How does it work? (70-200 f2.8 lens)

stu_the_flat

stu_the_flat

Messages
2,443
Name
Stuart
Edit My Images
Yes
I have become fascinated by these 70-200 f2.8 Lens. Mainly because I don't have a clue how they work!

Now my very basic understanding and perhaps assumption of how a zoom lens worked is as you zoom in you would be cropping out light? hence the f number would rise?

Is there a flaw in my understanding? more importantly how do these 70-200mm f2.8 lens over come this? (hopefully Zemax diagrams will not be needed in the explanation!) maybe I should by one and take it apart! :D

Cheers Stuart
 
would it help your understanding if I said I have an old [only 3 yrs old to be fair] sigma 70-300 f4.5-5.6 that fits inside the lens hood of my sigma 120-300 F2.8 - sideways! :eek: ;)


sorry, cant explain it technically, there will be plenty that can and will, I just know I lurve having 2.8 across the zoom range :thumbs:
 
Wikis maths is never very well explained just formula's lumped together! as far as I can remember the number of F stops can be calculated by F/# =a/f a = aperture f= focal distance. so I suppose the way a lens would keep a constant f2.8 is just to adjust a very wide aperture ring as the focal distance changes? but that would need large lens and that would introduce lots of complecated abberations?

YV your sigma 120-200 sounds cool you must be a substantial build to carry it around in your fit bag!. did a quick google and the prices made me feel that I needed a lie down!

Stuart
 
stu_the_flat said:
I suppose the way a lens would keep a constant f2.8 is just to adjust a very wide aperture ring as the focal distance changes? but that would need large lens and that would introduce lots of complecated abberations?
Click to expand...

Basically, yes. To get a constant f/2.8 over, say the 70-200 range then the front element of the lens needs to be 200/2.8 or roughly at least 72mm, whereas with a lens that has a max aperture of 5.6 at 200mm it only needs to be 36mm.

Yv's Siggy, on the other hand needs a 105mm filter to cover it's front element :bonk:
 
I'm not sure exactly what your question is, but it seems to be along the lines of: if the aperture is f/2.8 at 200mm, then in theory the lens should have a sufficiently large light gathering diameter to give a much lower f/number at shorter focal lengths. In fact, if the f/ratio held good with a typical 70-200mm f/2.8 zoom, then it would be around f/1.0 at 70mm!

This is the way the question is commonly posed, and most popular zooms at least conform to this theory to some extent by having a lower f/number at the short end, eg 70-300mm f/4-5.6. So why don't lenses like Canon's finest L range do something similar?

This is where the very basic optical theory of the f/ratio meets reality. In particular, optical life gets very difficult with low f/numbers, and it becomes impossible to deliver good edge sharpness, and light transmission that is spread evenly across the frame (vignetting).

The practical limits appear to be f/2.8 for a full frame zoom and any attempt to go beyond that at the shorter end of the zoom would result in very poor edge sharpness and strong vignetting which could not be effectively controlled. Lens manufacturers have elected to simply cut this out of the design brief, with the upside being a brighter, sharper, smaller, lighter and cheaper result.

Format also has a big effect, which is why Olympus can make a 35-100mm f/2 zoom which covers their much smaller 4/3rds format, but it's not easy. The reason why Canon and Nikon have chosen not to follow this route with an EF-S or DX lens for their crop format cameras is cost. Crop format is targeted at consumers who are unwilling to pay £2k for such a lens, which is the Olympus price.

Not sure if this this addresses your question! ;)
 
Hoppy,

That was a very useful explanation of something I have wondered myself. as I understand your explanation, that 70-200mm f2.8 could be an f1.4 or so on the short end, but they stop it being so because they couldn't carry off the quality there and it would effectively spoil the whole lens (as far as people's perception would be concerned).

interesting indeed.

dave
 
Your're welcome Stew/Dave :)

The really difficult bit is low f/numbers like f/2.8. If you compare the Canon 70-200mm f/2.8 L IS, an exceptional lens by any measure, with the Canon 70-300mm f/4-5.6 IS, also a very highly regarded lens, then the constant f/2.8 lens has significantly less reach, yet weighs 2.5x heavier and costs more than threee times the price. Not exactly an apples for apples comparison, but you get the drift.

Looking at the problems of edge sharpness and vignetting, obviously these are relieved if the lens doesn't have to cover the full frame format. And if this is combined with modest f/number requirements you get even more benefits, such as the EF-S 55-250mm f/4-5.6 IS which turns in a very good performance yet is a tiny little thing, only about half the weight of the 70-300 (under one quarter the weight of the 70-200 2.8) yet costs only £220.

You can cut it another way, and if you want wide focal length range and are prepared to pay more, and put up with more weight, and also a modest f/number, then the EF-S 18-200mm f/3.5-5.6 IS becomes possible. In all these alternatives the common denominator is a modest maximum f/number.

Looking at the f/number again, it is evident that the lower the f/number, less focal length range is possible. Taking this to its logical conclusion, if you want really low f/numbers like f/2 or f/1.4, then it's primes only.
 
There's a good description of this on photo.net,

f/stop is, for a simple lens, focal length/front element size.

A classical zoom lens, though, collimates the light (i.e., gathers it all and then stuffs it into a bundle of fixed width; sort of like piping it into an optical fiber). Then the "moving stuff" varies the magnification and effective focal length. The amount of light available to strike the film doesn't change. The iris is located in a portion of the lens where the light is nicely collimated. So a Canon 70-200/2.8 should really be thought of as a 70/2.8 lens with a magnifying system.

Anyway, the expensive and huge professional zoom lenses like the 70-200/2.8 and 80-200/2.8 Nikon are made using this system that goes back to 1946.

http://photo.net/bboard/q-and-a-fetch-msg?topic_id=23&msg_id=0000Wf

Note: Not disrespecting Hoppy's excellent article above, just posting an alternative description (which in any case feeds on from Hoppy's description i.e. Hoppy's is why, this is how).

HTH
 
pearce_jj said:
So a Canon 70-200/2.8 should really be thought of as a 70/2.8 lens with a magnifying system.
Click to expand...

I can see where this description is coming from, but I'm not sure if it's correct. if it was a 70mm f2.8 with a magnifier, I'd expect that as you magnified the image, you'd get less light coming in as you're kinda cropping into the centre of the image. it would be kinda like putting a variable teleconverter on the lens wouldn't it?[1]

but with a 70-200 F2.8 I'd expect to be able to maintain the same aperture/shutter speed combination at both ends. if it was a 70mm with a magnifier, surely this wouldn't work?

dave


[1] oh, am I the only person who thinks we ought to be able to buy TC's with built-in VR/IS?
 
pearce_jj said:
So a Canon 70-200/2.8 should really be thought of as a 70/2.8 lens with a magnifying system.
Click to expand...

I'll repeat the same quote :)

Not researched anything but I'd be inclined to think of it as a 200 2.8 with an image transfer system for shunting the light down the barrel when focussed to shorter lengths.
 
Hi Dave, as I understand it effective tc is on the front of the lens rather than the back, but a "normal" tc couldn't be designed that way and work across different lenses.
 
You must log in or register to reply here.
Back
Top