Fast lenses at a crossroads?

[YuengLinger]

I'm very sad to hear that Canon's new 11-24mm won't be a fast lens, and very concerned about the direction this suggests.

There are, I believe several strong reasons Canon is stepping back from fast lenses, reasons that would not have mattered as much before the omnipresence of smartphones with decent cameras.

First, the baby-boomer retirees who are big into photography join workshops and camera clubs where fast lenses are simply not seen as especially important. Why? Because very, very few of those type of hobbyists take pictures of people. They love landscapes and found still lifes and birds. And, though they have money, the economy scares them, they don't get anything from CD's anymore, so they save where they can. Why buy a 24-70mm 2.8 when the 24-105mm 4 costs half? Why buy anything in 2.8 when something "just a stop" slower costs less. And, finally for the babyboomers, weight is a legitimate issue, but they will obsess over saving a few ounces.

Then there is the learning curve for fast lenses. Not only is focus tricky below f/2, but effectively choosing how shallow to go takes a lot of trial and experience. I can tell from the great work on your website, you know that just having shallow DoF doesn't make magic; in fact, blurred out jaw-lines and other features, including hands, forearms, etc., can just look awful if overdone or with the wrong perspective. So people who do spend are often disappointed.

As for the user experience, for one example, the ergonomics of the 85mm 1.2 are challenging for many, as is the weird and pokey AF system. Once one is competent with this lens, it is magical, but otherwise, a real puzzle for enthusiasts who think it's all about the gear.

Canon seems scared and uncertain. They go for the fat middle of the remaining DSLR market. News agencies aren't spending for photojournalism much anymore, and the small local newspapers across the USA are nearing extinction. Small time portrait and wedding photographers with business sense have a very hard time justifying spending on fast L primes.

Please, Canon, remember how important fast lenses are for clearly elevating the dslr over smartphones.

 

[Lee Jay]

1) Making a lens as wide as 11mm with a pretty big zoom range also fast would be very, very difficult. It would probably double the price, if not more. The Nikon has a much longer wide-end and thus a much smaller zoom range (1.7 versus 2.2).
2) Canon is making new, fast lenses. 70-200/2.8L IS II, 35/2IS, for example. Sigma is as well with the Art series of primes and the 18-35/1.8 for crop.
3) An f/4 zoom on full-frame is still exceptionally separated from a cell phone camera with a 1/3" sensor and an f/2.4 prime. Around four whole stops separated, in fact.

 

[YuengLinger]

Good points, lee jay, but 12 or14mm at 2.8 would serve many needs.

 

[Lee Jay]

Good points, lee jay, but 12 or14mm at 2.8 would serve many needs.

Well, then buy a 14/2.8. There are more than one available. Wider could be cost prohibitive.

And that's why I just love my 15mm/2.8 fisheye, for which I paid $370 brand new.

 

[JonAustin]

Some good and interesting observations; not sure if we can know what conclusions Canon has drawn, vis-à-vis its corporate strategy for its SLR camera & lens product lines.

I am one of those (semi-retired) boomers, and I do a mix of paid and hobby photography. Much / most of the paid work is of the studio and environmental portrait variety, and -- even when I owned an 85/1.8 prime lens -- I rarely shot portraits at apertures larger than f/4.

I've seen some stunning wide-aperture portrait (and nature) work, but in my view, it's more "artsy" than commercially "viable." I now own a mixture of f/2.8 and f/4 lenses (both primes and zooms), and almost never feel the need for anything faster (except the rare occasion (for me) when I'm trying to AF in very low light).

 

[ajfotofilmagem]

No one has left a wider zoom lens 12mm (full frame) so far. ??? The Sigma 12-24mm F4.5-5.6 is already large and heavy. :

I imagine that a 11-24mm F2.8 lens would have a bulbous front element greater than 100 mm and would weigh more than two kilograms. I do not even want to imagine what would be the price of such a lens. :-X

 

[YuengLinger]

And the latest 16-35mm f4? Ok, replaces the 17-40mm wonderfully, but what about the v2 16-35 that was barely better than v1?

Where is the great fast UWA and updated 50mm?

 

[Lee Jay]

And the latest 16-35mm f4? Ok, replaces the 17-40mm wonderfully, but what about the v2 16-35 that was barely better than v1?

Where is the great fast UWA and updated 50mm?

Both low priority, IMHO. The 100-400L replacement was 10 years over due, the wide primes were a mix of acceptable and disaster, and they had to produce the STM lenses because of the new dual pixel focusing stuff, so that took some priority and the secondary benefit was that the high-volume slow consumer zooms got a pretty solid optical upgrade. The 55-250STM is quite excellent and the 18-135STM is really pretty good for a hyperzoom.

The 16-35/2.8 is already pretty good (personally, I think my 17-40L is pretty good) and the 50s are in the "who cares" category for most people. I'd put a 35/1.4, 135/2IS, and 85/2 IS all above a 50/1.4 replacement in priority order.

 

[c.d.embrey]

In the deep dark past, fast lenses were needed for low light. Now-a-days low noise high ISO sensors have replaced f/1.4 lenses for many people.

If you want/need fast lenses, buy primes, not zooms. Nikon has recently released the AF-S NIKKOR 20mm f/1.8G ED, for about $800.00. It weighs 12.6 oz (357 g). Can Canon be far behind ???

BTW f/1.8 is a good compromise between speed and weight.

 

[sulla]

f/4 on a zoom is not that slow. A great many zooms are f/5.6 (on one end).
What is unique in this zoom is the 11mm. Now, that is sensational. Not everything in every L lens can be sensational. Consider the 800 f/5.6. Now, this is also not a horribly fast lens, but to ask for a 800 f/1.4 would be just too much. Likewise, to ask of a 11-24mm lens to be fast, sharp, vignette-free and so on all at the same time would also be too much. Canon certainly pushed the limits of technology on this lens to unprecedented levels, lets not ask too much from the lens designers.
Perhaps in future, with better manufacturing techniques, all lenses aspherical, ultra low dispersion elements all around etc... but not today.

 

[pwp]

If the 11-24 was an f/2.8 lens it would be even more massive in both price and size. As an f/4, my hope is that it will be sharp wide open (not holding my breath on that one...). As Lee Jay points out, if you absolutely need f/2.8 and have the required budget, the 14mm f/2.8II is simply amazing. Most examples of the original 14mm f/2.8 are rubbish and should be avoided.

The brilliant value 17-40L f/4 which I had for years was sharper than my new 16-35 f/2.8II between f/5.6 and f/11. But images shot at f/4 were unusable mush.

Unlike the 17-40 f/4, the new 11-24 f/4 is not a budget lens and really needs to deliver quality output wide open.

-pw

 

[Ruined]

There is a much more simple explanation. It is extremely difficult to make an ultrawide zoom (i.e. starting at 16mm and below) that also has a wide aperture (i.e. f/2.8) without sacrificing image quality.

For instance, the 16-35mm f/2.8L II was Canon's 2nd attempt at making a lens in this focal range at f/2.8, and even with an increased front element size the lens cannot attain the sharpness of the recent f/4 lens and did not have tremendous improvement over its predecessor.

As an 11-24 will be used almost exclusively for landscape, f/2.8 is less important. Canon likely had the choice of a fast lens that was as sharp or had other artifacts, or a slower lens that was sharper with less artifacts - so likely they chose the latter.

With sharpness charts being the almost exclusive way some benchmark a lens (which is sad IMO), I am not surprised.

 

[neuroanatomist]

I'm amused by comments like 'it should be f/2.8' and 'just add IS' as if such things were as trivial as flipping a switch and came without tradeoffs in cost, weight, IQ, etc.

 

[3kramd5]

I see no evidence that Canon is abandoning fast lenses. Each design is a trade study. Wide apertures bring speed and shallow DOF, but add complications (particularly with edge sharpness), weight, and cost.

And the latest 16-35mm f4? Ok, replaces the 17-40mm wonderfully, but what about the v2 16-35 that was barely better than v1?

Exactly! It's a tradeoff. Could they have made the 16-35 2.8 II as good edge-to-edge as the 16-35 4? .... maybe. At what cost? At a thousand or two more a pop, how many fewer would they sell?

 

[Khalai]

...but to ask for a 800 f/1.4 would be just too much...

Why? Just over 570mm aperture opening. Who would not want a lens with 600mm filter thread?

 

[Lee Jay]

...but to ask for a 800 f/1.4 would be just too much...

Why? Just over 570mm aperture opening. Who would not want a lens with 600mm filter thread?

You know, if you really want it, you can get a 684mm f/1.9.

http://starizona.com/acb/HyperStar-Lens---14-Celestron-Edge-HD-P3115C773.aspx

 

[Sporgon]

An f/2.8 aperture on an 11 mm lens would be fantastic - if you could get reasonably sharp corners. At 11 mil you don't need small aperture for dof; I think f/2.8 on 11 mil would give everything from about 1 metre to infinity in focus at f/2.8.

The trouble is that it's not feasible to do this as the lens would have to be far too expensive, if indeed it were possible at all. Here's a link to the Tokina on TDP and this is on APS !

http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=947&Sample=0&FLI=0&API=0&LensComp=0&CameraComp=0&SampleComp=0&FLIComp=0&APIComp=0

Yuck.

Also a lot of modern lens tech has to go into making a fast lens sharp, even in the middle, even stopped down. 'Fast' doesn't automatically mean good, in fact, uncorrected 'fast' means bad. It is easier to make a slower lens with smaller diameter lenses sharper than a faster one with larger elements. Just off the top of my head a good example of this is the Voightlander 40mm pancake, f/2, compared with the Canon 40 mm pancake at f/2.8. The Canon is the sharper lens, even more so across the frame.

 

[jeffa4444]

Fast lenses are at a crossroads the biggest factor is cost. Yes you can make fast lenses with even field illumination, low disortion, low chromatic & lateral abberations but they are very hard to make. The situation was made more difficult when certain types of leaded crown glass were no longer available like Hoya ADC-1.
Professional lenses from companies like Angeniux, Fujinon, Cooke & Zeiss run many thousands of dollars and are covering a format similar to half-frame. Full frame with say the rumored Canon 53MP sensor makes acheiving high resolution wide lenses even more difficult.
Secondly as sensors become more sensitive with wider DR one of the reasons for fast lenses disappears the remaining being deliberate shallow depth of field for artistic reasons.