Olympus E-3 not a software update

You are correct that proper exposure depends only on f-stop, shutter speed and ISO, and the FOV of the 50mm on the Olympus is the same as 100mm on a full frame. And yes, at the same f-stop the DOF of the 50mm is NOT the same as the DOF of the 100mm. The 100mm at F2.8 has a more shallow DOF, the 50 will have to be stop down to F1.4 to match that. Also, if the two cameras has the same number of pixels, the full frame can be shot at 2 f-stops higher before becoming diffraction limited in resolution. Ever wondered why the point-and-shoots have f-stops that only goes up to f5?

Partly diffraction, as you say, but also partly that the lenses were cheap and hit confusion before diffraction, and partly that they are small and at some point a tiny hole is hard to make, even if with the rest of the lens geometry, it's only f8. And why go to f22 when they sync at 1/4000 and can compensate with fast shutter speeds for most of their users (with shaky hands).

As I mentioned in my responses to Roger's post. For the same FOV, DOF and light captured by the sensor. The full frame should have 2x lens focal length, 2 f-stops higher (smaller aperture), and shot at 2-stops higher ISO.

hooo, you lost me. But I want to figure this out, I might learn something.

Herb, put it simply for me. Let's stick with 100mm lenses, your 100mm and my 50mm, assume they have the same FOV, 24 degrees. Say the subject is at 2 feet. What are the relative fstops and DoFs? Are you saying that your aperture is larger, your plate scale means you put more photons into each sensor?

according to the DofMaster, at 2' and f2, I have a DoF of .03', 1.99 to 2.01 (must be some round off). At 2' and f22, I have a DoF of .31', 1.86 to 2.16.

http://www.dofmaster.com/dofjs.html

according to the DofMaster, at 2' and f2, I have a DoF of .03', 1.99 to 2.01 (must be some round off). At 2' and f22, I have a DoF of .31', 1.86 to 2.16.

 

http://www.dofmaster.com/dofjs.html

I posted something about this long ago starting with the Lens Tutorial in photo.net.

http://wetpixel.com/forums/index.php?showt...amp;#entry80286

Skipping to the bottom line it boils down to

DOF = ( c N / M ) * ( 1 + 1/M )

where c is the circle of confusion (about 0.03mm for FF), N is the F-stop, and M is the magnification ( Image size / Subject size ).

If we only consider non-macro cases where M is much smaller than 1 then DOF is simply c N / M^2. The most confusing part of this equation is of course c, the circle of confusion. That's roughly the distance on the sensor over which the image can blur and still be consider in focus, and should scale with the width of the sensor. That is a sensor twice as wide should have c twice as big for the same image resolution, and also for that sensor M is twice as big which implies that N must be a factor of 2 bigger also for this sensor.

Just checking with the DOF calculator you cited I get

for the Canon 5D

100mm lens

f/16

10ft

DOF = 2.89 ft

for the Olympus E-500

50mm lens

f/8

10ft

DOF = 2.94 ft

The difference from complete agreement, I suspect, is due to the author of this site scaling by number of pixels instead of sensor width.

Partly diffraction, as you say, but also partly that the lenses were cheap and hit confusion before diffraction, and partly that they are small and at some point a tiny hole is hard to make, even if with the rest of the lens geometry, it's only f8. And why go to f22 when they sync at 1/4000 and can compensate with fast shutter speeds for most of their users (with shaky hands).

Actually, if you ignore diffraction a smaller aperture would mask the imperfections of lousy lenses more than a larger aperture. Also, if you look back, maximum f-stop( minimum aperture ) has gone from ~f/8 -> f/5 as they put more pixels into the same size sensor, so I think it's unlikely it's f/5 because they can't make smaller holes.

Herb,

I appreciate that you have taken the time to learn all these formulas to explain the merits of a 35 mm sensor but I am a show me kind of guy and I have found that added DOF in macro and wide angle for underwater photography is a good thing not a bad thing. It is in fact the reason that medium format has limited use for underwater photography.

The fact is that Canon, Nikon and Olympus don't all use the same sensor or the same sensor design. It appears from your comments that you would expect that if we both shot an image at F/8 1/125th at ISO 100 with cameras of the same MP count that your image would be four times better and could be enlarged four times larger without less of quality because your sensor is appox. four times larger. That is just not the case.

I agree that if I want to limit DOF that I will need a lens that is one stop brighter than a 35 mm lens but I already have a macro lens one stop faster than any other modern macro lens.

I won't even get into the issues of corner sharpness with W/A 35 mm lenses.

Phil Rudin

I appreciate that you have taken the time to learn all these formulas to explain the merits of a 35 mm sensor but I am a show me kind of guy and I have found that added DOF in macro and wide angle for underwater photography is a good thing not a bad thing. It is in fact the reason that medium format has limited use for underwater photography.

I've never used medium format, but from what I've read, the problem of using it underwater is that the lens selection does not give it advantages vs 35mm. In my view, the same is not true of 35mm vs 4/3. I'll state my basic claim here again. If in the 35mm system you increase the focal length, f-stop and ISO by 2 you get the same image as the 4/3 if the sensor technology is comparable. For example, the 4/3 system would require an F/1.4 50mm macro to match the range of a 100mm F/2.8 in the 35mm.

Yes. Usually more DOF is a good thing in macro. Because the diffraction of light though a small aperture limits the resolution, there's a trade-off between DOF and resolution. 35mm has better DOF than 4/3 at a given resolution. It's acheived at 2 stops smaller aperture. The misunderstanding in this debate is that it's not DOF at a give aperture that's important but what's achievable at a give resolution. Have you seen world beating DOF and resolution from the point-and-shoots with a much smaller sensor?

The fact is that Canon, Nikon and Olympus don't all use the same sensor or the same sensor design. It appears from your comments that you would expect that if we both shot an image at F/8 1/125th at ISO 100 with cameras of the same MP count that your image would be four times better and could be enlarged four times larger without less of quality because your sensor is appox. four times larger. That is just not the case.

No. That's not what I wrote. In this case, if the resolution is not limited by the optics, the resolution should be the same. However, the noise in the 35mm sensor will be lowered in proportion to area (4x) if the sensor technology is the same. There are no direct data available for any of current sensors. We can only estimate from the camera output after much signal processing. You can look at the noise and detail and see if there's noise reduction that destroys details has been applied and so on.... So it's possible to argue that Olympus has found someone to make a sensor for them that perform better than the one that Canon has developed. In my opinion, that's not the case.

E-3 field report from Popphoto Editor in Chief, John Owens.

http://www.popphoto.com/cameras/4793/field...lympus-e-3.html

Phil Rudin