1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

SLR autofocus

Discussion in 'Everything Film' started by Lusty, Dec 20, 2004.

  1. Lusty

    Lusty Active Member

    Hi,
    My photography professor tried to teach us once how to set the camera so that everything between a few feet and infinity will be in focus. I didn't relaly follow it that time, and now more than ever i feel the need to set the camera in that way. Can someone help me understand how it works? Thanks,
    Craig
     
  2. TimF

    TimF With as stony a stare as ever Lord Reith could hav

    Sounds like hyperfocal technique. Ideally you need a lens with proper depth of field markings, which many modern lenses (especially zooms) simply don't have any more. If you have such a lens handy, set the infinity point to - or preferably just inside of the right side marked aperture you wish to use - as most are not that accurate, and depend in part on the manufacturer's estimation of Circle of Confusion, which can be quite slack; Canon I hear use 35 micrometers for EF glass, the normal standard being perhaps 30 micrometers and a more rigorous being 25 micrometers. The point of closest depth of focus will then be against the marked aperture on the left side.

    If you have a Palm or Pocket PC device, then there are very good Hyperfocal/DoF calculator programs available.

    Tim BSRIPN
    [​IMG]
     
  3. taxor

    taxor Well-Known Member

    Craig,
    First of all, you need a lens with a depth of field scale engraved on the lens barrel. This probably going to be on older manual focus model. It's quite an easy (though often misunderstood) technique to apply. Let's say (for example) you're shooting at f16 and focussed at infinity. One simply rotates the focus ring until the infinity symbol is below or above (depending on your lens) the f16 depth of field marker. Look in any good photographic book under hyperfocal distance or hyperfocal focussing and this should give you a clear understanding of the technique. It's a valuable means of ensuring maximum front to back sharpness in your pictures. If you have only a couple of lenses, it may be worth working out and memorising the hyperfocal distances for each aperture setting. Regards, TAXOR

    "I pink therefore I Spam" Arthur Mullard
    (someone nicked my last sig./img/wwwthreads/wink.gif)
     
  4. sirmy

    sirmy Well-Known Member

    If you can get hold of a copy of the Focal Encyclopedia of Photography there is a formula for working out hyperfocal distances for each focal lenght anf aperture combination. If you can bear o go through the process its possible to produce a table for each lens you use (or each focal lenght). Keep a copy in your gadget bag and you only need to be able to set aperture and focused distance and be confident in getting things acceptably sharp, even with out a depth of field scale on the lens barrel (this sounds strangely like one of theose adverts in a 70's photo mag.)
     
  5. TimF

    TimF With as stony a stare as ever Lord Reith could hav

    Dare I say the Palm / PPC way is a lot quicker - though obviously more expensive from initial set-up. Focus&#43, which I use, can be employed for Hyperfocal distances and depth of field with any user-defined lens and CoC value. My pretty old copy of the Focal Encyclopedia has lens focal lengths in inches, and is mind-boggling in comparison!

    Tim BSRIPN
    [​IMG]
     
  6. huwevans

    huwevans Not Really Here

    And dare I say that anything involving electronic computation is overkill for most practical purposes? :) Seriously, unless you struggle with mental arithmetic, the quickest method (in the absence of a proper DoF scale on a manual focus lens) is in your head. Of course, you have to know what to calculate, and that does vary depending on things like format, or chosen CoC.

    But for a good ready-reckoning method for determining the approximate hyperfocal distance for 35mm cameras (using a CoC of 0.025mm) try this - take the focal length of your lens in centimetres (eg. 20cm, for a 200mm lens), square it, and that gives you the rough HF distance in metres for f/4 (20² = 400m, in this example). Then to get values for other apertures, halve that figure for every two stops up the scale - so halve it for f/8, halve it again for f/16. If you want, say, f/5.6 or f/11, just take a mid point between the appropriate values - that's an approximation again, but it's fine for practical purposes. Remember that when you focus on the hyperfocal distance the near limit of your DoF is about half the HF distance, and you're good to go, with no palm-tops or tables or calculating aids of any sort. :)

    Huw Evans.
     
  7. SamsSauce

    SamsSauce In the Stop Bath

    Many moons ago, it was common practice to mark on the barrel of zoom lenses, a DOF scale which compensated for the focal length being used. Nikkors, and early Vivitars are classic example which come to mind, though many others had them. Sadly, the manufacturers don't bother marking them these days. I suppose that's technology moving forward ...

    Cynical Sam
     
  8. Benchista

    Benchista Which Tyler

    It's easier to do with "trombone" style zooms - few are, these days.

    Nick BSRIPN
    And why not?
     
  9. equiphoto

    equiphoto Well-Known Member

    I suggest you read Harold Merklingers book "The Ins and Outs of Focus". It's available for free download at http://www.trenholm.org/hmmerk/download.html

    It goes against a lot of the traditional wisdom on DOF but does deal specifically with methods to get a DOF from near to infinity (and also why the hyperfocal distance is a load of b***s).


    Peter
    peter-robinson.net
     
  10. huwevans

    huwevans Not Really Here

    why the hyperfocal distance is a load of b***s

    Ummm, not really, no. The hyperfocal distance most certainly isn't a 'load of b***s', and Merklinger's 'book' doesn't show that it is.

    Even he says (referring to the difference between hyperfocal focussing and his own method),

    'It would be foolish to suggest that the method photographers have
    been using for countless decades is not up to the job.
    '


    Merklinger's take on hyperfocal focussing is only valid if you happen to want the horizon to be the sharpest part of the photograph and are happy with a blurred foreground. Again even he says,

    'All-in-all, the two methods are really much the same. The differences lie
    in the selection of what is important.
    '

    Indeed, that is the nub. Merklinger's personal choice seems to be that the horizon should be sharp, whilst the foreground only needs to be 'recognizable'. Well, that's up to him, but you won't find many committed landscape photographers who are happy with the same criteria. The practical reality for rigid body cameras is that if you want two different elelments both to be 'adequately' focussed, even though they are at different distances, then you have to rely on depth of field. Hyperfocal focussing is nothing more or less than a means of making most efficient use of DoF when pushing the limits.

    In landscape photography (probably the most common genre in which HF focussing is employed) it is by no means usual to want the horizon to be the sharpest part of the landscape. Personally I'm much more concerned about my middle- and foregrounds being sharp - horizons seldom are anyway, simply because of atmospheric effects. Perhaps Merklinger lives 10000' up in the Rockies and can expect crystal clear air 300 days a year. But you don't get that in any part of Britain I've ever been to. No, here focussing on the horizon is usually just a waste of good resolving power, because once five or ten miles worth of pollution and haze have got in the way there's not much left to resolve. The foreground, OTOH, is almost always resolvable, and noticeable if it's left blurred.

    Finally, by way of example, in my own past landscape work on the 6x7 format probably nine times out of ten I would end up shooting at f/22 and hyperfocally focussing simply to get adequate sharpness from front to back. If I had focussed the lens at infinity I would have had to stop down to f/45 to get the foreground to the same degree of definition. Only one of my lenses on that format even went as far as f/45, and in any case diffraction would have ruined the resolution of the lens every where in the frame - not just at the extremes of DoF.



    Huw Evans.
     
  11. equiphoto

    equiphoto Well-Known Member

    Only just got back to this...

    The point is, as Merklinger shows, if you use the hyperfocal distance and assume that you are going to get everything sharply defined at infinity, you will be fooling yourself. For me, Merklinger is spot on as far as DOF beyond the point of focus is concerned, however, I agree with you though that for DOF in front of the point of focus, you are likely to end up with resolvable, out of focus, subjects if you stick to his method.

    I don't agree with Merklinger when he says the two methods are much the same. They are, in fact, totally different. The way you use them, the results you get and the definition of what's in and what's out of focus are all different. Merklinger's method depends on resolution while the traditional method depends on sharpness. You can have a subject which is, in theory, sharp, but is unresolvable and you can also have a resolvable subject which is clearly not sharp.

    Merklinger is true in saying it's a matter of choosing what's important. You may well decide that a hazy horizon is not important for you but that doesn't mean Merklinger's theory is wrong or should be ignored.

    IMHO, the hyperfocal distance theory, as blindly applied by most photographers, does not do the job it pretends to, and is, as far as I'm concerned, a load of...

    If you decide it does the job for you, that's fine. You obviously understand the theory and can make your own choice. But many others are blissfully unaware of it's shortcomings and would do well to read Merklinger's text.

    Peter
    peter-robinson.net
     
  12. huwevans

    huwevans Not Really Here

    Merklinger's method depends on resolution while the traditional method depends on sharpness.

    I'd like to know how you are distinguishing between these two concepts here. There is, AFAIK, no clear definition of 'sharpness' in the field of optics, or more generally in photograhy. Lens resolution has traditionally been measured in terms of line pairs per unit length and local contrast, and particularly that combination of the two which is the modulation transfer function (MTF). For more general measurement or comparison of what can be resolved in a final print lpmm is traditionally used. The nearest I have ever seen to a definition of the concept of sharpness is acutance, and that usually winds up back with a requirement specified in terms of a certain number of lpmm on the print.

    But in this case I'm not sure how you are using the term 'resolution' such that you think it applies to Merklinger's method and yet not to the traditional one. The MTF is not really relevant to subjects not in the plane of focus, but there's a definite general relationship between the limit of resolution in lpmm and the size of the circle of confusion in OOF subjects. A larger CoC places an upper limit on what can be resolved, even if the lens is capable of more. In practice it is this CoC limiting which is relevant to the resolving of fine OOF detail, except in the case where diffraction imposes greater restrictions. That isn't normally the case in most situations, but Merklinger's approach makes it more likely unless you are willing to sacrifice resolution in the foreground.

    In summary, the two methods essentially stack up as follows:

    Hyperfocal: Acceptable foreground, sharp mid-range, acceptable horizon;

    Merklinger: Unsharp foreground, acceptable mid-range, sharp horizon (if atmospheric degradation effects allow it).

    And the caveat to Merklinger's approach is that if you want to have an acceptably sharp foreground the only way to achieve it is by stopping down an additional two stops (relative to the hyperfocal method), which (depending on individual circumstances) runs the risk of imposing diffraction limits on the entire field.


    And I really have to comment on this:

    You may well decide that a hazy horizon is not important for you but that doesn't mean Merklinger's theory is wrong or should be ignored.
    ...
    IMHO, the hyperfocal distance theory, as blindly applied by most photographers, does not do the job it pretends to, and is, as far as I'm concerned, a load of...

    So you insist that Merklinger's alternative preference should be respected for what it is, but refuse that same respect for the more traditional preference? Sorry, but that won't do.

    Huw Evans.
     
  13. jchrisc

    jchrisc Well-Known Member

    Sounds of distant but distinct, loud cheers in the (hyperfocal) distance . . . /img/wwwthreads/smile.gif

    Chris

    [​IMG]
     
  14. huwevans

    huwevans Not Really Here

    You're sharp today, Chris. :)

    Huw Evans.
     
  15. jchrisc

    jchrisc Well-Known Member

    Always ready to applaud a six

    Chris

    [​IMG]
     
  16. equiphoto

    equiphoto Well-Known Member

    >I'd like to know how you are distinguishing between these two concepts here.

    For sharpness read acutance if you prefer.

    Resolution is as per Merklinger.

    The traditional method only considers how sharp (or acute if you want) the edge of a subject is. Merklinger considers whether a subject is recognisable, or resolvable, regardless of whether it's edges are sharp.

    >Hyperfocal: Acceptable foreground, sharp mid-range, acceptable horizon;
    >Merklinger: Unsharp foreground, acceptable mid-range, sharp horizon

    You just cannot make statements like that in any argument. 'Acceptable' is a subjective term, what is acceptable to one may not be for another. Personally, and for Merklinger as well it seems, the hyperfocal method does not produce 'acceptable' focus at infinity.

    >So you insist that Merklinger's alternative preference should be respected for what it is, but refuse that same respect for the more traditional preference?

    Yes I do think Merklinger's theory should be respected, or considered at least. As I suggested previously, you are free to make up your own mind but I am also free to put my opinion and mine is that the hyperfocal method doesn't deliver.

    >Sorry, but that won't do.

    And I'm sorry too Huw - that you can't accept that someone else's opinion doesn't agree with your own and that you should find it necessary to try to disprove it. Merklinger's theory is sound. Yes it's different and will give different results. It may not suit you, but that's fine with me. I would encourage anyone to read Merklinger's text and make up their own mind.

    I'm not defending Merklinger. I only suggested the original poster might like to read his work. If you want to argue the toss, email him.

    Peter
    peter-robinson.net
     
  17. huwevans

    huwevans Not Really Here

    And I'm sorry too Huw - that you can't accept that someone else's opinion doesn't agree with your own

    Errr, excuse me, but that is a grotesque and intolerable perversion of what I have said on this matter. You need to go back and read it again. You were the one who dismissed other opinion as 'a load of b***s', and that is the sole reason I responded. If all you had said was something like, 'here's an alternative view of the hyperfocal fucussing issue to which I subscribe,' I wouldn't have had any cause to reply at all. But you didn't - you dismissed the preferences of generations in a quite ludicrously bigoted way. All that I have done was simply to point out the salient facts, chief amongst which was that Merklinger's view involves a personal choice which happens to be different from that made by most on this issue down through the years.


    You just cannot make statements like that in any argument. 'Acceptable' is a subjective term, what is acceptable to one may not be for another.

    I was assuming that you understood the basic principles of depth of field theory, since you had presumed summarily to rubbish the whole concept of hyperfocal focussing. I am beginning to realize that that was a mistake. Yes, 'acceptable' is a subjective concept - but as the photographer you are supposed to decide for yourself what it means for your work, and even for the individual circumstances of a particular shot, or intended print size. This is the whole point of the Circle of Confusion involved in DoF calculations. You choose the CoC that gives you what you consider to be acceptable, and once you have chosen it it becomes an objective standard that you work with. If your choice isn't giving you the results you want then it either means that you have made a wrong choice, or that your choice is so exacting that some other limit (like diffraction, or the quality of the lens, or the resolution of the film, or just the limits of eyesight on the finished print) is proving to be a greater restriction.

    Merklinger understands these things, which is most certainly why he didn't dismiss the concept of hyperfocal focussing the way you did. I'll remind you again of what he had to say:

    'It would be foolish to suggest that the method photographers have
    been using for countless decades is not up to the job.'
    ...
    'All-in-all, the two methods are really much the same. The differences lie
    in the selection of what is important.
    '


    Rather than presuming to disagree with him on this (because from reading your posts I don't think you understand the theory well enough to pass such a judgment), or to accuse me of being unable to accept the different opinions of others, you should be a bit more receptive to what he says here.

    Huw Evans.
     
  18. equiphoto

    equiphoto Well-Known Member

    No doubt you will want to have the last word on this, you seem the sort of person who would insist on it, so, reply to this as you will, I won't be reading it.

    It is you Huw who are perverting what was said, In saying the hyperfocal theory was a load of b***s, I was quite clearly expressing an opinion. I can say whatever I like, it doesn't make it true. I can say Nikon cameras are a load of b***s, does that make it true? Of course it doesn't. It's an OPINION.

    Unfortunately, you seem unable to accept it and feel the need to write reams to try and disprove it.

    For the record, here's my OPINION:

    The hyperfocal theory does not deliver what most photographers assume it does, especially if used blindly as most do and even more so if they use the DOF scales engraved on lenses which probably use a totally unsuitable COC.

    Even if DOF is calculated using a realistic COC, it does not take into account whether a subject can be resolved.

    Merklinger's method overcomes both of the above failings while introducing a problem of it's own.

    Merkinger's method is the best for DOF beyond the point of focus while the traditional method works best for DOF in front of the point of focus.



    Peter
    peter-robinson.net
     
  19. BigWill

    BigWill Gorgeous oversensitive Nikon-loving cream puff

    I wish to take issue...........

    ...............with this line in your posting.

    "I can say whatever I like, it doesn't make it true. I can say Nikon cameras are a load of b***s, does that make it true? Of course it doesn't. It's an OPINION."

    .................when in fact we all know it should read:

    "I can say whatever I like, it doesn't make it true. I can say Nikon cameras are a load of b***s, does that make it true? Of course it <font color=red>does<font color=black>. It's an <font color=red>undeniable<font color=black> fact! /img/wwwthreads/smile.gif

    Apologies to Huw and Peter for trying to introduce a bit of levity into what was turning out to be a rather heated debate. God knows I'm no-one to preach on the subject as I frequently get "sucked into" heated arguments on here myself, as everyone knows, but if you can just take a "step back" sometimes I think it helps. Yes, we get het up about issues which are important to us, and rightly so, but sometimes we just have to "agree to disagree" and try to remain as civilized as possible towards each other. Maybe we should call a truce on this particular one and do just that. What say guys? Agree to disagree?

    BigWill



    <font color=blue>I'm sailing like a driftwood on a windy bay!<font color=black>
     
  20. huwevans

    huwevans Not Really Here

    I was perfectly happy for Merklinger to have the last word, as quoted above. You should have been too.

    It is you Huw who are perverting what was said, In saying the hyperfocal theory was a load of b***s, I was quite clearly expressing an opinion.

    Actually you claimed that Merklinger's book dealt with, inter alia, '...why the hyperfocal distance is a load of b***s'. It doesn't, and he himself says so. And if you had actually understood the principles involved here you would have recognized that yourself.

    But when you dismiss something in that way, with such sweepingly absolute and frankly insulting terms, you equally dismiss and insult the legions of people who, in full understanding of what they are doing, choose that way of working. I called you to account on that, quite properly, but rather than admit that you had been unreasonable in what you said, you first tried to manoeuvre round my objection by some sort of equivocating nonsense about 'resolution' and 'sharpness', and then by simply attacking me with increasing spite. That is not the response of a intelligent person defending a reasonable opinion.

    Your final analysis still suggests to me that you are deeply confused, but at least now you are saying something (in the very last paragraph) that is profoundly different from your initial highly inappropriate and unwise comment.

    Huw Evans.
     

Share This Page