[goatsNdonkey]

Isn't it really that we have to rely on the metering or auto-exposure system of the camera to accurately enough gauge what is getting through to the film or sensor plane, regardless of whether the numbers on the lens are precise? Or if we are using an ISO number and the sunny-16 rule, then we have to hope that any inaccuracies in the numbers on the lens cancel each other out or are close enough?

[Na Horuk]

There's no denying that 1.2 lenses are physically larger, heavier, and with larger diameter elements. You've got to get something for all that extra glass!

See, but I don't get why some lenses with the same focal length and aperture label are bigger than others. Like the Sigma 50mm f1.4, compared to the Pentax FA 50mm f1.4.. front element size is really different. So size is not necessarily faster aperture; and it really depends what you wanted to purchase. If you just wanted a bigger or more corrected lens, or if you want/need that extra light. Another example was given earlier with constant aperture zoom lenses - like the 16-50mm f2.8. How can it have exactly f2.8 when its entrance pupil doesn't change, but its focal length does? Lens designs are getting really complex, so there is wiggle room for weird interpretations, and I think that is what OP wanted to point out

Isn't it really that we have to rely on the metering or auto-exposure system of the camera to accurately enough gauge what is getting through to the film or sensor plane, regardless of whether the numbers on the lens are precise?

Yeah, but if you pay big money for f1.2 brightness, DoF, bokeh... but actually get practically the same as you would get from an f1.4 lens that costs way less.. then its a bit of a ripoff.
Its like if you buying a faster car only to find out its not really as faster as the specifications led you to believe. Then again, I'm sure that happens, as well lol advertising is a real pain in the ass

[Nicolas06]

I don't see use case where f/1.2 vs f/1.4 will make a real difference alone for noise or dof. We speak of at best 0.5EV noise level and around 17% less deph of field (31% is 1 stop and 2 stop is 50%). And that's before factoring in added vigneting, curved light rays that may not make it up to the sensor on some design or even the choice of f/1.2 lens so limited that you may not get the framing or working distance you want and then you see that indeed there no much to it.

What would get me interrested in a lens like that would be the rendering, not just what can be deduced from the math, but how lovely the lens truely render. f/1.2 for f/1.2 is not interresting by itself. It is more that such lenses are typically marketed as premium and tend to have the associated qualities.

[ScooterMaxi Jim]

When it comes to digital photography, the main culprit is the inability of the sensor to accept light efficiently from wide aperture lenses. Past f/2, the ability to receive the added light declines appreciably. See:
https://luminous-landscape.com/an-open-letter-to-the-major-camera-manufacturers/

Also keep in mind that f/1.2 lenses are more complex designs that result in lower t-values (greater light loss due to added elements).

So reception and transmission are the main culprits - typically robbing about half the added light between f/1.4 and 1.2.

[BigMackCam]

I don't see use case where f/1.2 vs f/1.4 will make a real difference alone for noise or dof.

I agree with you, Nicolas... but... how far do we take the comparison? f/1.2 doesn't offer an appreciable difference to f/1.4... then again, f/1.4 doesn't have much over an f/1.8, and that in turn doesn't offer a great deal over f/2 - but, an f/1.2 lens stopped down to f/1.4 might perform marginally better than the f/1.4 wide open, and stopped down to f/1.8 will probably perform a little better again. It should certainly perform better stopped down to f/2 than the f/2 lens wide open... In all cases, that f/1.2 lens lets in progressively more light than the slower lenses, and for some people, with some types of photography, any extra light can make a difference - no?

[Kerrowdown]

I would quickly dispose of this faulty lens !

Aye I thought it was supposed to be an F1.2, not this slow old thing at F1.8.

[Nicolas06]

I agree with you, Nicolas... but... how far do we take the comparison? f/1.2 doesn't offer an appreciable difference to f/1.4... then again, f/1.4 doesn't have much over an f/1.8, and that in turn doesn't offer a great deal over f/2 - but, an f/1.2 lens stopped down to f/1.4 might perform marginally better than the f/1.4 wide open, and stopped down to f/1.8 will probably perform a little better again. It should certainly perform better stopped down to f/2 than the f/2 lens wide open... In all cases, that f/1.2 lens lets in progressively more light than the slower lenses, and for some people, with some types of photography, any extra light can make a difference - no?

You make the hypothesis that faster lens perform better at narrower apperture, this might be true, not always the case and this depend what you mean by better.

As as said, I would consider an f/1.2 lens not because it is f/1.2, that I don't care, but because it would render better, if that the case of course.

But say for sharpness far example, many think their FA50 f/1.7 to be sharper than FA50 f/1.4 for example. DA35 f/2.8 definitely sharper than FA35 and DFA50 give a run for its money to equivalent FA50 f/1.4 and f/1.7. FA31 is sharper also than sigma 30mm f/1.4... The example could go forever.

Now as to when to stop, I would say price, size/weigh, quality.

An example for Canon as there more choice for 85mm focal length:

Mitakon 85mm f/2 = 199$
Samyang 85mm f/1.4 = 269$
Canon f/1.8 = 369$
Mitakon 85mm f/1.2 = 799$
Tamron 85mm f/1.8= 749$
Sigma 85mm f/1.4 = 869$
Sigma 50-100 f/1.8 = 1099$
Zeiss Planar T 85mm f/1.4 = 1283$
Zeiss Milvus 85mm f/1.4 = 1800$
Canon f/1.2 = 1800$
Zeiss Otus 85mm f/1.4 = 4490$

With AF, getting 0.6 stop from f/1.8 to f/1.4 cost 500$, but going from f/1.4 to f/1.2 to get a bit less 0.5 stop cost 1000$. By accepting to give away 1 stop you gain more than 1400$ and still get a lens that perform really really well, a lens that is sharp corners to corners at f/1.8 while the f/1.2 lens need f/4 to match that performance, at least from photozone review...

The f/1.8 lens is 425g 72x72mm and take 58mm filters.
The f/1.2 lens is 1kg, is 91x84mm and take 72mm filters.

For me that would make the f/1.8 lens from Canon a fantastic value and the obvious choice... Couting that anyway the subject, the light, the compo are all much much more important and the lens already perform really well it is quite difficult to justify the alternatives, at least to me.

But everyone can choose for himself.

And the thing is, if you wanted the best, I would not bet on the f/1.2 lens to best performer, even for rendering on the list.

[BigMackCam]

You make the hypothesis that faster lens perform better at narrower apperture, this might be true, not always the case and this depend what you mean by better.

As as said, I would consider an f/1.2 lens not because it is f/1.2, that I don't care, but because it would render better, if that the case of course.

But say for sharpness far example, many think their FA50 f/1.7 to be sharper than FA50 f/1.4 for example. DA35 f/2.8 definitely sharper than FA35 and DFA50 give a run for its money to equivalent FA50 f/1.4 and f/1.7. FA31 is sharper also than sigma 30mm f/1.4... The example could go forever.

Now as to when to stop, I would say price, size/weigh, quality.

An example for Canon as there more choice for 85mm focal length:

Mitakon 85mm f/2 = 199$
Samyang 85mm f/1.4 = 269$
Canon f/1.8 = 369$
Mitakon 85mm f/1.2 = 799$
Tamron 85mm f/1.8= 749$
Sigma 85mm f/1.4 = 869$
Sigma 50-100 f/1.8 = 1099$
Zeiss Planar T 85mm f/1.4 = 1283$
Zeiss Milvus 85mm f/1.4 = 1800$
Canon f/1.2 = 1800$
Zeiss Otus 85mm f/1.4 = 4490$

With AF, getting 0.6 stop from f/1.8 to f/1.4 cost 500$, but going from f/1.4 to f/1.2 to get a bit less 0.5 stop cost 1000$. By accepting to give away 1 stop you gain more than 1400$ and still get a lens that perform really really well, a lens that is sharp corners to corners at f/1.8 while the f/1.2 lens need f/4 to match that performance, at least from photozone review...

The f/1.8 lens is 425g 72x72mm and take 58mm filters.
The f/1.2 lens is 1kg, is 91x84mm and take 72mm filters.

For me that would make the f/1.8 lens from Canon a fantastic value and the obvious choice... Couting that anyway the subject, the light, the compo are all much much more important and the lens already perform really well it is quite difficult to justify the alternatives, at least to me.

But everyone can choose for himself.

And the thing is, if you wanted the best, I would not bet on the f/1.2 lens to best performer, even for rendering on the list.

Good post and reasoning, Nicolas.

As you say, it's a personal choice. For some, ultimate light gathering and/or ultimate DoF will be key - not cost or value-for-money, and perhaps not even sharpness... EDIT: or rendering

It's an interesting debate

---

Last edited by BigMackCam; 03-26-2016 at 04:57 PM.

[osv]

So I always was told that an f/1.2 lens was 1/2 stop faster than an f/1.4 lens (i.e. f/1.19) but it makes me wonder if such lenses aren't really closer to f/1.26 (only 1/3 stop faster than f/1.4). This could explain why so many people report very little difference between f/1.2 and /1.4 lenses in real world tests..

i've seen this lately with a bunch of 50mm 1.4 vs. 1.7/1.8 lenses that i've tested... i could barely get morning exposure on the same scene in at 1/8000th iso50, for all of 'em.

there probably are visible differences in the center exposure, i wasn't paying much attention to it, but what was noticeable was radical differences in wide open vignetting(legacy glass on ff).

so if you consider that 1/2 stop difference in terms of total light across the entire frame, it's probably at least a 1/2 stop, even if it's not so noticeable in the center area.

with slow glass, that light loss due to vignetting sometimes extends well into the aperture range... of course that'll be a lot less noticeable with ff glass on crop sensors.

[Nicolas06]

Slow prime lenses usually need 1 stop, zooms and WA sometime need more.

Fast primes tend to need 2 stops.

But yes if in absolute you need f/2.8 for an f/1.4 lens to have the issue solved, and the slowed lens need f/3.2 then sure it is solved at a narrower apperture. Just that you might have to consider most stop to have issues with a fest lens.

[Nicolas06]

On theses pictures:
- both lenses have terrible bokeh arround the lens box.
- the f1.4 lens look much sharper @f/1.4 than the the f/1.2 lens at f/1.2... we could think of focus error but if we check on the support, the f/1.4 shot are some part much sharper and constrasty than the f/1.2 shot.
- there no much difference is exposure even if one can see the difference... For noise/speed aspect this would be completely negligible difference.