Luminous Landscape Forum
Site & Board Matters => About This Site => Topic started by: Hans Kruse on September 05, 2014, 02:54:36 pm
-
I have just read the review and was pleased to see a comprehensive and detailed review resulting from real experience and use of the camera. Especially I appreciate the comments on ETTR which are often neglected in reviews and happy to see that Pentax has at least made a practical solution to avoid the need for bracketing. Neither Canon or Nikon has done that and I have the Canon 5D III and the Nikon D810. So to handle really high DR scenes without HDR blending true ETTR is essential as is written in the review.
There is one error: The size of the Pentax sensor is as mentioned 1.7x larger than 35mm full frame, but the difference between 35mm full frame and APS-C is even larger. It's 2.25x for Nikon and 2.56x for Canon. The difference between 35mm FF and 645 FF is 2.5x. And between 645Z and 645 FF as in Phase One IQ260 and IQ280 the sensor is 1.5x larger. So going from APS-C Canon to FF is about the same step as going from 35mm FF til 645 FF.
One could argue that the difference between 645Z and 35mm FF as in e.g. Nikon D810 is not that big. The comment on cropping 35mm 3:2 aspect ratio to 4:3 aspect ratio is a fair comment if this is common for the photographer. If this is the case then the area ratio is larger than 1.7x (1.66x) and becomes 1.87x. Although I like this review it would have been stronger for those interested in the real difference between e.g. D810 and 645Z by comparison shooting between the two. But I know this is a lot to ask.
I was surprised to see to see f/22 used that much and accepted. Lloyd Chambers at http://diglloyd.com/ has shot a series of comparison shots from f/2.8 up to f/22. And at f/22 details get really lost, f/16 is better, but really details start to get lost at around f/8. It's similar to 35mm FF where I find f/16 ok when needed with proper sharpening in Lightroom but f/22 is not.
So thanks for the review. All in all a good review!
-
Showing overexposure zebras or blinkies in live view is a terrific feature. Sony's A7(r) displays zebras in its EVF. It also lets you adjust the threshold, useful for optimizing exposure for video as well as stills. For me this alone wipes out mirror-based TTL viewing. I simply don't under/over-expose with the A7r.
Regarding aperture choice: if f/22 is the aperture I need to get the DOF I want, then f/22 is what I use. It makes no sense to choose a larger aperture for the sake of maximizing resolution in some areas of a photo if that results in undesired softness elsewhere. With the 645D I use f/16 a lot with the 35mm FA lens, and have used f/22 too. Results are fine & dandy.
-Dave-
-
Well, I enjoy Lloyd's site, but I don't take his word as gospel.
I have tested the 645z extensively in this regard and for my purposes diffraction doesn't start to show so that it's a concern until f/22. Much of my shooting is at f/16 and I am completely satisfied with the results. If I was super anal retentive I'd shoot at f/8 or f/11, but I'm not. :-)
Michael
-
Telecaster....
We must have posted at the same moment.
Agree 100%...obviously. Too many people become afraid of the diffraction boogyman. I'll take the DOF I want and need over a bit of diffraction lose any day of the week.
Michael
-
Hello,
in your review you judge the dynamic range of the sensor by looking at the bright and the dark parts of the image in Lightroom. As far as I know, one of the biggies of the latest process version of Lightroom is that Lr does an automatic optimization on raws shadows and highlights without showing it as a slider position other than the neutral point.
And, if I understood Jeff's words right, it depends on the image and no to the type of camera, so you will never see a complete us handled image in Lightroom.
Please, correct me if I'm wrong, but to me a judging on the dynamic range is pretty difficult using a raw file in Lightroom.
Nevertheless, I really enjoyed reading this review and I really appreciate this type of reviews which give us your personal feelings on a product's quality than showing curves and data !
Best wishes Robert
-
Showing overexposure zebras or blinkies in live view is a terrific feature. Sony's A7(r) displays zebras in its EVF. It also lets you adjust the threshold, useful for optimizing exposure for video as well as stills. For me this alone wipes out mirror-based TTL viewing. I simply don't under/over-expose with the A7r.
Regarding aperture choice: if f/22 is the aperture I need to get the DOF I want, then f/22 is what I use. It makes no sense to choose a larger aperture for the sake of maximizing resolution in some areas of a photo if that results in undesired softness elsewhere. With the 645D I use f/16 a lot with the 35mm FA lens, and have used f/22 too. Results are fine & dandy.
-Dave-
Regarding DOF shooting 35mm I never need to stop down to f/22. I do use f/16 when needed. Careful focusing and check of DOF can avoid excessive stopping down which does harm IQ. If course and I did intend to say there never would be a need to stop down to f/22, but it is not without cost in IQ.
-
Well, I enjoy Lloyd's site, but I don't take his word as gospel.
I have tested the 645z extensively in this regard and for my purposes diffraction doesn't start to show so that it's a concern until f/22. Much of my shooting is at f/16 and I am completely satisfied with the results. If I was super anal retentive I'd shoot at f/8 or f/11, but I'm not. :-)
Michael
Diffraction starts way before f/22. The questions is what is acceptable and what is not. If you are happy with f/22 that is fine, but to say there is no degradation is simply wrong.
-
Please note what I wrote... For my purposes it isn't an issue. I know when diffraction starts, i just prefer having more DOF to the small quality loss of a smaller aperture.
Michael
-
I think you have a minor error in the review. When comparing sensor sizes, you seem to use the ratio of the areas for the 645z to FF but the ratio of the linear dimensions for FF to APS-C. The ratio of areas for FF to APS-C would be in the range of 2.25 to 2.56.
This is just a nit, I enjoyed the review overall.
Edit: I see Hans Kruse has already made the same point, sorry for being redundant.
-
Hi,
My solid opinion, being analytic, is that the better lenses perform best at f/8 or below. Stopping down to f/16 on my P45+ largely eliminates colour aliasing, a good thing, but also a sure indication of sharpness loss.
My take may also be that if we buy MFD for sharpness, that advantage will be reduced/lost with stopping down to much, so I am reluctant to go below f/11 - f/16.
(http://echophoto.dnsalias.net/ekr/Articles/MFDJourney/Samples/Diffraction/MTF_of_aperture.png)
The example above is a Sonnar 150/4 on Hasselblad 555/P45+, near axis, but laws of physics are same for all lenses and all sensors.
In this case stopping down to f/16 corresponds to loosing like 16 MPixels of 52, 36 MP resolution remaining.
But, some lenses need to be stopped down to f/11 for optimum sharpness. I sometimes take a few pictures with different focus and blend, that often works reasonable well.
Best regards
Erik
Diffraction starts way before f/22. The questions is what is acceptable and what is not. If you are happy with f/22 that is fine, but to say there is no degradation is simply wrong.
-
In this case stopping down to f/16 corresponds to loosing like 16 MPixels of 52, 36 MP resolution remaining.
Loosing MP by stopping down the lens? I lack the knowledge to be able to follow this Eric. I was under the impression that stopping down introduces lack of sharpness/diffraction, but I never would have concluded that this equals loss in MP.
???
-
Erik,
With a due respect, but I believe that you (and many others) overstate the issue of diffraction losses at small apertures, especially with medium format.
Somewhat frustrated with this debate, after breakfast this morning I did a quick test, just to post here.
(http://www.luminous-landscape.com/articleImages/MR64/_MHR0263.jpg)
This is the set up, taken with the Pentax 645z and Pentax 120mm Macro, a lens of undisputed high resolution.
I did a series of exposures, keeping the ISO the same and simply varying the shutter speed relative to the aperture. The range was f/5.6, f/8, f/11, f/16, f/22, and f/32.
(http://www.luminous-landscape.com/articleImages/MR64/f56.jpg)
F/5.6
(http://www.luminous-landscape.com/articleImages/MR64/f22.jpg)
f/22
As I suspected, f/5.6 and f/8 were indistinguishable. F/11 was minutely less crisp and f/22 a bit more so. f/32 was noticeably softer.
The point though, is that a small increase in the amount of sharpening used is all that's need to bring the f/22 shot to the point that it's almost indistinguishable at 100% on screen, and completely indistinguishable in a 16X20" print.
So, while I have no problem with theory, I am much more interested in practice, and as I've written for years, make your own tests before believe so-called experts (including me).
The secret to making the f/22 shot look almost the same as the one at f/5.6 was simply a bit of extra sharpening. Nothing more.
By the way, I wrote about this back in 2002, in the film days, and the same thing applied then. http://www.luminous-landscape.com/essays/stop-d.shtml (http://www.luminous-landscape.com/essays/stop-d.shtml). Stop worrying and keep shoot!
Michael
UPDATE: My apologies. I mistakenly posted the f/5.6 shot twice. The correct f/22 shot is now there. Same difference.
-
Michael:
Point well made. By the way, what construction products do you serve with fruit? ???
Tom
-
Nice write up Michael - really enjoyed it, looks like a great camera at a sensible price point.
Having just watched the 'Mirrorless Revolution' video (again, very informative - especially for someone like me who's been ignorant of everything other than 35mm and MF) it would be interesting to hear the opinions of you both on the state of MF in general. To me that whole segment seems to be desperately trying to figure out where it's going - Pentax serve up the 645Z for less than $10k, Hasselblad may well be opening up their 'H' system again, and have pitched the new CFV-50c at ~$15k, while Phase One appear to be doing pretty much nothing (publicly). Strange times.
Jim
-
Please note what I wrote... For my purposes it isn't an issue. I know when diffraction starts, i just prefer having more DOF to the small quality loss of a smaller aperture.
Hi Michael,
That's fine if one does know what diffraction does with one's particular lens and sensor. I agree that in case of a (creative) trade-off one chooses the best option, even if it hurts the technical quality. However, others should be aware that from f/18 and narrower, effective resolution is reduced below maximum. Also high contrast detail will be unrecoverably lost, even deconvolution sharpening cannot restore it.
At f/22, 'only' some 80% of the maximum attainable resolution is left for restoration (similar to a low contrast 33 MP quality), and even then the amplitude of low contrast micro detail that could theoretically be resolved is reduced so much that it will be hard to see, because diffraction has the effect of a low rate AA-filter. That will effectively pull a 'veil of dullness' over the entire image.
Depending on subject matter, that creates a serious hit on image quality and will require superior quality postprocessing like deconvolution sharpening to restore a little detail from the drabness (and deconvolution sharpening, like with FocusMagic, does improve the quality of the detail that is left in such images).
Only when the creative options demand it, e.g. to achieve a very strongly composed image with important OOF detail contribution to the scene, or a unique event, will the trade-off for smaller apertures than f/18 be a good trade-off. One can boost the 'look' of such diffraction affected images with very specialized software like Topaz Clarity, to counteract some of the lost liveliness, and use Topaz Detail to trick our eyes in to seeing more detail than there actually is. That will work best on low noise images, because serious adjustments to tonality will be required.
I would personally be very careful, and avoid anything narrower than f/16, even with a very good lens, unless absolutely unavoidable. At f/18 the resolution is reduced to virtually zero at the Nyquist frequency, all aliasing is practically eliminated on the 645Z due to diffraction blur, which may be a justification to use such a narrow aperture (e.g. as an additional shot to repair aliasing artifacts with).
Of course this restriction can be relaxed a bit for small output sizes, but then it would also be adequate to use any of most other smaller cameras with lower resolution. An image from the 645Z deserves/demands large output sizes.
Cheers,
Bart
-
Loosing MP by stopping down the lens? I lack the knowledge to be able to follow this Eric. I was under the impression that stopping down introduces lack of sharpness/diffraction, but I never would have concluded that this equals loss in MP.
???
Have a look at Figure 3 in the excellent article (http://www.luminous-landscape.com/tutorials/resolution.shtml) by Rubén Osuna and Efraín García posted on LuLa. Megapixel equivalents vary more than you might think according to the wave length of the light used for imaging. These figures are for green light (λ = 550 nM) and a 36x48 mm medium format sensor:
f/5.6 = 120 MP
f/8 = 31 MP
f/11 = 31 MP
f/16 = 15 MP
f/22 = 8 MP
According to these figures, using f/22 on your MFDB reduces it to the resolution of an iPhone. Of course, the SNR of the MFDB would be much higher. That said, excellent images may be obtained from an iPhone as long as the viewing size is kept relatively small.
Bill
-
Hi,
Sorry, I see a lot of loss of IQ at f/22, this is my test (actual pixel crops):
http://echophoto.dnsalias.net/ekr/Articles/Aliasing2/Dollars.jpg
The image below is probably downscaled, so link above is a better reference.
(http://echophoto.dnsalias.net/ekr/Articles/Aliasing2/Dollars.jpg)
The A77 image is included as a reference, as far as I recall it was shot using same (or similar) lens at similar shooting position.
It is a myth that MF would be less prone to diffraction than smaller formats. It may be almst true that you loose less with "fat pixels" but the P645Z has relatively small pixels, smaller than a 24 MP (FF) DSLR but larger than a 36 MP (FF) DSLR.
BTW, I am he guy who says that there is no significant difference between P45+ and 24 MP DSLRs in A2-prints. I may need to revise that statement a bit, or may be not.
Best regards
Erik
Erik,
With a due respect, but I believe that you (and many others) overstate the issue of diffraction losses at small apertures, especially with medium format.
Somewhat frustrated with this debate, after breakfast this morning I did a quick test, just to post here.
(http://www.luminous-landscape.com/articleImages/MR64/_MHR0263.jpg)
This is the set up, taken with the Pentax 645z and Pentax 120mm Macro, a lens of undisputed high resolution.
I did a series of exposures, keeping the ISO the same and simply varying the shutter speed relative to the aperture. The range was f/5.6, f/8, f/11, f/16, f/22, and f/32.
(http://www.luminous-landscape.com/articleImages/MR64/f56.jpg)
F/5.6
(http://www.luminous-landscape.com/articleImages/MR64/f22.jpg)
f/22
As I suspected, f/5.6 and f/8 were indistinguishable. F/11 was minutely less crisp and f/22 a bit more so. f/32 was noticeably softer.
The point though, is that a small increase in the amount of sharpening used is all that's need to bring the f/22 shot to the point that it's almost indistinguishable at 100% on screen, and completely indistinguishable in a 16X20" print.
So, while I have no problem with theory, I am much more interested in practice, and as I've written for years, make your own tests before believe so-called experts (including me).
The secret to making the f/22 shot look almost the same as the one at f/5.6 was simply a bit of extra sharpening. Nothing more.
By the way, I wrote about this back in 2002, in the film days, and the same thing applied then. http://www.luminous-landscape.com/essays/stop-d.shtml (http://www.luminous-landscape.com/essays/stop-d.shtml). Stop worrying and keep shoot!
Michael
UPDATE: My apologies. I mistakenly posted the f/5.6 shot twice. The correct f/22 shot is now there. Same difference.
-
Hi,
My solid opinion, being analytic, is that the better lenses perform best at f/8 or below. Stopping down to f/16 on my P45+ largely eliminates colour aliasing, a good thing, but also a sure indication of sharpness loss.
My take may also be that if we buy MFD for sharpness, that advantage will be reduced/lost with stopping down to much, so I am reluctant to go below f/11 - f/16.
(http://echophoto.dnsalias.net/ekr/Articles/MFDJourney/Samples/Diffraction/MTF_of_aperture.png)
The example above is a Sonnar 150/4 on Hasselblad 555/P45+, near axis, but laws of physics are same for all lenses and all sensors.
In this case stopping down to f/16 corresponds to loosing like 16 MPixels of 52, 36 MP resolution remaining.
But, some lenses need to be stopped down to f/11 for optimum sharpness. I sometimes take a few pictures with different focus and blend, that often works reasonable well.
Best regards
Erik
Erik,
My results with the Nikon D800e with the Zeiss 135 mm/f2 ApoSonnar confirm your observations. Here are my results using Bart's test target (http://www.openphotographyforums.com/forums/showthread.php?t=13217). For those not familiar with the target, the 92 pixel circle indicates the Nyquist limit. The images were rendered by ACR and no sharpening was used.
Maximum resolution is at f/4 as determined by Imatest analysis of the slanted edges of the target. Resolution extends to the Nyquist limit and alaising is clearly present.
(http://bjanes.smugmug.com/Photography/Zeiss-135/i-3N7B4FK/0/O/f_4.png)
At f/16, the resolution extends nearly to Nyquist and alaising is nearly gone. Deconvolution can restore some lost detail.
(http://bjanes.smugmug.com/Photography/Zeiss-135/i-2njjtMW/0/O/f_16.png)
At f/22 there is severe loss of resolution and there is nothing to recover near Nyquist.
(http://bjanes.smugmug.com/Photography/Zeiss-135/i-zdB6NX6/0/O/f_22.png)
Here are the Imatest results:
(http://bjanes.smugmug.com/Photography/Zeiss-135/i-Qzk4CCg/0/O/GraphicalResults.png)
Bill
-
So there we have it. The experts on both sides have spoken.
Now people just have to decide if they want to do real-world photography or test chart photography.
Michael
-
Hi,
Could it be that that the guys doing chart photography can do real world photography on a shoestring just by making the best use of the equipment they choose to afford?
An interesting question, sort off…
Best regards
Erik
So there we have it. The experts on both sides have spoken.
Now people just have to decide if they want to do real-world photography or test chart photography.
Michael
-
Of course. And sorry if I seemed snarky. I didn't mean to.
I just find that fixating on that last 2% of IQ is a distraction. But if it's what turns ones crank, then who am I to nay say?
Michael
-
Another quote from Michaels Phlog shooting with the 645Z is the following:
The ISO just was what ended up being needed, and as I've discovered with the Pentax 645z, anything up to and including ISO 3200 shows hardly any noise or reduction in dynamic range.
There is always a (significant) reduction in dynamic range when cranking up the ISO. Just look at DxO graphs for various sensors.
-
Hi,
Now, I don't feel it that way.
In a way, I always think about your discussion with Ctein, he often used the term 'good enough'. You may ask him what he means, but he felt that 16 MP on his 4/3 was good enough for A2. I don't know what he means by good enough.
Another way to put it, I don't think small prints, like A2 are large enough to show a difference between well processed 4/3, APS-C, FF 135 or MFD. I have never used 4/3, but I do have experience with APS-C, FF (135) and P45+. I have little doubt that the latest and greatest MFDs can outperform what I have, but I don't know if they outperform the lower resolution stuff on real images, like a flower shot if printed at moderate size (like A2).
With the cost differential of MFD I may feel that making the best use of the 135 platforms may be a great way of saving money, compared to buying MFD stuff at 3-5X times the cost and not utilising it fully.
It is a little bit like the A7r. It seems to have a measurable loss of sharpness due to vibration, but it has not been observed by many authors. So you can waste something like 20-40% of image quality, without noticing. Probably has to with "good enough" and vision limits.
Best regards
Erik
Of course. And sorry if I seemed snarky. I didn't mean to.
I just find that fixating on that last 2% of IQ is a distraction. But if it's what turns ones crank, then who am I to nay say?
Michael
-
Hmmm… If I feel like printing 645D or A7r photos postcard size I'll darn well do it. And if that makes the pixel police squirm I'll stop down to f/32 and max out the ISO for good measure.
::)
-Dave-
-
Another quote from Michaels Phlog shooting with the 645Z is the following:
The ISO just was what ended up being needed, and as I've discovered with the Pentax 645z, anything up to and including ISO 3200 shows hardly any noise or reduction in dynamic range.
There is always a (significant) reduction in dynamic range when cranking up the ISO. Just look at DxO graphs for various sensors.
Arggggggg.
OK. I give up. The pixel police win.
Excuse me, I have to go out and do some photography.
Michael
<
-
Hi,
Message from squad car #2: I have been on a workshop with Hans and he shoots great images. But he was at Hewlet Packard, the company formerly known as Tandem Computers, so he knows a bit about bits, too…
Best regards
Erik
Arggggggg.
OK. I give up. The pixel police win.
Excuse me, I have to go out and do some photography.
Michael
<
-
Message from squad car #2: I have been on a workshop with Hans and he shoots great images.
Agree. I love the pictures on his website and might take a workshop with him sometime.
But sometimes his interpretation of DXO graphs overlooks some critical details in the fine print (http://www.luminous-landscape.com/forum/index.php?topic=92959.msg757749#msg757749), but hey even the police makes mistakes sometimes ;)
-
Arggggggg.
OK. I give up. The pixel police win.
Excuse me, I have to go out and do some photography.
Michael
<
I think it is key and important to distinguish between facts and what matters to you. I do that distinction myself and I will divert from the facts when I feel it makes sense. So what I wanted to say is: If a factual incorrect statement is made it is better to admit it than try to work around it. What matters in real photography is anyway much more complex and not just a matter of MTF curves etc. So may advice to you, if you can appreciate it, is to clearly distinguish. You do fall into that trap yourself rather often, so please don't come and call me the pixel police ;)
-
Agree. I love the pictures on his website and might take a workshop with him sometime.
But sometimes his interpretation of DXO graphs overlooks some critical details in the fine print (http://www.luminous-landscape.com/forum/index.php?topic=92959.msg757749#msg757749), but hey even the police makes mistakes sometimes ;)
I'm sorry I didn't come back to take up that thread ;)
-
Hi,
Message from squad car #2: I have been on a workshop with Hans and he shoots great images. But he was at Hewlet Packard, the company formerly known as Tandem Computers, so he knows a bit about bits, too…
Best regards
Erik
Thanks Erik, I have a feeling it's not about being right or wrong. Critique is not welcome it seems. You know I'm pragmatic and do not always do myself what I preach, but that does not mean that one should not appreciate what is factual correct and what is not. We are all humans and make mistakes, but when corrected one should be grateful.
-
Agree. I love the pictures on his website and might take a workshop with him sometime.
But sometimes his interpretation of DXO graphs overlooks some critical details in the fine print (http://www.luminous-landscape.com/forum/index.php?topic=92959.msg757749#msg757749), but hey even the police makes mistakes sometimes ;)
I'm not so sure about that. See my response. But surely you would be welcome any time on one of my workshops :)
-
Ok, I'm sorry if I offended anyone. The "police" comment was intended as humour.
I have respect for science, I have respect for facts.
I have greater respect for achieving an image that achieves what I want it to creatively in spite of breaking the rules. The rules and facts need to be understood, then they can be ignored as one wishes.
Michael
-
There is no denying that stopping down to apertures as small as F16 and F22 will reduce resolution at the precise point of focus in the image, and this is very easily demonstrated by shooting 2-dimensional scenes such as banknotes pasted to a wall.
However, most scenes we shoot are 3-dimensional. If the scene has great depth, and there are no parts of the composition we wish to de-emphasize by making them blurry, then stopping down can result in a sharper image. The image as a whole can be sharper. This is why I sometimes stop down to F16. I'm obsessed with resolution (to a degree) and find that F16 sometimes produces a sharper image, on balance, considering the whole image.
In other words, the parts of the scene which are furthest away from the precise point of focus, can be very significantly sharper at F16 than they would be at F5.6. In fact, such parts will be sharper at F16 and F22, than at F5.6, to a far greater degree than the precise point of focus would be sharper at F5.6, than it would be at F16 and F22.
Now, if I were even more obsessed with resolution than I already am, I would always try to use the sharpest aperture of my lens and employ focus stacking instead of stopping down. But alas, I'm not that obsessed. ;)
However, I should add that my own testing with 35mm format at F22 has disappointed. The fall-off in resolution at the point of focus, at F22, compared with F16, seems quite noticeable. Whereas the difference between F11 and F16, at the point of focus, is very slight. I imagine with MF format, F32 would be the equivalent aperture where resolution seems to take a dive.
-
Erik,
With a due respect, but I believe that you (and many others) overstate the issue of diffraction losses at small apertures, especially with medium format.
As I suspected, f/5.6 and f/8 were indistinguishable. F/11 was minutely less crisp and f/22 a bit more so. f/32 was noticeably softer.
The point though, is that a small increase in the amount of sharpening used is all that's need to bring the f/22 shot to the point that it's almost indistinguishable at 100% on screen, and completely indistinguishable in a 16X20" print.
So, while I have no problem with theory, I am much more interested in practice, and as I've written for years, make your own tests before believe so-called experts (including me).
The secret to making the f/22 shot look almost the same as the one at f/5.6 was simply a bit of extra sharpening. Nothing more.
Michael,
Your test images are shown at a very small scale, and it is impossible to evaluate contrast at high spatial frequencies where even deconvolution sharpening could not recover lost detail. For example, here is my test image of Bart's chart taken at f/4. The central blur diameter is about 104 pixels, which corresponds to a resolution of 90 cy/mm.
(http://bjanes.smugmug.com/Photography/Zeiss-135/i-3N7B4FK/0/O/f_4.png)
And here is the image at f/22. The blur diameter is about 129 pixels, corresponding to a resolution of 73 cy/mm. In addition, there is a large loss of contrast at even the low frequencies. The difference from the f/4 image is quite apparent at even a glance.
(http://bjanes.smugmug.com/Photography/Zeiss-135/i-zdB6NX6/0/O/f_22.png)
After deconvolution sharpening with Focus Magic using a radius of 2 pixels and amount of 150%, macro contrast is greatly improved but there is no change in resolution. The blur diameter is unchanged at 129 pixels. As Bart pointed out in his prior post, at frequencies near Nyquist, there is nothing to recover. Micro contrast is lost. The extent to which this would affect a landscape image would depend on how many fine details are in the image.
(http://bjanes.smugmug.com/Photography/Zeiss-135/i-r74Tdw3/1/O/f_22_Fm.png).
You could download Bart's chart and repeat the test. The nice thing about Bart's method is that, within limits, the distance from the lens to the chart does not affect the results. The optimal distance is 25-50 times the focal length of the lens.
Bill
-
Ok, I'm sorry if I offended anyone. The "police" comment was intended as humour.
I have respect for science, I have respect for facts.
I have greater respect for achieving an image that achieves what I want it to creatively in spite of breaking the rules. The rules and facts need to be understood, then they can be ignored as one wishes.
Michael
The humour didn't resonate with me unfortunately. Continous insisting that you are right made that rather difficult.
You respect facts, so which facts that have been discussed in this thread do you agree on?
-
The humour didn't resonate with me unfortunately.
[/sigh]
Humour is often an excellent antidote to excessive pernickety fastidiousness.
Michael was being gracious.
You should try it some time.
-
Have a look at Figure 3 in the excellent article (http://www.luminous-landscape.com/tutorials/resolution.shtml) by Rubén Osuna and Efraín García posted on LuLa.
Thanks Bill!
-
I have greater respect for achieving an image that achieves what I want it to creatively in spite of breaking the rules.
Exactly.
-
Ok, I'm sorry if I offended anyone. The "police" comment was intended as humour.
Hi Michael,
No problem. I do understand your reaction, which even triggered your response to double check what it was that you were seeing different from what physics tells us. One might ask why your example still looks reasonably good?
Well, there are a few things that can help to understand why your images still look acceptable, especially with a bit of increased sharpening.
Physics tells us (http://spie.org/x34304.xml) that for monochromatic green light, say with a wavelength of 555nm, a circular f/22 aperture will have a MTF=0% cut-off frequency at 1/(22 * 0.000555) = 81.9 cycles/mm. The physical (Nyquist) limit for the 5.3 micron sensel pitch of the 645Z sensor is at 1 / (2 * 0.0053) = 94.3 cycles/mm. So at 87% of Nyquist and higher spatial frequencies there will be 0% MTF response, that's what physics dictates. Depending on the chromatic corrections of the lens, red will blur more, and blue will blur less, but the majority of Luminance contribution comes from Green, and Luminance dominates our senses to detect sharpness.
This loss of resolution is of course mostly affecting the details in the focus plane, other detail will already be somewhat blurred, increasingly more as one gets further to the edges of the acceptable DOF zone. It also becomes more destructive for lower contrast micro-detail which vanishes beyond repair. Your high contrast bars and text input will still look relatively unaffected, because most of it is larger than the cut-off frequency (which is helpful for sharpening, but won't restore what is lost).
So Ray is correct in pointing out that 3D subjects appear to suffer less (because they are already blurred), but they do also suffer. And because the highest resolution detail is lost, it kind of blends in with the already blurred OOF detail, at a more comparable (lower) resolution level.
But as Bill showed, when we want to analyze the amount of loss, a better target should be used. Not to negate what we see in real life scenes, but to determine the best trade-off between our artistic DOF requirements, and loss of quality. Nothing wrong with making a trade-off that hurts technical quality, if it improves the overall image, as long as we make it a conscious trade-off. The image and its intended use should dictate our choices.
As I said before, for the 645Z, f/16 is about the narrowest aperture you can get away with without compromising technical quality too much. It will have lower resolution (especially low contrast structures with micro-detail) than at the lens' optimum resolution aperture, but in the balance of things there is an acceptable loss when it improves the image creatively. Hopefully the viewfinder/live view will allow you to judge that correctly, and I assume the camera offers a stopped down preview option.
Starting at f/18, you will start destroying detail (in the plane of best focus) beyond recovery, so this is where the better creative DOF must outweigh the technical loss of resolution/realism. Again, if the situation provides a creatively better image, by all means go for it, but do realize that it won't enlarge as crisp as wider aperture shots, and it will need a lot more advanced post-processing help to (realistically) fake that impression of sharpness.
I have respect for science, I have respect for facts.
I have greater respect for achieving an image that achieves what I want it to creatively in spite of breaking the rules. The rules and facts need to be understood, then they can be ignored as one wishes.
Agreed, although 'ignored' is a bit strong. Understanding the trade-offs as dictated by science helps us to make better informed judgement calls, especially if we have some wiggle room (maybe f/18 with better technical quality is also acceptable instead of f/22). Sometimes we need to accept lower resolution, if the image (DOF) dictates it (which assumes an informed technical vs creative trade-off).
Cheers,
Bart
P.S. The 81.9 cycles/mm at f/22, also roughly suggests a potential maximum enlargement factor of 16x if we want to achieve a 5 lp/mm printed resolution, thus the 43.8 x 32.8 mm sensor would produce 717 x 537mm (28 x 21 inch) output with the required resolution for uncompromised close inspection, while f/18 would allow 15% larger output with the same resolution in the plane of best focus.
-
What this entire conversation appears to miss, is that the reason we are stopping down is not to push our little photons through tiny, tight little holes, but because the subject of the photograph will otherwise not be in focus.
If key components of the subject are not sharp, the photograph is often pointless in landscape work. The selection of aperture is not a function of anything other than a desire to actually create the desired image. What resolution is obtained is a consequence of that choice, and irrelevant if the image is not satisfactorily captured at all.
- N.
-
What this entire conversation appears to miss, is that the reason we are stopping down is not to push our little photons through tiny, tight little holes, but because the subject of the photograph will otherwise not be in focus.
If key components of the subject are not sharp, the photograph is often pointless in landscape work. The selection of aperture is not a function of anything other than a desire to actually create the desired image. What resolution is obtained is a consequence of that choice, and irrelevant if the image is not satisfactorily captured at all.
- N.
For me at least this is implicit in the discussion. To get a large enough DOF can be achieved (as you know) in different means, but the cost of stopping down to f/22 was under debate and I think by now all should be in agreement that the cost is considerable in terms detail rendering. There are many things that can be said about how to avoid f/22. My personal approach is to use live view in such situations to get the optimal balance of where DOF is placed. I would typically shoot this on a tripod and I would also bracket the aperture and then judge in post processing which aperture with proper sharpening settings is the most pleasing picture. I seldomly use focus stacking. I have found that I can usually do my compositions in such a way that f/16 will be enough with the careful adjustment of focus. My default aperture is f/8 and f/11 for landscapes. I have used tilt and shift lenses but hardly do this any more.
If the details are rendered as mush I don't think is is irrelevant if the aperture was f/32 and the resolution was reduced a lot. In such a case one would be much better off either choosing a slightly different composition not needing f/32 or use focus stacking and manual blending if need be. In any case this is the decision of the artist and in my view this is not the discussion. The discussion is about facts.
-
What more needs to be said? In my world, reality trumps theory.
(http://www.luminous-landscape.com/articleImages/MR64/tomatoes.jpg)
In other words, I do what it takes to get the shot that I want and don't worry about not doing the theoretically correct thing.
Michael
-
I'd like to elaborate on the above....
If one is photographing 2 dimensional test charts, then of course f/5.6 would produce superior results. This would be the case for "flatlanders as well"
But in the three dimensional world that we inhabit DOF is a real concern, especially when shooting with larger format sensors. So, like everything in photography one must make trade off decisions. I choose DOF over minimizing diffraction. Why? Because it produces what are to my eye aesthetically superior images.
There really is no debate.
Michael
-
Reading the review and Michael's comments in this thread, I found the following statements from him concerning DR and diffraction.
From this thread:
".......................I have tested the 645z extensively in this regard and for my purposes diffraction doesn't start to show so that it's a concern until f/22. Much of my shooting is at f/16 and I am completely satisfied with the results. If I was super anal retentive I'd shoot at f/8 or f/11, but I'm not. :-)
Michael"
From the review:
".......................... If the ISO hits 12,800 a bit of noise reduction is needed, but otherwise there's hardly any concern about noise, and dynamic range holds quite well up to about ISO 6,400."
I don't see a denial of the effects of diffraction or reduced DR with high ISO in these statements, but rather that he finds the images acceptable despite those limitations. I do see in the posts of this thread a rather rude insistence that Michael, by being satisfied using high ISO and f/16, is denying these effects.
Tom
-
What more needs to be said? In my world, reality trumps theory.
In other words, I do what it takes to get the shot that I want and don't worry about not doing the theoretically correct thing.
Michael
Theory does encompass your macro situation, where image detail is determined by the opposing effects of diffraction and defocus. Both of these effects are well characterized by theory. Interested readers can refer to this article (http://en.wikipedia.org/wiki/Defocus_aberration) on Wikipedia. In a macro situation, depth field is very shallow and defocus is the limiting factor. In many landscape situations, diffraction would be the limiting factor. This assumes that other aberrations are well controlled. If you repeated your test in a landscape situation where there are no significant foreground elements, the results would be different. In both cases the laws of physics are not violated.
Bill
-
I do see in the posts of this thread a rather rude insistence that Michael, by being satisfied using high ISO and f/16, is denying these effects.
Hi Tom,
That's not what the comments were about, IMHO.
In principle, pushing an underexposed ISO 1600 exposure by 3 stops in post, should under normal conditions give better image quality (lower noise, more highlight headroom, better Dynamic Range) compared to an ISO 12800 shot. If this turns out differently with the 645Z because of High ISO noise reduction before writing the Raw data, then that's very interesting. Maybe it's even the start of a new trend.
Also, On a sensor with a 5.3 micron sensel pitch f/16 offers a much more solid basis for (capture) sharpening than f/22. Maybe this Raw data is also cooked, already somewhat diffraction 'corrected', which would make the differences smaller. If that's so, then it's another interesting observation by Michael if he picked that up without laboratory testing.
Also not challenged is Michael's creative choice for f/22 instead of something else, although the creative difference between f/16 (or f/18) and f/22 is relatively small for non-macro shots, especially in a viewfinder or Live View. So maybe in the longer run he will develop a preference for f/18 if a lot of DOF is required and large output is the goal? Focus stacking at the optimum aperture is also an option if technical quality is high on the list of choices.
Frankly, I see no rudeness, just honest concerns about how some people could potentially blindly follow Michael's ISO 12800 and f/22 remarks, without knowing Michael's creative options that led him to choose such settings, and being unaware of the trade-offs involved. No rudeness, but an exchange of info, an attempt to understand.
I for one, am intrigued by some of Michael's observations and look forward to someone doing a more technical test that will reveal what's going on under the hood of the 645Z. I'm sure Hasselblad and Phase One would also like to know.
Cheers,
Bart
-
Hi,
I don't think that tone in these postings were bad.
Regarding the ISO thing I feel that it is good that we have a high ISO capable sensor that still leaves reasonable DR when 6 stops underexposed, weather or not it is achieved with manipulation of raw data.
Regarding diffraction I am pretty sure that the stuff Bart, Bill, Hans and I discuss is for real, but it is also possible that stopping down to far is not the ethernal sin we may pretend. Tim Parkin has posted an article discussing this and indicating that f/22 is still a decent aperture, when used with adequate sharpening.
On the other hand, I am in doubt if a Pentax 645Z used at f/22 and 12800 ISO will yield superior results to an A7r ord D810 at f/16 and ISO 6400. So shooting small aperture and high ISO may negate the benefits of the larger format.
Best regards
Erik
Ps. Tim's article is here: http://www.onlandscape.co.uk/2012/07/diffraction-limited/
He is quite critical of this article published here: http://www.luminous-landscape.com/pages.php/tutorials/resolution.shtml
Reading the review and Michael's comments in this thread, I found the following statements from him concerning DR and diffraction.
From this thread:
".......................I have tested the 645z extensively in this regard and for my purposes diffraction doesn't start to show so that it's a concern until f/22. Much of my shooting is at f/16 and I am completely satisfied with the results. If I was super anal retentive I'd shoot at f/8 or f/11, but I'm not. :-)
Michael"
From the review:
".......................... If the ISO hits 12,800 a bit of noise reduction is needed, but otherwise there's hardly any concern about noise, and dynamic range holds quite well up to about ISO 6,400."
I don't see a denial of the effects of diffraction or reduced DR with high ISO in these statements, but rather that he finds the images acceptable despite those limitations. I do see in the posts of this thread a rather rude insistence that Michael, by being satisfied using high ISO and f/16, is denying these effects.
Tom
-
Sorry Bart* I don't agree. A polite, informative discussion concerning whether the use of high apertures and ISO on a camera like the 645Z is wise or if it reduces the output to a point and shoot could have occurred, instead Michael is accused of denying facts which, as far I can tell, he has not.
I am not referring to all the responses, but the ones that were confrontational in nature.
Tom
* and Erik, I didn't see your post.
-
I opened this thread by saying it was a good review but there were a few things that were not (factually) correct:
Larger Sensor
Some people may quibble about 51MP not being that much larger than 36MP. Well, they're off the mark, because the combination of larger sensels and more sensor surface real estate does add up to an appreciable advantage over 36MP. This size sensor is about 1.7X larger than full frame, which is an even greater ratio increase over APS-C's 1.5X or 1.6X difference from full frame 35mm.
This is incorrect as the sensor area is 1.7x larger than 35mm FF, but the difference between 35mm FF and APS-C, which is for Canon 2.56 and for Nikon 2.25. The difference is more like the difference between 35mm FF and APS-H (which is 1.7x).
The ISO just was what ended up being needed, and as I've discovered with the Pentax 645z, anything up to and including ISO 3200 shows hardly any noise or reduction in dynamic range. from Michaels Phlog shooting the 645Z.
The noise is well suppressed in what I see shot at ISO 3200, but the dynamic range is reduced significantly compared to shooting at ISO 100. Since there are no DxO data for the 645Z yet, looking at the D800 or D810, at ISO 100 (print) the DR is just under 15 stops, at ISO 3200 the DR is 9.5 stops. This is good, but decreased about 5 stops as expected from ISO 100.
Though f/22 seems like it would be into diffraction, in fact I have found that with medium format (even the smaller size of the Pentax sensor) it really isn't apparent even at f/22.
and again from Michaels Phlog.
This has been discussed during this thread and no need to repeat it here.
If somebody finds it insulting or rude to point out these flaws, well then, what can I say? Certainly Michael has not bowed an inch ;)
But it does point out one fact: There is no point in commenting on reviews and other postings by Michael if there are factual errors to report. So I will not waste my time in the future to do that.
-
Hi,
I think he bowed an imperial quibble, but probably not an imperial notch…
Just a point to make, Michael's reviews are subjective, he never said anything else. I have some respect for that, but it may not be the best base for purchase decisions. But I generally enjoy Michaels writings.
My own experience with my P45+ is closer to engineering stuff, like DxO-mark and Hasselblads own MTF data. A lot of myths and little reality. But I enjoy shooting with the stuff, and it can produce excellent results, at f/11 with the stars in proper alignment.
Best regards
Erik
If somebody finds it insulting or rude to point out these flaws, well then, what can I say? Certainly Michael has not bowed an inch ;)
-
Good Lord....... The sRGB / Adobe RGB color management thread must be contagious. ;D
We need some kind of cyber-disinfectant!
Rand
-
After some 25 years online (remember CompuServe?), all I can say is that it doesn't matter what you say or write, it's what people think you've said or written that is their reality.
The end.
Michael
-
After some 25 years online (remember CompuServe?), all I can say is that it doesn't matter what you say or write, it's what people think you've said or written that is their reality.
The end.
Michael
Quite right, Michael. And in my experience, what they "think" you said is a horse that is already in their barn saddled up and ready to ride. :-)
Rand
-
On the other hand, I am in doubt if a Pentax 645Z used at f/22 and 12800 ISO will yield superior results to an A7r ord D810 at f/16 and ISO 6400. So shooting small aperture and high ISO may negate the benefits of the larger format.
Hi Erik,
That's probably true. It would be interesting to see a comparison. However, hasn't it always been the case that the larger format tends to have an advantage mostly at a shallower DoF?
This also applies to the differences between the cropped format and full-frame 35mm. To get the equivalent DoF with full frame one has to stop down at least one stop. To get the equivalent shutter speed with full frame one has to raise ISO at least one stop. This effect tends to negate the advantages of full frame in terms of lower noise and higher resolution. It's usually unavoidable without employing focus stacking, except in specific circumstances when the desired DoF is so shallow that one has to use a very wide aperture with the cropped format lens, such as its maximum aperture of F1.4 or F2.8 which might be sub-optimal in terms of resolution.
In that situation, stopping down with the larger format might enable one to use a 'sharper' or optimal aperture, and if shutter speed is not a problem, there would be no need to raise ISO, in which case the larger format produces both higher resolution and lower noise.
One doesn't expect any camera to produce technically better images in all circumstances, unless one is comparing new technology with old technology. The 'best tool for the job' still applies. The main concern should be that the Pentax 645Z does not produce worse images at F22 than a 36mp D800E at F16, assuming general lens quality is comparable.
-
Another quote from Michaels Phlog shooting with the 645Z is the following:
The ISO just was what ended up being needed, and as I've discovered with the Pentax 645z, anything up to and including ISO 3200 shows hardly any noise or reduction in dynamic range.
There is always a (significant) reduction in dynamic range when cranking up the ISO. Just look at DxO graphs for various sensors.
Quite true. For a so called ISO-less camera, one loses 1 stop of DR for each doubling of ISO. If the base ISO of the 645Z is 100 then ISO 3200 is five doublings or 5 stops less dynamic range. Quite a significant hit. This can be seen with the Nikon D800e and other cameras using the most advanced Sony sensors. Bill Claff (http://home.comcast.net/~nikond70/Charts/PDR.htm) has extensive DR data on many cameras and for the D800e, he reports the photographic DR at ISO 100 as 10.61 EV and the DR at ISO 3200 as 5.84 EV. The 645z has a larger and more recent Sony sensor and would likely have about 1 EV more DR than the D800e, but a similar trend would be expected. We will see when DXO tests the sensor.
Bill
-
Hi,
Yes, that is what I expect.
On the other hand what Michael says is that DR at 3200 ISO was good enough.
When I was shooting Velvia I hade less than 5EV of DR, so I can see reason for that statement.
Best regards
Erik
Quite true. For a so called ISO-less camera, one loses 1 stop of DR for each doubling of ISO. If the base ISO of the 645Z is 100 then ISO 3200 is five doublings or 5 stops less dynamic range. Quite a significant hit. This can be seen with the Nikon D800e and other cameras using the most advanced Sony sensors. Bill Claff (http://home.comcast.net/~nikond70/Charts/PDR.htm) has extensive DR data on many cameras and for the D800e, he reports the photographic DR at ISO 100 as 10.61 EV and the DR at ISO 3200 as 5.84 EV. The 645z has a larger and more recent Sony sensor and would likely have about 1 EV more DR than the D800e, but a similar trend would be expected. We will see when DXO tests the sensor.
Bill
-
Hi,
Yes, that is what I expect.
On the other hand what Michael says is that DR at 3200 ISO was good enough.
When I was shooting Velvia I hade less than 5EV of DR, so I can see reason for that statement.
Best regards
Erik
Erik, this maybe what Michael thought he wrote but the actual text was as I quoted which is different (that there was hardly any reduction of DR). I cannot relate to what people think they wrote when they actually write somthing different. How can one discuss on that basis?
I have been writing online and in business for 25 years and this is not new at all, but I still find discussions where people try to change the subject when proven wrong is waste of time and I believe this will never change since this is so human.
-
One doesn't expect any camera to produce technically better images in all circumstances, unless one is comparing new technology with old technology. The 'best tool for the job' still applies. The main concern should be that the Pentax 645Z does not produce worse images at F22 than a 36mp D800E at F16, assuming general lens quality is comparable.
That's correct, although I wouldn't call it a concern. While f/16 will allow the D800E/D810 (in principle, disregarding the lens) to resolve all the way to it's Nyquist frequency, despite diffraction. The fact that for larger output the resolution must be divided by the additional magnification requirement (divide by 32.8/24) for equal size output, will pretty much equalize the playing field.
The 645Z does have the advantage of a larger image magnification on sensor from a longer focal length. Larger magnification will boost the MTF for same sized subject detail. Add to that some Pentax secret sauce by writing somewhat 'pre-cooked' data to Raw, and one could anticipate somewhat better image quality for the 645Z. How much better, remains to be seen in a side-by-side shoot-off.
Cheers,
Bart
-
Quite true. For a so called ISO-less camera, one loses 1 stop of DR for each doubling of ISO. If the base ISO of the 645Z is 100 then ISO 3200 is five doublings or 5 stops less dynamic range. Quite a significant hit. This can be seen with the Nikon D800e and other cameras using the most advanced Sony sensors. Bill Claff (http://home.comcast.net/~nikond70/Charts/PDR.htm) has extensive DR data on many cameras and for the D800e, he reports the photographic DR at ISO 100 as 10.61 EV and the DR at ISO 3200 as 5.84 EV. The 645z has a larger and more recent Sony sensor and would likely have about 1 EV more DR than the D800e, but a similar trend would be expected. We will see when DXO tests the sensor.
Hi Bill,
That's correct, although I do have a bit of a problem with Bill Claff's assumed acceptable shadow noise limit for the calculation of DR. The reason becomes clear with the Pentax 645Z for which we already know (from Phase One IQ250 samples) that it will probably best the engineering DR of the sensors used in the Nikon D800 series, by a small amount (maybe 14 stops in total). However, we do not know what the influence of noise reduction will be, which may be happening with the Pentax before Writing Raw, or by the use of dedicated software. I prefer to set my own lower limit of acceptable noise, after noise reduction. That's why I prefer pure Engineering DR numbers as an upper limit goal.
Michael reports very clean shadows at high ISO settings. That might add some low exposure region to the effective Dynamic Range that Bill Claff's previous measurements on other sensors would extrapolate to.
Cheers,
Bart
-
Regarding diffraction I am pretty sure that the stuff Bart, Bill, Hans and I discuss is for real, but it is also possible that stopping down to far is not the ethernal sin we may pretend. Tim Parkin has posted an article discussing this and indicating that f/22 is still a decent aperture, when used with adequate sharpening.
On the other hand, I am in doubt if a Pentax 645Z used at f/22 and 12800 ISO will yield superior results to an A7r ord D810 at f/16 and ISO 6400. So shooting small aperture and high ISO may negate the benefits of the larger format.
Best regards
Erik
Ps. Tim's article is here: http://www.onlandscape.co.uk/2012/07/diffraction-limited/
He is quite critical of this article published here: http://www.luminous-landscape.com/pages.php/tutorials/resolution.shtml
I do quote the tables from the article in LuLa in my workshops in order to say that there is a (severe) cost in fine details stopping down too much. So what I say is that although these numbers should not be taken as exact since they are dependent on the calculation parameters, it is is easy to see in test shots what the effect is. So I mention Lightroom sharpening to compensate for f/16 and beyond and also that the best thing to do is to take test shots with the given camera and lenses to see what is acceptable since every person has a different threshold for what is usable and acceptable (also related to print size). I also mention that often it is not needed to stop down as much if careful focusing adjustment is done using live view and possibly a slight composition change that still achieves the desired image but changes the DOF requirements just slightly. But if some feel f/22 is fine for them and it makes them work faster and don't like the live view adjustments then that is fine too. The key is to understand the consequence of the compromises. On 35mm I will stop down to f/16 when needed and change my standard sharpening parameters in Lightroom by increasing radius slightly and increase detail to 100. This helps to make the picture regain some of the crispness that it has lost at f/16 compared to e.g. f/8. For me it is not only about resolution but also about the look.
I do subscribe to the onlandscape monthly publications but have not seen this subject come up again. I do see many landscape photographers use f/22.
High resolution sensors on medium format does come at a collision with DOF in e.g. landscape shooting. I was instructor on two Phase One workshops last year and Phase One personel was very honest about this towards the participants. Basically saying that with large DOF the high resolution of the sensor could not be achieved, so if one wanted to have the full resolution in such shots then focus stacking was the recommended approach. I found this very good so that one can make the needed compromises with knowledge. I have shot many pictures at f/8 with the Phase One IQ160 I was using for a while and with very good resolution in all parts of the image. In some shots where I did not do aperture bracketing I found a lack of sharpness looking at 100% but downsized the picture was quite ok. The final is therefore: We all make mistakes even when knowing the facts ;)
-
Aloha,
I tend to shot the 645 35mm FA on my 645D
At f11-f16 the files are so rich, even when
Slightly off there seems to be tons of detail and sharpness.
I believe these are the apertures that lens is optimized for.
I could be mistaken though.
Paying a little bit/or a lot of attention to what is important
In the composition goes a long way to getting desired result.
Diffraction is an issue real world with the 35mm I can easily see it.
For what I do, I tend to avoid stopping down past f16 but sometimes I will click a few shots really stopped down just to compare the results later.
I guess focus stacking is an option if you are not shooting dynamic scenes.
Michael have you had the opportunity to shoot in bulb, with exposures over thirty seconds?
The 645D does a black screen noise reduction for shots over thrifty seconds which is pretty neat.
I guess in the Z you can turn that option on or off.
Also which cpl filters do you prefer for the camera?
-
Hi,
I don't think that tone in these postings were bad.
Regarding the ISO thing I feel that it is good that we have a high ISO capable sensor that still leaves reasonable DR when 6 stops underexposed, weather or not it is achieved with manipulation of raw data.
Regarding diffraction I am pretty sure that the stuff Bart, Bill, Hans and I discuss is for real, but it is also possible that stopping down to far is not the ethernal sin we may pretend. Tim Parkin has posted an article discussing this and indicating that f/22 is still a decent aperture, when used with adequate sharpening.
On the other hand, I am in doubt if a Pentax 645Z used at f/22 and 12800 ISO will yield superior results to an A7r ord D810 at f/16 and ISO 6400. So shooting small aperture and high ISO may negate the benefits of the larger format.
Best regards
Erik
Ps. Tim's article is here: http://www.onlandscape.co.uk/2012/07/diffraction-limited/
He is quite critical of this article published here: http://www.luminous-landscape.com/pages.php/tutorials/resolution.shtml
Erik,
I did take a look at Tim's article and found it provocative in a helpful way. He maintains that humans can resolve 117 cy/mm (lp/mm) rather than the usually quoted 73 cy/mm as the criterion for the diffraction limit. In the old days, lens testers used the USAF line bars in determining the resolution of lenses, but more modern thinking is that one must take the contrast into account as well as the resolution figure. When interpreting lens tests reported as MTF, contrast at lower resolutions is given more weight than high resolution at low MTF. This is taken into account by SQF (subjective quality analysis). Bob Atkins (http://www.bobatkins.com/photography/technical/mtf/mtf4.html) gives a simplified explanation of SQF. He notes that human perception peaks at about 6 cy/degree, which corresponds to about 1 cy/mm on a print viewed at 34 cm (about 13.5 inches). With these facts in mind, 117 cy/mm at 2-5% contrast may not contribute that much to perceived sharpness and we should place more emphasis on resolution at 50% contrast (MTF of 50%). Indeed, Norman Koren points out that it does not make sense to measure resolution where it is extinguished (at low MTF).
Our MTF 50 results at f/22 do indicate a severe loss of resolution.
Bill
-
Erik,
I did take a look at Tim's article and found it provocative in a helpful way. He maintains that humans can resolve 117 cy/mm (lp/mm) rather than the usually quoted 73 cy/mm as the criterion for the diffraction limit. In the old days, lens testers used the USAF line bars in determining the resolution of lenses, but more modern thinking is that one must take the contrast into account as well as the resolution figure. When interpreting lens tests reported as MTF, contrast at lower resolutions is given more weight than high resolution at low MTF.
Hi Bill,
While true, one needs to recognize that these MTFs are usually measured before (!) sharpening. As illustrated by your sharpened versions from shots of my test target, the higher spatial frequencies get boosted to levels that approach those of lower spatial frequencies. The bi-tonal USAF target and similar, are un-fit (due to aliasing and ringing artifacts from sharp edges that may fall in alignment with the pixel grid, or not, i.e. phase errors) for discrete sampling systems like our sensors (but then they were designed for film, and are perfectly suitable for that).
This is taken into account by SQF (subjective quality analysis). Bob Atkins (http://www.bobatkins.com/photography/technical/mtf/mtf4.html) gives a simplified explanation of SQF. He notes that human perception peaks at about 6 cy/degree, which corresponds to about 1 cy/mm on a print viewed at 34 cm (about 13.5 inches). With these facts in mind, 117 cy/mm at 2-5% contrast may not contribute that much to perceived sharpness and we should place more emphasis on resolution at 50% contrast (MTF of 50%). Indeed, Norman Koren points out that it does not make sense to measure resolution where it is extinguished (at low MTF).
Before sharpening. This is where additional tools like Topaz Labs Detail make a difference. They additionally address these lower spatial frequencies and can apply more or less contrast (even guided by luminance of color contrast) by changing the amplitude of that detail level, assuming it is there is sufficient quality to begin with. Deconvolution sharpening of the highest spatial frequencies also boosts lower frequency quality to a certain degree.
Our MTF 50 results at f/22 do indicate a severe loss of resolution.
And the spatial frequencies for those MTF50 scores are already pretty low, especially when enlarged for output.
Perceived resolution is a complex psycho-physical process, and therefore Michael's impressions based on shots of subject matter that's familiar to him, are interesting (although hard to quantify). And he is of course right that there is more to a successful image than resolution alone. It will take some objective testing to see where the perceived resolution is coming from. Based on objective facts, it's going to be easier to make balanced decisions between DOF (how much is enough, e.g. f/16 or f/22, in some situations) and resolution (how much do we limit magnification potential by our choices).
My DOF planning tool (http://bvdwolf.home.xs4all.nl/main/foto/dofplan/dofplan.html) attempts to address several of these issues, and also reports if diffraction affects the desired COC limits. Section 4 of my Slanted Edge tool (http://bvdwolf.home.xs4all.nl/main/foto/psf/SlantedEdge.html) web page, allows to independently (does not require the prior 3 steps but will use them if available) play with different scenarios, such as that diffraction adjusted COC limit. It reports the estimated MTF50 spatial frequency for adjustable sensel pitches and allows to copy and (e.g. with Excel) plot estimated MTF curves for diffraction combined with different levels of COC blur.
Cheers,
Bart
-
My DOF planning tool (http://bvdwolf.home.xs4all.nl/main/foto/dofplan/dofplan.html) attempts to address several of these issues, and also reports if diffraction affects the desired COC limits.
Wow! Just had a look at this Bart, perhaps it might be an economically beneficial idea to make an APP of this tool?
-
Wow! Just had a look at this Bart, perhaps it might be an economically beneficial idea to make an APP of this tool?
Hi Georg,
I know, it has been suggested here (http://www.luminous-landscape.com/forum/index.php?topic=81485.0) as well. If only I had more resources ...
It's not perfect (yet) as it is, but most of it works as intended, and obviously the Pentax 645Z (and siblings) is available as a pre-coded choice (as are most modern exchangeable lens cameras and backs). Of course an App would require (and be able to) use a different GUI, but might also require a slight compromise in one or two calculations. As it is, it doesn't require a connection after loading the webpage, so one could run it on it's own browser tab in the field without connection. But it's mostly a planning tool.
Cheers,
Bart
-
As it is, it doesn't require a connection after loading the webpage, so one could run it on it's own browser tab in the field without connection.
I have it added as an icon to the home screen on both iPhone and iPad - no loss of functionality, immediately accessible. Thanks (again), Bart.
M
-
I have it added as an icon to the home screen on both iPhone and iPad - no loss of functionality, immediately accessible.
Hi M,
That's why I designed it as a single webpage. An App could use Tabs and be less verbose with info icons and a how to use page or pages. Of course other optimizations could also be made, for smaller displays.
Thanks (again), Bart.
You're welcome. Enjoy it.
Back to the 645Z ... ;)
Cheers,
Bart
-
Aloha,
I was wondering if you had a chance to shoot in bulb more,
over thirty seconds, and if you liked the in camera
Black screen noise reduction that is a feature of the longer exposures?
Also
What brand of circular polarizer do you like?
Thanks, lee
-
Aloha,
I was wondering if you had a chance to shoot in bulb more,
over thirty seconds, and if you liked the in camera
Black screen noise reduction that is a feature of the longer exposures?
Also
What brand of circular polarizer do you like?
Thanks, lee
The longest I've shot with the 645z so far is about 5 seconds. I've seen some by others at 30 minutes. Very clean.
I use B+W polarizers mostly. Also Heliopan and Hoya. Many others are good, but you get what you pay for. Good ones have brass rings and won't bind. Cheep ones are aluminium and can stick.
-
The longest I've shot with the 645z so far is about 5 seconds. I've seen some by others at 30 minutes. Very clean.
Michael, the IS250 does a mandatory and automatic dark frame subtraction on every frame. Does the 645Z do this as well? Or is this optional?
The thing that concerns me is that at handheld shutter speeds and high gain settings, I'd expect to find abundant thermal noise. I'm curious to know if the Z can do dark frame subtraction on those settings, or whether one will have to do it manually.
-
The 645z has two settings for dark frame noise reduction...
1: For high ISO. This can be turned off if you like. When set to on it activates automatically depending on thermal conditions and whether it feels like it.
2: Long exposure dark frame noise reduction can not be turned off. It has an On and an Auto position. On will always do one (not sure when it cuts in) while on Auto it will do it when the length of time, temperature and who knows what else warrants it. I have mine set to ON and Auto, but I don't have enough long exposure experience yet to really understand its behaviour.
I am playing with multiple exposure star stacking, but without any guidance, and so I limit my shots to under 5 seconds to avoid trails, and under ISO 25,000, and have not noticed any DF delays.
Michael
-
The 645z has two settings for dark frame noise reduction...
1: For high ISO. This can be turned off if you like. When set to on it activates automatically depending on thermal conditions and whether it feels like it.
2: Long exposure dark frame noise reduction can not be turned off. It has an On and an Auto position. On will always do one (not sure when it cuts in) while on Auto it will do it when the length of time, temperature and who knows what else warrants it. I have mine set to ON and Auto, but I don't have enough long exposure experience yet to really understand its behaviour.
I am playing with multiple exposure star stacking, but without any guidance, and so I limit my shots to under 5 seconds to avoid trails, and under ISO 25,000, and have not noticed any DF delays.
Thanks for the informative reply!
In the IQ-250, the camera maintains a library of recent dark frames. It may re-use or re-generate them as it deems necessary. When the camera is using a stored dark frame, you would not experience delays.
I would guess that at ISO 25600 and anywhere under 1/40th, you will see the heat signature in the bluish-magenta cast in cases where DFNR is disabled completely. I'd also guess that the noise would be much worse after extensive use of live view.
-
Thanks for the informative reply!
In the IQ-250, the camera maintains a library of recent dark frames. It may re-use or re-generate them as it deems necessary. When the camera is using a stored dark frame, you would not experience delays.
Hi Luke,
While that is efficient, it may also result in unpredictable shooting intervals when the back decides that a new darkframe is required.
I was also wondering if the Pentax could switch it off, instead of deciding for itself. Maybe if Rawconverters would offer more support for dark-frame subtraction (like RawTherapee does), the camera manufacturers would reconsider their firmware options.
A benefit of CMOS is that it doesn't require as much power as CCDs, and thus can stay cooler in long exposure/sequence scenarios. If the supporting electronics are designed well enough they will have minimal effect on the heat transfer to the sensor.
There usually is a temperature field in the EXIF info that allows to check whether temperatures vary a lot, although it's not defined where that temperature is measured, and if every manufacturer populates that field. Anyway, it's often the exposure time that leads to 'dark current' build-up, i.e. temporal noise. Overall noise levels stay relatively constant with constant temperature.
Cheers,
Bart
P.S. The articulated LCD may provide better de-coupling of heat transfer from the LCD in case of prolonged use of Live-View.
P.P.S. According to Ming Thein (http://blog.mingthein.com/2014/07/15/the-pentax-645z-review-part-iii-sdm-lenses-and-long-exposures/), the 645Z rarely inserts a black frame exposure with exposure times below 4-5 minutes.
-
I noted the dig on the CCD v CMOS argument, from a technical point as recent CCD's have not been made (least not for most cameras) one wonders how they might stack up given equal tech in the modern era.
-
Hi Luke,
While that is efficient, it may also result in unpredictable shooting intervals when the back decides that a new Blackframe is required.
I was also wondering if the Pentax could switch it off, instead of deciding for itself. Maybe if Rawconverters would offer more support for black-frame subtraction (like RawTherapee does), the camera manufacturers would reconsider their firmware options.
A benefit of CMOS is that it doesn't require as much power as CCDs, and thus can stay cooler in long exposure/sequence scenarios. If the supporting electronics are designed well enough they will have minimal effect on the heat transfer to the sensor.
There usually is a temperature field in the EXIF info that allows to check whether temperatures vary a lot, although it's not defined where that temperature is measured, and if every manufacturer populates that field. Anyway, it's often the exposure time that leads to 'dark current' build-up, i.e. temporal noise. Overall noise levels stay relatively constant with constant temperature.
Cheers,
Bart
P.S. The articulated LCD may provide better de-coupling of heat transfer from the LCD in case of prolonged use of Live-View.
P.P.S. According to Ming Thein (http://blog.mingthein.com/2014/07/15/the-pentax-645z-review-part-iii-sdm-lenses-and-long-exposures/), the 645Z rarely inserts a black frame exposure with exposure times below 4-5 minutes.
Ola Bart!
I agree, the shooting intervals will become unpredictable.
What I got from my experience with the D800 in very high gain situations (> ISO 6400) at handheld shutter speeds, is that the extensive use of active electronics in a dense package produces prodigious thermal noise, as one can see from inspecting a dark frame.
I supposed that these characteristics would be similar in the MF Exmor. I am curious to see in more controlled tests how much this bears out.
It is hard to tell how accurate Ming Thein is about when his camera does or does not use DF noise reduction. If he is judging on the basis of delays, then he might not be taking the dark frame library into account.
As far as noise contribution from live view, I suspect that most of the noise comes from the heat generated by the continuous activation of the sensor and readout at many frames per second, and less from heat generated by the LCD itself.