Ad
Ad
Ad
Pages: « 1 [2] 3 »   Bottom of Page
Print
Author Topic: Shallow DOF and Sony F828 Comparisonhttp  (Read 7686 times)
dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #20 on: March 14, 2005, 03:25:40 PM »
ReplyReply

In any event, I'm still waiting on you professionals to get a picture given my parameters above to see how high dollalar cameras stack up given a large subject, very limited space to "move away" from the subject, and shallow DOF.

Lin, with all of the snow, you should be able to get a good shot of a large snow plow, huh?
Logged

Lin Evans
Sr. Member
****
Offline Offline

Posts: 269


WWW
« Reply #21 on: March 13, 2005, 03:30:52 PM »
ReplyReply

I don't have any images which would shed much light on this issue. Primarily I use the F828 for macros of highly detailed fine art which is actually at the opposite side of what you are trying to achieve - I shoot for maximum depth of field rather than minimum.

Today here in Colorado I'm covered in snow so won't get a chance to get out and shoot a sample until the weather improves. Perhaps some one else has a representative sample?

Best regards,

Lin
Logged

Lin
dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #22 on: March 09, 2005, 10:12:15 PM »
ReplyReply

Noted Howard. But regardless of the technical aspect of DOF, it seems that my camera is challenged to get shallow DOF, regardless. If you can get a shallow DOF using the F828, please post some images with settings and information so I may learn the technique.

Anyone else want to repsond and post some pictures?

PS--I think I took two pictures that demonstrate perfectly the purple fringe problem with the F828. I think someone would be hard pressed to get purple fringe as bad as I did in this particular picture--haha. I posted it at the bottom of the above page.
Logged

howard smith
Sr. Member
****
Offline Offline

Posts: 1237


« Reply #23 on: March 10, 2005, 08:38:45 AM »
ReplyReply

JW is correct. I mixed some things up. I'll try agian.

I think the question was how to reduce DoF. To get less, or shallower, DoF, you can:

1. Open up the lens. The larger the actual lens aperture the better to reduce DoF.

2. Use a longer focal length lens.

3. Decrease the focus distance, the distance between the camera and subject.

4. The apearance of in-focus (DoF) can be decreased on the print by making larger prints and/or viewing from a smaller distance.  (Of course, if the viewer is far sighted and this isn't corrected, then decreasing viewing distance may increase the DoF for that person.)

5. The apearance of in-focus (DoF) can be decreased on the print by using a sharper lens. Because DoF is the appearance of in-focus, it is easier for the eye to distinguish a difference if the difference is more apparent. The converse is easier for me to understand. It is harder to perceive a difference between an in-focus fuzzy image and a slightly out of focus fuzzy image. A "fuzzier" lens will appear to have greater DoF.

6. The apearance of in-focus (DoF) can be decreased on the print by selecting subjects with crisp edges. Clouds will usually appear to have greater DoF because their edges are fuzzy.

7. Display prints in bright light. This makes it easier for the viewer to see the image and detect focus differences.

8. Make prints that look sharper. Glossy usually appear sharper than matte of textured prints.

Items 1 through 3 are easy. They are hard, physical things on your camera that can "dialed in" or measured.

Items 4 through 8 are more subjective and involve a more personal acceptance of the appearance of in-focus.

There are probably more ways to increase DoF on the print. In my opinion, items 1 through 4 are the heavy weights, more dramatic and easier to see effects.

"... , you cannot seriously be suggesting that you can shoot the same subject, framing, FOV, and distance with an 828 and a 4x5 view camera and get anything remotely approaching similar DOF without using radically different exposure times, f/stop settings, or ISO, are you?"

All I am saying is, there is not different set of DoF facts or equations for an F828, a 35mm, a 6x7, a 4x5 and/or an 8x10 camera. They simply are the same. What ever you do to decrease the DoF on an 8x10 view camera will decrease the Dof on an F828. You do not need to relearn DoF just because you got a new camera.

There are other factors that do come into play that a direct comparison not physically possible. I have never seen a 10mm lens for an 8x10 camera, nor does the F828 have a 12" focal length lens. But that does not change the facts.

"Guys, I feel like my posts are being hijacked for a technilogical discussion each time I post anything on DOF." That is because much of DoF is technical. It is hard to talk about physics without introducing some "science."
Logged
Jonathan Wienke
Sr. Member
****
Offline Offline

Posts: 5759



WWW
« Reply #24 on: March 10, 2005, 09:07:56 PM »
ReplyReply

OK, Howard, that makes much more sense. There is one formula for DOF that applies to all cameras, but the range of practical and usable settings (focal length, physical aperture size, etc.) for different-sized camera formats dictate that the extremes of achievable DOF decrease as the size of the film/sensor increases. It's not solely due to camera format, but the range of useful focal lengths and apertures that go with the format.
Logged

didger
Sr. Member
****
Offline Offline

Posts: 2030



« Reply #25 on: March 11, 2005, 01:56:24 PM »
ReplyReply

Quote
you may want to play with the lens blur filter rather than gaussian blur
Yeah, I'm sure that all the blur options have some utility for different sorts of synthetic bokeh and, of course, you can use gradients along with all the other tools in quick mask to vary the effect with any blur tool.  However, for landscape images these fancy blur effects are of minimal use (to me at least).  I'm sure lots of commercial product shots and fancy magazine ad layouts are routinely done with such trickery.

For my part, I'll limit myself to ideas and let others do the hard work of implementation.  I've got thousands of images to work on and doing a lot of Photoshop experiments just for fun is not on the agenda.
Logged
DiaAzul
Sr. Member
****
Offline Offline

Posts: 777



WWW
« Reply #26 on: March 11, 2005, 04:36:50 PM »
ReplyReply

Quote
I understand that creating information violates the laws of thermodynamics.
I think you are describing entropy - the more hot air that is poured into a thread the more confusing it becomes. Entropy is a non reversible process, so any attempt to clarify the situation just stirs it up more.
Logged

David Plummer    http://photo.tanzo.org/
BJL
Sr. Member
****
Offline Offline

Posts: 5085


« Reply #27 on: March 11, 2005, 04:44:18 PM »
ReplyReply

From what I have seen, softening backgrounds in post-processing can work quite well in trained hands with typical portrait situations, where there is a clear separation between the main subject and background. It probably gets far messier in a situation where you have objects sloping off to the distance and need a realistic gradual fade. And it has little chance of adding "forground blur", such as blurring away branches or a fence in front of the main subject.

About enhancing DOF
a) It is not true that when an image is out of focus, information is irretrievably lost; information from nearby points is mixed together, but in principle it is mathematically possible to unravel this mixing. There is a new technology, being adopted by Zeiss for microscopes, where special lenses produce OOF effects that are uniform instead of increasng with distance from the focal plane. Then subsequent sharpening (akin to what Howard suggests) can achieve great DOF.

 Without this special optical treatment Howard's idea could work to sharpen the image at one particular distance. To apply it more generally, you would need details of the distance to each part of the image. I have imagined a camera that records this distance information by sweeping through all focus distances and effectively taking an AF reading at each pixel by local contrast comparisons. Method a) probaby kills of this idea and yet another chance for me to make my fortune.

c) didger's idea might work too: take a sequence of images at many focus distances and blend them, at each point favoring the image with highest local contrast as an indication of being most in focus.


I wonder what software already exists at microscope makers? In addition to Zeiss, Leica comes to mind, so maybe one of them could make a splash in specialized high end digital photography at last.
Logged
howard smith
Sr. Member
****
Offline Offline

Posts: 1237


« Reply #28 on: March 11, 2005, 08:44:30 PM »
ReplyReply

BJL, thanks for the info.  I won't waste a lot of time trying to focus after the fact.  Because out-of-focus spreads information non-randomly, it seems possible to retrieve it.  I'll wait for a PS plug-in.

I second Jonathan's comments on the F828 focal lengths and the use of the actual, not equivalent, for DoF.

Maybe (?) this would help dwdallam.  Add a supplimental WA lens to the F828.  It would allow the entire boat to be in the image and the focus distance greatly decreased.

DiaAzul's last image (stack of hay bales) is interesting.  But look at the ground just in front of the stack.  It looks to me that the ground looks "out of focus" but the hay just above there is in focus.
Logged
BJL
Sr. Member
****
Offline Offline

Posts: 5085


« Reply #29 on: March 12, 2005, 11:11:47 AM »
ReplyReply

Quote
"35mm equivalent" is meaningless in the contect of DOF, the actual focal length is what's important.
That and the fact that the degree of enlargement is typically different for these very different formats: for prints of the same size, enlargement goes up as the format size goes down.

So as far as DOF goes with the 828, the size factor of 4x means that
a) for a given f-stop, uncropped prints of a given size will have 4 times as much DOF as you would get with a 35mm lens giving the same Field of View. That is, using 35m foramt at the 28 to 200mm "equivalent focal lengths".
 Thus, you get the same DOF as if you used 35mm format with that equivalent focal length and four times the f-stop.

(Explanation: the 4x focal length change changes DOF by a factor of 4^2=16, the 4x enlargement change gives a factor of 4 in the opposite direction, for an overall factor of 4, and if you also make a 4x change in f-stop, that changes DOF by another factor of 4, getting back to the same DOF.)


P. S. to Howard: before you reply, note that I am explicitly addressing the case of making uncropped prints of the same size from the different formats. We know that you believe that this is a poor way to make DOF comparisons, but under the conditions stated, the conclusions follow from standard optical formulas, so can we leave it at that?
Logged
dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #30 on: March 12, 2005, 05:07:50 PM »
ReplyReply

Quote
Maybe (?) this would help dwdallam.  Add a supplimental WA lens to the F828.  It would allow the entire boat to be in the image and the focus distance greatly decreased.
What is a WA filter, and can I use it with my camera. The F828 is a fixed lens. However, I use a polarizer and lens protector with it.
Logged

dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #31 on: March 13, 2005, 01:41:48 AM »
ReplyReply

Quote
Quote
I have a question: how much would it increase the price of the Sony F828 to give it a larger sensor, like the Cannons here, while keeping the camera the same in all other respects? One thing I realy like about the Sony is the design. Everything fits perfectly in your hand. For those who have never handled it, you should just to see how it feels. Although I haven't owned the high dollar digitals, I've held them, and nothing feels like that F828.

It's not possible. The lens determines the circle of definition. A lens like the one on the Sony F828 can only produce a circle of definition to fit the sensor designed for it. If a larger sensor were placed in image plane, then the circle of definition would still describe what it does with the Sony except you would have a circle rather than a rectangle of image.

In addition the aperture would be all wrong. As you may know from optical physics the lens diameter plays the major role in determining the amount of light striking the film or sensor plane, thus defining the maximum F rating. The small sensor allows the small lens (relative to 35mm) to achieve the equivalence of a very fast 35mm lens. With a large sensor it takes more light to achieve the same results.

It's just not possible to have a relatively tiny camera with a large sensor unless it also has a lens properly designed to match the sensor. The lens can be larger than necessary, and the additional circle of definition wasted as with crop factor dSLR's, but it can never be smaller than what is required to achieve the needed aperture.

In essence this simply can't work.

Best regards,

Lin
Well, have you seen the F828? It's not a small camera. But more importantly, what I meant was I wonder how much more expensive the F828 would be if they could make it with a large sensor and matching fixed lens, as in 28-200mm Ziess lens it has now. In other words, I wonder what they could do pricewise to have the same design camera work with a large sensor?
Logged

dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #32 on: March 13, 2005, 02:35:12 PM »
ReplyReply

Quote
i
Yes, I've had an F828 since they became available. If you designed a similar camera with a large sensor it would be the average size of existing cameras with large sensors. The Canon 350D represents about as small of a camera body as will accommodate a 1.6x crop factor sensor. So if you can imagine a camera with a lens about the size of a Canon 70-200 L attached to the body of a Canon 350D then you will be somewhere close to what it would take. To get the F828's lens speed in a large sensor camera would require a lens with an objective of around 77 mm. To get the zoom would require a length similar to the 70-200 and would probably cost in the neighborhood of $4000. It's doubtful that anyone would buy it because the lens itself would weigh over three pounds.

In short, it won't happen and would be totally impractical...

Best regards,

Lin
Lin, why didn't you say so?!?!?  So what is the best shallow DOF you can get using my parameters above for the large subject (60' boat) in a marina type situation (scroll back up a little and see my parameters)?

That would probably be the best real world lesson I could get. No matter how many calculators I use and math models I comprehend, seeing the very limitation of the physical characteristics of the camera, shot by a professional, would allow me to "see' the F828's "best possible" DOF givern my parameters.

After that, if I can't achive the same results, I'll have to do my homework to reproduce what I see here. In other words, if it's impossible to get near what I want, given your superior abilities using the same camera, then spending time trying to get it is a fools errand.

Let me qualify this by saying that I understand knowledge of the technical limitations in conceptual form (math) is indispensible. And I will have a look at what Howard and Jonathan advise.

Jonathan, the portrait is what I would like to have as far as shallow DOF goes.

On the F828, could Sony make the same dedsign, as just a back, with a large sensor? I'd buy that. Like I said, I really like the feel of the Sony design. It feels like a part of my arm after a while.
Logged

dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #33 on: March 13, 2005, 08:47:27 PM »
ReplyReply

Quote
I don't have any images which would shed much light on this issue. Primarily I use the F828 for macros of highly detailed fine art which is actually at the opposite side of what you are trying to achieve - I shoot for maximum depth of field rather than minimum.

Today here in Colorado I'm covered in snow so won't get a chance to get out and shoot a sample until the weather improves. Perhaps some one else has a representative sample?

Best regards,

Lin
I did some more marina shots today. I'm thinking the F828 is pretty much always in semifocus, no matter what one does regarding DOF, except for macro when you do get some decent shallow DOF, and some close up shots, where it also gives some. But for large subjects, I'm not having any success at all.

 I'll be pleased when someone gets some pictures using my parameters above from the F828, and posts them. Then I'll know for sure.
Logged

dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #34 on: March 14, 2005, 03:21:01 PM »
ReplyReply

Quote
The answer is yes, it's possible but certainly not "practical". I could connect my F828 to my Swarovski ST-80 HD or my Meade ETX-90 - shoot from 1000 yards at F14 and produce a blurred background. But in the practical sense of using the camera wide open and shooting from a distance allowable within the native focal length for a reasonable frame it's just not practical.
Not only that, but how are you going to shoot a particular boat in a marina at 1000 yards? You can't becsaue you can only get 50' away from innermost boats, unless you can walk on water. THen you can only get so far becsaue you run into other boats that bloack the view of the boat you need to shoot. But your point is a good one. There is a way to get a shallow DOF using the Sony F828--just impractical, and in some instances, impossible. And your assertion on semantics, from what I can see, is right on too.
Logged

dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #35 on: March 14, 2005, 07:44:52 PM »
ReplyReply

Quote
Seriously, move some boats.  Get (buy, borrow, rent) a different, more flexible or suitable camera.  Get in a boat to get where you need to be.  Shoot on a foggy day when the backgound will be fuzzier.
Actually one of the boat captians said he'd move his boat for me in a week or so. Before I borrow or rent a camera, I want to make sure that I can get teh effect I need; hence, the request for some samples. Shooting on a foggy day is an ecellent idea, as we have fog up here frequently. In fact, that may do the trick.

What I have ended up doing for the interim, is taking pictures of interesting angles and objects and parts of the boats. This either fills the frame, or gives me some decent shallow DOF. So if I can't get the 30 degree bow to stern shots I really wanted with shallow DOF, I can get some interesting shots with decent shallow DOF anyway.
Logged

Lin Evans
Sr. Member
****
Offline Offline

Posts: 269


WWW
« Reply #36 on: March 14, 2005, 12:05:13 PM »
ReplyReply

Depth of field doesn't change "because" of subject size, it's determined by aperture and lens and indirectly by sensor size. Where the subject may lie relative to the capture within this depth of field is a different issue entirely. The camera, lens and aperture determine DOF. How the photographer uses this vis a vis the subject is a different issue.

Lin
Logged

Lin
Jonathan Wienke
Sr. Member
****
Offline Offline

Posts: 5759



WWW
« Reply #37 on: March 09, 2005, 10:38:35 PM »
ReplyReply

Quote
Look at the factors that concern depth of field. To increase depth of field, open the lens up, smaller f number. Increase the focus distance. Increase the focal length of the lens.
Ummm, Howard, only one of those three statements are correct. Opening up to a wider aperture will decrease DOF, not increase it. Using a lens of greater focal length will also decrease DOF. Moving the camera away from the subject to increase focus distance is the only factor you mentioned that will actually increase DOF.

As for the rest of your post, you cannot seriously be suggesting that you can shoot the same subject, framing, FOV, and distance with an 828 and a 4x5 view camera and get anything remotely approaching similar DOF without using radically different exposure times, f/stop settings, or ISO, are you?
Logged

howard smith
Sr. Member
****
Offline Offline

Posts: 1237


« Reply #38 on: March 11, 2005, 04:23:13 PM »
ReplyReply

I understand that creating information violates the laws of thermodynamics. What I was thinking was to increase the perception of sharpness, not actually add information. After all, DoF is just a perception of in-focus.

I was thinking of commenting earlier (and obviously didn't) that blurring isn't the same as out of focus.  A sharply focused image can be blurred, as in camera movement.  Have you ever panned as a biker rode by?  The bike is sharp, legs blurred.  An object behind the subject is very blurred, but might appear sharp if the camera was still.  But then the biker would be blurred instead of focused.

I have made images appear sharper (less haze) by burning shadows/dodging highlights of distant areas. Burning/dodging increases the local contrast. Like sharpening, it adds no information.
Logged
dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #39 on: March 11, 2005, 06:24:10 PM »
ReplyReply

1) Can someone take a pisture of a object, such as the boats I have posted, and show me a nice sdhallow DOF--any camera? I just want that for comparison. No Post processing, though.

2) Howard, I think your estimate on teh true mm size of the F828 is a little off. I think I read where it is actually 28 to 80mm effective?

3) OK, here is a great way to achive a nice shallow DOF, from a poster above. Take three pictures of the same object without moving the camera. Change the focus on each so the background is less in focus. Use Adobe to layer it and then use adjustment layer functions to show the blur and keep the focus where you want it. That could work. Still, it would be a masking/selecting nightmare. However, that would mean the objet could not move even a little. That means doing that with boats in a marina would prove virtually impossible. What do you all think?
Logged

Pages: « 1 [2] 3 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad