Help with focus please

14 posts / 0 new
Last post
youcantryreachingme's picture
Help with focus please

Despite all the ideas that I posted in a recent thread, I still struggle to get the entire bird in focus.

Below is a recent example, followed by the questions I put to a photographer. I'm sharing them here hoping that some of the contributors who seem to post fabulously sharp images can help. Thanks in advance!

Red-browed finch (Neochmia temporalis)

EXIF tells me this was f/7.1 at 1/400s and ISO 400. Lens is 120-300 f/2.8 at 300mm.

Admittedly I was some distance from the bird - the width of this crop is about 20% of the width of the full frame.

I can't recall for certain but I would expect that I got the camera to focus on the bird. I do recall I was using single focus (ie. single dot), as I say, almost certainly targeted on the bird.

You can see that the branch it's standing on is more in focus behind the bird than in front of the bird. (The left side of the branch is further from the camera than the right).

I got this lens because I got the impression that f/2.8 will really make portraits, in particular, pop out from their backgrounds. However I found when shooting wildlife that often the portion of the subject in focus was absolutely tiny. So in this case I've stopped down the aperture, yet even at f/7.1 the depth of field is tiny - maybe 3 inches at most?

I think there are two issues here - first is that the focus seems to be behind the bird. On the same day in different scenarios, photographing lizards I found similar - in those cases I know I focused on the eye. In some cases the eye is in focus but the rest of the lizard blurs even though we're talking just a few cm behind the eye.

The question is how to get consistently sharp images? Should I be stopping down the aperture even more? If so, then what's the point of having an f/2.8 lens? I might as well get a cheaper lens with more zoom range but smaller maximum aperture.

Another note - I found in the studio, photographing portraits (on a 24-70mm) that I could take 3 or 4 shots in a row; one might be absolutely crisp, while the others are not - despite same lighting, positioning, and perhaps only marginal movement in the model. Why?

To try and get crisp focus across the whole animal (in this case), I stop down the aperture. But the natural light means I have to increase the ISO. I find the shots to be grainy, and lacking detail. If I take the same shot with flash (for example, with the lizards that sit in the same spot), I find the shot with flash contains much finer detail on the lizard. Why can't I get that detail without flash? Or is that simply just how it is - you need to fully light up the subject to get the fine resolution of detail?

I suppose that's a few questions all rolled into one. Thanks for any tips you can provide!

Red-browed finch.

Rick N
Rick N's picture

Hi Chris, Interesting post.

I'm no expert but " simply just how it is" sums it up. You trade off one thing for another.

I deal with it in a few ways.

Get as close as possible but understand that the depth of field will narrow.

Stop down as much as possible depending on what level of bokeh your after.

Use fill flash as much as possible.

Most importantly me for me is to finish off the focus manually, I use the glint of the

sun on the eye as a good indicator, so using the auto focus for the initial focus

and fine tuning by eye.

Obviously this method is for static subjects.

Don't discount that the lens might need micro tuning as well.

Again I'm no expert but these things help for me.


youcantryreachingme's picture

Rick N wrote:

Get as close as possible but understand that the depth of field will narrow.

Interesting point I hadn't considered.

Another colleague suggested f/7.1 is still quite a shallow depth of field. He personally hates the appearance of flash on birds. I see where he's coming from, but it's hard to escape the idea that flash really brings out more detail. See lizards for comparison below.

Left = with flash, ISO 200, f/8.0, 1/250s.
Right = no flash, ISO 3200, f/5.0, 1/320s.

Mind you - every single variable is different, and "without flash" is underexposed, but I don't think bringing up the exposure would bring in as much detail as is visible in the flash shot. There is an enormous amount of extra detail with flash. The lizard was deep under a ledge.

How does one have a lens micro-tuned? Sigma lens, with Nikon body.

Eastern water dragon

Rick N
Rick N's picture

Will depend what body you have as to whether it will have the capability to micro tune.

When I am using fill flash it is generally at faster shutter speeds and close up using hi speed sync

so the flash is not so noticable and really is just fill.

Lachlan's picture

I think most of the extra detail with the lizard on the right is due to the massively lower ISO. Plus there's the whole ETTR thing. 

Don't get rid of the 300mm f2.8 lens because you're stopping down! Even if you're stopping down, you still get an extra stop of light over someone with a 300mm f4 lens (like me) who has stopped down to the same place. That's what you're paying extra for, the useability in lower light levels is a byproduct of that.  

To me, it looks like a problem with adjustment. Apparently, it's quite often a consideration with Sigma lenses that they need to be significantly adjusted. How you did that would depend on your camera, couldn't say without the model type. If that doesn't fix it you may end up having to send the lens in to an authorised repairer to get the lens itself calibrated. 

IMO, there's nothing wrong with the size of area you have in focus for f7.1. You do have a supertelephoto lens after all, and they're notoriously finnicky. I reckon if the focus was nailed onto the Red Browed Finch's eye, the photo would look perfectly fine. 

I'm not a fan of flash in wildlife photography. Ideally, it'd be really useful for wildlife photography, but flash is usually used in static situations like a studio where the effects of the flash can be calculate and the subject posed accordingly. It's just too easy to muck up an otherwise ok photo with a mistimed flash. You have to be a real expert to use it; I'm not. 

youcantryreachingme's picture

ETTR - wow. I actually hadn't heard of this (or totally dismissed and forgot about it if I have). Just read an introductory paragraph and it looks exciting. I've been working on a method for correct exposures using the in-built light meter (and that process seems to go out the window when I get low light scenarios like the lizard, or end-of-day shooting, or try to use flash) but I'm excited to think about a new approach. Should be fun to explore :)

I don't understand why this is true: "Even if you're stopping down, you still get an extra stop of light over someone with a 300mm f4 lens (like me) who has stopped down to the same place." - could you please explain it? I think you're saying if I take a shot at 300mm and f/4 that my lens will allow through an extra stop of light over your lens at the same settings. Is that right? How does that work?

"it's quite often a consideration with Sigma lenses that they need to be significantly adjusted. How you did that would depend on your camera" - I really need to read up on what people have called micro adjustment. It seems from your words that this is something I might be able to do myself? Can you expand a little on how that would be? Model is Nikon D800E.

It seems surprising to me that one invests a fair amount of money into a lens only to find they may well have to bring it in to a shop and (presumably) spend more to tune it. I suppose an analogy with a piano or car might be justified here (and help coping with the thought of even more money spent).

I'm still at an open mind with regards to flash for wildlife. As I say, the key thing seems to be difference in resolution / level of detail, but I need to investigate that more, and your suggestion that ISO is playing the bigger part (and that I could explore a technique for ETTR) too.

(PS.. Feel free to tell me to RTM! :D hahaha.. I just need more hours in every day!) (Don't we all?)

youcantryreachingme's picture

For anyone interested, this looks comprehensive (for "exposure to the right" - ETTR - shooting):

Ok, I think the method I was taught is actually ETTR while spot-metering the brightest part of the scene without actually using the term "ETTR".

That said, I haven't been conscientiously bumping exposure up by a few stops, and the article introduces appropriate post-processing for different dynamic ranges, which is new to me. I'm liking this - feel like I'm actually getting equipped with helpful information/knowledge (and it seems bird-lovers are more forgiving with technique questions than some other photography-specific forums, for sure! Thanks!)

Lachlan's picture

Hope this helps:

There is a fixed amount of light present in any given photo. However, the amount of light gathered by equipment varies. A smaller f-stop value (eg, f2.8 as compared to f4) lets a more light to fall on the sensor that a larger one, because the aperture (the hole the light goes through) of the camera lens is bigger. So if you stop your lens down from f2.8 to f4 (1 stop of light), it's letting the same amount of light through the aperture to the sensor as a lens with a maximum aperture of f4. However, because of the relationship between ISO, shutter speed and aperture, doing so means that you can increase shutter speed or decrease the ISO by 1 stop (eg 1/800-> 1600, ISO800-> 1600) before it's at the same level as a slower lens.

So, hypothetically, if you (300mm f2.8) and I (300m f4) we both photographing the same subject from the same spot with the same light (remember, this is hypothetical), I would have settings of f8, ISO800 and a shutter speed of 1/20, while you could have f8, ISO800 and a shutter speed of 1/40 (1 stop of light difference). In bad light conditions and a long focal lengths, the extra stop can make quite a lot of a difference (it's much easier to photograph a bird at 1/800 than 1/400). However, the lens has to be much bigger to capture the extra light, hence the price premium for f2.8 over f4. Same deal with the 50mm f1.2 or 1.4 lenses that cost heaps. They let in extra light and are bigger so cost more but are extremely useful for some applications (esp. low light situations where you can't wait for the light to get better (eg. a wedding)). 

youcantryreachingme's picture

Thanks Lachlan,

I believe I understand what you're saying, but I still don't understand how it makes sense.

300mm equal between us, means we are both viewing an identical scene with identical lighting.

ISO800 equal between us, let's assume, means our cameras are identically sensitive to the light.

f8 equal between us *should mean* our lenses are opening to the same extent, thus letting through the same amount of light. (I will get back to this later).

It does not make sense to me then that one camera can now use 1/20 while another uses 1/40 to record the same amount of light, just because the lens' maximum apertures are different.

So this led me to ask the question of what the f number is actually measuring and I found it is "the ratio of focal length to effective aperture diameter".

If it is true one camera can use 1/20 while another can use 1/40 to produce an equal exposure to the light, then the same aperture of f8 *must* be letting through a different quantity of light between the two cameras.

If aperture is a ratio involving the aperture diameter (which it is defined to be) then the absolute amount of light allowed through depends not on the aperture per se, but the diameter of the lens. You wrote "the lens has to be much bigger to capture the extra light [to support an f/2.8 value]" but I do not think that is right (or my assumption that you're saying "to support an f/2.8 value" is not right). A lens which has the diameter of a military cannon - whatever cannon you choose - can be given an aperture of f/2.8 as long as the ratio of the aperture opening relative to the focal length acheives that. (Note, I don't know what implication that has for focusing that image onto a sensor the same size as either of our cameras' lenses).

Think of it another way - a tiny point-and-shoot camera can have a super tiny lens and yet still might acheive the f/4.0 of your lens, despite *not* having all the extra glass that your lens has.

Hence it may well be the actual physical diameter of the lens that supports the extra stop of shutter speed at the same aperture.

What do you think?

If that's right, it means the fact my lens has a larger diameter is the reason why it could shoot at a faster shutter speed, all else being equal.

If that's right, then I still have my original question, although perhaps reframe it a little - what additional benefit does a maximum f/2.8 give me, if I am shooting at f/8? (ie. if it's not the maximum that gives me the additional stop for shutter speed, but rather the lens diameter).

youcantryreachingme's picture

Hi Lachlan,

I think I'm wrong. And I think something you said that I thought was wrong, is actually right. But I still think you're main point doesn't make sense too! :D hehehe.. (I hope you're laughing!)

I was suggesting that because my lens has a larger diameter that means f/8 on my lens is a physically larger opening than on yours. I think this is wrong. I think I was thinking of the opening (aperture) as a percentage of the lens diameter, but it's not. It's as a percentage of the focal length (ie. of 300mm), which is fixed between us. So for f/8, this means the aperture should have an average diameter of 1/8th of 300mm = 37.5mm.

So therefore what you said about my lens requiring larger-diameter glass to support f/2.8 is true! If I want an opening that's 1/2.8th of 300mm (that would be an opening of 107mm) I am not going to get it if my lens' diameter is only 90mm now, am I?

So yes, I agree, my lens needs to be fatter to support 2.8.

But I still don't see how that allows my lens to operate at 1/40 while yours can only manage 1/20 in our example. If I've stopped down to f/8 and you've stopped down to f/8, then the opening of the aperture is 37.5mm in both your lens and mine, no matter how fat my glass is or what my maximum aperture is.

So I still think the only benefit of f/2.8 is that I *can* shoot at f/2.8 if I want to. But if I choose to shoot at f/8 then I am not taking advantage of the lens' capability to shoot at f/2.8 and I don't see how the maximum aperture offers any other benefit at f/8.

I understand your point about stopping down, but let's do it in reverse. Assume we both shoot at 1/20 at f/8. At f/5.6 we both now could shoot at 1/40. At f/4 we both now could shoot at 1/80. Your lens is now at its limit, but *because my lens can shoot at f/2.8* I could now take a shot at 1/160, at f/2.8, while you could not. The gain in shutter speed is compromised by depth of field, but is also made possible because of the larger aperture (yes, which requires larger glass).


Lachlan's picture

I'm not very good with physics, so don't take anything I say by heart. 

Taking your scenario, AFAIK, yep, we'd both be at 1/20 f8, assuming identical conditions. However, while you were using ISO400, I'd be stuck at ISO800. You were collecting more light at 2.8, but have stopped down 3 stops of light to f8. I've stopped down two stops of light from f4. That light doesn't dissappear, and other settings have to be equally compensated to account for it.  

HendoNT's picture

I watched a UTUBE vid on microtuning lenses, very interesting,

it was how to calibrate a canon, but the theory should be the same, have a look.

miccro's picture

not sure what has been said is correct. Both lenses of same focal length but different maximum apature, stopped down to the same apature ie F8, would equate to the same amount of light in each case hitting the sensor. Thus both would need same ISO and shutter speed to achieve a similar exposure.

The extra 'stop of light' is the opening up of the lens to 2.8 that the f4 cannot do. All other apatures are the same????!

The advantage of the a faster apature and larger elements is that the camera stops down the lens only when taking the photo- it will meter and focus through at the maximum apature so low light focusing may be improved.

In this case 300mm at 7.1 should be fine to have entire bird in focus and agree you need the lens calibrated to body. 

Chris 333
Chris 333's picture

HendoNT wrote:

I watched a UTUBE vid on microtuning lenses, very interesting,

it was how to calibrate a canon, but the theory should be the same, have a look.

Great video. Would surely be worth doing if a camera has the settings for it.

I'm no expert - well, I'm an expert in having all the same focus issues as the OP, but I'm definitely not an expert in solving them... But theses are a few of the things that I've read about that may cause problems:

1. Dirty lens or dust in camera body. Very obvious yet probably sometimes overlooked. My wife had half her Bali holiday shots looking like they'd been shot through vaseline, thanks to a barely visible small finger smudge. oops..

2. Camera wobble. Hand held that's a given, but apparently the movement of the mirror vibrates the camera (sometimes more than others maybe??). Using Live View may improve results (memo to self:  must find out how that works on my camera...)

3. Imprecise auto focus. Unfortunately, my fading eyesight and need to wear glasses makes manual focus hard.

4. Insufficient or uneven light. Sometimes the light on the actual subject may just not be consistently good enough to get the crisp detail we hope for. Apart from forcing the settings out of the ideal range, it may make the autofocus less effective (wild guess, but sounds vaguely plausible...)

5. Lens needs calibrating to body, as described in that video.

6. Zoom lenses apparently usually don't perform all that well at full zoom.  I've read that there is a "sweet spot" for focus somewhere in the range, which may vary a bit even among identical lenses from the same maker. Of course, we all want that spot to be at full zoom but it usually isn't.  How much quality you lose at full stretch probably varies quite a bit. For instance, I've read reviews where the reviewer got poor results from a review lens, asked for another one, and then got different results. So maybe if we can calibrate the lens to the body we'd need to do it at full zoom, if that's what we want to use most??

Good luck finding answers. Chris.

Subscribe to me on YouTube