Fast lenses - are they faster all the time?

[fess67]

ok let me clarify.

I have a 70-200 f2.8 A 'fast' lens.

I have a 100-400 f4.5 - 5.6.....not so fast.

Question. If I am at 200 and F8 on both lenses, will the 70-200 outperform the 100-400 or will they be the same?

Does that make sense?

[Boo53]

Thats going to come down to lens quality, not lens speed, although some more experienced than I suggest that IQ drops off at the extreme zoom of some, if not most, lenses

[davidd]

If by 'perform' you mean transmit light, they will be the same. f8 is f8 on any lens. The 'fast' description of the 70-200 f2.8 refers to it's maximum aperture. It is also a constant f2.8 maximum, (at all focal lengths), whereas the 100-400 in your example can only open to f5.6 maximum aperture when zoomed out to it's maximum focal length.

[jjphoto]

...

Question. If I am at 200 and F8 on both lenses, will the 70-200 outperform the 100-400 or will they be the same?

Does that make sense?

Two things.

1/ You can't make sweeping generalisations about that kind of thing and always be accurate, there are always exceptions. A 'fast' lens may have been designed and built for a 'pro' market so it may in fact be a very high quality lens in many different ways. Whilst a slow lens may have been built for a cheaper market and to lower standards, also in various ways. Comparing a slow 'kit' lens with a Canon 'L' is an example of the above and is in some respects an unfair comparison.

2/ When you stop a lens down 2-3 stops (or close to diffraction) then there is often very little difference between the best and the worst, at least in terms of centre sharpness. Differences in performance tend to exist at the edges, in terms of flare, possibly distortion, field flatness, colour rendition, bokeh and mostly in wide open sharpness. But if you are looking at the performance on a crop camera then you are mostly seeing the centre performance anyway.

There's a bit more about that stuff here;

http://photocornucopia.com/1005.html

JJ

[arthurking83]

David is almost right on.
An f/8 lens is an f/8 lens all the time, and any lens with the same focal length with the same effective aperture should produce similar looking results.
This depends on focused distance too tho, as some lenses focus closely .. really close but do so at the expense of actual focal length.
They shorten the focal length to focus closer.


If by 'perform' you mean transmit light, they will be the same. f8 is f8 on any lens. ....
This part is theoretically correct, but almost always not in reality.
Each lens will have a T stop. That is, the actual amount of light passing through the glass within the lens itself.
A simple 2 element lens with 200mm and f/2.8 effective aperture will always pass through more light even at f/8 than a modern complex multi element lens will.
the simple two element lens may not produce high level of sharpness, but the t stop will be higher.

DXO's website has a good listing of lens performance for many lenses at given apertures.
And you may find that a 70-200mm f/2.8 lens may only have a T-stop of 3.4.
That is even tho the lens can produce a DOF of f/2.8, the actual transmission of light through the lens and onto the sensor may only be f/3.4.
ie. it may be underexposing by an amount.

Each lens is different, but some lenses perform similarly.

[ameerat42]

I agree with AK as the nearest definition to all these Qs. Transmission factor, or T-stops, theoretically matter.

Fess, the f-stop or f-ratio is the effective focal length you are using at the time divided by the amount of lens diameter diameter you are using at the time.

If you have 20 separate pieces of glass in one lens and only eight in another, then more light will be soaked up in the first than in the second, even at the same f-stop.

Also, as you focus closer and move the lens assembly forward a lot, as in some macro lenses, then you are effectively changing the focal length and so the f-stop.
Am.

[ricktas]

Another thing to remember is that auto-focus systems work with the lens at its biggest aperture. So even if you are set to f8, when the AF motors kick in, the aperture is opened up to the maximum for that lens. A 'fast' lens will in fact, focus faster (more light, bigger maximum aperture), means easier to lock onto something with AF systems. As soon as the lens locks focus, it is then stepped back down to f8, in a fraction of a second.

So a fast lens, not only (probably) has better quality glass elements in the lens itself, but it will generally be faster at locking focus using AF systems. I think this term 'fast lens' comes from the maximum aperture being bigger, thus auto-focus is therefore 'faster'.

[Steve Axford]

It is not quite true that a "fast" lens, ie one with a large maximum aperture, will always be fast focusing. It may be that it is quite slow to focus because the lens elements may be quite heavy and slow to move. Some fast lenses are very fast to focus, but lenses like the Canon 50mm f1.2 or the 85mm f1.2, which are both superb lenses and very "fast" are both quite slow to focus. The Canon 135mm f2.0 on the other hand is both fast and fast to focus. It depends what is important to you.

[fess67]

Thank you everyone for your responses.

I think Rick asnwered the question I was trying to ask, although Steve added to the response also. Had Rick not posted I think my next question was going to be 'why do they call it a fast lens?'

Both mine are L lenses so I am going to assume quality in the build but the 70-200 might be the go if I want faster focus. Of course at the expense of reach.

[rodw]

Thank you everyone for your responses.

I think Rick asnwered the question I was trying to ask, although Steve added to the response also. Had Rick not posted I think my next question was going to be 'why do they call it a fast lens?'

Both mine are L lenses so I am going to assume quality in the build but the 70-200 might be the go if I want faster focus. Of course at the expense of reach.

Um, one definition i found "If it can get wider (open), you can shoot faster, everything else being the same. Also, money tends to leave your wallet faster when you buy such lenses."

But really as you can let more light in with a fast lens, you can use faster shutter speeds. That makes sense to me.

[arthurking83]

They call it a fast lens simply due to the aperture.

These lenses have been called fast lenses since before the age of auto focusing camera gear.

Fast is simple .. large aperture allows faster shutter speed given a predefined film/sensor sensitivity(or ISO rating).

Rick and Steve are both right too tho.. fast lenses have come to mean all manner of things .. fast focusing, fast (as in wide) aperture .. and I suppose people will find other definitions of lens speed( a faster operating image stabilisation perhaps???)

Some lenses can be slow and yet still focus fast .... does that make it a fast lens?

Also note too .. for a specific lens design a faster focusing lens is usually a less accurate in it's focusing too.

That is, taking Steve's example of the Canon 50/1.2 which may focus not so fast .... due to the possibility of an ludicrously shallow DOF, there will be errors in focusing due to this thin DOF.

The lens may be designed deliberately with a very cautious auto focusing mechanism to achieve more accurate focusing(more precise helicoid movements within the lens barrel).
This can be a factor on top of the heavy lens element syndrome already stated.

Many macro lenses may have this 'long throw' focusing design.

[Mark L]

........
If you have 20 separate pieces of glass in one lens and only eight in another, then more light will be soaked up in the first than in the second, even at the same f-stop.

........

So you need to adjust your shutter speed to get the same result at f/8 ?

[ameerat42]

So you need to adjust your shutter speed to get the same result at f/8 ?

Theoretically. It would depend on how badly the (hypothetical) lens soaks up the light. But this does happen with reflecting surfaces, as in telescope mirrors.
If it's in need of re-aluminsing your "f/8" might even be closer to f11~16. In practical terms, reflectors with two mirrors are usually slower at any given f-stop than straight refractor lenses, as they reflect less light than lenses refract. T-values again.

[arthurking83]

So you need to adjust your shutter speed to get the same result at f/8 ?

Yep! To achieve the same exposure as the faster T-stop lens, you need either a slower shutter speed or higher ISO.

Camera tho may insist that the exposure is the same, but the picture tells the story.

I've written about the Nikon 18-105VR which requires about +0.7 Ev compensation on camera to achieve the same exposure as most of my other lenses .. Tammy's, Nikon's etc.

Even tho the camera's meter seems to think it's exposing as per the other lenses, the resultant images are underexposed by comparison.

**note that this could also be due to a less accurate aperture mechanism, but I don't it is in this case**


And strictly speaking the number of lenses is not always the determining factor.
Size of lenses has a large impact too, but more important is the quality of material of each glass element component.

Kit lenses may tend to use inferior quality glass(or plastics) and hence transmit less light.
Those faster f/2.8, or larger, complex zooms may have twice as many lens elements, and yet still transmit more light through them .... better grade lens elements

[ameerat42]

Troo! Troo.! In my hypothetical 8- and 20-element lenses "they" used the same glass.
Am.

[ving]

The 70-200 at 200 is f2.8 while the other @ 200 is likely to be at f5 or something. They focus at their max aperture so regardless the former will focus faster

Sent from my HTC Incredible S using Tapatalk 2

[arthurking83]

I'm not exactly sure that they focus faster due to the larger aperture .. this has yet to be shown to be true by anyone ... the larger max aperture simply allows the system to focus in lower light.

The problem say with comparing focusing between two similar lenses such as Canon's 70-200 f/2.8 and f/4 versions, is that they are not the same lens.
They may have similar specs, but they're not the same, and I'm sure Canon would have tweaked the f/2.8 version to help it perform better considering the price premium.


When you look at the specs of a lens and refer to it's speed .....

Focus speed can be a function of both lens AND camera. The body it's mounted on will contribute to the focusing ability, so AF is looked at as an entire system, not just the lenses ability. That's because the actual autofocusing is done by the camera, and the lens is quite dumb by comparison and is simply told what to do.
(OTOH, on a manual only lens, the focusing performance is referring to the lens only .. how fast, smooth or accurate it may be .. etc, etc)

Aperture speed is a lens only spec.

[Steve Axford]

Yep! To achieve the same exposure as the faster T-stop lens, you need either a slower shutter speed or higher ISO.

Camera tho may insist that the exposure is the same, but the picture tells the story.

I've written about the Nikon 18-105VR which requires about +0.7 Ev compensation on camera to achieve the same exposure as most of my other lenses .. Tammy's, Nikon's etc.

Even tho the camera's meter seems to think it's exposing as per the other lenses, the resultant images are underexposed by comparison.

**note that this could also be due to a less accurate aperture mechanism, but I don't it is in this case**


And strictly speaking the number of lenses is not always the determining factor.
Size of lenses has a large impact too, but more important is the quality of material of each glass element component.

Kit lenses may tend to use inferior quality glass(or plastics) and hence transmit less light.
Those faster f/2.8, or larger, complex zooms may have twice as many lens elements, and yet still transmit more light through them .... better grade lens elements

I've noticed that lenses do act differently at the same fstop, and it is sometimes the better lenses that let in less light. However, I have not noticed any underexposure or overexposure with specific lens. How would this happen as the metering is done after the light passes through the lens?

[arthurking83]

..... How would this happen as the metering is done after the light passes through the lens?

Yeah! it's weird.

I have this issue with the 18-105VR, and (maybe limited to this lens, or not) it could be a Nikon programming anomaly. :confused013

If I set this lens to f/8(where vignetting is about zeroed out) and 105mm, and compare the shot with one from the 105VR micro, the micro lens image is always brighter.

Setting the kit lens with +0.7 Ev balances the exposure just about right, give or take some differences in contrast and colour.

[Steve Axford]

Sounds a bit like a camera anomaly. Canon have the odd one as well, usually reported in great detail by the press.

[Wazza999]

If the actual amount of light transmitted by the lens wasn't what was metered to determine the exposure settings then a manual exposure adjustment would be needed when filters were added to the front of the lens as was the case 50 or so years ago before through the lens metering.

[Mark L]

^ But now we do have "before through the lens metering." So ....

[Wazza999]

So ... before the lens metering is?

[I @ M]

Sounds a bit like a camera anomaly. Canon have the odd one as well, usually reported in great detail by the press.

I can list a few lenses that behave differently to each other just as Arthur says and the differences are very noticeable in a studio situation with incident lighting. Most noticeable probably is a Tamron 90mm F/2.8 macro that is around 2/3 of a stop darker than a Nikkor 85mm F/1.4 at the same aperture and strobe light levels. Under "normal" conditions where the camera is metering the light there is less difference with those two lenses but still a difference in shutter speeds when used in aperture priority.


^ But now we do have "before through the lens metering." So ....


So ... before the lens metering is?

Think studio situations where strobes are used and the camera metering doesn't come into play at all. A hand held incident light meter is used to determine the aperture needed for a given power level or conversely to determine the power level of the strobes needed at a given aperture.
That is very much a case of before the lens metering to me.

[mikec]

That is very much a case of before the lens metering to me.

I'd agree with that statement, can't think of other instances either that you'd meter before the lens if you have TTL.

[fillum]

Perhaps worth reminding that lens ratings are generally (always ?) specified with the focus at infinity. So for example:-

- my Nikkor Micro 105mm f/2.8 AF-D actually has a max aperture of f/5 (from memory) at minimum focus distance (1:1)
- my Nikkor 18-200mm set @200mm has an effective focal length significantly less than 200mm when focusing close (from memory some reports claim as short as ~135mm)

In my experience neither of these characteristics has presented a problem in practice...



Cheers.

[Steve Axford]

I can list a few lenses that behave differently to each other just as Arthur says and the differences are very noticeable in a studio situation with incident lighting. Most noticeable probably is a Tamron 90mm F/2.8 macro that is around 2/3 of a stop darker than a Nikkor 85mm F/1.4 at the same aperture and strobe light levels. Under "normal" conditions where the camera is metering the light there is less difference with those two lenses but still a difference in shutter speeds when used in aperture priority.





Think studio situations where strobes are used and the camera metering doesn't come into play at all. A hand held incident light meter is used to determine the aperture needed for a given power level or conversely to determine the power level of the strobes needed at a given aperture.
That is very much a case of before the lens metering to me.
I haven't noticed it with Canon, but that doesn't mean it doesn't happen.

[Steve Axford]

Perhaps worth reminding that lens ratings are generally (always ?) specified with the focus at infinity. So for example:-

- my Nikkor Micro 105mm f/2.8 AF-D actually has a max aperture of f/5 (from memory) at minimum focus distance (1:1)
- my Nikkor 18-200mm set @200mm has an effective focal length significantly less than 200mm when focusing close (from memory some reports claim as short as ~135mm)

In my experience neither of these characteristics has presented a problem in practice...



Cheers.

Definitely not always. Many macro lenses are "tuned" to work best at macro distances. Some don't even work at long distance.