FX vs DX Image Quality Comparison (yes.. this again) Your Thoughts?

sevencrossing

when it comes to strict IQ given all other things remaining the same, there is virtually or possibly even no difference between the IQ of a FX or DX sensor.

Sorry and all that

But "other things" can never be equal
lets look at the AF-S 16-35mm F4G VR ( my favorite landscape lens)
there is no Nikon DX equivalent
Yes it will fit on a D7100 but at say 16mm, the edges are going to be cut off
compare the D7100 at ISO 6400 and the D750 at the same setting
and tell me there is no difference

spraynpray

when it comes to strict IQ given all other things remaining the same, there is virtually or possibly even no difference between the IQ of a FX or DX sensor.

Sorry and all that

But "other things" can never be equal
lets look at the AF-S 16-35mm F4G VR ( my favorite landscape lens)
there is no Nikon DX equivalent
Yes it will fit on a D7100 but at say 16mm, the edges are going to be cut off
compare the D7100 at ISO 6400 and the D750 at the same setting
and tell me there is no difference

Nikon 10-24 on a DX is 15-36 on FX.

sevencrossing

Nikon 10-24 on a DX is 15-36 on FX.

I don't have this lens. It has variable aperture and no VR it gets good reviews but not that good
I am pretty sure if you compared this lens @ 10mmm on a D710 ( Do X score 11) with the 16-35 @ 16mm ( do X score 19) on a D750 you would see a difference

JonMcGuffin

@seencrossing. We did cover this earlier as you could also put a 70-200 2.8 on each and match focal length, aperture and subject framing to easily produce the 'same' picture.

JonMcGuffin

And on that note, you could probably put a Sigma 18-35 1.8 on the DX and shoot it at 18mm and F2.8 and I would bet pretty strongly it would outperform the 16-35 at 24mm F4 which again theoretically should provide the same picture at the same shooting distance.

JonMcGuffin

I think I've mistakenly when referring to image quality inferred shadiness differences which I think is the first thing somebody think about when we talk about IQ.

When I first saw the D800 come to market a lot of people rushed out to see and test if it could "compete" with medium format. In most reviews and tests I saw the differences in my view were negligible in favor of medium format however I could barely tell that where it was improved was in fact not in sharpness but rather tonal gradation (right word?). For instance skin tones and the shift between maybe a highlight on the cheek and then moving down toward the chin.

This was telling me that the larger sensor might be capable of producing better pixels than a smaller sensor, in this case FX. Or maybe it was in fact the other variables we've discussed here that were at play.

What do you think? Are there objective methods to prove the opinion one way or another or because of the artistic nature of a photographic will it always be a subjective one.

sevencrossing

There is a problem with this thread

You cannot simply compare the IQ of a Dx sensor with the IQ of a FX sensor

The camera, lens, combination is always going to affect IQ,

One cannot just look at the sensors in isolation

If you are trying to decide whether to buy a D7200 or D750 you must compare all the features of and sample images from, both cameras; together with your chosen lens combination

WestEndFoto

There is a problem with this thread

You cannot simply compare the IQ of a Dx sensor with the IQ of a FX sensor

The camera, lens, combination is always going to affect IQ,

One cannot just look at the sensors in isolation

If you are trying to decide whether to buy a D7200 or D750 you must compare all the features of and sample images from, both cameras; together with your chosen lens combination

Yes I strongly agree, but practically, I over simplify this and go one step further. I am aware of the shortcoming, but it is a short term shortcoming, not a long-term shortcoming.

The image quality potential of a system depends on the format size. Once you strip out temporary marketing or technical advantages of manufactures offering certain systems, the image quality of a system improves as system size increases, from CX, to DX to FX to Medium Format. The D800 is strongly competitive with medium format because Nikon is so far ahead technically and medium format manufacturers are behind, not because "FX is almost as good as medium format".

This is a corollary of the the laws of physics and practical engineering considerations and will always remain true.

Further, I am investing in a system, not a camera. The camera is a side show.

Finally, weight or price differences, within the range of offerings typically discussed on this forum, are secondary considerations. A 400 2.8 is not a matter of if, but when. Image quality is my only real consideration and I am willing to suffer to get it, including packing 25 pounds of gear around Rome for ten days, the results of which are currently being presented in PAD (well, low resolution facsimiles of the real results).

And my style of shooting? You can look at my Flickr account in my signature, it is a bit of everything - including some eagle shots that I will post in the next two months that presented themselves when I was shooting flowers with my 400mm f/4 macro.

So unless I am shooting BIF exclusively, there is no real alternative between DX and FX, for me FX wins hands down.

heartyfisher

There is a problem with this thread

You cannot simply compare the IQ of a Dx sensor with the IQ of a FX sensor

The camera, lens, combination is always going to affect IQ,

One cannot just look at the sensors in isolation

If you are trying to decide whether to buy a D7200 or D750 you must compare all the features of and sample images from, both cameras; together with your chosen lens combination

+1
+ Dont forget to add the factor of a finite budget.. ;-)

spraynpray

Bringing this thread down from the esoteric heights it is climbing to, it is useful to remember these differences that are so plain to you D8XX owners (and which are so insignificant when you are comparing your D8XX's to an MF Pentax ;;) ) should be plain to see in the conditions that most people are viewing their images - on a PC screen. Otherwise they are not significant at all. If the images have to be blown up to bill board size to see them, well, how many of us print out any pictures bigger than 24 x 16 or so? Or come to that, even print out images using entirely colour managed PP system and top-end equipment and materials?

kyoshinikon

While I have preference to Fx, Dx isnt much inferior. Yesterday I went to my local camera store with the intention of walking out with a body (the D810 as I couldnt "afford" a D4s). I tried the D810, D7200, & D4s (I had ruled out the D610 and D750 which they were sold out of anyway due to both cameras seeming to be less professional and lackluster in the features I needed)

I tried all 3 cameras at ISO 10,000. For reasons I dont the D810 underperformed by about 2 stops on noise. The D7200 was much closer to the D4s than the D810... How about autofocus. I tried 2 lenses My new 70-200mm, & their 58mm prime. Once again the D810 underperformed the D7200. Now while the D4s blew both of them out of the department on AF, the D7200 held up pretty well against it. Now obviously this test was not in a lab and their demo D810 may have been a bad copy but it left an impression on me. Probably the fact that I was looking for features centric to low light sport photography didn't help.

The D750 while I had ruled it out, also outperformed the D810 based on the little shooting I did. In fact I liked the D7200 so much I felt the only real issue I had with it was with the Size. I pulled out my D7000 & D700 just to compare Raws at 100% and the D810 didnt seem to be enough for what I need it for...

I ended up walking out of the store with a refurb D4s (yep they had one that just came in) but when I can Ill go back and get the D7200 to replace my aging and trusty D7000...

Back to the topic, Dx I feel is no longer an issue. My D7000 performs almost as good as my D700 and that gap of performance is closing with every generation. I do agree with PB_PM that the D700 has a much better control layout than the D7200 and the buffer is still much better but I will say that the AF is on par (Similar 51 point AF), the build is comparable, and it has about every feature found on the D700 (as buttons too) but a PC sync & Viewfinder Lock. The whole point of me considering a D7200 over a used D700 was performance. I love my D700. It is a great steal and highly recommended, one of the best cameras nikon ever made in my book, but I was losing shots to its Iso performance at 6400. Dx is no longer an excuse.

spraynpray

Here-Here Kyoshi!

WestEndFoto

Bringing this thread down from the esoteric heights it is climbing to, it is useful to remember these differences that are so plain to you D8XX owners (and which are so insignificant when you are comparing your D8XX's to an MF Pentax ;;) ) should be plain to see in the conditions that most people are viewing their images - on a PC screen. Otherwise they are not significant at all. If the images have to be blown up to bill board size to see them, well, how many of us print out any pictures bigger than 24 x 16 or so? Or come to that, even print out images using entirely colour managed PP system and top-end equipment and materials?

Hmm.....I am not sure that I think of the Pentax as medium format. It almost seems like a turbocharged full frame. When I think of medium format I think about Hasselblad or Phase One.

Otherwise, I agree with you on your points Spraynpray. An hour ago I told a guy in a camera store that was admiring my gear that the D3500 with a few lenses that he was looking at would give him 90% of what I had from an image quality perspective. This was of course after determining that he was a casual enthusiast and not trying to achieve something particular.

WestEndFoto

And more to the point, I think that in terms of photographic quality, 5% of the variation is based on gear and 95% is based on the photographer. Sometimes I read my own stuff and the enthusiasm behind what I wrote and worry that I have been seduced by the gear. I certainly don't disagree with what I say (but there are lots of points that reasonable people can disagree with), but the gear that matters the most, lighting, gets the least amount of attention on this, or any, forum.

Hmm......

Paperman

If my aim were to make a comparison between DX and FX image quality, I would simply look at the available tests to see the " measurable " differences. Anything else depending on my/your or anyone else's observation will be subjective and more prone to error.

The answer is in the resolution/dynamic range/colour depth figures ; the only thing open to discussion is if they matter ( enough ) or not ....

heartyfisher

@kyoshinikon thanks for the mini/micro review .. ;-) . What you have said is consistent with the photographylife review. One thing different is that they even put the D7200 above the D4s in AF capability when dealing with poor marginal lighting and pushing the lenses (with TC). I think this is mainly due to the AF points coverage and the new -3EV sensitivity of the D7200. Thanks again for the "hands on" comments..

kyoshinikon

It is easy to judge a camera on paper but once you get it in your hands it may be a whole different experience. The D750 looks fairly good on paper and it is a fantastic camera but my experience with it told me it just wasn't optimized for what I do. The D7200 was more than impressive considering where we have come from. The D90 & D700/D3 was no comparison. The D7200 smoked it by more than a few stops. Times have changed since then tho and until you do a real world test of the latest Dx it is hard to judge it.

JonMcGuffin

Ok, well we've carried this debate to some extent back into the typical FX vs DX arguments.

I also understand that if you don't take the other FX 'advantages' into consideration, it's difficult to "prove" definitively one way or another if "FX" provides better IQ than DX. So I'd like to draw some kind of conclusion here, I realize many won't agree; such is life in the 21st internet world of armchair QB's and second guessers.

The original question was and I still I think it's a fair enough question to ask. Does a larger sensor have an inherent ability to actually resolve and capture better image quality defined as sharper, better pixels than does a smaller sensor - especially if they are current generation sensors (the D750 vs D7200 sensors here make most sense).

I think based on the feedback here we can draw some reasonable conclusions.

A) Strictly looking at the ability of the sensor to capture the image, when using modern sensors and capturing at base ISO sensitive's or at least ISO sensitive's very close to base ISO, there is virtually no FX benefit in terms of IQ specifically in regards to tonal ability or sharpness or "micro-contrast" of the image. The caveat here is output destination. This argument begins to fall down perhaps when venturing into very large printing where FX may provide some benefit, thought likely would be still insignificant based on normal viewing distances increases as prints become larger.

Lens choice/use will have a far greater impact on the quality of IQ than will the mere size of the sensor.

C) There is no real 100% effective way to do a true apples to apples comparison between the two, though there are ways you can get pretty close.

D) FX advantages do exist and typically come in the form of more flexibility and control over DOF, a clear benefit of higher ISO performance that is typically even more dramatic in low light scenarios. In other words, the higher you climb the ISO ladder the better IQ an FX sensor is going to give you over DX and that gap tends to get larger the higher you go. DX sensors have made up quite a bit of ground in this area, but the differences are evident and not really arguable. There are also tangible benefits not directly related to IQ of FX sensors that play a part in making "better pictures" which includes larger viewfinders, typically better camera builds, better availability of "pro" glass, better pro style controls, etc. Of course DX too than would have advantages depending on your intended needs which might be lighter, smaller body, less expensive prices, higher pixel pitch on the sensor which intern gives that "reach" people refer too, better AF spread, as well as a larger DOF - all else equal - which can be an advantage in many types of photography.

E) The size of the senor, given the literally 100's of equations that make up a completed image actually in the end plays a very minor role in determining the 'quality' of the resulting picture. A larger sensor may give the photographer a little more flexibility in terms of their ability to make that picture but the success or failure of that image wouldn't likely ever be a result of having one sized sensor over another.

Given the above, to answer the question somebody who will ask (as they ALWAYS do) "Should I buy FF or is a crop sensor alright"

I think the answers are maybe not as complex as we make it.
- IF you can afford FX, along with the professional level lenses with it, it's likely to reward you with greater flexibility than DX.
- If you require higher performance while shooting in the ISO 1600 -> 12,800 range, you will have more flexibility and better overall image quality with an FX sensor over a DX sensor.
- If making images with the shallowest of possible DOF is important to you, there is a benefit to FX over DX.
- If you want a larger viewfinder to compose your images in, FX will deliver you this.

In all other aspects, the balance of benefits of FX are rather small but the most important thing people would need to know is that an FX sensor alone does not make for "better pictures" or higher quality pictures over a DX sensor alone. The entire photographic process which includes so many variables and is inclusive of how the image will be presented in the end all play such large factors that this entire process from A to Z has far more impact.

sevencrossing

how many of us print out any pictures bigger than 24 x 16 or so?

I really dont know.

All I can say is. If you do want to make prints bigger than 24 x 16 ( and I do) you should seriously consider a D8xx

If you need, fast focussing and high ISO performance as kyoshinikon found out the D4s is a very nice camera

Both the D4s and the D800 are FX, but that does not mean FX is superior to Dx

It just means until Nikon announce the D400 there are no Nikon dx cameras with the specification of either the a D8xx or the D4s

JonMcGuffin I certainly don't mean to open up the proverbial FX vs DX can of worms

like it or not. The can is well and truly open. With the usual snide remark about D8xx owners and why don't they shoot MF

tcole1983

I don't know. I printed large prints with my d5000 and they looked fine. I have to assume the new 24 MP fx and dx bodies will do just fine also.
Glad I got my piece in early and ran :-)

Last point I will make is for the money I love my d5200. It gave me a quality product for money that I can still eat at the end of the day with. It is good enough that I don't feel like those shooting any other body out now are super far ahead of me in most aspects of what the body is producing. To me cost was one of the most important factors in the dx vs fx debate.

Ironheart

If you look at this purely from a scientific POV, you can see that the best FX sensors (nikon d810, D4s) are 1-2 stops below the theoretical maximum of a FX, and the best DX are 10-15% below the theoretical maximum for DX. I don't know why this is, perhaps manufacturers push the DX harder, but it is true. This is why we are down to1/2 to 1 stop difference. It comes down to personal choice and budget.

donaldejose

I don't have one of the 24mp DX sensor bodies so I cannot comment on their image quality. I still shoot DX with a 16 mp D7000. I have been waiting for either a 24 mp 10 fps DX body (matched with the new 300mm f4 for birding) or a 24mp DX mirrorless body (matched with 28 or 35mm "pancake" type compact lens for street scenes but I would prefer a 24mp FX mirrorless body if Nikon makes it in mirrorless) before I replace my D7000 with the latest 24mp DX sensor technology. I shoot FX with a 24mp D600 and D750 (purchased for its face recognition with OVF and high ISO ability to use for available light or constant light portraiture) and a D800 for landscapes (don't need the 36mp for portraits) so I don't need to have a D7100 or D7200 when I want image quality greater than the D7000 provides. I find I have gone "down sized sensor" to the Nikon 1 system for candid "snapshot" shooting because the small size of the lenses is amazingly convenient (Imagine a camera body in one suit coat pocket and an 80 to 300mm lens in the other suit coat pocket - how convenient is that? - you don't need to carry a camera bag if you chose your jacket and pants carefully) and the IQ quality is good enough for internet, computer monitor or 8x10 printing. I print 16mp DX files up to 16 by 24 inches and 24 or 36 mp FX files up to 24 by 36 inches. I don't see a difference in 24 or 36 mp FX files when printed up to 24 by 36 inches but I do see a difference between 16 mp DX and 24 mp FX when both are printed to poster size which is why I limit my 16 mp DX prints to 16 by 24 inches. I do not know if a 24 mp DX file would be indistinguishable from a 24 mp FX file when both are printed to poster size. I was hoping for some answer to that question by reading this thread but failed to find it. I am disappointed that no one with both a 24mp DX and a 24mp FX body mounted the same 70-200 lens on each of them and shot the same subject composed the same way by adjusting the zoom on the lens. Two such files available to view at 100% would have told us a lot about the state of the art DX vs FX sensor image quality as Jon originally requested. I would expect that if we cannot see a significant difference the latest DX sensor technology would be good now for poster size printing. Please, don't think this thread is over. If someone does have a current generation 24mp DX sensor and a current generation 24mp FX sensor and a 70 to 200 zoom please do the experiment. I want to see that objective data. Until then I will keep relying upon my old rules of thumb based upon my own experience to date: FX is about one stop better at higher ISO and can be printed clean to about one size larger. However, the most recent 24mp DX sensor technology may have closed the gap in ISO up to ISO 3200 or even 6,400 and in prints up to poster size. That is what the comparison Jon suggested could tell us.

I do agree with Jon's generalized summary but I would still like to see the data for the experiment he suggested because I do think the latest 24 mp DX sensor technology has changed the old rules of thumb I have been working with.

sevencrossing

If you look at this purely from a scientific POV, you can see that the best FX sensors (nikon d810, D4s) are 1-2 stops below the theoretical maximum of a FX, and the best DX are 10-15% below the theoretical maximum for DX.

There seems to be a gap in my understanding

Could you explain

1 What exactly is being calculated
2 How the theoretical maximum is derived
3 How is the 1-2 stops below the theoretical maximum worked out

Ironheart

http://home.comcast.net/~NikonD70/Charts/PDR.htm
I am talking ISO vs DR plotted on an XY graph. Plug the numbers in yourself. This is fact. We can't measure IQ, it is too subjective, however this is a pretty good stand-in.

spraynpray

VERY interesting link Ironheart.