[normhead]

This at least give some indication that for example if you want to print large you'd want at least some sharpness level.

If we view the level of sharpness in a 6 MP image, and with modern enlarging algorithms, to keep the jaggies at bay, there is no minimum standard that I know of when making a large size print. I've even seen artists deliberately use a small MP camera and blow the images up to huge sizes, just to achieve the effect that this technique creates, an almost comic book type poster effect. And those can be excellent images.

The "some level of sharpness" you'd need to achieve has to do with focus, not resolution. If the shot has at least part of the image in focus, you're good to go. Out of focus images are another question. Film images get fuzzier when enlarged, digital images, with the right algorithms, not so much.

[Nicolas06]

If we view the level of sharpness in a 6 MP image, and with modern enlarging algorithms, to keep the jaggies at bay, there is no minimum standard that I know of when making a large size print. I've even seen artists deliberately use a small MP camera and blow the images up to huge sizes, just to achieve the effect that this technique creates, an almost comic book type poster effect. And those can be excellent images.

The "some level of sharpness" you'd need to achieve has to do with focus, not resolution. If the shot has at least part of the image in focus, you're good to go. Out of focus images are another question. Film images get fuzzier when enlarged, digital images, with the right algorithms, not so much.

This really depend what you are after. You given the example of monet before. Now you speak of a special effect to get "an almost comic book type poster effect".

I don't think that something to do always or something that it should be that way because you have no choice. I think Monet didn't paint this way foremost because he addn't access to fine brush or because he was unable to do better. I don't think neither that all painting should look like the one of Monet or Van Gogh.

This is far less known but when in India they described something they describe as "mignature". This is painted with the finest brush and with a manifying glass. There lof of details in each painting, you can stare at the painting from near distance and discover very little details. That an interresting form of art that was really popular in India and indeed required lot of works. This art is present in many museum of India. You can search for it on the internet but it is not that great to see. Part of the interrested is to see the very fine details but typicall image shown are small on the web so you can't see anything of it.

So this is that. There no rule. Perfect eyes sight can see the detail up to 300dpi (even more in some occasion, but 300dpi is considered to be the standard). So you can print/display up to 300dpi and if the viewers are not so distant from the scene they'll discover details on the print they would not see if it was a 50/100dpi print. That's it. A print of a moderate painting size like 30x40" would be 105MP. 1/4 of it so 24MP would give 150dpi. A crop of a 6MP shoot would get maybe 30 dpi. Usefull? Useless, that your choice.

You can be more for Monet or more for mignature paintings. Or what ever else. And I would say, you don't have to be for one or another. You can enjoy the best of all form of art depending on the occasion. I think we can appreciate the fine details of some form of art an also appreciate the rough texture and lack of details of another form.

Or you may want to be limited to only Monet kind of paintings and picture... Because after all if Monet managed to make good painting with this, obviously there no need ever for another type of painting or any high resolution photographs.

[normhead]

An interesting post... but full of assumptions. We have evidence that most people can't even tell the difference between a 300 DPI print and a 100 dpi print, empirical evidence, not speculation. You don't really see more detail in a 300 dpi print in many cases. In some of your prints, some areas may look better in 100 DPI than in 300 DPI. The style of image you speak of, the little images within the big images, there is nothing that I know of that you would get that kind of difference between 16 Mp and 36 or even 51 Mp. In fact to use an argument like that you're going to have to do some research to find out at what resolution that type of effect is going become possible in photography. You're very creative in speculating what might be, but you have nothing to suggest whether or not these speculations have any merit.

You would have to establish that the type of person who like hi-def images would actually know he was looking at one, comparing a 300 DPI to a 100 DPI image. You assume that he would. I guarantee you it would not be anything like comparing a Monet to one of those Indian paintings.

[Nicolas06]

An interesting post... but full of assumptions. We have evidence that most people can't even tell the difference between a 300 DPI print and a 100 dpi print, empirical evidence, not speculation. You don't really see more detail in a 300 dpi print in many cases. In some of your prints, some areas may look better in 100 DPI than in 300 DPI. The style of image you speak of, the little images within the big images, there is nothing that I know of that you would get that kind of difference between 16 Mp and 36 or even 51 Mp. In fact to use an argument like that you're going to have to do some research to find out at what resolution that type of effect is going become possible in photography. You're very creative in speculating what might be, but you have nothing to suggest whether or not these speculations have any merit.

You would have to establish that the type of person who like hi-def images would actually know he was looking at one, comparing a 300 DPI to a 100 DPI image. You assume that he would. I guarantee you it would not be anything like comparing a Monet to one of those Indian paintings.

Just print text at 100 dpi with your printer, print it also at 300 and 600 dpi and ask for people to look for the difference.

That's funny how we have 16, 36 or 51MB while before the idea was that a crop of a 18-200 was enough... Something in the 2-3MP effective mega pixel at most.

Now almost all highend smartphone have arround 300dpi or more and people pay big money for it. And this is real pixels, each one with all colors not just 1 colors per photosite. I instantly saw the difference between the standard screen rez and the high resolution one. Even more funny because my eye sight is not even good enough to see the dof area properly on the view finder.

Last time I printed an A4 shoot of DA15 I was disapointed by the lack of sharpness I did see on the borders (not just extreme a good part of the image). From a 24MP K3 shoot. I didn't take any magnifying glass and wasn't trying to look at it at minimal distance eyesight that is for me arround 6" (I have some myopia), it was more normal distance of 20-30". This doesn't occurs with FA77 for example.

There more differences in lenses than many think. This include many aspect, colors, contrast, micro contrast, bokeh rendering, flare resistance and also sharpness.

We are more or less sensitive to each one, and depending of the shoot one characteristic or another or all become important. I for me a bit disapointed I can't print a DA15 shoot at 12x8" and get something that look sharp to me. And I do not think the 18-200 would do any better.

So if the scene need the focal length (for the perspective distorsion) and if there lot of dynamics the DA15 shine... I can't have the sharpness but I register the dynamic range and keep great colors and contrast.

But for border to border sharpness, something typicall of landscape, the FA77 shine. Even through I first one would think the DA15 was made more for landscape than the 77. But that's how it is.

Maybe the guy next to me can focus on the view finder just fine manually but can't see any difference between a 100 and 300dpi print. We are more different than we think. maybe you don't care of having more than 6MP and of high end lenses... Even through you think of a DA*200, have already a DA60-250 and a K3... Maybe you have no choice to get the other features you want. But because you and maybe many other are like that doesn't mean we are all like that.

---

Last edited by Nicolas06; 02-25-2015 at 05:18 PM.

[wildman]

Do the results please you? Do you get a good image? Then you have a good lens and camera.

The camera market would collapse within a year if people started to buy what they need rather then buying what they are told they need.

A common sense consumer is subversive to the interests of the American economic system.

[jsherman999]

...Regarding your last point, I believe sharpness is highly overrated as a comparator. It certainly matters to me a lot less than subject matter, composition, colour and contrast. Sharpness is valuable in some images and some types of photography and doesn't matter at all in a lot of images.

I fully agree, but truth be told, if an image I make is fantastic in all those categories but is soft... I'm not as happy as I could be.

I've printed enough images to know I'll lose my enthusiasm for anything that has missed focus or softness on the intended subject. I feel like it's a partially-missed opportunity.

Good subject matter, composition, color, contrast are not mutually exclusive to sharpness. I want them all available in an image I really care about!

[MSL]

Interesting observation... I would have though those two lenses would do well, but the are not only FF lenses, but lenses designed for film. It makes you wonder if there's a difference.

I tried to do that exact comparison, but DxO only had half the data. The lenses to compare are the DA50/1.8 and the FA 50/1.4, and the DA35/2.4 and the FA35/2. But they only have data on the first pair. If you compare the two lenses on various Pentax bodies, the DA50 comes out slightly better or equal to the FA50. I chalked it up to an optimization of the coating but the data was in agreement with what was shown in the video. I confess that I stopped looking after a while, because for most of the lenses I looked at, the Mpix seemed a lot lower than what similar lenses on cameras from other brands got.

[XMACHINA]

I'm not getting the anger being expressed by some people here.

Northrup didn't try to represent full-frame as a superior format to APS-C. He made the video to address the the widely-held belief that there is some advantage to using Full-Frame lenses on cropped-sensor camera bodies. He is completely transparent as to the way he arrives to the conclusion that no such advantage exists.


Furthermore, he specifically states that he himself owns at least one cropped-sensor body, and spent some time heaping praise on an ASP-C formatted lens. And those familiar with Northrup will know that he even shoots with Micro Four-Thirds cameras.


If someone wants to dispute his claims (they would mostly be arguing with DXO's data), that's fine. But people should ask themselves what's making them want to erect a strawman about Northrup's "apparent" belief in the superiority of 36x24mm format. He never stated anything close to that. If anything, his position is one that would serve to *discourage* buying full-frame gear.


It just looks like some people don't like the news and are groping for a justification to kill the messenger.

[yusuf]

Nice video but the way it's presented is misleading. Some math would have been helpful instead of general statements. Let me try.

Let's compare 24MP (6000x4000) FF & APS-C sensor. A FF sensor can resolve upto 83 lpm (best case, nyquist) but the APS-C sensor can resolve up to 128 lpm. The effective resolution of the sensor-lens combo will be decided by the worst of two. Hence, to fully utilize the sensor resolution, you will need a sharper lens for APS-C compared to FF which exceeds the sensor resolution else the penalty. For example, if you use the same lens, say Pentax M50 with approx 60lpm, will limit the max resolution to 60lpm on both the bodies. This will result in effective resolution upto 12MP on FF vs 5.7 MP on APS-C.

However, all the lenses will have more resolution in the center which works in favor of APS-C and this is no misconception as claimed in the video.

[audiobomber]

However, all the lenses will have more resolution in the center which works in favor of APS-C and this is no misconception as claimed in the video.

I checked your conclusion at DXOMark and Photozone.

1. DXOMark, Nikon 50mm f1.4G lens, on D3300 APS-C and D610 FF, both 24mp with blur filter. There is nowhere in the Sharpness Field Map measurements for each aperture that shows the APS-C with higher resolution in the center.

Nikkor AF-S Nikkor 50mm f/1.4G on Nikon D3300 versus Nikkor AF-S Nikkor 50mm f/1.4G on Nikon D600 - Side by side lens comparison - DxOMark

2. I couldn't find FF and APS-C bodies with the same resolution and same lens at Photozone but this should be close enough. Adding one million megapixels to the NEX7 would not make up this difference. The FF is still a bit sharper in the center:

• NEX7 (24mp APS-C), 85mm Zeiss, maximum resolution 3029 LW/PH
• A850 (25mp FF), 85mm Zeiss, maximum resolution 3453 LW/PH

Zeiss ZA Planar T* 85mm f/1.4 (SAL-85F14Z) on Sony NEX - Review / Lab Test - Analysis & Verdict
Zeiss Planar T* 85mm f/1.4 ZA ( Sony SAL-85F14Z ) - Full Format Review / Test Report - Analysis

[yusuf]

I checked your conclusion at DXOMark and Photozone.

1. DXOMark, Nikon 50mm f1.4G lens, on D3300 APS-C and D610 FF, both 24mp with blur filter. There is nowhere in the Sharpness Field Map measurements for each aperture that shows the APS-C with higher resolution in the center.

Nikkor AF-S Nikkor 50mm f/1.4G on Nikon D3300 versus Nikkor AF-S Nikkor 50mm f/1.4G on Nikon D600 - Side by side lens comparison - DxOMark

2. I couldn't find FF and APS-C bodies with the same resolution and same lens at Photozone but this should be close enough. Adding one million megapixels to the NEX7 would not make up this difference. The FF is still a bit sharper in the center:

• NEX7 (24mp APS-C), 85mm Zeiss, maximum resolution 3029 LW/PH
• A850 (25mp FF), 85mm Zeiss, maximum resolution 3453 LW/PH

Zeiss ZA Planar T* 85mm f/1.4 (SAL-85F14Z) on Sony NEX - Review / Lab Test - Analysis & Verdict
Zeiss Planar T* 85mm f/1.4 ZA ( Sony SAL-85F14Z ) - Full Format Review / Test Report - Analysis

Well, you missed the second para, FF will always have the advantage and those measurements are in agreement with what I said.

However, the last para merely says that it's incorrect to undermine the FF lens advantage on crop sensor, it was not a comparison statement with FF.

[normhead]

I checked your conclusion at DXOMark and Photozone.

1. DXOMark, Nikon 50mm f1.4G lens, on D3300 APS-C and D610 FF, both 24mp with blur filter. There is nowhere in the Sharpness Field Map measurements for each aperture that shows the APS-C with higher resolution in the center.

Nikkor AF-S Nikkor 50mm f/1.4G on Nikon D3300 versus Nikkor AF-S Nikkor 50mm f/1.4G on Nikon D600 - Side by side lens comparison - DxOMark

2. I couldn't find FF and APS-C bodies with the same resolution and same lens at Photozone but this should be close enough. Adding one million megapixels to the NEX7 would not make up this difference. The FF is still a bit sharper in the center:

• NEX7 (24mp APS-C), 85mm Zeiss, maximum resolution 3029 LW/PH
• A850 (25mp FF), 85mm Zeiss, maximum resolution 3453 LW/PH

Zeiss ZA Planar T* 85mm f/1.4 (SAL-85F14Z) on Sony NEX - Review / Lab Test - Analysis & Verdict
Zeiss Planar T* 85mm f/1.4 ZA ( Sony SAL-85F14Z ) - Full Format Review / Test Report - Analysis

My numbers showed about a 100w/ph advantage for FF based on about 2600 lw/ph, less than 4% and pretty much indistinguishable. Don't make me calculate what the differences would be in the size of the smallest line resolved, but it would be ridiculously tiny and practically identical.

[Pål Jensen]

There is obviously always an equivalence. !

That is a very different proposition than claiming two lenses of different focal lengths but same angle of view on their respective formats are equivalent based multiplying the max aperture.....

[Nicolas06]

That is a very different proposition than claiming two lenses of different focal lengths but same angle of view on their respective formats are equivalent based multiplying the max aperture.....

There a formula for deph of field. It take into account the magnification, viewing distance, sensor size, focal length, focussing distance, apperture.

If the most important is to keep the same exposure level and so to keep the same apperture, and to take into account the framing change to think that an iphone 5s is equivalent to roughly a 30mm f/2.2. Implictely printing size/viewing size is supposed to be the same and sensor size/focal length to keep the same framing. Focussing distance is not important as deph of field is not a factor. Here the idea is that keeping the same iso level as for the original shoot is really important to keep the expressiveness, artistic value and aspect of the image, more so than the deph of field.

If the most important is to keep the same focus area, but to not care of isos levels it is more logical to think of as roughly a 30mm f/20. Still implicitely printing/vieweing size is supposed to be the same and sensor/focal length to keep the same framing. Here the idea is that keeping the deph of field as of the original shoot is really important, more so than the initial iso value.

Some argue the same lense with same apperture/focal length would always give the same deph of field regardless of sensor size... That true as long as the focussing distance is keep but the framing is then completely different. To me this one is completely useless, even through it is true.

There just one generic formula and the understanding of the formula and what it imply and the consequence for a situation or another.

[ppppsssstttt]

Pentax has some rectilinear UWA lenses - FA 20, FA 20-35, M 20 to name a few...

i suppose you're right. i stand corrected. i just never think of rectilinear UWA's being as readily available or <$500 and the last time i really looked into it, it was into m42 lenses.