Landscapers... Full Frame vs Crop

Regardless of pixel size, it is true (even for film) that for a given size print, the optical image recorded by the camera will have been enlarged by a factor of 1.6 in each dimension more than the FF version of the same image. So optical imperfections will also have been enlarged that much more. Whether they become visible will depend a lot on the size of the print. The effect of diffraction is also subject to print size (and viewing distance etc) as to when it become visible - these are all related to the same geometry that drives the difference in DoF. (sensor : print size ratio). The pixel size comes in mainly when people think "I want to print at 300 ppi" for example, and compare images from cameras with pixels of different dimensions and forget that they are geometrically enlarging the image from the small pixel camera more than the large pixel camera, and aggravating any optical flaws at the same time. Like "I went from 13 MP to 21 MP in my FF so I should be able to print larger because I've got more pixels". True to a certain extent, but the same image is being enlarged more so flaws become more likely to be visible. The best large landscape prints come from the larger formats (8x10>4x5>MF>FF>crop>MFT>P&S).

Diffraction effects, DoF effects, motion blur (camera or subject) are all dependent on how much the image is enlarged after capture. Obviously other factors play their roles but this one is often overlooked So you need to specify print size to know how important all these factors are

Heya,

I do a ton of landscape on APS-C.

The main advantage full frame has, to me, having both full frame & APS-C, is that a better sensor has more room for pulling detail from a single exposure, reflective of dynamic range. When you try to lift light or pull data from the full frame image, you get a bit better result than from an APS-C exposure, usually.

But I've found that the advantage is mostly something that has culturally stayed relevant from the past. Modern APS-C is quite good, and better than some older full frame in many ways.

To me, the only reason to shoot full frame for landscape at this point is if it's truly your bread & butter and you need every potential advantage you can. Otherwise, frankly, I'd rather have several APS-C's shooting during that small time window to gather more exposures. Ultrawide on APS-C is a big difference for today versus years ago. There's really not a lens advantage anymore. It really just comes down to processing of single exposures (dynamic range), pixel density (full frame is more forgiving), and overall resolution (this only matters if you're doing really large prints really).

But I haven't seen enough of a difference for me to really care. I don't feel the need to buy a Sony A7r and a 21mm Zeiss lens to do landscape. But again, I don't make my living shooting landscape. Honestly if I were truly a big landscape shooter and wanted great equipment, I'd be shooting medium format instead of expensive 35mm that still is low resolution by comparison anyways.

Very best,

Good replies in this thread.

I agree with almost all that was said. Optimally you'd have as large a format as possible and the lenses you want.

But there really isn't a massive jump between the different systems and lenses - just incremental ones that determine where you want to be.

Personally I wanted to use the TS-E 17mm. So that meant FF. I have one shot that I have printed that I could have done with more resolution and in that case I don't even think I'd have got it with a D800 because it is a telephoto shot and the air quality was not stellar. Of course no-one but me stands 12" from a 36" print either.

However, even though I have a 5DIII and TS-E 17mm, I have several 18" prints of landscapes taken with my 7D + Sig 18-250 on my walls. Point being that even though I like the former IQ, that latter is still good enough for me to make a nice print with that I will hang up.

I don't think I've ever made an 18" print from my S-series cameras though. So I would say 1.6x crop + decent lens is my 'low point' for what I want to produce.

ejenner wrote in post #17161829
Of course no-one but me stands 12" from a 36" print either.

You are by no means as alone as you think. A good number of people who see a detailed large landscape print will take it all in from a distance but then get drawn in to examine the details or "explore" the landscape up close.
(I have some 30x40 Christoher Burkett prints that you need a loupe from a couple of inches to see all the details in (they did come from 8x10 transparency originals, though!) - Bigger is better if large prints are the goal).

I have done some large, poster-sized prints. I think my personal goal is always to capture images worthy of such print sizes. A normal print for me would probably be A3.

When I switched to full frame, I had the budget and the means to purchase exactly what I wanted, so I have no regrets there. This topic came about a few days ago for me when I was trying to explain, rather clumsily, the benefits of larger sensors in the context of smartphone cameras vs. DSLR's. This made me realize I didn't have a firm grasp of why a full frame sensor might really be advantageous over APS-C. I knew it had something to do with pixel size, and also something with resolution. Thanks to some truly wonderful replies so far I think it's a lot clearer in my mind now. I think this can also serve well for many contemplating a similar "upgrade" path.

I was curious about the same thing. So I did my own tests. If you can bear wading through all the irrelevant waffle, look here https://photography-on-the.net …read.php?t=1394​691&page=5

I use both for landscaping. As has been said the FF provides a bit more ability to pull some detail from the shadows. I have plenty of nice crop landscapes hanging on my walls too. Get decent glass, Lightroom and know how to use them and you can get great captures from a Rebel.

I aspire to print as large as possible. Shooting FF, I am comfortable that my collection of images will have better technical qualities, and withstand larger print sizes. Of course, if I could own a 80mp MF 6x4.5 cm sensor, I'de be all over it. Just not that much disposable income.......... yet.

AJSJones wrote in post #17161897
(I have some 30x40 Christoher Burkett prints that you need a loupe from a couple of inches to see all the details in (they did come from 8x10 transparency originals, though!) - Bigger is better if large prints are the goal).

So I don't know the photographer, but this reminds me of a print I saw in an office. Color, canyon country landscape, about 5ft tall by about 4ft wide and you could just walk closer and closer and see more and more detail.

I assumed it was 8x10 film, but I don't know for sure (maybe stitched MF digital?) it was unlike anything I have ever seen.

A lot of people will claim that the main advantage of full-frame is the ability to shoot at high ISO, and therefore that it's not really worthwhile for landscapes. That is quite misguided. Even at ISO 100, the difference in noise performance becomes quite apparent when you start pushing files hard (boosting contrast, recovering shadows). This is particularly true for smooth subjects which lack fine detail and texture. Unfortunately for landscape photographers, the sky is one of the smoothest subjects around, and it's a significant component of your shots.

To me, noise performance (at all ISO sensitivities) and dynamic range are the main reasons for a landscape photographer to use full-frame. And they're pretty compelling all on their own, in my experience.

On the other hand, the traditional pro-FF argument that the larger pixel pitch "is less stressful on lenses" tends to break down for ultra-wide FOV photography. Why? Because it's really hard to design lenses that are sharp in the extreme full-frame corners. Up until a few months ago when the 16-35 f/4L IS dropped, the 10-22 on crop was far sharper edge-to-edge than any Canon UWA zoom on full-frame. Today, the 16-35/4 at least makes UWA sharpness competitive between formats (it's probably sharper than the 10-22 in the image center and comparable at the corners). But, of course, it's also more expensive than the 10-22 by a decent margin.

All this is to say that if you're mainly shooting at UWA and want to use zooms, upgrading to full-frame won't necessarily give you a benefit in the realm of sharpness, especially if you're already using the fantastic 10-22. At mid-range and telephoto, it may be true that blinding sharpness is easier to come by averaged over all lens options, though.

brettjrob wrote in post #17162391
On the other hand, the traditional pro-FF argument that the larger pixel pitch "is less stressful on lenses" tends to break down for ultra-wide FOV photography. Why? Because it's really hard to design lenses that are sharp in the extreme full-frame corners. Up until a few months ago when the 16-35 f/4L IS dropped, the 10-22 on crop was far sharper edge-to-edge than any Canon UWA zoom on full-frame. Today, the 16-35/4 at least makes UWA sharpness competitive between formats (it's probably sharper than the 10-22 in the image center and comparable at the corners). But, of course, it's also more expensive than the 10-22 by a decent margin.

All this is to say that if you're mainly shooting at UWA and want to use zooms, upgrading to full-frame won't necessarily give you a benefit in the realm of sharpness, especially if you're already using the fantastic 10-22. At mid-range and telephoto, it may be true that blinding sharpness is easier to come by averaged over all lens options, though.

I agree with most of your points but my personal experience with the 10-22 and 17-40 mirrors that of the-digital-picture. The 10-22 is very good but even 17-40 can edge it out. http://www.the-digital-picture.com …omp=0&FLIComp=0​&APIComp=4

Even with the references above, I think when folks talk about the lens options for UWA on FF being an advantage, most aren't talking about the zooms but the wide primes like the ZE21, ZE15, the TS-e's, etc.

I think we have to always consider what generation FF vs what generation APS-C as well in these discussions. For example, shoot a 5DI against a 70D, both with comparable glass to get the same FOV, and you either have the option to resize the 70D down to the same size as the 5D3, or keep what you have with greater potential resolution of detail.

MNUplander wrote in post #17162877
I agree with most of your points but my personal experience with the 10-22 and 17-40 mirrors that of the-digital-picture. The 10-22 is very good but even 17-40 can edge it out. http://www.the-digital-picture.com …omp=0&FLIComp=0​&APIComp=4

Interesting. I personally owned a Rebel+10-22 for many years, followed by a 6D+17-40L for about a year, and shot both combos extensively. My experience was quite different from the samples shown there -- the 17-40L was simply horrid in the corners, even stopped down to f/8. In fact, it was somewhat soft even 60-70% of the way radially outward from the image center toward the corners.

I referenced that lens comparison tool on the-digital-picture frequently during my long hunt for an acceptable UWA zoom, and I always thought their 17-40L test shots seemed awfully generous -- compared against both my experience and its broader reputation among Canon FF owners. But it's quite clear that sample variation is a problem with that lens, so I don't doubt there are good copies out there.

I totally agree that a top-shelf prime on full-frame will get you unmatched sharpness in UWA photography, as long as you're willing to sacrifice the flexibility of a zoom.

brettjrob wrote in post #17163338
Interesting. I personally owned a Rebel+10-22 for many years, followed by a 6D+17-40L for about a year, and shot both combos extensively. My experience was quite different from the samples shown there -- the 17-40L was simply horrid in the corners, even stopped down to f/8. In fact, it was somewhat soft even 60-70% of the way radially outward from the image center toward the corners.

I referenced that lens comparison tool on the-digital-picture frequently during my long hunt for an acceptable UWA zoom, and I always thought their 17-40L test shots seemed awfully generous -- compared against both my experience and its broader reputation among Canon FF owners. But it's quite clear that sample variation is a problem with that lens, so I don't doubt there are good copies out there.


I totally agree that a top-shelf prime on full-frame will get you unmatched sharpness in UWA photography, as long as you're willing to sacrifice the flexibility of a zoom.

Which rebel did you use with the 10-22? Was it an older one where the sensor wasn't quite so dense? The samples on TDP are from the 60D with 18MP crammed onto APS-c.

Maybe it is an issue with sample variation with the 17-40. It can't match the corner sharpness of the wide primes we all talk about, but based on the copies I've used it did a whole lot better than what most folks give it credit for. However, I've owned 3 different copies and all of them performed similarly.

But, the 10-22 does often get brushed under the rug when compared to FF options when it shouldn't - it is a great little lens with outstanding IQ for it's price.

Some likes UWA, fisheye style landscape, where with landscape they could push it even more on FF, I guess. To me FF makes sense with T&S lens for landscapes, or 50mm prime, or 70-200.
All of these lenses are better on FF for landscapes, IMO.
While ago I posted blind test between crop and FF for cityscape. Some comments were interesting to read to understand where the difference is.
But it is for the case if you prefer naturally looking landscapes. If you into cheesy, overkilled in PP landscape pictures the tasteless crowd is after these days. the sensor size of the cropper is OK, as long as you are comfortable with lenses.
On serious note, I like how Sigma Merrill camera with fixed prime renders landscapes and even some Leica luxury compacts.
OK, I can't be serious for too long, yes, I'm planning to switch from FF to crop for landscapes!
In my 4x5 LF camera, by adding of 6x7 MF film back