Sensor size question

[Dan Brown]

I see Nikon has announced the D7100, DX, 24mpx. Would this DX size interpret to an FX sensor size of approximately 54mpx? 54 X 4/9 = 24 ??

[John Chardine]

Hi Dan- Nikon had already introduced a 1.5 crop 24mp camera in the name of the D3200. You are correct that if this sensor were FF it would be running around 54 mp- pretty amazing!

[Gary Kinard]

If Nikon ever makes a light weight lens like the Canon 500 with the image quality the same. I would be off to Nikon in a heart beat.
Those tiny pixels really give allot of detail. Amazing photos. Just look at the 1DX bird photos where they crop. Very bad compared to a good crop sensor. IMO.

[arash_hazeghi]

Those tiny pixels really give allot of detail. Amazing photos. Just look at the 1DX bird photos where they crop. Very bad compared to a good crop sensor. IMO.


Those tiny pixels give a lot of detail? No, not necessarily. There are many issues in pixel scaling including: diffraction, SNR, spectral response, fill factor, etc. Some of these issues are fundamental and cannot be corrected (e.g. diffraction) but some can be improved with sensor engineering although to a limited degree e.g. fill factor loss, dark current noise, etc. We have to wait and see how the new camera performs. The D7000 sensor is good (slightly better than 7D) and my guess is D7100 should have similar performance, which is great for an amateur model. Just like the D7K, the High ISO performance will be limiting though so it is not suitable for photographers who shoot low light or need fast shutter speeds. You also need the best super-telephoto optics to take advantage of the resolution. Most of the people who look at this model cannot afford the best Nikon super-telephoto lenses...

[Sabyasachi Patra]

I agree with Arash. The lenses are going to be the differentiating factor. I was not seeing that so much when I was clicking stills. Now that I am filming, when I project my film in the big screen, the sharpness in the scenes with and without TCs are becoming evident even with a high quality 400 2.8 lens.

[Jerry van Dijk]

I totally agree with Arash and Sabyasachi. I've been using the D7000 for a few years now and it is very sensitive to lens quality. There is a huge difference in IQ between some of the high end prime lenses I have been using and the somewhat simpeler consumer grade zoomlenses.
At its introduction, the D7000 was practically praised into heaven for it's high ISO performance. I do agree that it produces relatively low amounts of noise compared to some of its competitors at high ISO, but IQ becomes very low from ISO-800 and higher, even when processing from RAW. As a result, the usable ISO range for high quality images is effectively the same as with the competition and nowhere near the FF cameras.
Having said that, the D7000 is a very good camera for its price. The improvements Nikon have made on the D7100 make a lot of sense and will even further improve the camera's great field performance.

[Gary Kinard]

Strange, I have the 7D and just got the 1d mark IV. The difference is very noticeable The 7D resolves much more detail than the 1D mark IV. Not only that the 7D gives so much more detail that it is much easier to de-noise and get usable pics. So there must be a huge difference in sensor quality or something. I am no expert. But the difference between the 7D and mark IV is huge. And really big if cropping allot. I have been looking at the pics coming from the 1DX and for me it is a dog that wont hunt. Just not impressed what so ever. I like detail and the Mark 1V is a compromise for me until a good crop sensor comes out. I personally can deal with the noise.

Quote: Most of the people who look at this model cannot afford the best Nikon super-telephoto lenses... Hummmm I would say BS to that statement. It should read. Most people who look and buy this camera, or any other camera does not want big lenses, it is not there thing. But the 7100 IMO will be a big hit in birding for people who want detail.

www.birdsthatfart.com (http://www.birdsthatfart.com)
www.flickr.com/photos/avianphotos (http://www.flickr.com/photos/avianphotos)

[Roger Clark]

<tt>I'll try and clear up some of the confusion in this thread.
You can also search my previous discussions on this topic
here on BPN. The discussion is around pixel size and
packing more pixels into a given area (like full frame or
a crop sensor size).

First, for the sensor and pixel sizes currently in production,
the fill factors and pixel response are excellent. For example,
the 7D, 1DIV and 1DX all have very close sensor efficiency
in terms of photons recorded per square millimeter.

It is often said that small pixels are noisy, like the 7D.
Not really. It is the lens that delivers the light,
the focal length spreads out the light, the exposure time
limits the light, and the sensor chops up the light.
If one uses a sensor with smaller pixels, thus chopping the
focal plane into smaller pieces, it is no different
than adding a TC to magnify the image to cover the
same pixels on subject, and one will get the same number
of photons per pixel per unit time with both systems.
Scientifically, this is the Etendue of the system.
And diffraction and depth of field will be the same too, assuming
the TC is optically perfect. For more detail on this subject, see:
http://www.clarkvision.com/articles/telephoto.system.performance/

What people usually compare, however, is same f/ratio, ISO,
and exposure time between the large pixel versus the small
pixel system. This is an unequal comparison because the
small pixels get more pixels on subject, and effectively
act as a TC. What do you do when you add a TC? Increase
exposure time (assuming keeping the ISO constant). Well,
it is no different with smaller pixels. Think of it as
a permanent TC in the system. Treat it similarly.

Thus, if one does not have the focal length for a given subject,
then using a camera with smaller pixels effectively gives
you more focal length, and is functionally no different than
the camera with larger pixels plus a TC. Thus, use the exposure
times with the small pixel cameras as you would with the
large pixel camera +TC.

Attached is an example that shows smaller pixels do give images
with more detail. Note the 1DX has larger pixels than the
5DII, thus less detail. In most cases I'll take the smaller
pixels. In principle, for low light simply add pixels together
to get impressive low light performance. For example,
add 2x2 7D pixels together and get the light response of
8.6 micron pixels. (Though in practice, one is limited to
integer factors for adding pixels.)

Another difference between two cameras, one with larger pixels +TC
versus a camera with smaller pixels and no TC, where both
systems give the same pixels on subject, is that the
camera with smaller pixels will be operating the AF system
at a smaller f/ratio. If the AF systems were equal,
the smaller pixel camera will therefore perform better.

Gary, I have and use a 7D and 1DIV. I use the 7D when I need
more reach than I can get with the 1DIV. But the 1DIV AF
system is quite superior to the 7D (plus faster frame rate).
For example, I take a 300 f/2.8 +TCs on many of my trips,
like a recent trip to Africa. I take the 300 for its
lighter weight than my 500 and much faster AF speed.
Very occasionally (once on my Africa trip last month)
I use the 7D +2x +300 for the additional reach
(on some lovebirds). Workes like a charm.

Canon is rumored to be coming out with higher pixel count
cameras to compete with Nikon. I do look forward to
1DX class AF system with ~5 micron pixels (smaller than the
1DIV's 5.7 microns) with 10 frames per second, 16-bit
recording, and no banding noise. That would be
34.5 megapixels full frame.

7D-sized pixels would be 45.4 megapixels (that would be
great too).

Roger</tt>