@sounava: the sensors in dslrs like the d3100 or canon 550d use sensors that are actually smaller than that of 35mm film. A camera with a sensor the size of a 35mm film is called a full frame slr.
My statement was only in reference to the photo i mentioned where the magpie was barely a cm high. The cheaper slrs usually lose out on detail owing to the small sensor size, even if it has a high megapixel count, if you have to crop that much.
Cropping, in general, is not a problem no matter what camera you use, be it a point and shoot or an slr.
I know about APS-C and FF sensors. I only asked about the explanation to the statement of yours I quoted above - "Unless you have a full frame slr, there really won't be any point in cropping." Also, I kind of do not agree to what you say here. APS-C does not necessarily mean "cheap". And "entry level" DSLR does not necessarily mean it will lose out on detail because the sensor size in not 35mm. Do remember that D300s which costs 1lakh rupees is an APS-C camera and D5000 and D90 uses the same sensor. Before the D7000 came out that sensor was the best APS-C sensor, a spot now occupied by D7000. D5100 also uses the same sensor as D7000, the the D300s successor will also have the same sensor.
In the Canon field, EOS 7D is one of the best APS-C DSLR ever.
Anyway, moving to something technical -
*www.thinkdigit.com/forum/attachment.php?attachmentid=4446&stc=1&d=1303217570
See the illustration I created above. Suppose for sake of simplicity of explanation you are using an FX lens. You see the image circle which the lens will create and how the FF sensor is positioned edge to edge to the circle. Now imagine using the same FX lens on a crop sensor camera and all other image parameters remain same (position of the camera, focal lenght et al). Then since the sensor is smaller than a FF sensor, a "crop" of the actual image will fall on this sensor. That is why these sensors are called crop sensors and that is why phrases like "equivalent field of view", "focal length multiplier of 1.5" etc comes up.
So what is the net result? In a sense, you are getting a "crop" of whatever image you would have got from a FF.
But look at the bright side - The bird is now occupying a much larger area of the smaller sensor, whereas in the FF sensor, the bird is occupying a much smaller area.
So, the bird will appear larger on the cropped sensor than on the full frame. That is why a 200mm lens is said to give an equivalent field of view of 200x1.5=300mm in a crop sensor.
Now your objective is to crop the picture so that only the bird occupies the whole image. So in a FF you will have to crop more than in a APS-C. But the end result is the same - the bird occupies the whole image.
Q. Ok fine, but where does the number of pixels come into the equation?
Ans: Suppose you are comparing a FF camera with 12MP vs a APS-C camera of 12MP. So density of pixels in the latter sensor is more than in the FF sensor. When you crop to get only the image of the bird in the frame, the crop from the APS-C will ultimately have more pixels than the equivalent one from the FF. Of course comparisons will differ when the sensors have different megapixel count.
What is this hoopla regarding megapixels all about?
In general, more megapixels
should amount to more details in a picture. But that is not always the case. In case of point and shoots, the sensors are "tiny" and by tiny I mean 6mmX4mm. Yup
that small. Now if you pack 14 million pixels into such a small space, the individual pixels will have to be really small. They become so small that they individually can hardly capture enough photons to send "clear" signals of its own to the image processor. There is also some kind of "overlapping" of signals. So the images taken in point and shoots are not "sharp" in the true sense of the word when viewed at 100%. This escalates when the ISO is increased. ISO is nothing like amplifier - It amplifies the signals already received. So too much ISO leads to noise just as too much amplifying a sound signal leads to distortion.
If the sensor is large, the individual pixels are themselves large, they each occupy significant amount of photons and supply enough information to the processor to process the image. Result - clean and nice images with excellent colour rendition and almost no noise. Now this is already true in case of APS-C's. The images are really sharp. Nowadays technology have advanced so much that the new cameras produce noise free images even @ ISO 3200.
Ofcourse, if you use a cheap lens, you will really find that the images are not sharp at all. But this is true when the lenses are used with Full frames as well. Those cheap lenses simply do not have the resolving power to effectively take the advantage of these sensors. In this era, sharpness of an image in a DSLR, be it APS-C or be it FF, depends only on the lens. Period. The technology has advanced that much.
So where is the advantage of FF cameras? Why do they cost a bomb?
1. Stellar high ISO performance for the reasons mentioned earlier. Images taken in ISO 6400 will look like they were taken in ISO 400 with respect to noise.
2. Some really shallow depth of field work. (Can explain about this if asked by someone).
3. High dynamic range of the sensor.
4. Excellent focus systems
5. Weather shielded body
6. Stellar continuous shooting mode (like say 11fps).
7. Pro features (excellent flash commander mode for example).
8. The viewfinder! Once you see with this viewfinder, you will cry when using a entry level camera's viewfinder
9. Sharp images (though it does not mean that APS-C images are not sharp)
10. Many more such features.
Nowadays DSLR users have become so sensitive, that topmost priority of manufacturers is image quality. When they increase the pixel count in a DSLR, they do so only when they are sure that the technology they possess will give better results than the previous generation sensor. In DLSRs, at the present age, more pixels = less sharp or noisier image do not hold true.
But beware of point and shoot - This market is consumer gimmick oriented. They pack more pixels because "aam admi" thinks that more pixels = better camera. But the results come out actually worse when viewed at 100%. Thats why higher end point and shoots like Canon S95 or G12 have not gone for the megapixel spree (16MP) and uses modest megapixels. But in point and shoots also, there can be voices in support of more megapixels - After all, more pixels = more information. More information = better resized pictures, better overall noise reduction. I said noise "reduction" - the camera processor (or any Post processing software like lightroom) will be more efficient in reducing the overall noise of an image the more is the number of pixels. Of course, more number of pixels will mean worse performance with respect to per pixel noise. But when noise reduction is applied the applied result will be better. If all you do is upload the image on Facebook or take small prints like 6x4 inches, you will actually find the newer cameras performing better even thought when viewed at 100% they look definitely worse.
So what is the gist?
1. I do not agree with the statement regarding cropping
2. I pointed out the basic differences between APS-C and FF and the advantages of going towards FF with some references from point and shoots thrown in.
