Why smaller sensors beat full-frame sensors for wildlife photography

[480sparky]

pixel pitch means very little unless you first define how large your pixels are.

Pixel pitch = how large the pixels are.

He means 'quantify the difference between the pitch of two sensor pixels'.

 

[RVT1K]

pixel pitch means very little unless you first define how large your pixels are.

Pixel pitch = how large the pixels are.

The pitch is how far apart they are spaced in the grid that they are arranged in, not how big they are.

 

[Designer]

pixel pitch means very little unless you first define how large your pixels are.

Pixel pitch = how large the pixels are.

I think you are mistaken about your definition of the term; "pitch".

The normal definition is more like;

Dot pitch (sometimes called line pitch, stripe pitch, or phosphor pitch) is a specification for a computer display, computer printer, image scanner, or other pixel-based device that describes the distance, for example, between dots (sub-pixels) on a display screen. In the case of an RGB color display, the derived unit of pixel pitch is a measure of the size of a triad plus the distance between triads.

(from Wikipedia)

 

[RVT1K]

When we expose wafers, the pitch has nothing to do with the size of the feature being written. It is how far apart they are spaced. For example in a grid pattern with 5nm dots on a 25nm grid, the dot size is 5nm and the pitch is 25nm.

 

[freixas]

I think you are mistaken about your definition of the term; "pitch".

Pixel pitch is the distance between pixels, but since sensor designers don't put dead space between pixels, it's also treated as equal to the pixel size by most of the articles I've seen. The pixel is the entire circuit, not just the light collection area. It is the microlens, Bayer filter, support circuitry and whatever else they've placed on it.

In looking for a reference, I ran into this page: http://www.clarkvision.com/articles/does.pixel.size.matter/

It's a long article, but covers the topic I introduced about 10,000 times better. So, go there, read it and we can end the discussion here. A couple of extracts from the Conclusions section:

"Small versus large pixels matter less in modern sensors: with the low noise (read noise and camera electronics noise) available in many of today's cameras (circa 2014+) one can synthesize large pixels from a sensor with small pixels obtaining similar (or even better) performance."

"When choosing between cameras with the same sized sensor but differing pixel counts, times have changed. A decade ago, I would have chosen the camera with larger pixels (and fewer total pixels) to get better high ISO and low light performance. Today I would choose the higher megapixel (thus smaller pixels)."

So, go for a smaller pitch and, if you need the higher image quality of the larger pixels, do it in post.

My other point was that, for wildlife photographers, ignore the crop factor. Look instead at the pixel pitch ratio. A 50 MP FF and 20 MP APS-C have about the same pixel pitch. All other things being equal, the APS-C sensor doesn't provide any crop factor advantage--it just means you have fewer pixels to throw away. For a 20 MP FF and 20 MP APS-C, you do indeed have a 1.6x advantage in resolution.

 

[Original katomi]

someone asked about real life comparisons would a canon 600d ver canon 70d shots of birds be of use here
18 mp ver 20mp
Pics posted on mid flight thread.

 

[freixas]

someone asked about real life comparisons would a canon 600d ver canon 70d shots of birds be of use here
18 mp ver 20mp
Pics posted on mid flight thread.

The difference in pixel pitch is very small (70D: 4.11μm vs. T3i: 4.31μm). I would guess that the majority of the difference would be in the electronics. My ideal comparison would be around a 2x difference, but I'd be happy with 1.2x or so. The Canon 7D Mark IV vs Canon DX1 Mark II would be interesting (1.22x difference and the electronics might be similar). Among the Nikon and Sony cameras, there might be comparable cameras with even larger pitch differences, but I'm not familiar with those lines.

I could post 70D vs 80D, but it would be the same story. Similar pitch and very different electronics.

 

[Derrel]

I alway looked at "pixel pitch" as being number of pixels per millimeter, or "pixel density", rather than pixel SIZE.

I liked Clark's drawing at http://www.clarkvision.com/articles/does.pixel.size.matter/http://www.clarkvision.com/articles/does.pixel.size.matter/

 

[freixas]

I alway looked at "pixel pitch" as being number of pixels per millimeter, or "pixel density", rather than pixel SIZE.

Pixel pitch is indeed the spacing between pixel centers. This is close enough to pixel size for practical purposes and for most discussions, including this one.

 

[coastalconn]

In the real world a typical two camera setup for top wildlife shooters today is the Nikon D850 and the Nikon D500, And while the Nikon D500 is a good camera, to me I personally think that the 850 shots tend to look better

A few years ago several members here compared Their D800's which is 36MP to their D 7200 . As I recall, in all cases the cropped-down D800
files typically looked just a little bit better than the D 7200 photos.

On the Nikon side, the high-end cameras for over 10 years have offered smaller than full sensor captures. For example in 2005 I bought the Nikon D2X which offered a 12 megapixel 1.5 DX sensor,as well as a 2.0 8.2 frames per second high-speed crop mode. In 24 x 36 mm or FX format cameras, Nikon has long offered would it calls the DX crop mode

That is exactly where I ended up. D850 and D500. I'm still in the honeymoon period with the D850 and only around 7k clicks. But from what I can tell, the D850 does hold up better in IQ. The AF still is slightly better on the D500. In good light and fairly low ISO you can not tell a difference between the 2 cameras when you shoot the same subject at the same distance with the same lens and crop both to the same size. The D850 shines though when you would overstuff a DX viewfinder though.

 

[freixas]

hat is exactly where I ended up. D850 and D500. I'm still in the honeymoon period with the D850 and only around 7k clicks. But from what I can tell, the D850 does hold up better in IQ. The AF still is slightly better on the D500. In good light and fairly low ISO you can not tell a difference between the 2 cameras when you shoot the same subject at the same distance with the same lens and crop both to the same size. The D850 shines though when you would overstuff a DX viewfinder though.

The two cameras have almost identical pixel pitches. The D850's pixels are about 1.03x the size of the D850's. Viewed at 1:1, the same object (shot with the same lens, etc.) should appear about the same size. If the D850's images look better, I would point to a difference in the electronics (which includes the sensor design). The D500 was released in 2016, the D850 a year and a half later. Both are listed as "pro"
cameras. Perhaps someone with more Nikon experience could provide more detail on the potential electronic differences in the two cameras.

Because the pixel pitch is so close, there is no point seeing if the smaller pixels yield a superior image when down-sampled.

How about a Nikon D850 vs a Nikon D5 (close to a 1.5x difference)?

 

[coastalconn]

hat is exactly where I ended up. D850 and D500. I'm still in the honeymoon period with the D850 and only around 7k clicks. But from what I can tell, the D850 does hold up better in IQ. The AF still is slightly better on the D500. In good light and fairly low ISO you can not tell a difference between the 2 cameras when you shoot the same subject at the same distance with the same lens and crop both to the same size. The D850 shines though when you would overstuff a DX viewfinder though.

The two cameras have almost identical pixel pitches. The D850's pixels are about 1.03x the size of the D850's. Viewed at 1:1, the same object (shot with the same lens, etc.) should appear about the same size. If the D850's images look better, I would point to a difference in the electronics (which includes the sensor design). The D500 was released in 2016, the D850 a year and a half later. Both are listed as "pro"
cameras. Perhaps someone with more Nikon experience could provide more detail on the potential electronic differences in the two cameras.

Because the pixel pitch is so close, there is no point seeing if the smaller pixels yield a superior image when down-sampled.

How about a Nikon D850 vs a Nikon D5 (close to a 1.5x difference)?

The answer is simple. The D850 is a BSI sensor. However on paper if you view Bill Claff's photons to photos, the D500 has a greater PDF than the D850 or D5 in crop mode.

However I was referring to the title of this thread. I was a die-hard DX shooter however the D850 changed that because I loose 1 FPS and the AF drives my 500/4 and 300/2.8 slightly slower with a slightly lower hit rate, the IQ is better when you have to lift shadows and in general with PP. I always felt the D500 could get a little "crunchy" at times with IQ.

And if your curious what Nikons I have shot with the list is very long starting with a D40 up to a D4 with all D8xx models also. I think my flickr page is still in the link and I tag all images with camera and lens, fyi...

 

[freixas]

The two cameras have almost identical pixel pitches. The D850's pixels are about 1.03x the size of the D850's. Viewed at 1:1, the same object (shot with the same lens, etc.) should appear about the same size. If the D850's images look better, I would point to a difference in the electronics (which includes the sensor design). The D500 was released in 2016, the D850 a year and a half later. Both are listed as "pro"
cameras. Perhaps someone with more Nikon experience could provide more detail on the potential electronic differences in the two cameras.

The answer is simple. The D850 is a BSI sensor. However on paper if you view Bill Claff's photons to photos, the D500 has a greater PDF than the D850 or D5 in crop mode.

Thanks for the insight.

However I was referring to the title of this thread.

The title could use some work. A more appropriate title might be "Why smaller pixels beat larger pixels for wildlife photography--all other things being equal and for smaller pixels that aren't too small".

The D850's pixels aren't really all that big. The thing I learned from the article at ClarkVision (link is now at the top of my OP) is that, if you want bigger pixels, you might be able to do that in post. By choosing smaller pixels, you can opt for either more resolution or less noise. Photon noise is unavoidable--it increases as pixel size decreases. The S/N ratio for the electronics, though, is being raised every year.

Someone like Roger Clark could probably calculate (or maybe has already calculated) the theoretical minimum pixel size (assuming downstream noise is 0) as a function of maximal acceptable noise and available light.

 

[Original katomi]

The tech speak went over my head about five pages ago. I decided that if when I ever change from canon 600d then it would have to be to med format as I have pushed my kit to the limit.. and it’s pushed back

 

[480sparky]

Pixel pitch. Noise. S/N ratio. Crop factor.

Pffft... just grab a camera and go shoot something.