What's the point of making constant aperture lenses?

[RedTele]

Yes, I know it helps having constant maximum aperture so you can keep the same exposure settings as you zoom in/out. My question is why design a lens in this way?

Take a 70-200 f/4.0 lens for example. Max aperture of f/4 @200mm means that the physical size of the aperture diameter is 200/4 = 50mm. While f/4 @ 70mm you're using 70/4 = 17.5mm. You're using less than quarter the area. Now I'm guessing that because of angle of view etc you can't exactly use the same size max aperture at wide end as the long end, but surely you could use more than what is specified as the maximum aperture for the wide end? Say you used 25mm diameter at 70-mm FL, you get f/2.8, a full stop faster!

I think a f/2.8-4.0 lens is more versatile than a constant f/4 lens and a f/2.0-2.8 than a constant f/2.8 lens. No one is forcing you to shoot at maximum aperture, it just gives you extra flexibility. So why do lens manufacturers design lenses this way? It's like artificially imposing a speed limit on the lenses. What am I missing here?:shrug:

 

[arad85]

At a guess, it will be more difficult/costly to produce glass that can handle wider angles at lower f numbers. If you think about the light rays and the bending of them a lens does, you'll see that a 70-200 has to work harder at bending those light rays on the shorter focal lengths than the longer. Lenses are built to a cost point and you see that even small decreases in f number cost a lot...

 

[Mahoneyd187]

I have no idea about how the lens design would differ, but rather than just a help,I'd find it a necessity that whilst working in manual exposure,in a fast paced environment,the exposure settings remain the same

 

[strumstrum]

it maybe just me but one of the reasons I like wider aperture lenses is I feel they perform better at a 'normal' aperture... IE a f4 200mm lens will produce a better result at f5.6 than a f5.6 200mm lens will at f5.6..?!?!:shrug:

 

[stylgeo]

I see RedTele's point but I have to agree with arad's argument on cost and R&D. What's more is that companies in this market, as in all markets, use price discrimination to attract a wider range of customers, maximising their sales, so their net profit. For example, if Canon only produced a 70-200/2.8 IS, it would only attract a certain number of customers that can afford that particular lens. But taking into consideration that not all people can afford that lens, it offers 4 different lenses in that range, the 2.8 IS and non-IS and the f/4 IS and non-IS. The same principle applies to all markets. For example cinema tickets. They offer student prices, premier prices, cinephil prices etc. This way they maximise their profit by offering higher prices, above their marginal cost. Take for example books. There is a new Harry Potter book coming. The publisher will first publish a first edition signed hardback copy and charge it £80 for the hardcore fans and collectors. At the same time they release the hardback copy for £25 for the real fans. There are however customers that aren't willing to pay that much for a Harry Potter book, so they will wait for the softcover at £6.99 a few months later. Same applies for lenses.

There is however the chance of product overlap. If a cheaper product has similar specs to a more expensive substitute, then they will lose the market for the expensive one, so they differentiate a lot to ensure that there is no market overlap and that they will indeed attract all members of the paying public.

 

[Brachytron]

Calculating f-number is not simply dividing focal length by aperture diameter, you have to consider the "entrance pupil", which is the physical projection of the aperture - if you look down the front of your lens while zooming in, the aperture will become magnified as the elements in front of act as a magnifier. This is why fast telephotos have bulbous front elements, to magnify the aperture to make it collect more light. Thus the physical size of the aperture doesn't actually change when you zoom from 70 to 200mm. Its just the magnification of it that changes. Optically it is easier to have a variable aperture as it allows the lens desigher more flexibility in the glass used. I suspect it as much a marketing thing or an excuse to make a lens more expensive as it would be no good if performance was not up to scratch.

 

[RedTele]

I think some of you misunderstood my point but stylgeo's marketing argument and Brachytron's explanation on f-number derivation mostly answered my question.

Price discrimination makes sense, but as I said in the original post, why not make both ranges faster? Still get the discrimination between the more expensive and the cheaper model...

Brachytron, where did you get the information? Is there any good websites I could read up on?

 

[Oggy]

Less variation in DOF as you zoom in or out?

 

[boyfalldown]

Calculating f-number is not simply dividing focal length by aperture diameter, you have to consider the "entrance pupil", which is the physical projection of the aperture - if you look down the front of your lens while zooming in, the aperture will become magnified as the elements in front of act as a magnifier. This is why fast telephotos have bulbous front elements, to magnify the aperture to make it collect more light. Thus the physical size of the aperture doesn't actually change when you zoom from 70 to 200mm. Its just the magnification of it that changes. Optically it is easier to have a variable aperture as it allows the lens desigher more flexibility in the glass used. I suspect it as much a marketing thing or an excuse to make a lens more expensive as it would be no good if performance was not up to scratch.

Think about what you're saying there. The f-number is correctly written as f/2.8 for example - ie focal length divided by (admittedly nominal number) to give diameter

Fast telephotos haven't bulbous front elements, they tend to be seen on ultra wides. Its is easier to build a variable lens, but the design of fixed appetures is much more complicated then you make out

 

[stylgeo]

Price discrimination makes sense, but as I said in the original post, why not make both ranges faster? Still get the discrimination between the more expensive and the cheaper model...

Well in this point, even if you avoid product overlap, there's the point of cost and overall weight. If Canon again made a 70-200 f/2.8-4 then to completely avoid product overlap as you suggest they would have to also make a 70-200 f/1.8-2.8. That would be a very difficult lens to produce, it would weight a tonne (the 200/1.8 weighs more than 3kg), and it would cost a fortune. Besides, with these lenses, they know that most of the times are used from mid range to full tele, so they will also gain more profit by selling their fast shorter primes. If I really need the f/2.8 at 200mm and f/1.8 or faster at 70mm, I would have to buy both the 70-200/2.8 and the 85/1.8 (or 1.2). This is a very clever market strategy, and they know what they are doing. I would imagine that if they could make an 70-200/1.8-2.8 they would make it but offer it at a very high price, but still keep the one with the constant 2.8 quite cheaper. That would be a luxury product, one that would ensure that you won't need to buy their shorter fast primes, so you'll avoid the hassle of changing lenses, and for that reason you would have to pay for it.

 

[HoppyUK]

I don't think there is any marketing discrimination or any other smoke and mirrors going on. The simple fact is that it doesn't work like that in practise, or at least it would not result in a lens that delivered acceptable performace at a lower f/number. It would not be possible to produce it at an acceptable price, and/or size and weight. F/2.8 zooms are big and heavy enough as it is, expensive, and generally not at their best even at f/2.8, let alone f/1.8. If it could be done, it would be.

It seems like the practical limit for a zoom is f/2.8, no matter how you cut it. Low f/numbers are extremely challenging optically, and if you want to go lower than that, you have to forget the zoom function completely and go for fixed focal length. The variable aperture trick only works for low-f/number zooms, like f/3.5-5.6.

Olympus make a couple of f/2 zooms, for a price, but they only have to cover a 4/3rds sensor which is one quarter the area of full frame.

 

[StewartR]

If it could be done, it would be.

I agree. If it were possible then one of the manufacturers would do it. Sigma would do it if the big boys didn't. There is no global agreement between all the manufacturers.

It seems like the practical limit for a zoom is f/2.8, no matter how you cut it. Low f/numbers are extremely challenging optically, and if you want to go lower than that, you have to forget the zoom function completely and go for fixed focal length.

Spot on.

Incidentally I think it *might* be possible to make a 70-200mm f/2.8-4 zoom. But you wouldn't make it by starting with an f/4 zoom and opening it up at the 70mm end. You'd start with an f/2.8 zoom (big, heavy, expensive) and restrict the aperture at the 200mm end. And it would be completely pointless.

 

[stylgeo]

If it could be done, it would be.

I would imagine that if they could make an 70-200/1.8-2.8 they would make it but offer it at a very high price

My point exactly, if they could they would, but it would be very expensive and unwieldy. But in the case of a 2.8-4, that would overlap with the constant 2.8, that's my case. After R&D reaches to new products and new innovations, then it is up to market forces and strategies to decide what comes out of the factory.