I wrote "proprietary" I did not mean to suggest a new algorithm that must be licensed separately.
But I do think that it's still possible that the GFX MS software may be better or worse than the A7r4 MS software, and I also believe my earlier statements hold true. (Let me know if not.)
Whew!
-
That sounds about right, except for one critical argument: you don't really need a 150 MP camera for beautiful images, good colours, good dynamic range and leaf shutter lenses. If you're buying a 150 MP camera, the primary value proposition is the 150 MP. Otherwise, you do have lower MP count options, offering the exact same "aesthetic contributors" for a fraction of the cost.
And for wedding photography, 150 MP is overkill anyway.
I do need 150 MP for my large gallery prints (I would need a single shot 400 MP if one existed). So squeezing every single pixel of every camera I had is a critical requirement, for me. While a non-stabilised camera restricted my freedom of movement and creativity, tying me to a tripod in at least 50% of the situations, the IBIS allows me to access more creative angles, more difficult locations, more rapid light changing situations, while still realising the main value proposition of the camera I bought: currently, the 100 MP of the GFX 100. Even if pricing was identical (which is far from being the case) I would still have to choose as my 1st option the 100 MP stabilised camera vs the non stabilised 150 MP. The 150 MP would be my second option / camera, and I would consider it primarily for those extra MPs (leaf shutter lenses and tech lenses are another reason, but for my shooting they are maximum 20% of the shooting scenarios).
Simply because the GFX 100 allows me to focus more on creating images, than fiddling with my tools. It gets out of the way faster, like any good tool should. I've been on the "slowing down will make your shots better" path, and that while it may be true for beginning photographers who need to get out of the spay-and-pray mode of photography, for those of us who know what they need from their tools and their images, the IBIS-mirrorless-modern AF combination is really unlocking new creative possibilities. I would rather slow down to explore more angles, more fleeting light or movement scenarios, than slow down just because I need to focus on operating 3 metal legs and multiple knobs with all the paraphernalia associated with that. And even in those situations where I need to commit to doing that, I will do it primarily because those extra MPs count, because in my 1.80 by 1.20 to 3m by 2m gallery prints, every single bit of properly captured and processed detail counts.
It's hard to consider seriously the argument that even if you're spending the money for 150MP it shouldn't really matter if you're only getting 50MP out of that. The 10x price difference (German car level of investment) is hard to swallow based on "other aesthetic contributors" - especially when those same "contributors" can be realised with different tools for a fraction of the cost, especially considering the final display medium. While that argument may work for rich dentists and snobs, I wouldn't trust a real image creator making that argument. Sorry if I come across as harsh, but I spent too much time, energy and money feeding the marketing machines of camera makers and sellers, who kept convincing me that I always need more and more and more diminishing-returns-"special"-"magic'-"unquantifiable aesthetics"-gear, while in reality I only needed ... 400 MP stabilised single shots, which nobody can deliver (small laughing break allowed) :))) And yes, while I'm in the target audience for that, I will really need 400 realised MPs not 50 MP with "added aesthetic benefits". So please, let's try a better sales pitch, next time :)
You seem to have misunderstood my post. My only point is that this is far less common than most people seem to think because the range of situations in which you can hand-hold the IQ4 150mp is far greater than I see commonly written/assumed. If you trusted the internet, then an IQ4 150mp without a tripod is useless, which is just empirically false; most of our users use their gear handheld at least much of the time.
And for what it's worth, I don't agree that any camera, compared to any other camera can provide "the exact same aesthetic contributors for a fraction of the cost" (color, lens look, dynamic range, tonality etc). Every camera and camera system (unless it's just a rebrand or such) is unique. The color, lens look, tonality of a Fuji GFX100 and IQ4 150mp is very far from "the exact same"; reasonable people can disagree on which one is better in these regards – I'm sure you know my opinion (nothing touches the color and tonality of an IQ4), but I recognize it as an opinion. They are very different camera systems: they use different RGB filters on the sensor, different IR filters in front of the sensor, different sensor sizes, different lens systems, different shutters, different resolutions, different color profiles, and handle very differently (menus/features/focus/tools/ergo etc); really the only thing that is the same is that they are both larger-than-35mm and both have higher-than-other-cameras sensor resolution. Again, all of those differences will favor either camera in different scenarios and for different users.
In terms of dynamic range, the IQ4 150mp is demonstrably better than the GFX100, even before accounting for the Frame Averaging feature that takes the IQ4 dynamic range to an entirely new level. Again, this may or may not matter to a given photographer; if you're shooting low-contrast scenes the dynamic range is irrelevant; if you're only shooting moderate-high contrast scenes the difference between the GFX and IQ4 will not be important; if you're shooting very high contrast scenes this can matter quite a bit.
It sounds like the GFX might well be the best camera for you. That is, succinctly, if you very frequently shoot in a situation where IS is mandatory for your results, then you're best served by a camera with IS.
Just checking though... have you had the chance to actually use an IQ4 150mp in the field, on an XF or XT?
-
Okay, let me try again, because it feels like we're talking past one another now.
Well anyway, the three of us (Jim, you, me) all likely agree on one thing: none of us can know exactly what the benefits, limitations, and relevant comparisons will be until they can be directly tested.
Real-world tests show that which theory only speculates on.
Should anyone want to test a GFX (in a rumored pixel-shift mode or otherwise) against an IQ4, we (DT) are glad to arrange for an IQ4 to be available to you in our NY or LA locations or in San Fran, Phoenix, San Diego, Dallas, Houston, or Austin, all of which we'll be to in the next couple months (see our events calendar for more detail (https://www.dtcommercialphoto.com/events/)).
-
Well anyway, the three of us (Jim, you, me) all likely agree on one thing: none of us can know exactly what the benefits, limitations, and relevant comparisons will be until they can be directly tested.
Real-world tests show that which theory only speculates on.
Should anyone want to test a GFX (in a rumored pixel-shift mode or otherwise) against an IQ4, we (DT) are glad to arrange for an IQ4 to be available to you in our NY or LA locations or in San Fran, Phoenix, San Diego, Dallas, Houston, or Austin, all of which we'll be to in the next couple months (see our events calendar for more detail (https://www.dtcommercialphoto.com/events/)).
Ship them both to Nantucket and I will test them thoroughly. Might take me a year or two.
8)
Sharon
-
And for what it's worth, I don't agree that any camera, compared to any other camera can provide "the exact same aesthetic contributors for a fraction of the cost" (color, lens look, dynamic range, tonality etc). Every camera and camera system (unless it's just a rebrand or such) is unique.
Just checking though... have you had the chance to actually use an IQ4 150mp in the field, on an XF or XT?
1. "Any camera" and "exact same aesthetics" are definitely not my words - my whole post was along the line of diminishing returns and real creative output, with the final display medium in mind, based on the whole "even a degraded 50MP out of a 150MP back is amazing"
2. I haven't had the chance to test an IQ4 150 in the field, it's way too rich for my blood, I'm not even going to consider it until the price comes waaaay down - the XT is the interesting platform, even though the lack of EVF support makes it a bit crippled for location work.
But I'm stitching and stacking in all possible combinations, so I am no stranger to "ultra-high" resolution final capture files (with all the dynamic range, micro-contrast and colour tonalities associated with them).
And please don't tell me, again, that this is not about resolution, it's about how beautiful a 150MP shot, degraded to 50MP of details due to camera shake, is vs a properly shot 50MP smaller format option, because I will never be able to stomach the 10x price argument based on that. I'm definitely not the target audience for that sales pitch.
-
IBIS has been around for some time and its value as a feature is likely well known to many. Multi-Shot (pixel shift) has also been available for decades and the value of capturing full RGB color data for every pixel, for those that have made comparisons, is also understood. That manufacturers have, in recent years, combined these two features to provide multi-shot image quality at much lower price points than previously available is of great interest and value to me. I hope to see even wider adoption of it.
Reading thru this thread and seeing the various attempts to diminish the quality or value of one or more manufacturer's products or features relative to another reminds me, for some strange reason, of a famous philosopher's quotation.
"Well, Art is Art, isn't it? Still, on the other hand, water is water. And east is east and west is west and if you take cranberries and stew them like applesauce they taste much more like prunes than rhubarb does. Now you tell me what you know."
Groucho Marx
Animal Crackers - 1930 (https://www.youtube.com/watch?v=d1-iqfSOaMQ)
-
I really don't understand the obsession/overstatement
No comment.
-
Okay, let me try again, because it feels like we're talking past one another now.
Doug suggested that an algorithm that reconstructs MS from a given camera might look great relative to an inferior algorithm that reconstructs SS on the same camera, and that to make the best decision we should compare the best MS only to the best SS end-to-end implementation, inclusive of the software processing.
I suggested that we will soon be able to compare GFX MS to SS and of course Phase SS.
You suggested that the improvement will look roughly like the A7r4 MS, presumably since the GFX and A7r4sensors are approximately the same per unit area.
I responded in an attempt to play devil's advocate by saying that Doug's statement suggests that any improvement may not look quite the same, since the A7r4 MS reconstruction algorithm may be different than the GFX MS one. (I also said that I doubt that it will be the case that it is much different.)
You said a bunch more things, some of which add a lot more information (who has written better algorithms, what is available today, etc.) that is interesting, but I believe none of which invalidate what I wrote. You ask me to speculate about manufacturers' intentions, when all I was trying to do was leave open the theoretical possibility that the software could, indeed, make a meaningful difference.
Yes, I think it is unlikely that Fuji will have some special algorithm that is much better. No, I do not know who writes the GFX algorithms that are in LR, C1, Iridient, Phocus, dcraw, or whatever else. I would have assumed that Fuji would at least help one or more of these dev teams optimize one or more of these algorithms, so when I wrote "proprietary" I did not mean to suggest a new algorithm that must be licensed separately.
But I do think that it's still possible that the GFX MS software may be better or worse than the A7r4 MS software, and I also believe my earlier statements hold true. (Let me know if not.)
Whew!
My wife used to be the NP for a bunch of surgeons. They had a saying: that, if there are lots of ways to do something, none of them are very good.
Demosaicing a single shot Bayer-CFA image is hard. There are a huge number of ways to do it. Adobe just invented a new one last year. There are problems with them all. So establishing the single-shot baseline may be tricky.
Except for camera and subject motion, demosaicing a 4-way or 16-way multishot image is easy, and there aren't a lot of ways to do the basic demosaicing. Except for some gratuitous post processing done by Sony, pixel shift images from the a7RIV produced by IE look a lot like those from PixelShift2DNG. Any test images should be captured without much camera or subject motion, so they all ought to look about the same (except for the extra PP that SOny adds, which is not tamable). And I've never seen a multishot algorithm that impressed me with how it dealt with camera motion.
There will continue to be improvements in single shot demosaicing (and I'll bet some of them will turn workstations into room heaters like Adobe's new one does). But, except for motion, i don't expect much in the way of such improvements in multishot reconstruction.
Jim
-
Demosaicing a single shot Bayer-CFA image is hard. There are a huge number of ways to do it. Adobe just invented a new one last year. There are problems with them all. So establishing the single-shot baseline may be tricky.
Right. The computation reminds me a lot of the inverse of dithering algorithms for print. Plenty of tradeoffs to be made, different goals to be achieved, and of course, subjectivity.
Except for camera and subject motion, demosaicing a 4-way or 16-way multishot image is easy, and there aren't a lot of ways to do the basic demosaicing.
I understand what you're saying now. I didn't realize the demosaicing was so straightforward for these two modes.
-
Right. The computation reminds me a lot of the inverse of dithering algorithms for print. Plenty of tradeoffs to be made, different goals to be achieved, and of course, subjectivity.
I understand what you're saying now. I didn't realize the demosaicing was so straightforward for these two modes.
Glad we're finally on the same page.