D800e

[D.P.]

Depends on which way around you want to think about it. D300 and D700 have the same MP count, yet the D700 has a lower pixel density. Each pixel on the d700 uses a larger area of the lens. As the pixel covers a larger area of the lens, the lens is able to deliver more LP/mm, this effectively increases the resolution that can be obrained from that lens.

Or a short way of putting it, is for any given output size, FF will provide more resolution from a lens than when on a crop camera.

Obviously there are caveats that can affect this, like corner softness etc. but the above is the general rule most of the time.

I agree with this, and is what I was saying but from the other direction.


To illustrate my point, if we take a lens say an 85mm f1.4 stopped down to f/8. Clearly the lens doesn't change when putting it on a full frame camera, it will always output the same resolution image from its rear element.

Now if we create 2 sensors using 100% identical technology so all pixels are equal and identical, and we create 1 sensor the size of a 35mm FF and the other an ASP-C crop. All pixels are the same size and shape, same quantum efficiency, same ADC readout, etc. if we then put black masking tape on both sensors so that only a single area 1x1cm square is exposed on each sensor. if we were to take a photo with the APS-c sensor and the FF sensor we will end up with 100% identical images of the same resolution, the same sharpness, the same detail, the same noise performance, the same magnification, colour depth, DR, FoV, DoF. Hence the FF camera has not magically change any way the lens operates, it has not made the lens become sharper by magic.

That is the only point I was making, but some people (not you) seem to think by magic putting the lens on a FF camera will increase the sharpness, where sharpness is measured at the per-pixel level.



even if we remove the black asking tape the difference between the image would be the FF camera image is wider. If this FF I age was cropped down to give the same FoV as the crop camera then we would end up back with 2 identical images with the same sharpness, resolution and detail. Many people don't seem to understand this.


The FF camera will only gain when trying to photograph the same scene with the same FoV and using a lens with a longer focal length equivalent to the amount of cropping required, so about 130mm compared to the 85mm on crop. If one did use this longer focal length lens on the FF camera with the sensors described as above then the FF camera will offer a higher resolution image of the same scene, the pixel level sharpness may be the same, better or worse depending in the lens, but when printed to the same size will offer greater sharpness and detail.

 

[An Exception]

The FF camera will only gain when trying to photograph the same scene with the same FoV and using a lens with a longer focal length equivalent to the amount of cropping required, so about 130mm compared to the 85mm on crop. If one did use this longer focal length lens on the FF camera with the sensors described as above then the FF camera will offer a higher resolution image of the same scene, the pixel level sharpness may be the same, better or worse depending in the lens, but when printed to the same size will offer greater sharpness and detail.

Yes I agree. I think allot of people get confused by this as there are so many different ways to describe the same thing. I used to find the same thing when talking about DOF differences...

 

[Ksanti]

That is the only point I was making, but some people (not you) seem to think by magic putting the lens on a FF camera will increase the sharpness, where sharpness is measured at the per-pixel level.



even if we remove the black asking tape the difference between the image would be the FF camera image is wider. If this FF I age was cropped down to give the same FoV as the crop camera then we would end up back with 2 identical images with the same sharpness, resolution and detail. Many people don't seem to understand this.

I understand all that perfectly, but the only way you could possibly argue that full frame cameras don't give sharper images through using more of the imaging circle is if you print and display every full frame image at 1.5x the size (for each side ie 2.25x area) of a crop camera. You magnify/crop a full frame camera's image 1.5x less to get to any given display size and as such apparent sharpness is 1.5x greater.

Graph I made a few months ago to illustrate the point. For the sake of argument I said imagine a lens projects 'light pixels' (a hypothetical, arbitrary unit just given by how tight a lens can focus light e.g. 0.2mm) rather than light onto a sensor as it made it easier to demonstrate. If a lens focuses down to 0.2mm 'pixels' then you can fit 21,600 light pixels onto a 35mm sensor versus 10,625 (assuming ASP-C of 25mm*17mm) on a crop camera. Assuming no further cropping, and ignoring any corner softness etc. you then have a situation where whenever you display an image from a crop camera, it will be made using less than half the 'light pixels' you would have used to display a full frame image at that same size:

EDIT: I see we agree on the physics and how the images change, just not on how useful that change is. IMO the points you're making about per-pixel sharpness and equivalent sensor crops are, well, pointless, as there aren't any real world situations other than at the extreme telephoto end where you're going to need to crop so extremely that you go past an ASP-C crop size, unless you're doing something very, very wrong.

 

[An Exception]

^^^
I agree with the above, 1.5x or 1.6x more resolution on FF, which is obviously the best case, but even towards the corners in the vast majority of cases FF is still as sharp or even a fair bit sharper. Also it's not just sharpness, but any lens faults like CA etc. appear reduced on FF, all these little differences can add up to a notable advantage...

 

[D.P.]

I understand all that perfectly, but the only way you could possibly argue that full frame cameras don't give sharper images through using more of the imaging circle is if you print and display every full frame image at 1.5x the size (for each side ie 2.25x area) of a crop camera. You magnify/crop a full frame camera's image 1.5x less to get to any given display size and as such apparent sharpness is 1.5x greater.

Graph I made a few months ago to illustrate the point. For the sake of argument I said imagine a lens projects 'light pixels' (a hypothetical, arbitrary unit just given by how tight a lens can focus light e.g. 0.2mm) rather than light onto a sensor as it made it easier to demonstrate. If a lens focuses down to 0.2mm 'pixels' then you can fit 21,600 light pixels onto a 35mm sensor versus 10,625 (assuming ASP-C of 25mm*17mm) on a crop camera. Assuming no further cropping, and ignoring any corner softness etc. you then have a situation where whenever you display an image from a crop camera, it will be made using less than half the 'light pixels' you would have used to display a full frame image at that same size:


EDIT: I see we agree on the physics and how the images change, just not on how useful that change is. IMO the points you're making about per-pixel sharpness and equivalent sensor crops are, well, pointless, as there aren't any real world situations other than at the extreme telephoto end where you're going to need to crop so extremely that you go past an ASP-C crop size, unless you're doing something very, very wrong.

no, my situation arises all the time. Say you own the 24,35 and 85mm f/1.4 trio of primes and you have a d800 and a d7000 and you wish to photograph a subject at distance using the 85mm prime. The d800 will not provide any increase in sharpness or resolution over the d7000.
A personal example, I own a 300mm f/4.0 lens and a 1.4TC which is the minimum for birds on a crop body really. The different in pixels per feather of the epic battle i photogrpahed last week between an Osprey and a Bald Eagle between a d800 and D7000 would be more or less zero. Actually, the d7000 would provide slightly higher resolution crop. And hence my 12Mp D90 provides much higher output resolution than if I was to buy a 12MP D700, hence I didn't but the d800 will provide an increase in pixel density for wildlife and total resolution for my landscape work.

Comparing 2 different camera using 2 different lenses is not a fair comparison. You need a longer focal length on FF to achieve the higher resolution given equal pixel density, and if you are going to use a longer lens on the FF camera why not use that longer lens on the crop camera, in which case you gain the same resolution and detail over the subject as you need to crop less. Similarly, even if you put the same lens on a ff and a crop camera and claim the final printed image from the FF has a higher resolution is false because it has a wider FOV- crop it to the same FoV as the crop camera and you are back to square 1.


the price difference between a 300mm f/4 and a 500mm f/4 is huge and is a critical factor when considering a FF setup. A d800 won't help a wildlife photographer above a d7000 if they cannot afford the bigger lenses necessary to take advantage of the larger sensor. Having the same grizzly bear have the same resolution and sharpness as the crop camera but with lots of empty space surrounding the subject is not a benefit.




Anyway, this is just one aspect. FF is clearly better than crop if you can normalise differences in lenses and the larger area can benefit light gather capabilities or shallow DoF. But you still need the good quality lenses to get these advantages. If you one of the cheap variable aperture f4 to f5.6 lens on an expensive FF camera then you will get better low light performance using a f2.8 lens on a crop camera, much better performance actually. Even using some thing like the canon 24-105mm f4.0 lens simply means you need to use a canon/tamron 17-55mm f2.8 lens to get similar noise performance and shallow DoF.

Where FF leaves a crop camera in the dust is if you use something like a 24mm f1.4 wide open at close focus distances. You just cannot replicate that on a crop camera. But that is a very special effect that hardly anyone will use and even for pros it is a special trick. One of the horrible things with the rise of DSLRs is the over use of shallow DoF photography when many scenes benefit from an increased DoF, it's like the HDR fad. Just because you can get paper thin DoF doesn't mean you should. Personally I'm usually fitting to get as much DoF as possible before having diffraction or shutter speeds limit me.

 

[Ksanti]

no, my situation arises all the time. Say you own the 24,35 and 85mm f/1.4 trio of primes and you have a d800 and a d7000 and you wish to photograph a subject at distance using the 85mm prime. The d800 will not provide any increase in sharpness or resolution over the d7000.
A personal example, I own a 300mm f/4.0 lens and a 1.4TC which is the minimum for birds on a crop body really. The different in pixels per feather of the epic battle i photogrpahed last week between an Osprey and a Bald Eagle between a d800 and D7000 would be more or less zero. Actually, the d7000 would provide slightly higher resolution crop. And hence my 12Mp D90 provides much higher output resolution than if I was to buy a 12MP D700, hence I didn't but the d800 will provide an increase in pixel density for wildlife and total resolution for my landscape work.

Comparing 2 different camera using 2 different lenses is not a fair comparison. You need a longer focal length on FF to achieve the higher resolution given equal pixel density, and if you are going to use a longer lens on the FF camera why not use that longer lens on the crop camera, in which case you gain the same resolution and detail over the subject as you need to crop less. Similarly, even if you put the same lens on a ff and a crop camera and claim the final printed image from the FF has a higher resolution is false because it has a wider FOV- crop it to the same FoV as the crop camera and you are back to square 1.


the price difference between a 300mm f/4 and a 500mm f/4 is huge and is a critical factor when considering a FF setup. A d800 won't help a wildlife photographer above a d7000 if they cannot afford the bigger lenses necessary to take advantage of the larger sensor. Having the same grizzly bear have the same resolution and sharpness as the crop camera but with lots of empty space surrounding the subject is not a benefit.




Anyway, this is just one aspect. FF is clearly better than crop if you can normalise differences in lenses and the larger area can benefit light gather capabilities or shallow DoF. But you still need the good quality lenses to get these advantages. If you one of the cheap variable aperture f4 to f5.6 lens on an expensive FF camera then you will get better low light performance using a f2.8 lens on a crop camera, much better performance actually. Even using some thing like the canon 24-105mm f4.0 lens simply means you need to use a canon/tamron 17-55mm f2.8 lens to get similar noise performance and shallow DoF.

Where FF leaves a crop camera in the dust is if you use something like a 24mm f1.4 wide open at close focus distances. You just cannot replicate that on a crop camera. But that is a very special effect that hardly anyone will use and even for pros it is a special trick. One of the horrible things with the rise of DSLRs is the over use of shallow DoF photography when many scenes benefit from an increased DoF, it's like the HDR fad. Just because you can get paper thin DoF doesn't mean you should. Personally I'm usually fitting to get as much DoF as possible before having diffraction or shutter speeds limit me.

1) I said quite explicitly that the only situation where you crop so extensively that a crop camera becomes better due to pixel density is at the extreme telephoto end.
2) If you're trying to shoot something at 85mm that needs to be shot at 135mm or you needed to move forward, that's your own fault, not the camera's. 99.99% of the time people are using the appropriate lenses for their format anyway, e.g. people moving from a Sigma 30 1.4 replace it with a 50 1.4 on full frame, not a 35 1.4, so they are using that longer focal length to reap the resolution benefits.
3) No, you don't (need the quality lenses for the benefits). Just look at the comparisons Rockwell did. You might lose some low light performance if the lens is slow enough but all the sharpness etc. still stands.
4) Nor can you replicate the shallow DoF on any crop length. Not a 35 f/1.4 look, not a 50 f/1.2 look, not an 85 f/1.2 look. This is not some really specialised scenario, it applies wherever you might want to use shallow depth of field. In any case, whether you think it's a fad or not, it's popular at the moment and is the most immediately noticeable differentiation between proper portraiture and something taken on a camera phone.

 

[Bluelion]

Plenty of healthy discussion I see, but where are the D800e bodies for sale? I'm itching to make a purchase, but haven't found anyone selling them yet.

 

[xaldub]

Not quite on topic, as it involves the D800 rather than the "E" version, but it looks like Amazon UK has cancelled all its D800 pre-orders this morning. Lots of unhappy folks this fine Sunday morning ! I'm speculating here, but it looks as if Nikon UK has refused to supply Amazon ( especially pre-orders with the lower then MRRP ).

 

[xaldub]

Plenty of healthy discussion I see, but where are the D800e bodies for sale? I'm itching to make a purchase, but haven't found anyone selling them yet.

You can only pre-order them from Nikon registered pro-dealerships. And, as they are being made at approx 8% of the production rate of the D800 it could be a long wait.

 

[D.P.]

1) I said quite explicitly that the only situation where you crop so extensively that a crop camera becomes better due to pixel density is at the extreme telephoto end.
2) If you're trying to shoot something at 85mm that needs to be shot at 135mm or you needed to move forward, that's your own fault, not the camera's. 99.99% of the time people are using the appropriate lenses for their format anyway, e.g. people moving from a Sigma 30 1.4 replace it with a 50 1.4 on full frame, not a 35 1.4, so they are using that longer focal length to reap the resolution benefits.
3) No, you don't (need the quality lenses for the benefits). Just look at the comparisons Rockwell did. You might lose some low light performance if the lens is slow enough but all the sharpness etc. still stands.
4) Nor can you replicate the shallow DoF on any crop length. Not a 35 f/1.4 look, not a 50 f/1.2 look, not an 85 f/1.2 look. This is not some really specialised scenario, it applies wherever you might want to use shallow depth of field. In any case, whether you think it's a fad or not, it's popular at the moment and is the most immediately noticeable differentiation between proper portraiture and something taken on a camera phone.

1) And this is exactly the point I was making which you were arguing against. Pixel density is an important factor in determining the resolution of a subject for a given lens and apparent sharpness of images. A sensor with a lower pixel density needs a lens of relative lower inherent resolving capability to render the same pixel sharpness.

2) If you are changing the lens or other factors then it is is not a fair comparison between sensors. The fact that to get the same reach on a FF camera and to try to exploit the larger sensor requires a longer lens is very important. Someone might own a 70-200mm f/.2.8 lens on a crop as a modest wildlife lens and have some success, especially combined with a 1.4xTC. If the same person upgraded to a FF camera their 70-200 is going to be far too short for them, the cost of a 300mm f/2.8 is a massive problem, let alone its size and weight.

3) I never said anything about image quality. I was only referring to the aperture. An f/5.6 lens on a FF camera will give much worse low light performance than an f/2.8 lens on a crop camera, and much less ability for shallow DoF. This is an important point, someone might want to 'upgrade' from a 17-55mm f/2.8 lens on a crop sensor to a 24-105/120mm f/4.0 on a FF camera, in doing so they have lost the 1 stop of low light performance and end up getting the same ability for shallow DoF.

4) DoF is as much a function of the distance to the focal length as it is the aperture of the lens. If you are going to be changing the lens or the camera position on a FF camera then you can do the same on a crop camera to replicate most of the same shallow DoF effect. A 50mm lens shot at f/1.4
at a subject 10 feet away gives a DoF of 1ft, a 35mm lens on a crop body at f/1.4 will give the same 1ft of DoF at a distance of 8.5 feet. Equivalence is typcially easy to achieve, especially if on a FF you are shooting at f/2.8 As I said previosuly, the one area that is impossible to replicate is a lens like the 24mm f/1.4 shot at 1.4 at close focal distance. There just isn't a fast wide prime in existence.

If the only reason you don;t use a camera phone is to get a shallow DoF then I feel sorry for you and I would be very disjointed in your camera and lens setup. The large prints of sweeping landscapes, feeding grizzly bears, and the rack of a huge bull moose against the setting sun is justification for me.

 

[D.P.]

Not quite on topic, as it involves the D800 rather than the "E" version, but it looks like Amazon UK has cancelled all its D800 pre-orders this morning. Lots of unhappy folks this fine Sunday morning ! I'm speculating here, but it looks as if Nikon UK has refused to supply Amazon ( especially pre-orders with the lower then MRRP ).

It was an error by Amazon apparently.

 

[D.P.]

Plenty of healthy discussion I see, but where are the D800e bodies for sale? I'm itching to make a purchase, but haven't found anyone selling them yet.

The E model will be in tight supply for a number of months, bodies will continue to trickle through. I think you just have to be very vigilant and as soon as you see one in stock order there and then.

 

[Bluelion]

Sheesh just read the update on Nikon-rumours about the jungle company cancelling the UK orders. Perhaps the D4 will be a little easier to get hold of after all, at its ridiculous price :/

 

[An Exception]

^^^
I thought that article had been updated to say it was an error with the jungle companies system, and all orders are back on?

 

[peter]

Having some time off to try it out ( results with high iso's are fantastic ) going for a walk with it now..

 

[Rojin]

Nice, enjoy

 

[D.P.]

very nice, very jealous!

 

[An Exception]

It's a very small thing to concentrate on to just spot meter or ETTR, photography is hardly the most mentally taxing of pursuits that humans can go through. It's just instinctive now, even if I was shooting on a d800 or D7000 I would do the same nowadays.

The only time I've ever clipped highlights to any meaningful degree has been in studio-style shots with black backgrounds where I can go back and reshoot. Location stuff I've never had a problem with exposures being blown out, nor shadows being too noisy.

Below is a DR comparison that may (may not) interest you.
http://www.digifotopro.nl/content/canon-5d-mark-iii-vs-nikon-d800-dynamisch-bereik

Basically with the D800, you can pretty much toss out your ND filters...

 

[D.P.]

Also here:
http://www.fredmiranda.com/5DIII-D800/index_controlled-tests.html

There is just no comparison, the DR of the D800 is phenomenal.

 

[Ksanti]

Below is a DR comparison that may (may not) interest you.
http://www.digifotopro.nl/content/canon-5d-mark-iii-vs-nikon-d800-dynamisch-bereik

Basically with the D800, you can pretty much toss out your ND filters...

That's interesting and impressive, and I'm not saying for a second that the 5D3 is better than the D800 (were it not for the mounts I'd choose a D800 any day), but tbh the only time I ever use ND filters is my 9 stopper, though I do appreciate that many people use Grads etc. a lot more than myself. Again, with my shooting the majority of stuff where that sort of dynamic range is useful is when I get the exposure wrong in camera on manual stuff.

The fact for me is that the 5D3/D800 generation of FF cameras are so far off from being buying propositions for me that I may well be considering a move to Sony if I do move away, given the fantastic 135 and 85, if they end up using the D800 sensor in the a99.