Aperture Choice

[jenkins]

<p>I seem to be asking myself the same question when i am out and about, what aperture to choose?<br>

I understand it in a portrait sense, blurring backgrounds ect, but in say a landscape setting i think i am losing the plot and i am guessing, with an assigned shutter speed from my camera and checking the histogram for exposure i am not really sure why i picked f16 over f22 :(<br>

Is there a way to choose your f number within reason or is it just an artistic choice along with allocated shutter speed and exposure?<br>

I just can't see the difference in DOF in small increments in my mind.<br>

Also can Aperture affect sharpness of landscape shots?<br>

Thanks.</p>

<p> </p>

[Matt Laur]

<p>Depending on the camera you're using, you absolutely will lose some sharpness when you get up to and past f/16. At that point, diffraction will cost you - sometimes very noticeably.<br /><br />Other than that... it comes down to how aperture, focal length, distance to your subject, and the look you're after all intersect. I seem to spend most of my time shooting lenses either wide open, or at f/8. Really!</p>

[nealcurrie]

<p>Have a look at http://nealcurrie.com/t-dof.html for examples of how changing aperture changes depth of field. The effects are not quite as obvious when you are further from your subject.</p>

<p>In a portrait sense, people usually use a very wide aperture. In a landscape setting, I usually use f/8 for 35mm (I don't shoot digital, but it's the same for these purposes). I strongly recommend avoiding f/16 and f/22 for shots unless you really need the extra DOF, as it does affect sharpness.</p>

<p>Here's how <em>most</em> lenses work...<br>

Around f/2, they are usually ok in the middle, and soft in the corners.<br>

As you get closer to f/8, then get sharper everywhere.<br>

As you go past f/8 or f/11, they get softer everywhere due to diffraction. The reason for it isn't really important unless you like stuff like math & physics. Because of this, I try to shoot my landscape shots from f/5.6 to f/11 unless it can't be helped.</p>

 

[colin carron]

<p>Simon., most lenses are at their sharpest at f5.6 -f8.. So for maximum sharpness that is the area to start with. If you can get everything in focus that you want in focus at f5.6 and have a decent shutter speed then that will be the best aperture. If you can't get the depth of field then reduce the aperture until you can. However lens performance will drop off as you get into the smaller apertures. <br />Lens sharpness is given by two factors which work in opposite directions. The smaller the aperture, the fewer prnlems you get with optical aberrations (lenses are not optically perfect but the bit in the middle works best) . But the opposite s true of diffraction (the tendency of light aves to bend round a sharp edge - in this case the lens iris which forms the aperture). So the best sharpness occurs at about f5.6 or so depending on the individual lens.<br />If you then throw shutter speed into the mix you can see that a bigger aperture will give a faster shutter speed and so reduce camera shake. A trpod takes this cnsideration out of the equation.</p>

[allardk]

<p>Simon, does your camera have DOF preview? Combined with a good focusing screen or Live View that will give you an idea of what DOF you'll get with a certain setting. You can also use a DOF calculator. My favourite is Barnack, its's free, and you can get it <a href="http://stegmann.dk/mikkel/barnack/">here</a> (windows only). It won't show you exactly how a picture will look but the blur circle vs. distance graph gives you an idea of not just what things will be out of focus, but also by how much.<br>

The difference in DOF between 16 and 22 on a wideangle lens focused far away is minimal.<br>

Most lenses start to loose sharpness of in-focus areas after f/8 (ie 11 or larger numbers). That doesn't necessarily mean the DOF does not increase.<br>

Other reasons I can think of for choosing a particular aperture except DOF and shutter speed:</p>

<ul>

<li>The shape of star-like features around lights on night shots.</li>

<li>Bokeh</li>

<li>Flare</li>

<li>Vignetting</li>

</ul>

 

[bruce_margolis]

<blockquote>

<p>can Aperture affect sharpness of landscape shots?</p>

</blockquote>

<p>Yep, and here's a fun way to see how a change in aperture affects sharpness of objects in your photos. Assuming you have a tripod, select a landscape shot with a nice foreground and background with your fastest lens. Focus on the foreground and take photos at every aperture. For most people, this means setting a camera on aperture priority or whatever they call it in your camera.</p>

<p>The resulting photos will show you the range from largest aperture (shallow depth of field) to smallest aperture (deep depth of field). It is unlikely you will see any diffraction on a small image but you certainly will see how increasing the f/stop also increases the depth of sharpness in objects in your image, at least up to a point. For more subtle differences, zoom in on each photo or crop a selected area and compare the differences.</p>

 

[jenkins]

<p>Very clear answers, really glad i asked them, this has been bothering me for some time.<br>

What prompted it was i read on the forums that you should never use an f number that is half your focal length, that sounded alarm bells with me. My camera is a Nikon D40, with 18-55mm and i have a Nikkor 50mm Nikkor F1.8 and yes it does have a preview Allard.<br>

I am not trying to get you to look at my pictures, but if you like, have a look at Untitled, the picture of the big cloud reflecting in the Creek, these are the wide open spaces i try to shoot, so i need an overall sharp picture, i don't think it looks that sharp.<br>

So the sunny 16 rule is pretty close to the absolute limit for aperture and sharpness.<br>

<strong>If you can get everything in focus that you want in focus at f5.6 and have a decent shutter speed then that will be the best aperture. If you can't get the depth of field then reduce the aperture until you can</strong><br>

Colin how can you tell you have 'everything' in focus on a landscape scene, with the small view finders that cameras have, so take a hypothetical situation where you thought it was not 100% would you just tweak the aperture upwards of F5.6?<br>

One last question, is it never a good idea to have your lens at full focal length for the same reasons, as in diffraction?<br>

This has actually made the operation of the camera so much easier for me, as in less unwanted choices, really appreciate it chaps, I will look at those links.</p>

<p> </p>

[jcoffin]

<p>Just one minor addition: many of the responses have mentioned f/5.6 to f/8 as the sharpest aperture. This depends heavily on the lens though. A bit easier way to remember things is that almost no lens is at it best wide open (there are exceptions, but they're fairly unusual). You usually gain quite a bit by stopping down one stop, and some more (though not as much) by stopping down one more stop. From there it's anybody's guess -- some lenses will gain a little more by stopping down further, and others won't.</p>

<p>For a fast prime lens, that could be at f/2.8, but a slow zoom f/11. While it's true that the fast prime <strong>might </strong> gain a bit more sharpness by being stopped down to f/5.6 or f/8, 1) it might not, and 2) even if it does, the improvement is likely to be restricted (at least primarily) to the corners. In the other direction, nearly no lens will gain any appreciable amount of sharpness by being stopped down past f/11 -- if you're lucky, f/16 might be almost as sharp as f/11, but it'll almost never be sharper. By f/22, nearly all lenses are essentially equal (all quite poor).</p>

<p>On the other hand, there's a reason for the old saying: "f/8 and be there". Even with a lens that's really sharp at (for example) f/1.4, it's easy to miss focus a bit, or simply lack the DoF to get everything you want looking sharp. Even in a portrait, you need DoF from at least the tip of the nose to the eyes, and preferably to the ears, or it can look blurry even when/if the part that's in focus is deadly sharp.</p>

[allardk]

<p><em>What prompted it was i read on the forums that you should never use an f number that is half your focal length, that sounded alarm bells with me.</em><br>

Simon, go back to where you read that. Look at who wrote it. Look at the rest of that discussion. If you read it anywhere else than <a href="../casual-conversations-forum/00SEeY">this thread</a> I'm going to fall on the floor laughing and when I get up I'll post a seperate thread about it to let more people know it's complete nonsense.</p>

[nealcurrie]

<p>You again Allard! I couldn't tell that it was sarcasm until you'd answered my question requesting a mathematical expanation (I have a math degree). I also like the one about the noise being inversely proportional to the number of modes on the dial :).</p>

[jenkins]

<p>Allard i really hope you take up Blues Harmonica one day and join a forum i partake in God help you lol.<br /> I went to bed after reading that thinking how can a 200mm lens have an F stop of F/100, or 400mm F/200 lmao.<br /> Remember i only have an 18-55mm lens so what you said made perfect sense to me, still it prompted this thread in which i have learnt alot, so in a back handed way cheers, now where are your pictures, oops my finger is stuck on the 3 button ;)</p>

 

[colin carron]

<p>'....how can you tell you have 'everything' in focus on a landscape scene?'<br>

Simon, the short answer is that it can be very difficult especially with the small pentamirror viewfinders on many of the APS-C size DRLRs. Using the depth of field preview is the number one way but then the v/f image can get even dimmer.<br>

One a 'quick and dirty' suggestion that is practical with DSLRs is to take several identical photos starting with your best guess for focus and then changing the point of focus to get the camera to focus at a range of distances. <br>

You can also get help with estimating the 'hyperfocal distance'. Thie hyperfocal distance is a rather fuzzy idea where you can predict what will be in focus given the focal length, the aperture and the distance to where the camera is focused.<br>

Here is a (Nikonian) calculator <a href="http://www.nikonians.org/html/resources/guides/dof/hyperfocal2.html">http://www.nikonians.org/html/resources/guides/dof/hyperfocal2.html</a><br>

but there are lots of others and you can get pocket claculators to carry around if you want. :<br>

<a href="http://fotosharp.com/depth_of_field.html">http://fotosharp.com/depth_of_field.html</a></p>

<p>The problem with the whole idea of the hyperfocal distance is the subjective one of 'when do things look in focus and when are they out of focus?'. There is no definite line here and the hyperfocal distance is calculated from a thing called the 'circle of confusion' (good name!) which tries to put a number to the limit of what is in focus. This depends on the print size, the viewing distance etc so it all gets a bit murky.</p>

<p>But it is nontheless a useful guide. In particular it is useful in predicting how much of the image is in focus in front of, and behind, the point where the lens is exactly in focus. (the 'plane of focus'). Usually you will find that the area in focus in front of the point of exact focus is smaller than the area behind the point of exact focus.</p>

 

[colin carron]

<p>...sorry...duplicate post ...gateway timeout trouble.</p>

[allardk]

<p>Simon:<em> Allard i really hope you take up Blues Harmonica one day <br /> </em><br>

Mmm, I might. Never thought about it, but it sounds fun.<br>

<em>now where are your pictures</em><br>

<a href="http://www.allardkatan.net/photo/index.html">here</a><br>

Neal: (<em>I have a math degree)</em> That's good to hear. I was afraid that Fourier transform stuff would sound like posing but it probably made some sense to you.<br>

I just posted <a href="../casual-conversations-forum/00SMaa">this thread</a> . See if you can beat me at making confusing statements...<br>

Just to make sure I'm not misunderstood: it was not the purpose of my first post in the other thread to fool people, and certainly not to make fun of them. I found it sort of funny when it turned out people were taking it seriously, but I think I did my best to make clear how stuff really works. I guess I just underestimated how easy it is to misjudge someone's intention in a forum post.</p>

[jenkins]

<p>Colin thanks for your time posting this, as the old adage goes the plot thickens, this thread has been very helpful, Allard i am happy i made you laugh makes the world go round and i can see now you made a point of making a joke of it in the later post, i tend to speed read sometimes, this time to my detriment ;)<br>

No more speed reading.</p>