Lens choice - Why?

[Warb]

I take photo's for fun, or to remember events and people. I've ended up with a selection of lenses to help me achieve that, ranging from wide angle through to telephoto. Last year I tried to photograph my daughters netball games, and found great difficulty using any lens I had! I have searched the web, and by far the most commonly recommended focal length for netball is a 70-200mm zoom. This lens is recommended, whether the question relates to full frame or "crop" camera's, seemingly regardless of any other factor. But I can't see why!

My daughters games are outside, in the morning, and light is therefore plentiful. Spectators (parents!) are courtside, minimal distance to the players. Movement with a camera is hard, simply because someone will step in front of you or want to chat. The best option seems to be to pick a place on the sideline and camp for a quarter. The best place on the sideline is dependent on the position your subject is playing - for example there seems no point in taking pictures of the back of a shooter at the other end of the court!

Assuming that the above makes sense, I did some tests. I measured out a fake court, and used my daughter and her practice "hoop" to take some pictures. No regard to exposure etc., just framing. Using old 75-300mm and 28-80mm zooms on my 7D, 70mm is far too big for using under the goal, unless you just want a face. 200mm is fine for tight framing when the players are "half a court" away, but 100mm is still usable with some cropping on the computer. 70mm is not quite enough reach.

As far as I can see, 70-200mm would be a good focal length for those conditions on a full frame camera, though I'd probably still want something shorter under the hoop. But on a 7D I really can't see it as a good solution. So why is it so often recommended?

I already have a 100-400mm zoom, a great lens, but when I tried it for netball it wasn't fun. Too long for the baseline, too long for the sidelines unless the play was at the other side of the court. Step backwards and the viewfinder is filled with the back of someones head!

I also have a 17-55mm, which it my most used lens. Great under the hoop, but too short everywhere else. So if your subject is a goal shooter you're OK, but not so much for a more mobile player.

From my backyard experimenting, it looks like the 24-105mm f/4L would be the best choice as a single lens. In fact a touch longer would be nice, but the available lenses aren't weather sealed and it's either dusty or (rarely) raining! I also realize that lens choice is personal, that f/4 is slow for indoor sports etc., but disregarding all that I still think that 70-200 is too long on a 7D. Yet people asking the question with most or all of the same restrictions I have (crop camera, daytime outside, courtside etc.) are still told "70-200mm".

So what am I missing? Or does it relate to the mystique of the lens in question, rather than the focal length?

[Kym]

For sports, and wildlife for that matter, shutter speed is what is needed if you want sharp images.
Put the ISO up until you are shooting 1/500 or better
Aperture at around f/8 is a good start
The focal length is what you need to frame the subject plus a bit more (i.e. zoom out a bit); the exact FL will vary as needed

Consider learning how to pan and also use continuous autofocus

Crop vs 35mm is not that relevant as it's what is in the frame that matters, see http://www.ausphotography.net.au/forum/showlibrary.php?title=Indexes:Appendix_D_-_Full_Frame_vs_APS-C_DX_crop_factor_w_example

[ameerat42]

Forget the "recommendations". It would be hard to impossible to evaluate such information. You've already ascertained a number of empirical points about shooting, so a few (hopefully relevant) observations:
1. No particular focal length range would do the lot, BUT...
2. It would appear that being able to change your position relative to the action is important.
3. Depending on the type of shot you want to get, (almost half a field of play to a tightish goal shoot) the minimal focal length may have to be of the order of 17mm. You could determine the max focal length from your analyses. (Maybe 200mm?)
4. If you are able to get up high and shoot down into the court then perhaps a 70-200 would do the job.
5. A fairly responsive AF lens (fast and accurate) would be the order, and good OS would not hurt.

And lastly...
6. I do not know of such a lens. Somebody will, though.

So, my main point: move around.

Am.

[Warb]

For sports, and wildlife for that matter, shutter speed is what is needed if you want sharp images.
Put the ISO up until you are shooting 1/500 or better
Aperture at around f/8 is a good start
The focal length is what you need to frame the subject plus a bit more (i.e. zoom out a bit); the exact FL will vary as needed

Consider learning how to pan and also use continuous autofocus

Crop vs 35mm is not that relevant as it's what is in the frame that matters, see http://www.ausphotography.net.au/forum/showlibrary.php?title=Indexes:Appendix_D_-_Full_Frame_vs_APS-C_DX_crop_factor_w_example

Great tips for sports shooting, but that doesn't help me understand why people recommend a lens that seems too long. If you looks at the example in the thread you quote, you'll see that the bird fills far more of the frame on the crop camera at the same focal length. That is exactly my point, to frame a shot the same way on a crop camera with the same lens, I need to be much further away. So from a fixed vantage point a shot that is tightly framed on a full frame camera with a 70-200mm lens (at 70mm) will show only the shirt number when taken on a crop camera. I exaggerate but I hope you see what I mean.

ISO, aperture, shutter speed, depth of field are not in question (though for interest typically I can shoot 1/500 f/5.6 at ISO200, but normally 1/1000 with auto ISO f/4 or f/5.6). I simply wonder why people recommend a lens that (from what I can see) cannot be zoomed "out" far enough to frame a picture. For example, from the corner of a netball court to the goal is 8.5m including the run-off. The kids are (say) 1.4m tall, and with the ball just leaving their outstretched hands my 75-300mm will frame them portrait with a few inches of court/air top and bottom at 75mm. Moving any closer to the goal instantly crops feet or ball, the same if they are closer to me than the goal. Standing 8.5m behind the goal (for straight-on pictures) results in four people standing in front of me, but moving closer crops the image again, so straight-on pictures will always be waist and up at maximum.

Now I do like the tightly framed pictures of players concentrating, but I'd like to be able to choose something else too!

[Kym]

Great tips for sports shooting, but that doesn't help me understand why people recommend a lens that seems too long.

It depends. It seems you are right on the edge of the netball court. Most sports you are not that close.
Consider if you are at one end and want to get a shot down the other end? You'll need length.
If you are going ti be close then obviously a short lens is needed.
I know someone who shoots at Cricket and Football games, 600/4 is what he uses

[Warb]

4. If you are able to get up high and shoot down into the court then perhaps a 70-200 would do the job.
5. A fairly responsive AF lens (fast and accurate) would be the order, and good OS would not hurt.

And lastly...
6. I do not know of such a lens. Somebody will, though.

So, my main point: move around.

Am.

Unfortunately the area is flat, with about 15 or more courts back-to-back in a grid. There is a small rise (up to the changing rooms and cafe!) at one end, I tried shooting from it when my daughter was on an adjacent court but the "eye level" shots seem nicer, and it only works at all for a few of the courts. The courts being in a grid is also the reason why moving about is troublesome. Parents line up on the run-off line 1m from the edge of the court (can't remember what it's actually called, netball is new to me), which we're not allowed inside, and are back to back with spectators on the next court. The ground is littered with kids, drinks, balls, buckets of hot chips (healthy Saturday morning sport!), bags etc. Moving between quarters is fine, but trying to move to follow the play or frame a shot would likely be suicide.

My other option is to swap both position and lens each quarter, and use 70-200m from mid-court, and a shorter lens from behind the goal. The choice would be dependent on what position my daughter is playing - their coach is more "have fun" than "win at all costs", so they swap players/positions each quarter so the entire squad gets a game and a chance at attack/defense. Being a rural area they need a few spare players in case people don't make it to the game! Maybe I'll spend some more time on the fake court and see what I can do!

- - - Updated - - -


It depends. It seems you are right on the edge of the netball court. Most sports you are not that close.
Consider if you are at one end and want to get a shot down the other end? You'll need length.
If you are going ti be close then obviously a short lens is needed.
I know someone who shoots at Cricket and Football games, 600/4 is what he uses

I absolutely 100% agree! And that's why I don't understand the 70-200mm recommendation for netball on a crop sensor camera. When I photograph my kids surfing, I use a 100-400mm zoom (still on the 7D, so frames like a 600mm+ on a full frame, if you see what I mean) and curse it for being too short. But a netball court is 30m long and 15m wide, and I can't see myself taking too many pictures of the far end.

Netball (forgive me if you know, I had no idea until my daughter started playing) isn't like basketball. The players have allocated areas of the court, so an attacking player will always be at one end, a defender (edit: on the same team) will always be at the other etc. So if you are concentrating on a particular player, you are only working with (for arguments sake) 50% of the court. Taking "goal in the background" shots of a goalkeepers face would be end to end, of course, but most other shots would (?) be better from the sides, corners or behind the baseline. So most shots would be 20m maximum, and with the photographer behind the baseline and the kids under the goal, perhaps 2 or 3m. Perhaps a step or two further if there aren't other parents waiting to leap in to the frame just at the wrong moment, and no adjacent courts (see my last post).

A 70-200mm lens on a 7D is framing like a 110-320mm on a full frame body. That doesn't give much room at the distances involved, yet I've read countless recommendations to use that lens when the question has been "which lens for kids netball, from courtside, on a crop sensor camera". I thought I must be missing something!

[MissionMan]

If I was to guess offhand, there are probably two reasons why it gets recommended:

1. Focus speed - The 70-200 is a very fast lens
2. f/2.8 - It's goes f/2.8 which would be needed for indoor shooting (a lot of netball in Australia and other countries is done indoor)

I would say that netball is about the same court size as basketball and I've seen people recommended the same for that. Having never done it, I can't comment on whether it's ideal though.

[ameerat42]

Is it a Canon lens - that one? - the OP referred to? I know it's a Canon 7d he's got.

Back to your nub, Warb, I think it's sounding like one or all of:
a) conventional wisdom
b) urban mythology
c) arm waving.

Am(known to be skeptical sometimes).

[arthurking83]

.....

Netball (forgive me if you know, I had no idea until my daughter started playing) isn't like basketball. The players have allocated areas of the court, so an attacking player will always be at one end, a defender (edit: on the same team) will always be at the other etc. So if you are concentrating on a particular player, you are only working with (for arguments sake) 50% of the court. Taking "goal in the background" shots of a goalkeepers face would be end to end, of course, but most other shots would (?) be better from the sides, corners or behind the baseline. So most shots would be 20m maximum, and with the photographer behind the baseline and the kids under the goal, perhaps 2 or 3m. Perhaps a step or two further if there aren't other parents waiting to leap in to the frame just at the wrong moment, and no adjacent courts (see my last post).

.....

This is obviously a factor if you are concentrating only on the one player.
If a pro(or semi pro) was photographing the game, they would be required to get many more shots of individual players.

I have no idea on the rules of netball(until you'd mentioned them), but your points all seem to make perfectly good sense if you're focusing on the one player.

Sigma used to make a 50-150/2.8 for APS-C(now discontinued), and Tokina make a 50-135/2.8(not sure if it's still a current product tho).

I reckon a 24-120mm lens sounds about ideal (in terms of framing) for APS-C sensors. Only problem of course is that it's certain to be f/4. As Rick mentioned, focus speed is a relevant feature, and most of these f/4 type lenses will not be quite as fast or accurate to focus as a pro level f/2.8 type lens design.

Sigma and Canon both make 24-105's at f/4. Nikon's 'equivalent' is a 24-120 @ f/4.

[ricktas]

re f2.8. It is not about shooting at f2.8. When people talk about a fast lens, the lens is no faster than any other lens in regard to shutter speed. 1/500th of a second is the same on every camera/lens. Where f2.8 comes in and a lens being called fast, is that a lens auto-focuses at its widest aperture. So even of you have f8 chosen, when the lens is auto-focusing, it opens the aperture up to f2.8, letting a lot more light in, focuses, and then goes back to f8 for the photo to be taken.

So that gives you the reasoning behind using a fast lens for sport (or for many other genre).

As Kym; says, 70-200 is a good all-round focal length for sport in general. Motorsport, footy, and many other sport you cannot get right up to the edge of the playing field, or the playing field is bigger. Netball uses quite a small playing area compared to most other sport.

But you can use whatever lens you want to, whatever works. Macro lenses for instance, make great portrait lenses as well. Just cause a lens is marked Macro, doesn't mean it cannot be used for lots of other genre. Sport is the same, just use the lens that is needed to get the job done.

[Shane H]

If I may offer a suggestion, go to flickr (it would be my choice. Others may suggest something else), search for Netball and, if it's there, have a look at the exif data of any photos that interest you. The focal length and lens used may help. Of course it won't tell you how far away from the subject the camera was but with your experience with netball courts, it could still be of some use.

Cheers
Shane

[Warb]

Where f2.8 comes in and a lens being called fast, is that a lens auto-focuses at its widest aperture. So even of you have f8 chosen, when the lens is auto-focusing, it opens the aperture up to f2.8, letting a lot more light in, focuses, and then goes back to f8 for the photo to be taken.

So is there a measurable difference in focus speed that is attributable solely to maximum aperture? I realise that some lenses (and some bodies) focus faster than others, but assumed that was simply a design feature. But from what you say it would appear that if I put a ND filter on, to effectively lose a stop of light, then the same lens will focus more slowly? Or have I misunderstood?

If this is correct, is there a light level at which it starts to have an impact? For example, regardless of the lens in use there is a huge amount of light at midday in summer, far less in the evening in winter. Under both conditions the lens will have the same maximum aperture, but an enormous variation in light to focus by. If the focus speed changes with maximum aperture due to the amount of light, can we deduce or calculate the brightness required to attain top focus speed? Presumably there is a point where the mechanical components can't go any faster, thus limiting the maximum speed of focus, and presumably there is a point where the reducing light levels cause the "brain" speed or increasing number of try/fail cycles (not sure how the focus mechanism works!) to reduce below the speed of the mechanicals? So the next question becomes "what level of light is required"? Clearly a badly lit church might be be dark enough that (in isolation) a reduction in wide open aperture for focussing might cause a reduction in focussing speed. But does the same apply at midday, midsummer on a sunny day on Bondi Beach?

Perhaps it was easier when I just assumed the 70-200mm f/2.8 Canon lens was just inherently fast to focus because it was designed better!

I've spent some more time testing, and also looking at "pro" photo's of netball, and I'm starting to understand. It seems that the vast majority of "pro" pictures of netball are portrait, and very tightly framed - often just waist up, single player. It is rare for a shot to include more than two players (in a close attack/defence face-off). The ball (having left the hands) appears to be of no great significance, and is often not included in the frame. Shots of two players in a battle under the net, concentrating their hardest, face-to-face, whilst a third is sitting down retying her laces don't make the cut!


This has been helpful, thanks to everyone who has contributed so far! I now have some idea why the pro's use 70-200mm, from both technical and "artistic" perspectives. I'm still not 100% sure it would work for me on a 7D, but more time with my daughter at the practice hoop will no doubt answer that!

[ameerat42]

A 2ple of points and suggestions:
1. Don't forget that the aperture you set (f-stop) for any given light conditions will (transmission factors aside) yield the same intensity at the sensor.
2. You could try to determine a minimum useful illumination (for AF on your cam/lens) by progressively darkening a scene, say indoors with variable lights.
3. You'll know when the AF causes the lens to "hunt" for a set position. I think there has to be a fairly great drop in light levels before problems arise.

[Dug]

It makes more sense that a lens like the 70-200 being say 90% useful on a full frame (as you would use it going by your experiments) will be far less useful on a crop for focal length.
Perhaps it is a case of the crop users just wanting it to work, or they take less variety of shots when using one, or it is just what they have as their go to sport lens and are less focused on just one sport even though netball is one they say they use it for.

Seldom does one lens work for an entire application, there will be some compromise in its feature set.
The pro might say the 70-200 is the best on their FF for the bulk of the work , but they could well have a second body with a lens for end court and crowd shots as a secondary lens.

Ignore the status quo if you have a particular need, like a one lens solution for all the angles you want at a netball game.

Your experimenting is a good methodical approach to this and there appear to be a number of lenses suggested here that will fill your focal length needs.

[ricktas]

Yes if you put ND filters on you lose some of the ability to focus. In fact if you have ND filters on as the sun is setting, as it gets darker you may just find the lens hunting to try and find focus. You will hear the AF motors whirring away but the focus does not lock. At what point this occurs depends on the lens, the camera AF system, the subject and more. AF works by finding contast edges in a scene and locking onto them. This is why you cannot sometimes lock onto a plain blue sky, but put the AF point on the edge of a cloud and it can lock on quickly. There are so many factors that affect how quickly AF lock occurs, that the lens is just one part of the whole.

As for the pro netball photographers. They may actually not use AF at all. Set camera to manual focus. Pick a point on the court, manually focus on that. Then just wait for the action to take place at that point of the court. Once you watch quite a few games you start to learn there are specific places on the court that more action takes place and there are dead spots were not much happens at all. Once you know where those 'active' spots are you can manually focus on those points and wait till the action comes to you (so to speak).

[arthurking83]

So is there a measurable difference in focus speed that is attributable solely to maximum aperture? I realise that some lenses (and some bodies) focus faster than others, but assumed that was simply a design feature. But from what you say it would appear that if I put a ND filter on, to effectively lose a stop of light, then the same lens will focus more slowly? Or have I misunderstood?......

A lens's ability to focus quickly and accurately will be a factor of at least three variables(in this situation).

light levels, design(of the lens) and the body it's used on.

Other factors that can impact on focusing ability is the subject matter at the point of focus and of course the lens's ability to focus at certain distances.
(I doubt that in a netball situation such as you're referring too, this is going to be relevant at all)

I doubt that you would notice any appreciable drop in focusing ability on a Pro model 70-200/2.8(say from Canon or Nikon) with only 1 Ev of light reduction(ie. light levels, or as you proposed, the use of an ND filter).

Note that while Sigma and Tamron also make 'pro' level 70-200/2.8 models too, they don't focus as fast as the much more expensive CaNikon equivalents. But in most circumstances they're probably still acceptable.
I have Tamron's older model 70-200/2.8(non VC) model, and it uses a standard DC type focusing motor.
In comparison to the Nikon AF-S version(both new and old), it's slower. But this is only really noticeable when a side by side comparison is made.
When compared to the newer VC model Tamron lens of the same type too tho, the older non VC model also feels a lot slower too.

But in isolation, on an old Nikon D70s or the more capable D300 and D800E, my older model Tammy 70-200/2.8 still focuses 'fast enough' .. where fast enough means I can't ever remember a situation where it left me wanting for focus speed.
I'd still want the faster focusing model tho, as the speedier focusing speed advantage is better to have, then not to have it.

Things to note too: just because the lens has a fast maximum aperture, doesn't automagically imply that it is going to focus fast or accurately!
as an example of this: I have Nikon's 1052.8 VR macro lens as well.
This macro lens is revered by some, and it is in Nikon pro lens lineup(gold ring).
But if I needed a fast focusing lens at this focal length, then I'd prefer to use the Tamron 70-200/2.8 as it's appreciably quicker to focus than the 105/2.8 Nikon lens.
The Nikon 105 has their ultrasonic/AF-S ring type focusing system, yet it's quite slow to focus and generally erratic if it isn't kept close to the correct focusing distance.

Overall tho, the general usage of a lens like the Canon 70-200/2.8 will probably demand very quick focusing from such a lens type. So Canon(and Nikon) will place emphasis on this ability on those lenses.

If you take the example of a lens like an 85/1.4, or in Canon terms 85/1.2 .. these lenses probably don't focus as quickly as the 70-200/2.8's from their respective makers can do.
Absolute focusing speed probably isn't as high a priority from those lens types, as would be focusing accuracy from such lenses.
Yet these lenses are obviously letting in more light with their much larger apertures.

I think your assumption that lens design has more to do with it is still a easier way to look at it.

[shaneando]

There's a few reasons why a 70-200 would be often recommended.

First, many professional photographers would use a full frame camera so the angles of view would be more suitable. Second, most 70-200 lenses are designed with sports in mind. They are usually f/2.8 for high shutter speeds in lower light, and have very fast focussing mechanisms. Also they are normally robust designs with good weather sealing.

Also bear in mind that most pro sports photographers will be carrying and using two bodies, with different lenses, to cover action over a wide area. For example they may have a 24-70 and 70-200 for netball. And often they are restricted to certain areas so don't have the freedom of moving each quarter.

If as you say most of your daughter's games are outside, there is little need for an f/2.8 lens which opens up a lot of possibilities. For sports, shutter speed is the key parameter and in daylight you should be able to get high enough speeds with even an f/5.6 lens.

There is nothing wrong with your approach at all. You know the conditions you will be shooting in - right on the sideline, ability to move easily, mostly daylight - so pick the lens that suits you best and don't worry what others are using.

Good luck!

- - - Updated - - -


re f2.8. It is not about shooting at f2.8. When people talk about a fast lens, the lens is no faster than any other lens in regard to shutter speed. 1/500th of a second is the same on every camera/lens. Where f2.8 comes in and a lens being called fast, is that a lens auto-focuses at its widest aperture. So even of you have f8 chosen, when the lens is auto-focusing, it opens the aperture up to f2.8, letting a lot more light in, focuses, and then goes back to f8 for the photo to be taken.

So that gives you the reasoning behind using a fast lens for sport (or for many other genre).



Rick I'm going to have to respectfully disagree with this part of your post. Shutter speed is the main reason for using fast (large max aperture) lenses for sport and wildlife.

On Nikon cameras at least, the autofocus sensors are positioned in a way that corresponds with an aperture of f/5.6, or on some of the latest bodies there are a few with f/8 spec. A lens with a larger max aperture, say f/4 or f/2.8, will certainly let more light in, but no additional light reaches the autofocus sensors. Maximum apertures smaller than f/5.6 will result in less light being received and will impact on AF performance. I think Nikon has started adding f/8 max aperture to some of its latest bodies in recognition that many people use teleconverters which often reduced the max aperture to f/8.

I believe some Canon bodies do have AF sensors that are positioned for f/2.8 to take advantage of a larger aperture, but this is a camera body AF system feature, not a lens feature.

AF systems (on the body) are normally specified with a minimum light level at which they will work, for example on the D750 it's -3EV. This has more of an effect on focus performance in low light than the lens aperture.

Regards

Shane

[ricktas]

But shutter speed is shutter speed Shane. 1/1000th of a second is the same no matter what lens or camera. my comment re a fast lens does not relate to shutter speed, but to a fast lens being 'faster' to focus due to using a wider aperture to do so. Saying that a fast lens for wildlife etc is due to shutter speed, is incorrect, cause the shutter speed of any lens is the same as every other lens. 1/1000th isnt faster on one lens than another.

The term 'Fast Lens' refers to focusing benefits, not shutter speed.

[shaneando]

I have to admit that most of the time when I've heard the term "fast glass" it was in reference to the maximum aperture, not the focus speed, so maybe some confusion about that particular term.

In any case, for the same lighting conditions, an f/2.8 lens will allow me to select or achieve a shutter speed that is four times faster (two stops) than an f/5.6 lens. That is the advantage of a lens with a larger max aperture.

And I still disagree about the lens being faster to focus due to the wider aperture. I've tested this myself, admittedly not scientifically, but good enough to show the principle. On my D750 I tried my 85mm f/1.4D vs the 16-85mm (which is f/5.6 at 85mm) in a dark room - no noticeable difference in focus performance. I think what often happens is that high end lenses with f/2.8 apertures are normally built with very fast AF motors and electronics, and focus faster anyway, but this is not related to the aperture itself.

There is a very good, but somewhat technical, discussion on this subject at dpreview at the moment - http://www.dpreview.com/forums/thread/3713509

[ameerat42]

Shane. That ability to focus in low light that you refer to may be for your camera only. Certainly on mine it hunts some in low light.

To the term "fast" in lenses... Nowadays it mostly refers to the greater aperture a lens has. Historically, this "fast"-ness was linked to exposure time.
So, the "faster" the lens, the "faster" the exposure - ie, the less the exposure time.

And now a caution: TECHNICALLY, it's how low an F-stop the lens can attain, because "aperture" simply means "opening". So,

- if a f=50mm lens had an "aperture" of 50mm diam, then it would CERTAINLY be "fast" at f/1.0.
- but if a f=300mm lens had an aperture of 50mm diam, then it would not be as "fast" at f/6.

But you probably already knew this:D

PS: Oh! Read about it here (http://en.wikipedia.org/wiki/Lens_speed)

[peterv]

.....Re: link you posted.....

Excellent,though like you said a bit technical.

A couple of weeks ago I would have had no idea what was being said. Now,I actually figured out what was being explained.

[arthurking83]

..... On Nikon cameras at least, the autofocus sensors are positioned ......

Shane


Shane I have the impression you may have misunderstood how Nikon's AF system works.

The sensors aren't positioned for a specific aperture .. they are sensitive to a specific aperture value.

If you have ever used an old film camera that has a split prism feature in the focusing matte, you will basically understand how the PDAF AF system works.

If not, then in brief, the image that is formed on the sensor is split into two parts. As you view a specific subject and place the split prism area onto that subject the subject appears split into two parts .. disjointed!
As you focus closer to the correct plane, the subject becomes one .. ie. formed again. This is now in focus. PDAF sensors work the same way(CDAF works differently).

Again, if you've ever used a split prism, you know that at about f/5.6 the split prism is useless(on the whole) .. and at f/8 is totally useless. Looking through the viewfinder with an f/8 lens .. all you see is black in the split prism area. At f/5.6 you get a partial blackness .. in comes in and out dependent on the angle you're eye comes into the viewfinder.

PDAF works in the same way.
Nikon's f/5.6 sensitive focus points aren't positioned in a special way. That I can remember, they are all basically positioned in a central area of the focus point group. As Nikon have many and varied AF systems types for many of their camera, it's hard to describe it in a singular manner.
But (lets say) on a D750, the central vertical column of 15 AF spots are all f/5.6 sensitive. the rest may need f/2.8, and the central AF spot is sensitive to f/8 maximum values.

So you're comment that:


... the autofocus sensors are positioned in a way that corresponds with an aperture of ...

is incorrect.

They are sensitive to f/5.6 only .. this sensitivity is probably dependent on both the hardware tech, as well as the software used to make sense of the hardware.

Also you said that:


A lens with a larger max aperture, say f/4 or f/2.8

No matter the sensitivity of the AF sensors, a larger aperture lens always allows more light into the system, which includes both vf, and AF sensors!

An f/4 lens allows less light through to the AF system than does an f/2.8 lens. Whether this translates to AF performance differences is is dependent on so many factors.
Light being the predominant one.

an example!
if you had two basically two identical lenses, other than the point that one was f/2.8 model and the other was an f/4 model .. otherwise all other design factors were identical.

We had this lens on a D750, which we know is good down to -3Ev ambient lighting for AF.
-3Ev lighting is basically a similar experience to the lighting from a full moon .. so lets say we're shooting portraits under moonlight only .. out in the sticks with only a full moon for the light.

In this situation, don't you reckon the wider aperture f/2.8 lens would AF more reliably(if at all!) than the f/4 lens would.

[shaneando]

Arthur, I think you are confusing the AF points that are visible in the viewfinder vs the actual sensors inside the AF module, which is what I'm referring to.

I highly recommend you read this thread on dpreview - http://www.dpreview.com/forums/thread/3713509. If you haven't already come across her, Marianne Oelund is a very well credentialled photographer as well as an expert on the design and operation of photography equipment and theory. She uses Nikon equipment.

Using the actual components of a Nikon D300 AF module set up on an optical test bench, she shows that as you stop down a lens from f/5.6 to f/8 to f/11 etc that the amount of light hitting the AF sensors decreases, and ceases altogether at f/11. However opening up to f/4 to f/2.8 does NOT increase the amount of light hitting the AF sensors. It seems counterintuitive, but it is a real physical characteristic of the AF module.

My own summarised understanding is this (although again I recommend reading the above link for a full correct description) - the sensors inside the AF module are located at the edge of the exit pupil when the main lens is set to f/5.6. I don't really know why f/5.6 was chosen, but I suspect it's a good compromise between getting enough light, having a large enough distance for the triangulation calculations, and giving the ability to work on almost every lens available.

BTW I realise that we have gone way off topic here. This subject interests me so if it's preferable we can start a new thread.

Cheers

Shane

[arthurking83]

Bluddy 'ell!
I started reading that thread, and didn't get out of it until about 2AM! :D

Anyhow .. I've read Marianne's threads on DPR previously, but don't frequent those boards due to the general nature of them.

Very illuminating information on the inner workings of the AF system, and so far she seems to have presented a very good argument for her case.

Two points I find unresolved as yet.
It's still a bit inconclusive that the amount of light at the AF module begins at an f/5.6 aperture equivalent, even tho much data has been posted to prove the idea.
The images captured by snellius appear to not only prove the point, but also to disprove it as well.
(I'm not a DPR member and won't join in that forum every again(unless they seriously consider some level of moderation, so I can't post these questions directly to snellius or Marianne)

The images snellius shows from underneath the AF module definitely indicate that light levels are reduced starting at about f/5.6.
But, the images taken from the front of the camera/lens, don't correlate as well to support the arguments made.

Looking at the images coming from the front, at f/1.4 it's clear that the entire AF module is visible .. most notably the centre cross pattern portion.
At f/2.0, the periphery AF points are still visible, but at f/2.8 they are obscured .. even tho we can clearly see the centre cross pattern AF pattern.

While it appears that Marianne is probably completely correct, the images presented seem to leave a small level of doubt about the actual workings, as opposed to the implied workings.

The second point not touched on is that as the AF modules most important component is a imaging sensor(just like the sensor that captures the image) software allowances haven't been made. This would be either improbable, or impossible to determine and Nikon wouldn't be willing to reveal such info either.
That is, what level of gain does the AF module sensor receive at differing aperture values .. and different lenses mounted.
Considering all this electronic communication between camera and lens, I seriously doubt that Nikon haven't used some software tweaking in the AF system to cater for various lens types!

Anyhow, as you said, the last few replies are going a bit OT(although still pertinent to the OP's original point).

It is interesting and on topic to note that, if true the AF module can only see apertures of f/5.6 or so. I can't imagine Canon's and other brands using completely different PDAF module methods. While they may have differences .. the basic processes will be similar/same.

So that f/2.8 lens may not focus as quickly as it may otherwise do! :D

I'm happy to continue in a new thread if the interest to do so exists.

[shaneando]

Arthur

I have started a new thread - http://www.ausphotography.net.au/forum/showthread.php?137214-DSLR-autofocus-system-design-and-operation

Agree on the dpreview forums - I do frequent there for technical information but you have to filter out a lot noise, and there are only a handful of regular posters that I really bother to read at all.

Cheers

[shaneando]

To the OP...

We got off track with a discussion about the effect of aperture on autofocus. However there are some undeniable reasons why 70-200 lenses tend to be recommended for sports in general:

- image quality tends to be at or near the best for each respective manufacturer
- having a fixed aperture of f/2.8 allows for good shutter speeds even in relatively low light.
- they do focus quickly if for no other reason than they tend to have powerful focus motors
- they have a reasonable zoom range of nearly 3:1
- usually they are solidly built with good sealing.

Having said all that you need to determine which lens suits your own circumstance best, which it seems like you are doing!

[Warb]

To the OP...

We got off track with a discussion about the effect of aperture on autofocus. However there are some undeniable reasons why 70-200 lenses tend to be recommended for sports in general:

I completely agree with the "general sport" recommendation, and the reasons for it. What had me concerned was the recommendation of that lens in the specific example, including all my own parameters. It seemed (and still does!) to be equivalent to being recommended a Ferrari when the question included the statement "I live at the end of a 10km rough dirt road". Sure the a Ferrari might be a great car, but is it really suitable for driving on dirt roads?

After much research and thought, I suspect that the majority of recommendations for the 70-200mm in the situation in question are gut reactions attributable to the "general sport" suitability of this lens and do not take the details of the situation in to consideration. In fact if I was considering using two bodies, I would certainly be using a 70-200mm on one of them. But for a single lens solution on a 7D, for the average "kids photo album" photographer, I'm still not convinced that a 24-105mm or 18-135mm (or something in that range of focal lengths) isn't a better choice. When the new season starts, I'll have my chance to find out!!

[shaneando]

I completely agree with the "general sport" recommendation, and the reasons for it. What had me concerned was the recommendation of that lens in the specific example, including all my own parameters. It seemed (and still does!) to be equivalent to being recommended a Ferrari when the question included the statement "I live at the end of a 10km rough dirt road". Sure the a Ferrari might be a great car, but is it really suitable for driving on dirt roads?

After much research and thought, I suspect that the majority of recommendations for the 70-200mm in the situation in question are gut reactions attributable to the "general sport" suitability of this lens and do not take the details of the situation in to consideration. In fact if I was considering using two bodies, I would certainly be using a 70-200mm on one of them. But for a single lens solution on a 7D, for the average "kids photo album" photographer, I'm still not convinced that a 24-105mm or 18-135mm (or something in that range of focal lengths) isn't a better choice. When the new season starts, I'll have my chance to find out!!

Sigma make a 50-150 f/2.8 zoom lens, which on a crop sensor body is very close to 70-200mm on a full frame body. I've never used one, although several times I've considered getting one for my D7100. It is reportedly a very good lens, it might be worth checking out.

[Warb]

Sigma make a 50-150 f/2.8 zoom lens, which on a crop sensor body is very close to 70-200mm on a full frame body. I've never used one, although several times I've considered getting one for my D7100. It is reportedly a very good lens, it might be worth checking out.

I agree, this lens was on my shortlist and on paper I thought it looked ideal. Its only possible downside is that it's an APS-C lens, ruling out upgrading to a full frame body - though, of course, such an upgrade would probably make the lens "too short" for the job anyway! But it looked exactly right, in fact as an f/2.8 image stabiliser lens it certainly appeared that Sigma had designed it specifically as a 70-200mm "L" equivalent for APS-C.

Sadly it appears to be discontinued. Whilst the Sigmaphoto.com.au website states "Sigma mount discontinued" and lists Canon under "available mounts", I was unable to find a dealer who had one and most online sellers either don't list it or have it flagged as discontinued. Second hand there is/was one on Gumtree, but I found it odd that it was "purchased earlier this year", and "used mainly in the studio", yet was also stated as "just been serviced". Other than damage or fault, why would a lens less than a year old require service?

But if I ever find one on a shop it will be hard to leave it there!

[ricktas]

I agree, this lens was on my shortlist and on paper I thought it looked ideal. Its only possible downside is that it's an APS-C lens, ruling out upgrading to a full frame body - though, of course, such an upgrade would probably make the lens "too short" for the job anyway! But it looked exactly right, in fact as an f/2.8 image stabiliser lens it certainly appeared that Sigma had designed it specifically as a 70-200mm "L" equivalent for APS-C.

Sadly it appears to be discontinued. Whilst the Sigmaphoto.com.au website states "Sigma mount discontinued" and lists Canon under "available mounts", I was unable to find a dealer who had one and most online sellers either don't list it or have it flagged as discontinued. Second hand there is/was one on Gumtree, but I found it odd that it was "purchased earlier this year", and "used mainly in the studio", yet was also stated as "just been serviced". Other than damage or fault, why would a lens less than a year old require service?

But if I ever find one on a shop it will be hard to leave it there!

Contact Sigma direct. Ask them which stockist might have one. They will check their inventory and give you the contact details of the stockists who last had some shipped. You might get lucky! I have done this. Meant a call to a shop in Melbourne, ordered over the phone, got it the next day. Otherwise it was a wait of 6 weeks or more, if I ordered one from my local store.

[shaneando]

I agree, this lens was on my shortlist and on paper I thought it looked ideal. Its only possible downside is that it's an APS-C lens, ruling out upgrading to a full frame body - though, of course, such an upgrade would probably make the lens "too short" for the job anyway! But it looked exactly right, in fact as an f/2.8 image stabiliser lens it certainly appeared that Sigma had designed it specifically as a 70-200mm "L" equivalent for APS-C.

Sadly it appears to be discontinued. Whilst the Sigmaphoto.com.au website states "Sigma mount discontinued" and lists Canon under "available mounts", I was unable to find a dealer who had one and most online sellers either don't list it or have it flagged as discontinued. Second hand there is/was one on Gumtree, but I found it odd that it was "purchased earlier this year", and "used mainly in the studio", yet was also stated as "just been serviced". Other than damage or fault, why would a lens less than a year old require service?

But if I ever find one on a shop it will be hard to leave it there!

Personally I wouldn't worry too much about whether it's been serviced recently, but it depends on what was done. It could actually be a positive - if the service was done by an authorised Sigma repairer and the seller has the receipts from the service then you could have a bit more confidence that the lens is in good condition. Chances are it was just a basic service to check if everything was OK.

My local camera shop has a used version on the shelf for $795, but it's Nikon mount!!