Photo Forum / General Photo Topics / Australian Photography / September 2007
landscapes and hyperfocal distance
|
|
Thread rating:  |
Troy Piggins - 07 Sep 2007 14:53 GMT I'm thinking about taking some dusk cityscape pics to test out my new tripod, and I got to thinking about landscape photos and focussing distances etc.
I've been reading that the hyperfocal distance is where you focus at a point and everything from half that distance to infinity is acceptably focussed. I got to thinking and I have a bunch of questions...
Say I'm taking a photo of Brisbane's Casino from over the river at Southbank. Normally I would've just auto-focussed on the building and let her rip, so to speak. Is that what most of you would do?
Or would you use the hyperfocal distance to get everything in the depth of field?
Do you do that for nature landscapes?
Say the lens I'm using is 24mm at f/11. On my 30D the hyperfocal distance is 2.7m. So if I manually focus at 2.7m, everything from 1.35m away to infinity is in focus? Is that how you would do it? Manually focus? Is that accurate enough?
I have downloaded a freeware little program for my XDA that is a depth of field calculator and gives the hyperfocal distance which is handy and portable. It is very similar to the one found online here:
http://www.dofmaster.com/dofjs.html
Please offer your thoughts.
 Signature Troy Piggins Gallery: http://piggo.com/~troy/gallery Please feel free to provide constructive criticism on any photos I post. I'm always learning and appreciate feedback.
PixelPix - 07 Sep 2007 23:39 GMT > I'm thinking about taking some dusk cityscape pics to test out > my new tripod, and I got to thinking about landscape photos and [quoted text clipped - 9 lines] > building and let her rip, so to speak. Is that what most of you > would do? It depends if you have some foreground interest that needs to be in focus also eg. a park bench, a section of the boardwalk or perhaps even a boat. If you have foreground elements like these, then maximising DOF is important and hyperfocal focusing is the best option.
If there is no foreground element that needs to be sharp, then focusing on the building itself may be the better option. Remember when we talk about hyperfocal focusing, we refer to "acceptable" sharpness.... there will still be a field of "best" sharpness and that will always be at the point of focus. So if there is no other image element that needs to be sharp, then it is good to give the best possible sharpness to our main point of interest.
> Do you do that for nature landscapes? I include foreground elements in my landscapes the majority of the time, so I am more often than not using hyperfocal focusing.
> Say the lens I'm using is 24mm at f/11. On my 30D the hyperfocal > distance is 2.7m. So if I manually focus at 2.7m, everything > from 1.35m away to infinity is in focus? Is that how you would > do it? Manually focus? Is that accurate enough? Manual or autofocus is fine, just remember that most standard screens are not designed for accurate manual focus these days. In many cases an "auto focus and recompose" technique is fine, as the DOF spead will hide any small miss in focus. Personally I often use a manual focus Olympus lens and a dedicated manual focus screen.... but not all DSLR variants have such luxury.
> I have downloaded a freeware little program for my XDA that is a > depth of field calculator and gives the hyperfocal distance which > is handy and portable. It is very similar to the one found > online here: > > http://www.dofmaster.com/dofjs.html DOF Master is a great resource and IMHO everyone should have it bookmarked.
Cheers
Rusty http://www.pixelpix.com.au/gallery
Troy Piggins - 08 Sep 2007 00:20 GMT * PixelPix is quoted & my replies are inline below :
>> I'm thinking about taking some dusk cityscape pics to test out >> my new tripod, and I got to thinking about landscape photos and [quoted text clipped - 23 lines] > element that needs to be sharp, then it is good to give the best > possible sharpness to our main point of interest. I see. I hadn't appreciated the difference between "acceptable" and "best".
>> Do you do that for nature landscapes? > > I include foreground elements in my landscapes the majority of the > time, so I am more often than not using hyperfocal focusing. Cool. I'm glad to know it is used regularly. I was thinking it may be just a theoretical type phenomenon that you know is there, but doesn't get used in practice.
>> Say the lens I'm using is 24mm at f/11. On my 30D the hyperfocal >> distance is 2.7m. So if I manually focus at 2.7m, everything [quoted text clipped - 7 lines] > Olympus lens and a dedicated manual focus screen... but not all DSLR > variants have such luxury. Hmm. So to autofocus, the only way you could do that in my example above is to AF on something 2.7m away and recompose? Or to manual focus, would you use the distance markings on the focus dial on the lens, or is that too inaccurate for this? Or by manual focus do you mean manually focus on something 2.7m away and recompose, but you don't need to hold down the shutter button as you would with AF?
>> I have downloaded a freeware little program for my XDA that is a >> depth of field calculator and gives the hyperfocal distance which [quoted text clipped - 5 lines] > DOF Master is a great resource and IMHO everyone should have it > bookmarked. I do. And now I have a version I can carry with me always :)
Thanks for your tips Rusty.
 Signature Troy Piggins Gallery: http://piggo.com/~troy/gallery Please feel free to provide constructive criticism on any photos I post. I'm always learning and appreciate feedback.
D_Mac - 08 Sep 2007 03:20 GMT > * PixelPix is quoted & my replies are inline below : > [quoted text clipped - 77 lines] > Please feel free to provide constructive criticism on any photos I post. I'm > always learning and appreciate feedback. You do realize Troy, that Hyperfocal distance is a mythical location? The presumption is that before and behind the subject can be made sharp really means "less unsharp" than closer or further elements of the picture. A couple of years ago I revisited the issue when I did my first 100% digital wedding and discovered the charts I had been using for 10 years with my Mamiya and 35mm cameras didn't work with a 1.6 crop camera.
The widest hyperfocal distance will be obtained with the smallest sensor size. The conundrum in that is the largest sensors will yield the most detail. Some yeas back, manufacturer's began dropping distance scales from lens barrels. I presume now they did this because DSLRs were arriving and they had broader distances where less of the picture was unsharp.
In practice, a number of other elements will impact the sharpness of a landscape. The most significant is the aperture. Once you get below about F/12 with a 1.6 crop, a whole new set of variables surface. Not the least being light fall off towards the edges and the phenomena of losing sharpness as the aperture gets smaller. This itself varies with the size of the sensor too.
An average quality lens on your camera (including "L" zoom lenses) in the range suitable for landscapes will have a sweet spot at about F/ 8.0 and begin to lose sharpness (and therefore definition) at about F/ 12. By the time you get to F/22 there will be noticeable light fall off at the edges and considerable softness in the picture.
In 2005 I used the "sweet Spot" of a 28 - 70 "L" to take this shot with a 20D. http://www.weddingsnportraits.com.au/POD/01-09-07. I eventually enlarged the picture to 48 inch high and although you can't see it in this clip, everything from the bride to the Casino is inside the "Hyperfocal distance". I manually set the lens to a distance I guessed to be the edge of the river. There is never any EXIF data in these images, incidentally.
Although Russell's information is always reliable, there is nothing like a bit of practical experimentation to figure out what works best for you. The quality of your lens will dictate how well any of the information you gather will actually relate to your circumstances.
Doug
cmyk - 08 Sep 2007 04:38 GMT > The widest hyperfocal distance will be obtained with the smallest sensor size. The hyperfocal distance is a point and has no width, so there's no such thing as the "widest hyperfocal distance".
> Some yeas back, manufacturer's began dropping distance scales from lens barrels. > I presume now they did this because DSLRs were arriving and they had broader distances where less of the picture was unsharp. The manufacturers began dropping distance scales from lens barrels long before DSLRs arrived on the scene". Those scales were based on work done nearly 100 years ago for viewing a typical 15*10cm (6*4in) picture at a distance of 35cm (14in). They were probably dropped because the manufacturers realised the underlying assumptions those scales relied on were fundamentally flawed. They were flawed because they disregarded the final image size, viewing distance and what the user might regard as "acceptable sharpness".
In 'Camera Lens News. No.1', published in 1997, Carl Zeiss AG made note of the fact that changes in expectations had led to a perceived lack in Depth of Field as being the most common complaint about its more recent lenses and concluded "Those who use depth of field scales, tables, and formulas (e. g. for hyperfocal settings), restrict themselves - most probably without knowing why - to the image quality potential of an average pre-World-War-II emulsion".
That criticism applies equally to programs like DoFMaster (f-Calc is another) that disregard the final image size and viewing distance. In effect, those programs assume that if you double the height & width of a print, you'll view it from twice as far away. In reality, people tend to view a picture of 60*40cm from less than twice the distance they'd view a 30*20cm picture, just as they'd view a 30*20cm picture from less than twice the distance they'd view a 15*10cm picture. The fact that the larger pictures tend to be viewed from relatively closer distances reduces the DoF for the larger prints.
> Once you get below about F/12 with a 1.6 crop, a whole new set of variables surface. Since a 20D/30D already gives as much DoF at f14 as a 135 format (35mm) camera at f22, why use anything smaller and run into diffraction issues? In fact, if you want to do,say, a 30*20cm print that'd be viewed from around 45cm away, I'd suggest keeping the aperture no smaller than f9.5 (f15 for the 135 format), otherwise diffraction will start degrading the image. The smaller the sensor, the sooner diffraction exceeds the CoC.
> Not the least being light fall off towards the edges This is far less of a problem with a APS-C format DSLR using 135 format lenses than it is for a 135 format camera using the same lens.
> By the time you get to F/22 there will be noticeable light fall off at the edges See previous comment. In event, the light fall-off for most lenses decreases as the aperture is stopped down.
 Signature cmyk
Troy Piggins - 08 Sep 2007 11:30 GMT * D_Mac is quoted & my replies are inline below :
> [---=| Quote block shrinked by t-prot: 74 lines snipped |=---] >> [quoted text clipped - 3 lines] > > You do realize Troy, that Hyperfocal distance is a mythical location? Mythical is not the term I would use - mathmatical maybe? I saw a formula for calculating it somewhere where it's a function of the focal length, f-number, and circle of confusion limit. So I presume that it is an actual distance that for a given lens focal length can be calculated.
> The presumption is that before and behind the subject can be made > sharp really means "less unsharp" than closer or further elements of [quoted text clipped - 22 lines] > 12. By the time you get to F/22 there will be noticeable light fall > off at the edges and considerable softness in the picture. I have read about lens sweet spots. I didn't realise they related to what I was talking about.
> In 2005 I used the "sweet Spot" of a 28 - 70 "L" to take this shot > with a 20D. http://www.weddingsnportraits.com.au/POD/01-09-07. I [quoted text clipped - 8 lines] > for you. The quality of your lens will dictate how well any of the > information you gather will actually relate to your circumstances. Yeah, thanks. I'm gonna have a play with it.
 Signature Troy Piggins Gallery: http://piggo.com/~troy/gallery Please feel free to provide constructive criticism on any photos I post. I'm always learning and appreciate feedback.
k - 08 Sep 2007 03:19 GMT | I'm thinking about taking some dusk cityscape pics to test out | my new tripod, and I got to thinking about landscape photos and [quoted text clipped - 14 lines] | | Do you do that for nature landscapes? if I have something within 1.5m of the lens I want included in the landscape then yes, af gets turned off, I use the distance scale on the lens and fire away :)
but..
I generally try to shoot at the sharpest aperture and that's different for each lens, so sometimes I won't maximise my DOF (which is what shooting at the hyperfocal distance is all about)
| Say the lens I'm using is 24mm at f/11. On my 30D the hyperfocal | distance is 2.7m. So if I manually focus at 2.7m, everything | from 1.35m away to infinity is in focus? Is that how you would | do it? Manually focus? Is that accurate enough? absolutely! do you have distance markers on your lens?
gawd.. I shot across 4 days at a festival once without ever looking through the viewfinder - slapped a 24mm on the F1 set to the hyperfocal and duct taped in place, IR filter over the front, set the shutter speed /aperture to the appropriate settings, pointed the camera wherever I felt and fired away like a loon - all good :)
that was THE way to do things in pre AF days, especially when shooting on bright sunny days -
k
Troy Piggins - 08 Sep 2007 10:47 GMT * k is quoted & my replies are inline below :
>| I'm thinking about taking some dusk cityscape pics to test out >| my new tripod, and I got to thinking about landscape photos and [quoted text clipped - 24 lines] > each lens, so sometimes I won't maximise my DOF (which is what shooting at > the hyperfocal distance is all about) Ok - thanks very much for your tips.
>| Say the lens I'm using is 24mm at f/11. On my 30D the hyperfocal >| distance is 2.7m. So if I manually focus at 2.7m, everything >| from 1.35m away to infinity is in focus? Is that how you would >| do it? Manually focus? Is that accurate enough? > > absolutely! do you have distance markers on your lens? Yes - they all do. I wasn't sure how accurate they were for this purpose, that's all.
> gawd.. I shot across 4 days at a festival once without ever looking through > the viewfinder - slapped a 24mm on the F1 set to the hyperfocal and duct [quoted text clipped - 4 lines] > that was THE way to do things in pre AF days, especially when shooting on > bright sunny days - Thanks again. I don't have the luxury of that experience.
 Signature Troy Piggins Gallery: http://piggo.com/~troy/gallery Please feel free to provide constructive criticism on any photos I post. I'm always learning and appreciate feedback.
Scott W - 08 Sep 2007 12:00 GMT > I'm thinking about taking some dusk cityscape pics to test out > my new tripod, and I got to thinking about landscape photos and [quoted text clipped - 28 lines] > > Please offer your thoughts. The easiest way to think about hyperfocal distance is to start out with the idea that if you focus at some distance x then things at a distance of x/2 will be blurred the same amount as things at infinity. The idea then is that at some f/stop you will have an acceptable focus at both x/2 and infinity, with everything in between being in better focus.
A lens with a DOF scale on it makes this a very easy thing to do, sadly only one of my lenses has a DOF scale on it. Even with the DOF scale you need to be careful, I find that using a DSLR with a 1.6 crop factor I need to use about one stop slower then what the DOF scale tells me I need.
For landscapes I often just focus at infinity for a couple of reasons, I don't want to compromise the sharpness of the distance stuff and I often shooting with a fairly short lens and as such I have plenty of DOF.
Scott
Troy Piggins - 08 Sep 2007 13:38 GMT * Scott W is quoted & my replies are inline below :
> [---=| Quote block shrinked by t-prot: 23 lines snipped |=---] >> [quoted text clipped - 12 lines] > then is that at some f/stop you will have an acceptable focus at both > x/2 and infinity, with everything in between being in better focus. That's what I'm understanding.
> A lens with a DOF scale on it makes this a very easy thing to do, sadly > only one of my lenses has a DOF scale on it. Even with the DOF scale > you need to be careful, I find that using a DSLR with a 1.6 crop factor > I need to use about one stop slower then what the DOF scale tells me I need. Bugger - none of my lenses have that. Oh well.
> For landscapes I often just focus at infinity for a couple of reasons, I > don't want to compromise the sharpness of the distance stuff and I often > shooting with a fairly short lens and as such I have plenty of DOF. Thanks Scott.
 Signature Troy Piggins Gallery: http://piggo.com/~troy/gallery Please feel free to provide constructive criticism on any photos I post. I'm always learning and appreciate feedback.
mark.thomas.7@gmail.com - 09 Sep 2007 12:06 GMT > I'm thinking about taking some dusk cityscape pics to test out > my new tripod, and I got to thinking about landscape photos and [quoted text clipped - 4 lines] > acceptably focussed. I got to thinking and I have a bunch of > questions... My only additional comment has pretty much been said, so I guess it is just reinforcement... Nowadays it is hard to do *properly*, unless you have high end equipment with decent distance scales, bright viewfinders, d-o-f preview etc.. I do it sometimes, but less often than when I was shooting more manual equipment.
However... once you get to know your equipment/lens, and have a 'feel' for the depth of field you have at different f-stops, there is nothing you can't do with manual focus (or a-f, if there are objects at suitable distances) and a bit of patient experimentation - and if you're doing land/cityscapes you should have time to play, although it sounds like you will be shooting in that magic 15 minutes or so when the light is right.. Perhaps do your experimenting in daylight, then apply the knowledge later..?
And as an interesting sidenote and trivia question - dusk shooting is easier at more southerly (or northerly for north hemispherians) latitudes..
Now why is that?
(No prize for first correct answer, just kudos. (O;)
PixelPix - 09 Sep 2007 12:22 GMT On Sep 9, 9:06 pm, mark.thoma...@gmail.com wrote:
> On Sep 7, 11:53 pm, Troy Piggins <usenet-0...@piggo.com> wrote:> I'm thinking about taking some dusk cityscape pics to test out > > my new tripod, and I got to thinking about landscape photos and [quoted text clipped - 27 lines] > > (No prize for first correct answer, just kudos. (O;) Umm? Cause it lasts longer?
mark.thomas.7@gmail.com - 10 Sep 2007 11:35 GMT > > And as an interesting sidenote and trivia question - dusk shooting is > > easier at more southerly (or northerly for north hemispherians) [quoted text clipped - 5 lines] > > Umm? Cause it lasts longer? Correct! But you weren't exactly convincing. (O;
As you go lower (or higher) and therefore further from the equator, the sun becomes increasingly angled as it rises/sets. Which means that it is closer to the horizon for a much longer period, and therefore casts usable light for a much longer period. I recently moved from lat 34 to about 27, and found it VERY noticable. It's almost like the lights go out suddenly at dusk, no long lingering twilight... sigh.
If you take it to the extremes (at the poles) you can end up with 24 hour daylight/night... No problems with rushing to catch the "magic hour" there..
This may also explain in a small way why daylight saving is not as popular as you get closer to the equator, as there isn't so much variation winter/summer, nor is there much usable twilight.
Just call me Dr Karl... (O;
PixelPix - 10 Sep 2007 13:44 GMT On Sep 10, 8:35 pm, mark.thoma...@gmail.com wrote:
> > > And as an interesting sidenote and trivia question - dusk shooting is > > > easier at more southerly (or northerly for north hemispherians) [quoted text clipped - 7 lines] > > Correct! But you weren't exactly convincing. (O; Yeah... the sarcasm was lost in the text. ;-)
Troy Piggins - 10 Sep 2007 20:44 GMT * mark.thomas.7@gmail.com is quoted & my replies are inline below * :
>> > And as an interesting sidenote and trivia question - dusk shooting is >> > easier at more southerly (or northerly for north hemispherians) [quoted text clipped - 15 lines] > almost like the lights go out suddenly at dusk, no long lingering > twilight... sigh. Yeah, I lived in Cairns for a while. One day I was outside and it was bright daylight, went inside for a cup of coffee, went outside again and it was pitch black. :)
> If you take it to the extremes (at the poles) you can end up with 24 > hour daylight/night... No problems with rushing to catch the "magic [quoted text clipped - 3 lines] > popular as you get closer to the equator, as there isn't so much > variation winter/summer, nor is there much usable twilight. Exactly.
> Just call me Dr Karl... (O; No, that name is reserved for someone who wears louder shirts than you do.
 Signature Troy Piggins Gallery: http://piggo.com/~troy/gallery Please feel free to provide constructive criticism on any photos I post. I'm always learning and appreciate feedback.
Mike Warren - 10 Sep 2007 20:51 GMT > Yeah, I lived in Cairns for a while. One day I was outside and > it was bright daylight, went inside for a cup of coffee, went > outside again and it was pitch black. :) Tell me about it! I call the golden hours the golden 10 minutes.
 Signature Digital Fake Book Free chord/lyric display software for Windows http://www.mike-warren.net/digitalfakebook
-- Posted via a free Usenet account from http://www.teranews.com
Troy Piggins - 09 Sep 2007 13:59 GMT * mark.thomas.7@gmail.com is quoted & my replies are inline below * :
>> I'm thinking about taking some dusk cityscape pics to test out >> my new tripod, and I got to thinking about landscape photos and [quoted text clipped - 19 lines] > the light is right.. Perhaps do your experimenting in daylight, then > apply the knowledge later..? Thanks for your tips.
> And as an interesting sidenote and trivia question - dusk shooting is > easier at more southerly (or northerly for north hemispherians) [quoted text clipped - 3 lines] > > (No prize for first correct answer, just kudos. (O;) I'm thinking further from equator, sunsets take longer?
 Signature Troy Piggins Gallery: http://piggo.com/~troy/gallery Please feel free to provide constructive criticism on any photos I post. I'm always learning and appreciate feedback.
mark.thomas.7@gmail.com - 10 Sep 2007 11:35 GMT > I'm thinking further from equator, sunsets take longer? Correct - see above!
Colin_D - 11 Sep 2007 03:17 GMT > I'm thinking about taking some dusk cityscape pics to test out > my new tripod, and I got to thinking about landscape photos and [quoted text clipped - 28 lines] > > Please offer your thoughts. Doesn't your 30D have a DEP position on the mode knob? Just angle the camera so the outer focus marks are on the nearest and furthest points you want in focus, and take a half-pressure. Read off the aperture in the viewfinder and then set mode to Av and the required aperture if you're using metered exposure, or set mode to manual and set aperture and shutter to the indicated values.
Colin D.
 Signature Posted via a free Usenet account from http://www.teranews.com
Troy Piggins - 11 Sep 2007 04:37 GMT * Colin_D is quoted & my replies are inline below :
> [---=| Quote block shrinked by t-prot: 23 lines snipped |=---] >> [quoted text clipped - 13 lines] > you're using metered exposure, or set mode to manual and set aperture > and shutter to the indicated values. I could do that, but seems like a lot of frigging around. I'd have to change my AF from centre to all, then do as you say, then change back; adjusting modes etc... I do most in full manual anyway.
 Signature Troy Piggins
|
|
|