ingcel
wrote:
sherwoodpete
wrote:
Coldamus
wrote:
I don't disagree with that. The large sensor of the Canon is bound to have better high iso performance than the small sensor of the EX1, though I will be surprised if it is enough to make up for the difference at the wide end between f5.8 and f2.4.
What I disagree with is the assertion that the longer focal length of the canon somehow compensates for the slowness of its lens.
If sherwoodpete's logic were correct, then on any particular camera a 300mm lens at f2.8 would give 10 times better performance in low light than a 30mm lens at f2.8. That is clearly not the case. Both produce the same illuminance level and result in the same shutter speed for correct exposure at a given iso. It is only the f-number that matters, not the focal length. Of course the sensor performance matters too but that is another argument.
phototransformations
wrote:
The illumination may be the same, but if you can get, on the Canon at ISO 3200 or ISO 6400, about the same level of noise as on the EX1 at ISO 400 or 800, the Canon may still outperform the EX1, and it will almost certainly produce cleaner images in good light at base ISO. My Panasonic G3, with kit lens, beats my EX1 by at least a stop in low light because it has relatively clean high ISO images even though the largest f/stop is a couple of stops larger than the EX1 at equivalent focal lengths. I would be surprised if the Canon was noiser than the G3.
Apologies, I abbreviated my explanation.
The logic is really that the sensor size dictates the focal length, to achieve the same angle of view. Thus when I referred to focal length previously, the context is the use of two different cameras, having different sensor sizes, and thus different focal lengths, as well as of course the different f-number.
Whichever way you juggle the numbers, noise in low-light situations is dictated mostly by two things:
So far ok
Thus, whichever camera is putting more photons onto the sensor will be the one with the lowest-noise image, provided, as I said previously, the two sensors are of similar efficiency. And the larger physical size of the aperture on the larger-sensor camera can mean that it is capturing more photons than the compact camera.
No. The number of photons is not related to the physical aperture, but the f-number. Consider two systems with the same f-number but different apperture (i.e f-number=2.0 aperture 10mm and 20mm) By definition of f-number the focal lengths will be different (20mm and 40mm respectively) The system with the largest focal legnth will have twice the magnification and therefore will be looking at half the length of the target.
I believe you are describing a different scenario. I assume this is due to a lack of clarity on my part.
In the situation I referred to, an identical portion of the target is captured in each case. This is achieved by keeping the ratio of sensor size to focal length constant. In turn this means the angle of view of each camera will be the same.
To give a real-world example, a 5.2mm focal length on the EX1 and a 24mm focal length on full frame, or a 12mm focal length on four-thirds or 16mm on APS-C will all have the same angle of view.
I'm sure the remainder of your explanation below is coherent and makes sense, but it is describing an altogether different situation. As such, there is no contradiction.
The two effects cancel each other. The largest aperture would collect four times the number of photons - if the photon flux were the same, but as each pixel receives light for one quarter of the area, the photon flux is one quarter and the number of photons in the pixel is the same (That is why the f-number is used and not the aperture).
It makes as little sense to quote just an f-number on its own (without specifying sensor size) as it does to specify focal length alone.
No. It makes a lot of sense to quote only the f-number.
For example, which gives the widest angle of view, a 5mm or a 10mm lens? If the two lenses are used on two different cameras having different sensor sizes, the answer might not be obvious. For example comparing a compact camera against a DSLR, it may be that the 10mm on the DSLR gives a considerably wider view than 5mm on the compact. Or considering telephoto capability, things tend to work the other way around, with the compact having the advantage .
In this case , the field of view depends on the focal legnth AND the detector size. The photons collected in a pixel depend ONLY on the f-number (and the shutter speed, of course).
So, rephrasing what you said above:
Whichever way you juggle the numbers, noise in low-light situations is dictated mostly by two things:
Regards.
That last part was of course off topic here, but was to illustrate that taking individual figures in isolation can be misleading.
Regards,
Peter
Thanks for the interest. It's probably not worth extending this discussion since it is moving in different directions and fragmenting, rather than focussing on the original point.
Regards,
Peter
