[SunnyOctopus]

I've looked at the charts and read the articles, but I still don't have a clear answer!



Bokeh, focus, and build quality aside, what is the absolute sharpest lens/aperture combo on a crop body Canon camera?



Is it 50mm 1.8 II at 4.5?



Please help! Thanks for your time! ^^



I HOPE A GENIE GRANTS YOU ETERNAL HAPPINESS.

[albi]

[quote name='SunnyOctopus' timestamp='1303537453' post='7859']

I've looked at the charts and read the articles, but I still don't have a clear answer!



Bokeh, focus, and build quality aside, what is the absolute sharpest lens/aperture combo on a crop body Canon camera?



Is it 50mm 1.8 II at 4.5?



Please help! Thanks for your time! ^^



I HOPE A GENIE GRANTS YOU ETERNAL HAPPINESS.

[/quote]



Quite strange question, especially because after a certain level eyes cannot perceive slight differences.

Prime lenses (especially macro lenses) stopped to F/5.6-8 are probably the sharpest ones.

[Rainer]

[quote name='SunnyOctopus' timestamp='1303537453' post='7859']

Bokeh, focus, and build quality aside, what is the absolute sharpest lens/aperture combo on a crop body Canon camera?

Is it 50mm 1.8 II at 4.5?

[/quote]



First of all ... it is certainly not the 50/1.8 (regardless at what aperture).



Second ... the question is relatively irrelevant, since the answer bears no

vissible effect on the resulting images. Quite a number of lenses (including

the 50/1.8) are capable to produce extremly sharp images. What makes an

image really good, however, is more that just the sharpness ... Bokeh,

contrast, flareresistance are at least as important to the result as the sharpness

itself is.



The question for maximum sharpenss is one that beginners often come up with

in order to find out what lens to buy to get good images. But this is really mounting

the horse from the wrong end.



Just my 2cts ... Rainer

[wim]

Actually, the greatets lenses available for FF are sharpest at around F/4, when resolution is diffraction limited. This means that for APS-C, you need to look for lenses that are sharpest at arounf F/2.3 or thereabouts, as that is where they hav ethe potential to be equally sharp at images of the saem size.



However, as others already indicated, this is totally meaningless. Often we want to stop down further, or less, because that is the image we want to create, or we want to use a specific lens for the special way it renders under certain circumstances. And it may well be that that is not at its sharpest aperture either.



HTH, kind regards, Wim

[popo]

Just quickly skimming various reviews here for Canon APS-C, fast primes seem to peak around f/4. The 85/1.2 is the only one I've found so far that has a relatively good performance at f/2.8, but even then it is only comparable to f/4 not better. It is curious that full frame centre results also peak around f/4, but border performance seems to lag a stop or two behind.

[wim]

[quote name='popo' timestamp='1303634001' post='7878']

Just quickly skimming various reviews here for Canon APS-C, fast primes seem to peak around f/4. The 85/1.2 is the only one I've found so far that has a relatively good performance at f/2.8, but even then it is only comparable to f/4 not better. It is curious that full frame centre results also peak around f/4, but border performance seems to lag a stop or two behind.

[/quote]

That's actually inherent to lens design limitations... <img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/biggrin.gif' class='bbc_emoticon' alt='' />



And this has become clear again with 14+ MP APS-C cameras, and 17+ MP FF cameras. The artificial cut-off in resolution plays much less of a factor with the generations of cameras which have such high resolution sensors.



Kind regards, Wim

[popo]

But if you take away the practical limitations, then bigger aperture = more potential resolution. Anyone want a sensor capable of out resolving a diffraction limited f/1.4 lens or faster? <img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/biggrin.gif' class='bbc_emoticon' alt='' />



I just found it curious that in practice, the "sweet spot" for both formats happened to be around f/4 even with their density differences. Assuming diffraction was the limit when stopping down, intuitively you'd think the FF ones would be much further down from the relative pixel sizes.

[wim]

[quote name='popo' timestamp='1303720305' post='7890']

But if you take away the practical limitations, then bigger aperture = more potential resolution. Anyone want a sensor capable of out resolving a diffraction limited f/1.4 lens or faster? <img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/biggrin.gif' class='bbc_emoticon' alt='' />



I just found it curious that in practice, the "sweet spot" for both formats happened to be around f/4 even with their density differences. Assuming diffraction was the limit when stopping down, intuitively you'd think the FF ones would be much further down from the relative pixel sizes.

[/quote]

Not really. If a lens is best at F/4, it really will, or should be, with any medium behind it, whatever the size. There is a physical limitation to resolution in the first place, about 1/2 the wavelength of light used, and there also is diffraction, which sets in sooner with APS-C than with FF, sure, but that is due to the smaller CoC required to provide an equally sharp picture. At F/4, an excellent lens still has a maximum resolution of around 400 to 450 lp/mm at F/4 for air images. At larger apertures resolution is lower (except with highly specialized and extremely expensive optics) due to teh (im)possibilities of correcting optical faults for edge rays well enough, and at smaller apertures it is less due to diffraction. It is impossible to create lenses which are diffraction limited at larger apertures at a price point the market will bear.



And then there is the medium or sensor - system resolution is defined by the lens formula, which involves both lens resolution and medium (or sensor) resolution, where the latter incorporates CoC, and hence automatically incorporates the effect of diffraction caused by smaller sized mediums. The higher resolution of smaller sized sensors is automatically offset by the larger effect of diffraction, resulting in more or less the same sweet spots at the same apertures, with as only difference more (APS-C) or less (FF) DoF. All this, provided, as mentioned above, that the AA-filter does not have an effect, i.e. flatten the MTF curve.



BTW, I'd love a sensor capable of resolving about the same as an F/1.4 lens which is diffraction limited. However, this sensor would have to be as good as modern large MP FF sensors, the lens would have to be not much bigger and heavier than the current crop of lenses (which i tlikely will be), and my computer for processing images would have to be faster by a factor of 1000-10000 in my guesstimates to process the images as fast per image as my current system does for my smallish 25 to 40 MB images now <img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/biggrin.gif' class='bbc_emoticon' alt='' />.



I did work things out a few times, and in reality, I'd settle for less. I can easily do 60 cm X 90 cm prints at amazing quality, in colour, now, which would have been impossible in the past with colour film on a similar analog camera.



Kind regards, Wim