Specific Landscape System Question

[gnaztee]

Hi, first post for me on here, though I've been visiting the site for years. I realize that I could post this in the gear/technical forum, but the landscape shooting aspect is integral to my question. I'm thinking about a system change in order to simplify what I use, but I'm wondering if said change is a good opportunity to "upgrade." The particulars:

- 95% of my photography is travel and landscape
- I currently have a Pentax K-5 setup with a collection of both quality zooms and primes, which I use for landscape photography
- I also use a Fuji X100 for travel (35mm-e is perfect travel length for me), and generally bring the K5 or K-01 for more deliberate shots when traveling (X100 for large prints from RAW doesn't match my K5)
- I don't need shallower DOF
- I don't need better high ISO performance (but I'd use it if it's there)
- I don't need more resolution
- I'm not a professional, but I take it very seriously
- I'm considering selling it all to get 1) Sony RX1, and 2) Nikon D600 with a 17-35 (or so) and a 70-200

My question is this: For landscape photography, where will I really see the difference between my K5 and the FF sensor in the two cameras I'm looking at?

I'm pretty sure I see it in the photos when I compare, even online, but I can't verbalize the difference I see. Something in the tone/color depth definitely looks different to me. Am I seeing this or can it simply be chalked up to subject and post-processing?

I'm not trying to turn this into a FF vs Crop sensor discussion in all aspects, just in this particular aspect that I can't specify as it applies to landscape photography.

I hope I'm making myself clear, please ask for more info if I'm not. And I thank any/all of you for your help.

Todd
www.naskedov.zenfolio.com

[fike]

For online viewing, unless enlarged and cropped to 50% or 100%, you can't see the difference, unless....
1) you are dealing with very low light conditions
2) you are dealing with very wide dynamic range (most commonly seen in pics that include direct view of the sun and shaded areas or interior images where you are seeing subjects with indoor lighting against windows, etc...)
3) you are dealing with extremely saturated colors, particularly in the yellow and purple range (I think this is more debatable than the two above)

Online publishing is very forgiving if you have some idea of how to post process to minimize noise.  The downsampling needed to publish any images to the web is essentially a very effective noise reduction through averaging.  So this issue is mitigated by online media. 

There are real differences between these cameras, but it requires a bit of pixel-peeping to see the differences.  There are some applications where that is critical, but for travel and landscape, you won't see dramatically improved capabilities with the full frame gear.

[gnaztee]

Thanks for the response. Okay, so practical application question: How much difference will I see at 24"x36" on a wall?

[fike]

We are firmly in the land of opinion here, but I think it is reasonable to say an uncropped 24"x 36" image that compares full frame with a cropped sensor in ideal (brightly lit) conditions will have a noticeable difference when viewing distances are less than two feet.  Outside of close viewing, a well-exposed image under good conditions will generally be difficult to differentiate between full frame and cropped sensor. 

Simply put, I think at 24x36 you have crossed the line into a place where full frame or large format (or panoramic/mosaic stitching) is desirable.  If you said 13x19, I would have said it didn't matter, but at this larger size, you will notice the difference.

[gnaztee]

I appreciate your opinion, thanks. One more: is the perceived difference at that size and distance based on something quantifiable, in your opinion? Like color depth, or something else?

[fike]

I appreciate your opinion, thanks. One more: is the perceived difference at that size and distance based on something quantifiable, in your opinion? Like color depth, or something else?

I think color depth is really hard to ever detect with the naked eye.  There may be some gut feeling that one is better than an other, but with color depth, the differences are hard to quantifiably detect.

Low-light performance is visible based on the presence or absence of noise....the more objectionable of which is low frequency and chroma (mottled color) generally in shadows.

Under ideal conditions sharpness is the most readily identifiable problem with large prints made from smaller sensors.  Get 6 inches from a print and the eye can detect very minor variations in sharpness that will result from different sized sensors.  All of this is, of course, assuming you are following a decent printing workflow.

[Slobodan Blagojevic]

... How much difference will I see at 24"x36" on a wall?

None.

None that matters, that is.

I've printed 24x36 on canvas from a point & shoot (Canon G10, 15 Mp, no interpolation) and even from a nose-length close inspection it looks great. I've printed dozens of images on large-size canvas from 8 Mp to 21 Mp, from ISO 80 to 1600, hand-held and tripod, and you can't determine which one comes from which combination.

Sharpness, 300 ppi resolution, etc. are highly overrated.

[Roman Racela]

Ditto!

We are firmly in the land of opinion here, but I think it is reasonable to say an uncropped 24"x 36" image that compares full frame with a cropped sensor in ideal (brightly lit) conditions will have a noticeable difference when viewing distances are less than two feet.  Outside of close viewing, a well-exposed image under good conditions will generally be difficult to differentiate between full frame and cropped sensor. 

Simply put, I think at 24x36 you have crossed the line into a place where full frame or large format (or panoramic/mosaic stitching) is desirable.  If you said 13x19, I would have said it didn't matter, but at this larger size, you will notice the difference.

[robdickinson]

I think color depth is really hard to ever detect with the naked eye.  There may be some gut feeling that one is better than an other, but with color depth, the differences are hard to quantifiably detect.

It was one thing I noticed first moving from 7d to 5d2. You cant quantify it but it shows in images especially with foliage.

A well shot 7D at 18mp in good light at base iso with a quality lens vs the same on a 5d2 is quite close in terms of look and you have to print quite large to see any real difference.

As soon as you push the ISO up over base you can notice more difference.

Also IMO lenses with bigger sensors tend to produce sharper results given similar levels of quality. EG the 10-22 IMO is a getter all round lens than the 17-40L but because of the ff 'advantage' (aka larger photosites etc) the results are better on the ff.

[gnaztee]

Okay, good stuff. Thanks guys.

Fike: I guess by "depth," I really mean something more like tone gradients...if Im using the terms correctly. In other words, there seems to be a smoother quality in color changes...I'm thinking about multi colored skies/clouds as an example. Not highlight clipping, but I'm not sure if what I'm describing is correct.

Slobodan: I've printed several of my photos on canvas (16x20 up to 30x40) and while I love it, it seems like a medium that would blunt sharpness a bit. I don't say that to question your points at all, only wondering if canvas might negate the sensor size difference more than a paper print.

And just so I don't sound like I've made my mind up and I believe FF is better, I'd really rather the consensus be that there is no difference. It would be cheaper. Sensor size doesn't matter to me much for travel shots, but I'm  wondering about the ultimate (affordable) image quality for landscape stuff. I just want to ask the folks who have experience with both sensors (I don't).

I'm also not trying to drag this out, making you guys answer Qs over and over. I'm just intensely curious. Tanks again for your time and effort folks.

[fike]

Okay, good stuff. Thanks guys.

Fike: I guess by "depth," I really mean something more like tone gradients...if Im using the terms correctly. In other words, there seems to be a smoother quality in color changes...I'm thinking about multi colored skies/clouds as an example. Not highlight clipping, but I'm not sure if what I'm describing is correct.

First, I like to review my terminology at a reliable site like http://www.cambridgeincolour.com/tutorials/bit-depth.htm .   One note about the article.  It is often confusing when people talk about bits per pixel.  Sometimes 24-bits per pixel is referred to as 8-bits per channel.  There are red, green, and blue channels at each pixel, so multiply those 8 bits by three to get 24-bits per pixel.  Some photographers choose to operate their workflow in 16-bit per channel mode or 48-bits per pixel.  If you are shooting jpeg, I think you are unlikely to notice a difference, (besides jpg only supports 24-bit per pixel or 8-bit per channel) but if you intend to shoot raw and occasionally really push the pixels around very aggressively, more data is better and the extra bit-depth will serve you well to avoid posterization in deep color gradients like you find in blue skies or vibrant flowers. 

Most importantly, this bit depth isn't, an attribute of the camera or sensor.  This  attribute can be configured during post processing of RAW files. 

There are also 12-bit and 14-bit raw capture options that ARE attributes of the sensor (or board design), and I am a bit out of my element in that area except to say that the differences is really marginal and exists at the extremes of performance. There isn't much consensus about how critical this is. 

And just so I don't sound like I've made my mind up and I believe FF is better, I'd really rather the consensus be that there is no difference. It would be cheaper. Sensor size doesn't matter to me much for travel shots, but I'm  wondering about the ultimate (affordable) image quality for landscape stuff. I just want to ask the folks who have experience with both sensors (I don't).

Trust your own judgement.  Unfortunately the only way to decide for yourself is to try out the gear and make up your own mind.  Perhaps you might like to rent some of this gear before you decide for yourself.

[luxborealis]

Perhaps you might like to rent some of this gear before you decide for yourself.

Renting is the best way to see for yourself if there really is a difference and if the difference is really worth the additional outlay. And don't just look on screen - the proof is in making 24x36s to really compare. "Can you see the difference. No I can't see the difference."

When I moved from Olympus 4/3s to FF the difference was remarkable. But, I was using an E30 at the time and moved to a D800e. The K-5 sensor is far superior to the E30, so the difference might be quite a bit less than what I saw. BTW - if you do decide to go FF, then consider ponying up the extra for a D800. "In for a penny, in a for a pound."

[Paulo Bizarro]

I think that you already have a system that comprises arguably one of the best APS-C sensors and some of the best prime lenses available. I don't know if you already have some Limited lenses, but that is where you should put your money in.

[gnaztee]

Thanks for all of your thoughtful responses. Much appreciated 

[xocet]

If you are sure it's not just a question of GAS.... 

The best bet is to see if you can rent or borrow some different camera types. - FF, a Fuji Xtrans camera, and maybe even one of the Sigma Merrill sensored cameras.

Another, free thing you could try is a different image editor.  While I'd think you should (in theory) be able to end up with a similar finished product using any of the well known tools, you may find it easier to get 'the look' using a different one (compared to what you use today).