D800E's power to resolve.

[Geoffc]

In reality my original point was not really about pixel pitch, it was about pixels actually being used. If you take a D800 and a D7000, they both have a very similar number of pixels covering the DX area of the sensor. If we're being picky the D7000 has the edge as it has more and as such could resolve detail at a more granular level. If you look at the original uncropped image of the bear it was a lot smaller than the DX area so any pixels in the area between DX and FX don't count. It doesn't matter if you shoot a bear in FX mode with lots of space around it as we are throwing all of the space part away when we crop. I'm still waiting for an answer to my question regarding the dimensions of the cropped image. . The DR and colour depth of the D800 is something else and not what I call resolution. Now when I get a bear filling my D800 frame, that's what you call serious resolution.

Anyway I think I'll drop this now as I appear to have caused upset by saying the D800 is not the be all and end all when it comes to megapixels.

 

[Scott Murray]

I am just unsure how we got from the original post of Silven posting a bear taken with his D800 onto this debate. I am sure you have not upset anyone me included but I for one do not see the need to debate something like the pixel density or what ever in this thread as that's not why it was started in the first place. Surely we can have another so we can debate it there.


Sent from my iPad using Tapatalk HD

 

[Geoffc]

I am just unsure how we got from the original post of Silven posting a bear taken with his D800 onto this debate. I am sure you have not upset anyone me included but I for one do not see the need to debate something like the pixel density or what ever in this thread as that's not why it was started in the first place. Surely we can have another so we can debate it there.


Sent from my iPad using Tapatalk HD

Scott

We got here because I responded to the original post by saying that a D7000 would have achieved the same result. I was only highlighting that the D800 is only lots of pixels/resolution compared to other cameras if you actually use all the pixels.

I actually own a D800 so I'm hardly questioning that it's a good camera.

The pixel density wasn't an issue for me, it was just a way of highlighting that the D800 and D7000 sensors have similar numbers of pixels per square millimetre.

I will post some proper high res comparisons in a few days so we judge for ourselves how the D300/800/7100 images compare. Unfortunately I don't own a D7000 anymore as I would have liked to include that as well.

 

[Ironwood]

I think all the newer cameras are pretty amazing. I am happy with my D7100 at this point in time. Would I swap it for a D800/E ? You bet I would. But I dont have the spare cash to justify buying one, but that may or may not change down the road.

​Every time I zoom in on one of the images from my D7100 I am amazed at the detail it captures, providing the focus is spot on of course. I am glad it does, because it is helping me resist the urge for a FX camera.

 

[Geoffc]

I think all the newer cameras are pretty amazing. I am happy with my D7100 at this point in time. Would I swap it for a D800/E ? You bet I would. But I dont have the spare cash to justify buying one, but that may or may not change down the road.

​Every time I zoom in on one of the images from my D7100 I am amazed at the detail it captures, providing the focus is spot on of course. I am glad it does, because it is helping me resist the urge for a FX camera.

Everyone seems to view FX as an upgrade but i have mixed feelings. Unless you can afford long fast glass FX may not be an upgrade depending on what you're doing. In fact if you're a sports/action shooter the frame rate of the D800 may not be up to the job. When I press the trigger on my D300s it fires in raw at 6-7 fps until I get bored of the noise. The D800 doesn't. When I want to shoot high res landscape, portraits or high ISO low light shots the D800 wins hands down. It's the old thing of the right tool for the job.

My wife got the 7100 a few weeks ago but it's too early yet to know what we think of it compared to the D300 or D800. I suspect it will be better at some things and not at others. Apart from the buffer it seems like a very nice camera and most importantly it's got a decent autofocus system. It's also soooo quiet when the shutter operates. Great for wedding ceremonies. The D800 sounds like a door slamming!!

 

[Silven]

I wasn't upset at all, not in the least. I was however engaged and very curious as to your reasoning. So I did some re-search and then I did some more. I'm still not "getting" your reasoning for saying that if I crop the the D800/E image to DX size,(ie, 1.5 crop) that the D7000 or D7100 would have as good or better IQ. The D800/E has a sensor that is 2.35 times larger. It surpasses the D7100 in every test from ISO to Dynamic range. If you crop it after shooting by 1.5 and not in camera, you will have more and larger pixels then the D7100. You'll forgive me but I just don't understand why why why the D7100 should theoretically be better if you crop the D800/E to the same size.

This is an example of what I found on-line. "True resolution is effected by the physical restrictions of sensor size, and it may differ from the megapixels announced by manufacturers. Nikon D800 has 51% higher resolution" That was at http://kiyas.la/en/digital-camera/compare/Nikon-D800-vs-Nikon-D7100

Maybe I'm pretty dense (actually, probably, most likely.) but I just don't get it.:bi_polo:

Roy1961, I hear what your saying about getting a prime 300 or better lens. I've been spending my money on shorter less expensive primes. I don't do wildlife photography for a living nor do I sneak around trying to find some Hollywood so and so's sunbathing nude on some secluded beach from miles away so I can't really justify the $8000 for the prime I'd love to have. If I won the lottery I'm pretty sure I'd go nuts but until and if that ever happens I'm gonna have to sneak up on my wild subjects. I do have a Sigma 50-500 but I'm sure that's light years from a Nikon glass in the same length. I haven't had a chance to try it on my D800E as it's still out for repair. I damaged some pins so it wouldn't talk to my camera body.

 

[Geoffc]

I wasn't upset at all, not in the least. I was however engaged and very curious as to your reasoning. So I did some re-search and then I did some more. I'm still not "getting" your reasoning for saying that if I crop the the D800/E image to DX size,(ie, 1.5 crop) that the D7000 or D7100 would have as good or better IQ. The D800/E has a sensor that is 2.35 times larger. It surpasses the D7100 in every test from ISO to Dynamic range. If you crop it after shooting by 1.5 and not in camera, you will have more and larger pixels then the D7100. You'll forgive me but I just don't understand why why why the D7100 should theoretically be better if you crop the D800/E to the same size.

This is an example of what I found on-line. "True resolution is effected by the physical restrictions of sensor size, and it may differ from the megapixels announced by manufacturers. Nikon D800 has 51% higher resolution" That was at http://kiyas.la/en/digital-camera/compare/Nikon-D800-vs-Nikon-D7100

Maybe I'm pretty dense (actually, probably, most likely.) but I just don't get it.:bi_polo:

Roy1961, I hear what your saying about getting a prime 300 or better lens. I've been spending my money on shorter less expensive primes. I don't do wildlife photography for a living nor do I sneak around trying to find some Hollywood so and so's sunbathing nude on some secluded beach from miles away so I can't really justify the $8000 for the prime I'd love to have. If I won the lottery I'm pretty sure I'd go nuts but until and if that ever happens I'm gonna have to sneak up on my wild subjects. I do have a Sigma 50-500 but I'm sure that's light years from a Nikon glass in the same length. I haven't had a chance to try it on my D800E as it's still out for repair. I damaged some pins so it wouldn't talk to my camera body.

Silven,

I'm glad I haven't upset you as that was not my intention. Can I make a few points. Firstly your original post was talking resolution not IQ. This has crept in during the posts. I may have been guilty of referring to it as well.

Secondly I referred to the D7000 not D7100 in my original post very deliberately due to the pixel density similarity.

Thirdly the FX sensor may be 2.35 times larger but you're only using a teeny bit of it in your bear pic. For example if you employ 5sq mm of your FX sensor for the bear and 5 sq mm of the D7000 sensor you have about the same number of pixels/resolution. Also a lot of the D800 high IQ scores are as a result of down sampling for comparative tests not just because it has wonderful pixel technology. The 800 has pixels in abundance and noise is lost in this process.

I'm going to be as good as my word and do a proper documented test that will allow everyone to objectively decide what they think the answer is. Maybe I'm just not being clear enough in my explainations.

 

[Scott Murray]

Here is a test thats already done for thsoe that are interested D7100 vs D800E - Nikon D7100 vs. Nikon D800E - Sensor Comparison

 

[papa2jaja]

I understand where Geoff comes from but I wonder, since the D7000 is DX, if the same area of the world is photographed (same size in the real world), wouldn't it occupy a larger area (not relative to the sensor but in terms of absolute dimensions) on an FF sensor than on a DX sensor, simply because the FF sensor is larger?

If this is true, wouldn't an identically sized crop contain the same number of (or maybe even more) pixels on the D800's FF sensor as on the DX sensor? And even if this cropped area on the DX sensor would contain a comparable number of pixels, since the pixels on the DX sensor are smaller, wouldn't this very fact (the smaller pixels) cause the DX pixels to be able to capture less colour nuances as the larger pixels of the D800's FF sensor? With the consequence that the DX sensor's image would have less nuances of colours and hence less visual detail?

A confusing topic, apologies for jumping in with my layman's questions.

 

[aroy]

In my opinion if you are able to fill the FX sensor area with object of your interest (in this case the bear), then D800 will score over all the DX format cameras, due to its higher MP. As such the D800 will be as good or better if the crop area contains more than 24MP. It is only when the image is tiny and occupies a small portion of the FX sensor area, that DX cameras offer better utilization.

The idea that the DX format multiplies the focal length by 1.5 has come from the notion that given same MP - D3200 and D600, the DX camera will have a larger image (same distance, same lens), because it is smaller, while the image due to the lens is same, hence you get an illution of more magnification.

Another point to note that for same technology, bigger pixels will have better DR and smaller S/N ratio. As per DXO tests both the D800 and the D600 are much better on these parameters compared to the competing DX sensors.

What it comes down to is what has been advised - ensure that the area of interest fills up your view finder (thus the sensor/film). Unless you can fill the sensor with area of interest, you are wasting the sensor real estate.

 

[papa2jaja]

So what it would take to even things out for small objects is for FF to use a lens with a longer focal range?

 

[Geoffc]

So what it would take to even things out for small objects is for FF to use a lens with a longer focal range?

​Yes. About 50% greater FL when comparing D800 and D7100. Don't confuse this with crop factor, it's just the pixels per mm difference between the two.

 

[aroy]

I understand where Geoff comes from but I wonder, since the D7000 is DX, if the same area of the world is photographed (same size in the real world), wouldn't it occupy a larger area (not relative to the sensor but in terms of absolute dimensions) on an FF sensor than on a DX sensor, simply because the FF sensor is larger?

If this is true, wouldn't an identically sized crop contain the same number of (or maybe even more) pixels on the D800's FF sensor as on the DX sensor? And even if this cropped area on the DX sensor would contain a comparable number of pixels, since the pixels on the DX sensor are smaller, wouldn't this very fact (the smaller pixels) cause the DX pixels to be able to capture less colour nuances as the larger pixels of the D800's FF sensor? With the consequence that the DX sensor's image would have less nuances of colours and hence less visual detail?

A confusing topic, apologies for jumping in with my layman's questions.

The image projected on the sensor is of the same size irrespective of the sensor size for a given lens. The extent is limited by the image circle of the lens. So if an projected image is 24mm wide it would be of the width of a DX sensor, while it will fill up only 24/36 width of an FX sensor. (I am taking the FX sensor to be 24x36mm and the DX sensor to be 24x16mm - the actual dimensions are 23.5x15.6mm for D7100 and 35.6x24mm for D600). There fore for same MP in each sensor (24MP for example) an image of 15mmx15mm in the centre will look bigger and have more pixels in DX.

The DR and colour range are directly related to the pixel size. As FX has larger pixels for the same MP, FX will have better DR, better colour range and lower S/N ratio. With modern technology the difference; though there; is quite small and for all but highly critical work can be ignored.

As discussed in many articles, the peak FPS is today limited by the CPU performance rather than the shutter, so lower MP cameras can have higher FPS compared to higher MP cameras.

 

[Silven]

The DR and colour range are directly related to the pixel size. As FX has larger pixels for the same MP, FX will have better DR, better colour range and lower S/N ratio. With modern technology the difference; though there; is quite small and for all but highly critical work can be ignored.

While this statement is generally true, it doesn't hold fast in all aspects. When put through all of DXO's testing Nikons newest 2 FX sensors were superior to everyone else's. Yes I know no one has the exact Mega pixels to compare apples to apples but the actual brains of the camera is it's processor. The sensor is just the eyes. If the brain can't interpolate what it sees to it's maximum potential then IQ won't be all it could be. While pixel size affects the dynamic range, dynamic colour range, and noise, just having larger pixels does not automatically guarantee better IQ. I personally feel that the D7100 and the D800/E have the same sensors. The D7100 just has a cropped version of the D800/E's. That's why when you crop the D800/E's sensor mathematically you basically come up with the D7100. It would make sense from a strictly business point of view too. Why develop two separate sensors at an enormous cost when you can downsize your flagship sensor to fit 3 other price point models and still keep your stellar reputation for high IQ? I know that the automobile industry for example do this a lot. I saw a stat once on Motor Trend magazine that a certain North American car company had vehicles in it's line up that shared as much as 68% of their parts. As in they were the same part, same part number etc etc.

What I still don't understand is why people assume the D7100 or D7000 for that matter would have the same or better IQ as the D800/E if you only used a tiny part of their respective sensors. This makes no logical sense to me.

 

[Geoffc]

While this statement is generally true, it doesn't hold fast in all aspects. When put through all of DXO's testing Nikons newest 2 FX sensors were superior to everyone else's. Yes I know no one has the exact Mega pixels to compare apples to apples but the actual brains of the camera is it's processor. The sensor is just the eyes. If the brain can't interpolate what it sees to it's maximum potential then IQ won't be all it could be. While pixel size affects the dynamic range, dynamic colour range, and noise, just having larger pixels does not automatically guarantee better IQ. I personally feel that the D7100 and the D800/E have the same sensors. The D7100 just has a cropped version of the D800/E's. That's why when you crop the D800/E's sensor mathematically you basically come up with the D7100. It would make sense from a strictly business point of view too. Why develop two separate sensors at an enormous cost when you can downsize your flagship sensor to fit 3 other price point models and still keep your stellar reputation for high IQ? I know that the automobile industry for example do this a lot. I saw a stat once on Motor Trend magazine that a certain North American car company had vehicles in it's line up that shared as much as 68% of their parts. As in they were the same part, same part number etc etc.

What I still don't understand is why people assume the D7100 or D7000 for that matter would have the same or better IQ as the D800/E if you only used a tiny part of their respective sensors. This makes no logical sense to me.

Silven,

Shrinking the D800 to DX does not end up with the D7100 it's nearer the D7000. Also, this debate keeps drifting onto IQ when it was supposed to be about resolution. Whilst I appreciate that the IQ of a 4mm point and shoot sensor might impact resolving power, it just doesn't have that impact with an APSC like the 7100.

To keep the whole thing on track, go back to your original bear and measure how many megapixels your cropped image was. If the answer is less than 16 mp I think my original point stands. From those images I can't imagine it would be bigger.

At least this whole thing has got everybody thinking which is always a good thing

 

[Dave_W]

If we're just talking resolution, I saw an ad for a Nokia 808 cell phone with a 41 mp camera in it. Does that mean it's "better" than a D7100? I'd have a hard time saying yes based solely on resolution. Frankly, resolution is one of the least important aspects to the overall performance of a camera. In fact, it's not unlike comparing the different sizes of the gas pedal of a Maserati vs. a Honda Civic and drawing conclusions form the results. But hey, that's just me.

 

[Silven]

If we're just talking resolution, I saw an ad for a Nokia 808 cell phone with a 41 mp camera in it. Does that mean it's "better" than a D7100? I'd have a hard time saying yes based solely on resolution. Frankly, resolution is one of the least important aspects to the overall performance of a camera. In fact, it's not unlike comparing the different sizes of the gas pedal of a Maserati vs. a Honda Civic and drawing conclusions form the results. But hey, that's just me.

I agree cause the Honda pedal would win every time right?
​

 

[BackdoorArts]

I've posted this link several times before with regard to DX vs. FX and how it relates both to resolution and other IQ factors, but it bears coming back to when these discussions go this way and that. There are many factors to take into account when discussing the differences between the two sensor types and how pixel density impacts IQ. Resolution, which is what we are discussing, is one, and the D7100 wins that battle - it's purely a pixel count thing. But as good as that sensor is, there are disadvantages that come with the DX crop factor that make the D800 operating at 16MP's a potential winner in many other IQ arguments, including high ISO, low light, under/over exposed image correction, etc. A pixel's job is to capture light, and the smaller it is the more difficult it is to do it in extreme circumstances. That's why it's important to get the right tool for the right job.

DX or FX for Sports and Wildlife Photography

 

[Geoffc]

If we're just talking resolution, I saw an ad for a Nokia 808 cell phone with a 41 mp camera in it. Does that mean it's "better" than a D7100? I'd have a hard time saying yes based solely on resolution. Frankly, resolution is one of the least important aspects to the overall performance of a camera. In fact, it's not unlike comparing the different sizes of the gas pedal of a Maserati vs. a Honda Civic and drawing conclusions form the results. But hey, that's just me.

Dave,

Did you deliberately ignore my point about the 4mm sensor?

 

[joenmina1]

IMHO, the end game of successful cropping is to maintain IQ and DR with minimal noise. I too have a D7000 and a D800. While the D7000 is a great camera, it is not even in the same zipcode as the D800 in terms of IQ, DR or noise control...not to mention autofocus speed/accuracy. Almost all the pics in my gallery are 100% crops and I wouldn't have a snowball's chance in hell of getting that quality with my D7000, crop factor notwithstanding. Again, IMHO.