continuous lighting

[Mike_E]

OOPS, we may all be wrong... http://www.luminous-landscape.com/tutorials/dof2.shtml

I guess that the real issue is apparent subject size. By my moving back I was decreasing the size of the subjects to get them all in and thinking that it was a function of only the subject to focal plane distance.

they don't call it the circle of confusion for nothing.

 

[Helen B]

Mike,

The Luminous Landscape article is good, but there are two particular limitations to it:

Catch 1 -it deals with an object significantly closer to the camera than the hyperfocal distance. In this case it is true that depth of field is only dependent on subject magnification and f-number (unless the next catch applies). When you are focusing further way, and the focus point of the shorter lens approaches its hyperfocal distance, the depth of field for the short lens gets greater than that for the longer lens.

Catch 2 -it doesn’t consider pupil magnification. This is what must be applying in your case. Telephoto lenses appear to have smaller exit pupils than normal lenses, and this causes them to give greater depth of field at the same f-number. The converse is true for retrofocus lenses. It isn’t usually a very big effect, but it can be. At the sort of distances we are talking about here, catch 1 trumps catch 2.



All the above applies only to depth of field, not to the apparent amount of background blur.

Best,
Helen

 

[Mike_E]

Hi Helen, I will agree that a wide angle lens has a deeper DoF when you are speaking of the total area in that the shorter the focal length the nearer the point of hyper focus (that few feet plus infinity = more area ).

However, relative to the subjects position in space, I think the article may well be correct. He was talking about the toy dog and not the tower after all.

Visualizing the lens and it's cone of focus, it seems to me that the shape and angle of that cone will not change as long as the area of the image striking it remains the same. Having synced the area of the subject image all of the other near objects' angles (near to the subject that is) should remain the same as well. So, if the CoC is the same then the DoF shuold be as well.

Here is another page with some numbers to compare. The 12th paragraph may be what you are thinking of? http://www.cambridgeincolour.com/tutorials/depth-of-field.htm

 

[Helen B]

I’ll stick with my story that the Luminous Landscape article (‘LL’ hereafter) is misleading with respect to Catch 1 (it ignores Catch 2, but it is a beginner-level article).



For the same object magnification (ie the same magnification of the object the lens is focused on) a lens that is focused near to its hyperfocal distance will have greater depth of field than a longer lens (ie there will be a greater distance between the nearest point that appears to be in focus and the farthest point that appears to be in focus). Fortunately for me, this is easy to demonstrate.

As I mentioned in my previous post, the LL illustrations are of an object that is significantly closer than the hyperfocal distance. In that case my Catch 1 does not apply. The dog is three feet from the gremlin. He states that the camera was between the dog and the gremlin when the 28 mm lens was in use – ie closer than 3 ft. If he used a full-frame DSLR the hyperfocal distance for a 28 mm lens at f/5.6 is about 15 ft.

Supposing that the 28 mm lens is 3 ft from the gremlin, and a full frame DSLR was used. These are the calculated depths of field for constant gremlin magnification (CGM), gremlin distance (GD) and the hyperfocal distances for the lenses he used at f/5.6:

17 mm: 1.28 ft; 1.82 ft; 5.6 ft
28 mm: 1.20 ft; 3 ft; 15.2 ft
50 mm: 1.18 ft; 5.4 ft; 48.5 ft
100 mm: 1.15 ft; 10.7 ft; 194 ft
200 mm: 1.15 ft; 21.4 ft; 774 ft
400 mm: 1.15 ft; 42.8 ft; 3094 ft

There is an 11% increase in DoF going from 400 mm to 17 mm - ie negligible.

As you can see, in none of these cases is the object distance close to the hyperfocal distance. Catch 1 does not, therefore, apply. The DoF is pretty much unaffected by lens focal length.

Here is an example for a D80 (CoC of 0.02 mm assumed) using a 50 mm lens at f/2, focused on an object 60 ft away as the basis. If you want to do the same calculations, see DofMaster. The figures in brackets are the hyperfocal distances at f/2. As you can see, the 25 mm is focused comparatively close its hyperfocal distance, hence Catch 1 applies.

25 mm lens (51 ft), 30 ft object distance: DoF = 53 ft
50 mm lens (205 ft), 60 ft object distance: DoF = 38 ft
100 mm lens (820 ft), 120 ft object distance: DoF = 36 ft
200 mm lens (3280 ft), 240 ft object distance; DoF = 35 ft

There is a 39% increase in DoF going from just 50 mm to 25 mm, and a 51% increase in DoF going from 200 mm to 25 mm.

In each case the magnification of the object in perfect focus is the same. The magnification of the object at the far limit of the DoF is different:

25 mm lens, object at 72 ft, magnification 0.11%
50 mm lens, object at 85 ft, magnification 0.20%
100 mm lens, object at 140 ft, magnification 0.23%
200 mm lens, object at 259 ft, magnification 0.25%

This does not, however, affect the result that the distance from the near to the far points that appear to be in focus is greater for the 25 than for the 50 mm and 100 mm. As the focused object gets further from the hyperfocal distance, the difference in DoF between the lenses decreases.

The statement in LL "In fact, if the subject image size remains the same, then at any given aperture all lenses will give the same depth of field" is incorrect, because of Catch 1 and Catch 2. The experiment was flawed because they did not do their homework. They should have repeated it at a greater distance to show Catch 1, and closer with telephoto and retrofocus lenses to show Catch 2.

 

[chrisk121]

want white so i can easily cut out the people, or blend it into the white background on the cover.

Why do they need to be jumping if you are going to cut out the background? Wouldn't it look the same if they were standing? You also have the problem of timing the jumps.

 

[Mike_E]

OK, a quick test with my camera. Using the internal grid to measure the apparent size of the subject and 24mm and 60 mm as the focal lengths. The distances were (approximately as I didn't use a caliper) 0.76M @ 24mm and 2.0M @60mm.

Using the DoF calculator graciously provided by Cambridgeincolour.com the DoF for 24MM @0.76M is 0.064M and the DoF for 60mm @ 2.0M is 0.058M. 0.006M is well within the limits of error for my measurements.

Not only does Luminous landscapes maintain what I just tested, so does cambridge-in-colour.
A site by this Gentleman, Sean McHugh http://www.cambridgeincolour.com/about.htm who seems to have quite the resume.

Helen, I don't mean to be argumentative and I really think we may be talking about two different things.

In the internist of clarity for anyone else reading this who may be unfamiliar with DoF Please read the above links and make up you own mind (you could always test too. )

ChrisK, great question!!

 

[doobs]

Why do they need to be jumping if you are going to cut out the background? Wouldn't it look the same if they were standing? You also have the problem of timing the jumps.

If they were standing, it would look like they are standing; not jumping. :\

 

[spyder]

Thank you all for you advice.

If you are making a composite by stitching together five images and removing the background, why not do it as a composite with one camera and only a few people at a time...

i was planning on everyone holding hands because the theme is connections. although now after all of the replies i may want to do a simpler route.

How much electrical power is available? What f-number did you have in mind for DoF? What ISO will you be able to get away with? What is your budget? Can you rent?

-i was planning to shoot wide open at f1.8 so i could have the max shutter speed, but now i realize that that would be very hard to have everyone in focus.
-i dont want any noise, so i was planning on around 400ish
-i am not exactly sure what my budget is, but i dont think i can buy a bunch of strobes.
-i doubt i will rent. it is just for yearbook, so if it was impossible to do i would probably just split up the people

how important it is to you that all the photos are taken at the exact same moment.

it is pretty important because the image will wrap around the bottom quarter of the book in a ring around the cover. i was planning on having it on burst, and taking a bunch of shots with all the cameras. i am very comfortable with photoshop so i figured i would be able to piece the people together even if it wasnt perfect, but the closer they are to perfect the easier it will be.

PS Are you using auto white balance for your volleyball shots? Have you tried it with custom white balance?

i was using auto that game because i had forgotten how to do the preset white balance, but when i went home that night i looked it up.

Spyder, having 5 cameras to work with is great, perhaps you should do the shot outside so that lighting won't be an issue- a painted wall somewhere?

thats not a bad idea. i should probably go scout out some better shooting areas because the gym is notorious for bad lighting

Do you have a theatre? That could solve you lighting right there.

BRILLIANT! we dont have a theater per say, but it is a church school so we have a chapel thingy that has lots of light. in fact countless speakers have commented on how bright the spotlight is on them. it would be a bit harder to set up the background because there is no wall to stick it to, but there is a whitish curtain behind it that may suffice. i will explore this option.

Here is a rough estimation of distances.
Wireless triggering at the 'same time'? To what accuracy? Do you know for sure that the timing variation between D80s following simultaneous wireless triggering is small enough not to be a problem with fast-moving objects?

i know that they are not perfectly synced, but very close. i did think about this being a problem, but i figured i would take a ton of shots so i have a lot to work with in photoshop.

Indeed, one camera, one shot would be ideal, and I wondered why that wasn't your first thought - I guessed that you must have a good reason to discard the obvious (it should have been Plan A). Resolution was my first guess.

this was my first idea, but as you stated, resolution would be a problem. i want a really crisp, sharp picture because it will be the focus of the yearbook page. the people wil be about 3 inches tall on the cover. i figured the 30 feet by using a ratio. 5ft(average height) to 3 inches equals a little under 30 feet to 17 inches.

Why do they need to be jumping if you are going to cut out the background? Wouldn't it look the same if they were standing? You also have the problem of timing the jumps.

-some will have their legs tucked up, and their hair etc will be mid air and cool looking.
-as to the timing, i am going to have my friend who is asb pres be in charge of telling everyone what to do and when. he is really funny, but commands attention. plus he is friends with everybody. i figure al long as about 75% of the people are in air, ad everyone looks happy, it will be a good/interesting shot.





again, thank you all for the input, it is invaluable. and sorry for the huge, huge reply.

 

[Helen B]

OK, a quick test with my camera. Using the internal grid to measure the apparent size of the subject and 24mm and 60 mm as the focal lengths. The distances were (approximately as I didn't use a caliper) 0.76M @ 24mm and 2.0M @60mm.

Using the DoF calculator graciously provided by Cambridgeincolour.com the DoF for 24MM @0.76M is 0.064M and the DoF for 60mm @ 2.0M is 0.058M. 0.006M is well within the limits of error for my measurements.

Not only does Luminous landscapes maintain what I just tested, so does cambridge-in-colour.
A site by this Gentleman, Sean McHugh http://www.cambridgeincolour.com/about.htm who seems to have quite the resume.

Helen, I don't mean to be argumentative and I really think we may be talking about two different things.

Mike,

You've missed something, I'm afraid. I wrote

"Catch 1 -it [the Luminous Landscape article] deals with an object significantly closer to the camera than the hyperfocal distance. In this case it is true that depth of field is only dependent on subject magnification and f-number (unless the next catch applies). When you are focusing further away, and the focus point of the shorter lens approaches its hyperfocal distance, the depth of field for the short lens gets greater than that for the longer lens."

I later gave examples of two cases, one with a close object (the LL example), and one with a more distant object. I pointed out that they result in different relationships. You don't say what f-stop you used, but it sounds like your 24 mm was nowhere near the hyperfocal distance if it was 0.76 m away. Catch 1 does not apply in that case.

Here is a quote from Cambridge-in-colour:

"We describe depth of field as being virtually constant because there are limiting cases where this does not hold true. For focal distances resulting in high magnification, or very near the hyperfocal distance, wide angle lenses may provide a greater DoF than telephoto lenses. On the other hand, for situations of high magnification the traditional DoF calculation becomes inaccurate due to another factor: pupil magnification. This actually acts to offset the DoF advantage for most wide angle lenses, and increase it for telephoto and macro lenses. At the other limiting case, near the hyperfocal distance, the increase in DoF arises because the wide angle lens has a greater rear DoF, and can thus more easily attain critical sharpness at infinity for any given focal distance."

He is describing Catch 1 and Catch 2, albeit not very accurately - but it is just an introduction, not the whole works (and he is a Cambridge man, so one shouldn't expect too much). Apart from anything else, he refers to 'the traditional DoF calculation'. I think that he means 'the simplified DoF calculation', because the rigorous one is every bit as 'traditional' as the simplified one, and the rigorous calculation does not become inaccurate in the situations he describes.


Best,
Helen

 

[Mike_E]

(and he is a Cambridge man, so one shouldn't expect too much).


Best,
Helen

Yep, I must have missed something.