Why aren't zoom lenses faster than 2.8?

[Sporgon]

Reading these last few posts reminds me of how surprised I am that my 135/2 isn't larger, especially compared with my 200/2.8

[infared]

There are now:
http://www.dpreview.com/news/2013/04/18/Sigma-announces-worlds-first-F1-8-constant-aperture-zoom-lens

Equivalent of f2.7 FF on a Nikon 1.5x crop, or f2.9 FF on a Canon 1.6x crop. So on Nikon it is faster than an equivalent of f2.8 FF, but not by much. I'd still prefer a Canon 24-70 II on FF, even if it is 'only' f2.8

Agreed...I only shoot FF and MFT...but it is great to see Sigma pushing the envelope.....and if it has great IQ it will be an incredible lens for crop sensor cameras!

[jcns]

200-500 2.8
simply
http://gizmodo.com/5217989/the-photographer-king-needs-no-tripod
and only 35lbs.
you want 1.8 with that?
you want IS with that?
have you been curling 45 lb dumbbells?

[neuroanatomist]

Reading these last few posts reminds me of how surprised I am that my 135/2 isn't larger, especially compared with my 200/2.8

135 / 2.0 = 67.5
200 / 2.8 = 71.4

Only 4mm different...

[woollybear]

Reading these last few posts reminds me of how surprised I am that my 135/2 isn't larger, especially compared with my 200/2.8

135 / 2.0 = 67.5
200 / 2.8 = 71.4

Only 4mm different...

Is that different than 67,5 and 71,4?
Easy now....just adding a little humour.

[RLPhoto]

Muhahaha, If this is possible, I believe uber-fast zooms are possible.

http://www.43rumors.com/the-need-for-speed-zoomatar-75mm-180mm-and-250mm-all-f1-3/

[Sporgon]

Reading these last few posts reminds me of how surprised I am that my 135/2 isn't larger, especially compared with my 200/2.8

135 / 2.0 = 67.5
200 / 2.8 = 71.4

Only 4mm different...

Yes now you mention it I can see that based on the same maths the 200/2 and the 300/2.8 are similar diameter.

But if the formula is as simple as focal length / f stop, why does my 135/2 vignette much more at f2 than my 200/2.8 does at 2.8 ?

[K3nt]

Reading these last few posts reminds me of how surprised I am that my 135/2 isn't larger, especially compared with my 200/2.8

135 / 2.0 = 67.5
200 / 2.8 = 71.4

Only 4mm different...

Yes now you mention it I can see that based on the same maths the 200/2 and the 300/2.8 are similar diameter.

But if the formula is as simple as focal length / f stop, why does my 135/2 vignette much more at f2 than my 200/2.8 does at 2.8 ?

The simple formula is just that.. simple and a guideline. A lot more goes on in the world that is precision optics that can produce(or not) the vignetting you're experiencing.

[Sporgon]

Reading these last few posts reminds me of how surprised I am that my 135/2 isn't larger, especially compared with my 200/2.8

135 / 2.0 = 67.5
200 / 2.8 = 71.4

Only 4mm different...

Yes now you mention it I can see that based on the same maths the 200/2 and the 300/2.8 are similar diameter.

But if the formula is as simple as focal length / f stop, why does my 135/2 vignette much more at f2 than my 200/2.8 does at 2.8 ?

The simple formula is just that.. simple and a guideline. A lot more goes on in the world that is precision optics that can produce(or not) the vignetting you're experiencing.

My questiion is essentially, what goes on ?

So to reply 'a lot more goes on' is hardly an answer.

I suspect it has to do with the physics of passing light through a group of lenses: the closer the design comes to perfect 100% transmission of the light value the more disproportionate the fall off at the extreme of the image circle.  So to avoid more vignetting the faster lenses would have to produce a larger image circle, which would increase size, weight and cost in a similar vein to the OP's original question on zooms.

So given that the 135/2 and 200/2.8 are of a similar design and price, I should have expected the faster lens to have more vignetting when wide open.

[ksagomonyants]

http://www.canonrumors.com/2013/04/sigma-announces-18-35-f1-8-dc-hsm-art-for-aps-c/

[Lawliet]

But if the formula is as simple as focal length / f stop, why does my 135/2 vignette much more at f2 than my 200/2.8 does at 2.8 ?

Basically: The lenses have the same diameter, but at longer focal lengths they need less curvature.  The latter is the cause of most flaws in a lens. More a problem with wide angle lenses would be the cos^4-law, i.e. the kind of vignetting that comes from rays hitting the aperture from far off the optical axis. Think of a disc, if you look at it from lets say 45° it looks like a cats eye, less area to get light through then the full circle (or just look at bokeh-discs, round in the center, recompose and they get squashed)

[jrista]

Muhahaha, If this is possible, I believe uber-fast zooms are possible.

http://www.43rumors.com/the-need-for-speed-zoomatar-75mm-180mm-and-250mm-all-f1-3/

Hah! That is one crazy lens! I wonder what the IQ was like wide open...that thing must have had some beastly CA!

[jrista]

Reading these last few posts reminds me of how surprised I am that my 135/2 isn't larger, especially compared with my 200/2.8

135 / 2.0 = 67.5
200 / 2.8 = 71.4

Only 4mm different...

Yes now you mention it I can see that based on the same maths the 200/2 and the 300/2.8 are similar diameter.

But if the formula is as simple as focal length / f stop, why does my 135/2 vignette much more at f2 than my 200/2.8 does at 2.8 ?

The simple formula is just that.. simple and a guideline. A lot more goes on in the world that is precision optics that can produce(or not) the vignetting you're experiencing.

Just to be specific, the relative aperture as a ratio couldn't be called a "guideline". It is a mathematical fact. The "entrance pupil" diameter, which is the diameter of the aperture as viewed through the front of the lens at "infinity distance", quite literally IS the focal length divided by the relative aperture number. There is no guideline here, that is quite specifically EXACTLY how to compute the size of the entrance pupil, which puts a limit on the minimum size the front element can possibly be. If the front element were smaller than that, then the entrance pupil would have to be smaller as well.

Sometimes manufacturers "fudge" the design a little. For example, the 100-400mm L lens is actually more like 390mm, and the entrance pupil and the diameter of the front element are just a little smaller than would actually be necessary for a lens that was truly 400mm long (I've actually measured it myself.) For a ~390mm focal length, the numbers add up and seem to be correct, probably because it was a matter of manufacturability vs. cost to shorten the lens just a little. Not that it matters much, a few mm difference in focal length aren't going to matter (less than 5% difference in subject size relative to the frame), but it could mean quite a bit from a cost standpoint.

The size of the front element of a lens cannot be smaller than the entrance pupil, however it can be larger. I guess you could say it is a "guideline" that the front element has to be at least as large as the entrance pupil. Wide angle lenses tend to have front elements that are significantly larger than their entrance pupils, less so because of the needed light gathering power and more so just so they can gather incident light from an appropriately wide angle. In my 16-35mm f/2.8 L lens, for example, the front element is HUGE, while the aperture is quite tiny in comparison. I'd say the front element is dozens of times larger in area than the entrance pupil, maybe a dozen times larger in diameter.

[K3nt]

Reading these last few posts reminds me of how surprised I am that my 135/2 isn't larger, especially compared with my 200/2.8

135 / 2.0 = 67.5
200 / 2.8 = 71.4

Only 4mm different...

Yes now you mention it I can see that based on the same maths the 200/2 and the 300/2.8 are similar diameter.

But if the formula is as simple as focal length / f stop, why does my 135/2 vignette much more at f2 than my 200/2.8 does at 2.8 ?

The simple formula is just that.. simple and a guideline. A lot more goes on in the world that is precision optics that can produce(or not) the vignetting you're experiencing.

Just to be specific, the relative aperture as a ratio couldn't be called a "guideline". It is a mathematical fact. The "entrance pupil" diameter, which is the diameter of the aperture as viewed through the front of the lens at "infinity distance", quite literally IS the focal length divided by the relative aperture number. There is no guideline here, that is quite specifically EXACTLY how to compute the size of the entrance pupil, which puts a limit on the minimum size the front element can possibly be. If the front element were smaller than that, then the entrance pupil would have to be smaller as well.

Sometimes manufacturers "fudge" the design a little. For example, the 100-400mm L lens is actually more like 390mm, and the entrance pupil and the diameter of the front element are just a little smaller than would actually be necessary for a lens that was truly 400mm long (I've actually measured it myself.) For a ~390mm focal length, the numbers add up and seem to be correct, probably because it was a matter of manufacturability vs. cost to shorten the lens just a little. Not that it matters much, a few mm difference in focal length aren't going to matter (less than 5% difference in subject size relative to the frame), but it could mean quite a bit from a cost standpoint.

The size of the front element of a lens cannot be smaller than the entrance pupil, however it can be larger. I guess you could say it is a "guideline" that the front element has to be at least as large as the entrance pupil. Wide angle lenses tend to have front elements that are significantly larger than their entrance pupils, less so because of the needed light gathering power and more so just so they can gather incident light from an appropriately wide angle. In my 16-35mm f/2.8 L lens, for example, the front element is HUGE, while the aperture is quite tiny in comparison. I'd say the front element is dozens of times larger in area than the entrance pupil, maybe a dozen times larger in diameter.

I know the f-stop and focal length formula is definite. What I was referring to is the magic that happens beyond that. Compensating for CA, flare, coatings, lens material they all contribute to the final resulting light that eventually ends up saturating the sensor behind the lens. And yes, like you say, some lenses are using a lot larger front elements that would, based on the maths alone, be overkill.

I enjoy this thread. Excellent stuff in here and good discussion. Have a great week all of you! 

[Einstein333]

I wish they would make a really fast standard zoom lens like a 30-90 f2.0 OS...