Focusing on 1/3rd of scene

[proust]

<p>I came across this article that talks about focusing on a subject that's 1/3 into the scene to achieve best overall DOF.<br>

http://www.apogeephoto.com/feb2005/jaltengarten2_2005.shtml<br>

I realize this is more of guidance than a strict rule but I'm trying to understand what it means to focus on 1/3rd into the scene if I'm using AF. (I have a Canon 6D + 24-105 lens). Do I select one of the 11 AF points such that one of the points happens to lie 1/3rd into the scene? I can understand using MF and focusing 1/3rd into the image but not sure how it works for AF. This question is more directed towards images where there isn't a distinct point of focus...if there is a distinct point that I want to concentrate on, I can likely get that by using a specific focus point.<br>

As an example, here's one image where I'd like to apply this 'rule' but didn't: <br>

-<br>

If I wanted to achieve best DOF, should I have been focusing approximately near the woman at the left? (seems to be 1/3rd into the image).<br>

Thanks.</p>

[q.g._de_bakker]

AF is good. It doesn't know about such rules and does what it should do: focus on what is supposed to be in focus, instead of focus on something else, because some rule says you shouldn't focus on your subject, but at some arbitrary point in space.<br><br>But if there isn't a distinct point of focus (you could ask yourself why not), and you want to maximize DoF, the general idea is that, as the rule states, it is distributed such that at long distances, about 2/3 extends beyond the point where the lens is focused at, and 1/3 in front of that distance.<br>The article is wrong in quite a few things it says. One of them is the proposed way of finding that '1/3 of the way in' distance. It is not relative to the frame of the image: even though she may be 1/3 up from the bottom of the image, the woman on the left is not standing at 1/3 of the distance from the camera to the horizon.<br>In the example, there is no particular distance to focus at, and you would want DoF to extend to, but not beyond infinity. For that, the hyperfocal distance is invented. You could guess where 1/3 of the distance from you to infinity is. You could also use one of those DoF calculators you find on the internet. Manual focus lenses did have scales on them that helped a lot.<br>You would also need to know what aperture to use, whether it stretches DoF to where you think it does or not. Just using the smallest but one, as the article suggests, is no good.<br><br>DoF is only distributed in (approximate) thirds at long focus distances, gradually changing to an even distribution (halves) when coming close®. So the rule of thirds does not hold in the example given in the article of a flower a few inches from your lens (and a short lens does not produce more DoF than a long one either).

[jon_shumpert2]

<p>John, I found a free depth of field calculator app for my phone. Essentially, you choose the camera , lens, aperture and distance and it will give you the approximate depth of field in feet or meters. It will also give the near and far limits of the depth. I agree with Q.G that the grid used as an example for rule of thirds is not useful for determining distance.</p>

[Spearhead]

<p>I have spent a lot of time on observing how people respond to sharpness and depth of field. What my fairly unscientific study revealed years ago, and I have followed ever since, is that what you want to be most in focus should be most in focus. Using depth of field to try and keep some level of apparent sharpness in the foreground, for example, means that if the foreground is important, people find the photo "unsharp." On the other hand, if what's in the front is most important, it doesn't matter much if what behind it becomes unsharp faster than "desirable." </p>

[William Michael]

<blockquote>

<p>As an example, here's one image where I'd like to apply this 'rule' but didn't: <br /> <a title="Link added by VigLink" href=" -<br /> <strong>If I wanted to achieve best DOF</strong>, should I have been <strong>focusing approximately near the woman at the left?</strong> (seems to be 1/3rd into the image).</p>

</blockquote>

<p>No. As previously noted, she is not one third into the scene. Only for the sake of addressimg this specific question - I'd reckon that one third in is a bit beyond the tree on the left hand side - but note I am not advocating that method.<br /> <br /> Also as previously noted the article has several flaws and or omissions of relevant facts: one flaw is the “method” described for calculating where ‘one third into the scene’ is located.</p>

<p>***</p>

<p>I think that you need to concentrate on the outcome that you want and then employ the easiest way, for you to achive it. I don't think that using this "one third focus technique" is the easist. <br /> <br /> I also think you have stated that outcome that you want and <strong>I have assumed</strong> what you mean by <em>“I wanted to achieve best DOF”</em>, you mean something like this -<br /> <br /> <em>“in this scene there is a lot of foreground grass that I want to appear reasonably sharp, but I also want the horizon to appear reasonably sharp too – and I want to achieve that all in one shot - in simple technical terms: I need to have a really big Depth of Field to achive this outcome - but I also want to use a suitable ISO and Shutter Speed - so I want to have a simple technique that I can employ in the field to suit a quick mental compromise of Tv, Av and ISO, once I have assessed the EXPOSURE that I will use for the shot" <br /></em></p>

<p>*</p>

<p>I could not retrieve the technical details from your image – but it appears to be made with a moderate wide angle lens – so maybe this working example below will be of help to you.<br /> <br /> The image below was made with a 28mm Lens on an EOS 5D (that is a ‘full Frame’ camera) and I think this is a reasonable approximation of the Field of View of the lens that you used, if anything perhaps your lens was a tad wider, I think, and that will assist:</p>

<p><img src="http://d6d2h4gfvy8t8.cloudfront.net/17855284-md.jpg" alt="" width="680" height="416" /><br /> <strong>Bordeaux, near the Garonne River, 2012</strong></p>

<p>Similar to your image, there is a lot of foreground, and I wanted that to be in reasonably sharp focus – but I also wanted the middle ground (the main building)to be in reasonably sharp focus – and I wanted the horizon (far right) to also be in reasonably sharp focus.</p>

<p>Obviously those points in the scene CANNOT ALL be ”IN SHARP FOCUS”, that’s impossible if you only have one shot.</p>

<p>*</p>

<p>What I find handy for this shooting scenario is my <strong>knowing</strong> the Hyperfocal Distance for a 28mm Lens (on 5D Format Camera) at F/11 is about 10ft – so that’s about where I focussed using the camera's CENTRE POINT Auto Focus and targetting the grid lines made by the pavers at about 12ft away from the camera and the shot was pulled at: <strong>F/10 @ 1/1250s @ ISO200.</strong> Maybe, with 20/20 hindsight, I should have sacrificed shutter speed and moved the Aperture to F/16. As you should appreciate, it is all a set of compromises.</p>

<p><strong>I also knew my image’s USES - </strong>it would be for screen display of my ‘holiday shots’ locted in an informal area on my home and at the screen’s resolution and size (10 x 8 inches) this image looks quite fine and is quite sharp throughout the foreground, middle ground and background – so it is important to know what the uses are to be for the image that you are making. (i.e. how big the final image will be and how it will be viewed or scrutinized later on).</p>

<p>I do ‘know’ a few Hyperfocal Distances for different Focal Lengths, mainly just the wide angle to normal lenses – 24, 28, 35, 50, and mainly only at F/8 and F/11. That knowledge comes NOT from shooting landscapes but from shooting doorstops and similar situations employing, ‘hail mary’ technique: that is to say, with the camera held over the head and without using Auto Focus.</p>

<p>I don’t think that there is much relevance for using Hyperfocal Distances except for some rare situations: and this shooting scenario is one where you might find it handy. A post graduate PhD in rocket science or pure maths is NOT required to memorize a few H.F. distances at F/11 for the few different focal lengths that you regularly use - and one can write them on a bit of paper and stick them on the lens turret if need be.</p>

<p>*</p>

<p>Another technique that you might use to achieve what I have <strong><em>assumed</em></strong> the goals are for your image: is to use the DISTANCE MARKINGS against the APERTURE MARKINGS on the lens barrel. But if you used a ZOOM LENS then it is likely that there will be no such markings and many modern prime lenses don’t have them either.</p>

<p>Below is a 35mm lens which has an overlay of clear tape and markings, because I was using it for Zone Focus shooting, but if you use this method you need to set the “∞” (infinity) marker at the Aperture that you will be using and note approximately where the corresponding <strong><em>same</em></strong> Aperture marker aligns, on the distance scale:</p>

<p><img src="http://d6d2h4gfvy8t8.cloudfront.net/15671930-md.jpg" alt="" width="680" height="486" /></p>

<p>Another issue with using the distance markings / aperture markings technique is that modern lenses tend to NOT have a large rotation from closest focus to infinity, which renders this method relatively inaccurate when compared to using the same method with older lenses, whose rotation might have traversed 270° or more.</p>

<p>*</p>

<p>What you can also use (for a static Scene) is “Focus Stacking”. This tecnique requires several shots t be made and also has some limitations; requirements and a few issues – you will need a tripod and a bit of time for each shot, but I will leave that for you to investigate that option, if it interests you.</p>

<p>WW</p>

[Wouter Willemse]

<blockquote>

<p>what you want to be most in focus should be most in focus</p>

</blockquote>

<p>This. And this, And this again. The aim of the article linked to, and of "hyperfocal" distances is the assumption that you want everything in a photo look sharp. People tend to repeat this as some short of rule for landscape shooting: hyperfocal, f/11 or f/16 - and that's how you shoot a landscape.<br /> Strangely, this is not how my eyes work. I do not see sharp at infinity and close simultaneously, and it never hinders me in appreciating landscapes or similar.</p>

<p>Now, I'm not going to say that is fully wrong, but it's worth thinking twice about that. It is saying that everything in the image is equally important: the image has no specific point you want the viewer to focus on. Instead, they'll glance the image, hovering over it - it won't "catch" them as easily. And frankly, often a composition does get a bit stronger is there is a clear point of focus. It helps the viewer making sense of the image. Like Jeff, it's just based on some personal experiences and basic theories on how people percept images, so what I am saying it not "the truth"... but Jeff's short version has an awful lot of merit: if you want your viewer to focus on a part of the image, focus on that part of the image. Adjust aperture to taste, and go from there.<br>

I think the other WW's image of Bordeaux makes a fine example: there is no need at all to have the trees and buildings on the far right impeccably sharp, so DoF does not need to stretch to there necessarily. By default the bright-lit wall facing us on the left third takes most attention (as it's considerbaly brighter) - so, that'll be the focus point. Then why not simply focus there, to ensure that's as sharp as can be done, and the rest will fall into place automatically.</p>

[Palouse]

<p>I agree with Wouter, selectively and artistically throwing the foreground/ background out of focus by means of wider aperture and/or focus point makes for natural and creative pictures, sometimes. Where do you want people to focus their attention?</p>

[William Michael]

<p>The contents of the commentaries in this thread basically fall into addressing TWO topics:<br>

</p>

<ul>

<li>

<p> From an <strong><em>artistic perspective</em></strong> discussing whether or not it a good idea to have a big DoF, without much consideration for focussing on <strong><em>a particular</em></strong> Subject in the shot</p>

</li>

<li>

<p><strong><em>Methods</em></strong> to achieve the above, without consideration for the artistic worth of the resultant image.</p>

 

</li>

</ul>

<p>For clarity my comments <strong><em>only</em></strong> addressed the latter. <br>

<br>

To the particular point and answering the question asked by Wouter regarding the Bordeaux shot – <br>

There was a reason to make that shot how I did, that reasons is: mainly because because I wanted to. <br>

My reasoning doesn’t make that particular shot a good landscape shot, nor a better shot had than if I had made the shot with the sharp focus on the building: I simply wanted to do, what I wanted to do. <br>

By the way, I did take two more shots from that same vantage point and for those other two I used AF on the building as Wouter indicated he would have done. So I came away with a choice.<br>

The reality is, when comparing all the three on my 10x8 display monitor, the image above has noticeably sharper grey pavers in the foreground, and I like that on that little monitor, as the building look quite crisp too: as I mentioned I think it is very important to also know the USES for the image.<br>

In the unlikely event that I would make a print of that scene, I’d most likely use one of the shots where I focussed the main building. <br>

<br>

***<br>

<br>

Commenting on the worth of a Photographer addressing <strong><em>all</em></strong> Landscape shooting with the idea of employing the H.F Distance and having an overriding need to make all shots with a very large DoF – my view is: <strong>that’s just silly</strong>. <br>

Photography is not a formula – it is about achieving one’s goal or the vision - for each shot.<br>

<br>

WW </p>

[q.g._de_bakker]

So long live AF, which doesn't even know about DoF.

[Wouter Willemse]

William, I hope my post didn't read as an attack or even contradiction of what you posted; that certainly wasn't my intent. My comment on your photos are no more than "what I would have done", and that 's very much "for what it's worth". Nothing but another opinion in a sea full of opinions... And I certainly agree that in the end it's about realising a vision.

[proust]

<p>Thanks all for your responses. William W - you correctly articulated what my goals were :).<br>

I realize that no one-size rule/guidance works for all photos and there is a subjective artistic decision to be made about the DoF/focus point...I'm more interested in learning the technical aspects of how best to use DoF in the right context so I have options available to experiment with. I hadn't used the hyperfocal DoF calculators and will try that out. Thanks again for your replies!</p>

[William Michael]

<p>Wouter,<br>

You worry way WAY <strong>WAY</strong> too much.<br>

The intent and meaning of your commentary was crystal clear to me.<br>

I was merely clarifying a few points in regard to the (two) lines of discussion which are happening here. </p>

<p>*</p>

<p>I think that John should come away from this conversation with information and opinions that will allow him to critically question the <strong>premise</strong> and the <strong>content</strong> of the article to which he linked - and also allow him to critically question the relevance and application of DoF, for <strong>each</strong> shot that he makes. </p>

<p>WW</p>

<p> </p>

[William Michael]

<blockquote>

<p>"I realize that <strong>no one-size rule/guidance works for all photos</strong> and there is a <strong>subjective artistic decision to be made about the DoF/focus point</strong>...I'm more interested in <strong>learning the technical aspects</strong> of how best to use DoF in the right context <strong>so I have options available to experiment with.</strong>"</p>

</blockquote>

<p>Good. That is very good. That is excellent. You have made my morning.<br /> You were posting at exactly the same time as I.</p>

<p>WW</p>