- This topic is empty.
DELETED
-
Went out this morning for a couple of hours before the NEXT :x storm came. :cry: :x :cry: :x
Took some hand held shots at f20 and because they were rested on my car door were pretty sharp but then found a bridge and set the tripod up at f22 at 24mm and 70mm–a series of shots and they arent sharp at all even though i used the timer and no wind at all either.
The metering was the whole 9 points and nothing is sharp anywhere in the whole pic in about 5 pics.
Lens is a canon 24m-70m 2.8L and a canon 20d. and 9 point metering was used on both sets of shots.
Anybody any ideas what could be going wrong?
You don’t say what type of bridge but I suspect there may have been traffic passing over the bridge causing the bridge to vibrate resulting in images with symptoms identical to camera shake. Or, if the bridge was one over traffic, the traffic from below can be enough to cause this too.
Could it be that the lens just isnt that sharp at f22 ?
Unrelated, but just wondering why you were shooting at f22 instead of say f16 ?
Mark
No bridge is an antique bridge that people walked across made out of stone hundreds of years old..
At f22 purely because the bridge more than filled my lens and i was to be honest experimenting a bit.
Do you think that this lens isnt that sharp on distant objects at f22 then ? I think i have only ever done a few close up shots at this aperture, the bridge was about 20 – 100 yards away.
I’m a Nikon user so can’t say anything about this lens. But most/every lens would have some degree of sharpness
problems at each extreme. Just a guess. :)The difference between f20 and f22 shouldn’t be that noticeable though. It could be diffraction because f22 at 24mm is a pretty narrow aperture. Why not post up some example photos?
I cant Pete i cant post straight from my pc, i could maybe send them to somebody who can stick them up for me or any other way please?
PeteTheBloke wrote:
The difference between f20 and f22 shouldn’t be that noticeable though. It could be diffraction because f22 at 24mm is a pretty narrow aperture. Why not post up some example photos?
I’m with Pete on this. Most cameras (not lenses) are going to be diffraction limited at apertures this small. It sounds like this is your problem to me.
tara1 wrote:
I cant Pete i cant post straight from my pc, i could maybe send them to somebody who can stick them up for me or any other way please?
Tara, email me with the photo, I’ll stick it up online. Or alternatively use the gallery feature that comes with this site.
ciaran wrote:
Most cameras (not lenses) are going to be diffraction limited at apertures this small.
Ciaran, could you expand a bit on this comment?
PeteTheBloke wrote:
ciaran wrote:
Most cameras (not lenses) are going to be diffraction limited at apertures this small.
Ciaran, could you expand a bit on this comment?
I’ll try :) This is my understanding of it. (Skip to last bit if you already know what diffraction is)
Diffraction is related to your lens in that it’s related to the aperture you have your lens at. As light passes through the lens, it’s squeezed through your aperture diaghragm. Normally light travels in straight lines, but during this squeezing it’s deflected/disperses (which is called diffraction). The smaller the whole it gets squeezed through, the greater the degree of diffraction. So at smaller apertures, we have a greater amount of diffraction.
As the light diffracts, rather than getting parallel light waves hitting your sensor directly and at the same time, they diverge. So some have to travel different distances before hitting the sensor. One’s that have diverged will interfere with parallel rays of light, sometimes adding to and other times subtracting from them, with softness in the picture being the result. It get’s a bit “mathsy” after this, but here goes… Diffraction interference appears as a shape called an airy disc, which is effectively a solid dot/peak of a certain diameter surrounded by rings spaced apart of ever decreasing thickness/amplitude. (basically a sync function). If objects are close to eachother, their airy discs will overlap, causing softness. A complete overlap will mean the image at that point can not be resolved, where as overlaps less than this leading to softness.
To calculate the diameter of an airy disc for any given aperture, just divide your fstop by 1500. So for a lens at f8 we have an airy disc of .0533mm. If we stop down to f11 it becomes 0.0733mm and stopping down again to f16 we have a diameter of .01066. It can be seen that stopping down increases the diameter of the airy disc. For objects to be resolved sharply they have to be twice this limit apart, anything less will result in softness.
Ok.. here’s how it’s not just about optics
So far I’ve talked about optics in general and just focused on what diffraction is. What I meant by my statment, where cameras are effected by diffraction is as follows:
The resolution of your system (body/sensor/lens) is limited by diffraction if the circle of confusion for your sensor is less than twice the diameter of the airy disk, which is calculated by the above equation. Taking the Nilon DX sensor for instance. It’s circle of confusion is thought to be about 0.02mm. (35mm film is 0.026mm). For sharpness, this means that the largest airy disk should be .01mm, which means it (the sensor.. not the optics) becomes diffraction limited at around f16. In fact on my D2X I’ve found it’s closer to f11.
Note that diffraction is always there, but it only becomes an issue when stopped down past the diffraction limit of your sensor. For my D2X any perceived added DOF I think I may get past f11 is actually lost due the softness introduced by diffraction.
Ciaran, that’s as good an explanation as to why hyper-focal focusing for maximum depth of field, is effectively a complete waste of time. I guess it’s a bit like the duality of light – depending on the conditions, light can either be a wave or a photon. So too, depending on certain parameters, hyper-focal focusing either works or it doesn’t. FWIW, I’ve decided to give up on hyper-focal focusing and now just focus at a point one-third into the scene. Also, the above should give a clue as to why f/8.0 – f/11.0 is generally considererd to be the “sweet spot” of a lens.
written whilst in a rush at work so may not be factually totally correct – but you get the idea :wink:
Thanks for that Ciaran. I had a fair understanding of diffraction – small hole acing like a point source etc., but you threw me when you brought the camera into the equation.
How did you arrive at the info on your D2X? Is it subjective, or did you do a series of tests? I’d be keen to hear more because I currently have two zoom lenses that “overlap” and I want to test them against each other to help me decide whether to flog one (Tamron 17-35mm and Pentax 18-55mm – the Tamron is a full frame lens and the Pentax is strictly for digital).
PeteTheBloke wrote:
How did you arrive at the info on your D2X? Is it subjective, or did you do a series of tests?
It’s a combination of a few things. It’s not often I get out and shoot landscapes, but a while back I broke out the tripod and off I went. Rather than calculating DOF, I went with the lazy method of focusing a third into the frame and shot at the narrowest aperture I had (f32). Disapointingly, the shots were soft. I simply put this down to the lens been soft in the extremes, so the next time, I did a few test shots opening up the lens, whilst taking shots. The results were still pretty disappointing, so I did a bit of research into this diffraction lark.
There’s an “expert” Nikon user, Thom Hogan, who produces books and reviews on pretty much everything nikon (http://www.bythom.com) and I purchased one of his books, which is where I got most of the information. From the maths the DX sensor should be diffraction limited around f16, so I took a series of shots from f22 through to f8 (landscape) and decided myself that f11 was the sweetspot. It’s by no means exhaustive or scientific, just something I landed at myself. But since then, a lot of reviews I have read and threads on various forums, seem to support this.
Ciaran thanks for your help, i have tried twice and have a problem with my email now, will sort it out over the next day or two and resend.
You must be logged in to reply to this topic.
