D7100 vs D700 and a basic full frame question

[posiview]

All things taken into consideration and having gone from D300s to D750, I have to say, overall, the camera places a distant second to the photographer.

I could spout on about reasons, but in reality is the person behind the camera that matters, not the machine in front of him/her.

Bit of a boring answer......take it or leave it

Cheers.

 

[sk66]

Probably means the ISO, DR and colour advantages haven't been taken away.

Well, that would be wrong...
Resolution isn't "exactly" tied to DR... it's more the light sensitivity/full well capacity and technology applied to it.

 

[odd jim]

To see the differences compare the DXO results in screen mode (100% view) instead of the print mode (both full sensors/images printed at 8x10). If the D800 is cropped to DX then the 100% view (and prints) is the same between them.

http://www.dxomark.com/Cameras/Compare/Side-by-side/Nikon-D800-versus-Nikon-D7000___792_680

But you're cropping a better quality image, where the advantages of the larger photo sensors still hold weight, such as better detail, contrast and sharpness, and that's just at base ISO. Add high ISO to that and the advantages of course are still relevant even more so when cropping.

I don't tend to look at DXO though, it's far too lab orientated and their data for me really doesn't add up or bear weight in the real world (this is a site that says my RX100 has better DR than my 6d - absolutely no chance!)

 

[sk66]

But you're cropping a better quality image, where the advantages of the larger photo sensors still hold weight, such as better detail, contrast and sharpness, and that's just at base ISO. Add high ISO to that and the advantages of course are still relevant even more so when cropping.

Not really, between the two the photo sensors (pixels) are essentially the same size (I think the D8xx is actually 15.6MP DX)... Now, the D8xx DX crop *would* have some potential advantages over a D7200 (and a possible negative; resolution) for the reasons you note.

I agree that DXO is "lab oriented"... and a lot of people put way more importance on the measurements/results than they should (and often misunderstand it). But it *is* empirical/standardized/factual...

 

[snerkler]

Well, that would be wrong...
Resolution isn't "exactly" tied to DR... it's more the light sensitivity/full well capacity and technology applied to it.

Why would it be wrong? How I understand it colour depth and dynamic range are largely dictated by the pixel size as well as the tech and processing behind the sensor (which makes the DR of the D810 even more impressive when you consider the pixel size). If you then crop some of these pixels away it doesn't suddenly make the DR and colour depth of the remaining pixels reduce, therefore one would expect a cropped D810 image to 16mp to still hold more DR and colour depth than a 16mp D7100 image.

 

[sk66]

Why would it be wrong? How I understand it colour depth and dynamic range are largely dictated by the pixel size as well as the tech and processing behind the sensor (which makes the DR of the D810 even more impressive when you consider the pixel size). If you then crop some of these pixels away it doesn't suddenly make the DR and colour depth of the remaining pixels reduce, therefore one would expect a cropped D810 image to 16mp to still hold more DR and colour depth than a 16mp D7100 image.

Depends on which color/DR/etc comparison you are looking at... if you compare the equalized (same size print/display, DXO's default report) it is more dictated by sensor size (total light collected). If you compare the sensors at a 100% relationship (DXO's screen report) it is more dictated by the light per pixel.

Equalized works if you do not crop and you use them for the same final result (print/display size). 100% works if you do crop or if you use the higher resolution sensor for larger print/display size.

 

[odd jim]

Why would it be wrong? How I understand it colour depth and dynamic range are largely dictated by the pixel size as well as the tech and processing behind the sensor (which makes the DR of the D810 even more impressive when you consider the pixel size). If you then crop some of these pixels away it doesn't suddenly make the DR and colour depth of the remaining pixels reduce, therefore one would expect a cropped D810 image to 16mp to still hold more DR and colour depth than a 16mp D7100 image.

Agreed. Its akin to printing and (relatively) cropping a 35mm at 10x8 and say 110 film to the same size. The 35mm print would always be better. The size advantages are even more comlex in the digital world.

 

[sk66]

Agreed. Its akin to printing and (relatively) cropping a 35mm at 10x8 and say 110 film to the same size. The 35mm print would always be better. The size advantages are even more comlex in the digital world.

If you "relatively" crop the two size sensors, then yes the larger sensor remains "relatively" larger and has an advantage... that is *not* what I was talking about. I was talking about cropping a larger sensor down to the same equivalent size as the smaller sensor (D800 DX crop).

 

[odd jim]

If you "relatively" crop the two size sensors, then yes the larger sensor remains "relatively" larger and has an advantage... that is *not* what I was talking about. I was talking about cropping a larger sensor down to the same equivalent size as the smaller sensor (D800 DX crop).

But you'll still retain the IQ advantages of the FF sensor (detail, contrast, colour, DR etc), you're just equalising effective pixel resolution give or take. It's only the overall "size" of the image that is affected?

 

[snerkler]

If you "relatively" crop the two size sensors, then yes the larger sensor remains "relatively" larger and has an advantage... that is *not* what I was talking about. I was talking about cropping a larger sensor down to the same equivalent size as the smaller sensor (D800 DX crop).

I agree with odd jim tbh, you are not changing the physical property of the pixels you are merely removing the outer ones.

 

[jamiebonline]

wow this is getting technical I follow it to some extent but .. just a couple of things. I hardly ever crop. If I crop it is, at most, 5 percent of the image, I would say. I don't want huge files on my poor little laptop so even if I could afford a D810, it wouldn't be suitable at all.

interesting about comparisons with DOF. bokeh does concern me but i shot with a D7000 and a 50 1.8 lens when I started taking photography more seriously, almost exclusively and found the bokeh to be ''good enough''. I thought 85 1.8 was even better. I know other factors can be considered like distance from subject and subject from background. I am just coming to an average conclusion.

being able to see the difference between full frame and crop from same generation cameras. i accept it must be true if you have a large display and especially if they are side by side, but 90 percent of my pictures end up on the internet in low resolution being looked at by ''normal'' people on normal monitors or... phones! So it wouldn't be for them. It would only be for me.

low light performance is not so important. again, i am drawing the conclusion from looking at how many times i have shot at very high ISOs and the answer is... never. My highest ISO setting ever was 800. It's just my style. I am mostly at 100. As I said I shoot portraits using natural light, mostly outdoors, some window light. I also take some street photography pics during the day but I am picking up a little Olympus xz-2 secondhand for that. Far more convenient.

 

[snerkler]

wow this is getting technical I follow it to some extent but .. just a couple of things. I hardly ever crop. If I crop it is, at most, 5 percent of the image, I would say. I don't want huge files on my poor little laptop so even if I could afford a D810, it wouldn't be suitable at all.

interesting about comparisons with DOF. bokeh does concern me but i shot with a D7000 and a 50 1.8 lens when I started taking photography more seriously, almost exclusively and found the bokeh to be ''good enough''. I thought 85 1.8 was even better. I know other factors can be considered like distance from subject and subject from background. I am just coming to an average conclusion.

being able to see the difference between full frame and crop from same generation cameras. i accept it must be true if you have a large display and especially if they are side by side, but 90 percent of my pictures end up on the internet in low resolution being looked at by ''normal'' people on normal monitors or... phones! So it wouldn't be for them. It would only be for me.

low light performance is not so important. again, i am drawing the conclusion from looking at how many times i have shot at very high ISOs and the answer is... never. My highest ISO setting ever was 800. It's just my style. I am mostly at 100. As I said I shoot portraits using natural light, mostly outdoors, some window light. I also take some street photography pics during the day but I am picking up a little Olympus xz-2 secondhand for that. Far more convenient.

From what you have said then it won't make a great deal of difference if you go for FF or crop. What might be worth concentrating on is how they render images, how they render colour (even if shooting RAW) and what dynamic range they have.

Also, FF doesn't change DOF per se, it's just to get the same framing as crop body you have to move closer or use a longer focal length which of course alters DOF.

 

[Swanseajack]

wow this is getting technical I follow it to some extent but .. just a couple of things. I hardly ever crop. If I crop it is, at most, 5 percent of the image, I would say. I don't want huge files on my poor little laptop so even if I could afford a D810, it wouldn't be suitable at all.

interesting about comparisons with DOF. bokeh does concern me but i shot with a D7000 and a 50 1.8 lens when I started taking photography more seriously, almost exclusively and found the bokeh to be ''good enough''. I thought 85 1.8 was even better. I know other factors can be considered like distance from subject and subject from background. I am just coming to an average conclusion.

being able to see the difference between full frame and crop from same generation cameras. i accept it must be true if you have a large display and especially if they are side by side, but 90 percent of my pictures end up on the internet in low resolution being looked at by ''normal'' people on normal monitors or... phones! So it wouldn't be for them. It would only be for me.

low light performance is not so important. again, i am drawing the conclusion from looking at how many times i have shot at very high ISOs and the answer is... never. My highest ISO setting ever was 800. It's just my style. I am mostly at 100. As I said I shoot portraits using natural light, mostly outdoors, some window light. I also take some street photography pics during the day but I am picking up a little Olympus xz-2 secondhand for that. Far more convenient.

For what you're saying I would just stay with the D7000 and possibly look to buy more quality glass, e.g.. 35mm f1.8G lens. I have had the D7000 & D7100 and for me there wasn't much between them apart from the cropping of the 24mp files of the D7100. If you look around there are many reports of the people preferring the files from the Sony sensor on the D7000 over the D7100 sensor which is made by Toshiba, me included. Nikon have gone back to Sony for the D7200.

If your really want to give FF a try, buying a D700 or D600/D610 (SH) for that matter, you wouldn't lose much, if any money, if you found they weren't suitable for you.

Myself, I love my D750 and I feel the camera is on a different level to the DX bodies I have owned, in terms of Handling (important to me) ISO, DR, Af etc.........

 

[sk66]

But you'll still retain the IQ advantages of the FF sensor (detail, contrast, colour, DR etc), you're just equalising effective pixel resolution give or take. It's only the overall "size" of the image that is affected?

I agree with odd jim tbh, you are not changing the physical property of the pixels you are merely removing the outer ones.

You're not understanding how the DR/ISO noise/Color/etc are determined/achieved...
Almost every comparison/rating you will find compares the different *full* sensors to generate "the same sized image." And that's fair if that's how you are going to use the camera. This is DXO's "print" comparison (the default view).
And you're not understanding how a high resolution/small pixel sensor achieves comparable results... it's called "oversampling." You take a whole lot more pixels than are required and combine them into a smaller output. In this way a displayed/printed pixel is actually the information from many pixels, and that increases color/DR/etc and reduces noise. The same thing is done in astrophotography where you take many images and stack/combine them for the same type of effect, increased DR/color/etc and reduced noise.

But if you throw those extra pixels away, they can't be combined and the results are less in every aspect. And if you are going to use that higher resolution sensor to do something different, like larger prints for more critical evaluation (large fine art prints), then you also loose the oversampling advantage (the significant factor here is viewing distance/size). This is DXO's "screen" comparison.

If you just follow the link I posted (choose the measurements tab and select "screen") you can see how the D7000 and D800 compare at the pixel/100% level w/o the oversampling advantage. And if you use the D800 in DX mode you have equalized the size/removed the oversampling advantage and therefore the "screen" comparison is *also* (effectively) the "print" comparison.

Look at it this way.... it doesn't matter what size image you start with, if you crop it hard it will be worse in every way. If you crop one, and not the other then the crop one is being put at a disadvantage, or it is being "equalized" if it started with an advantage.