seany65
Well-known
Hello all, I've decided to bother you with another question (or more if they occur), which I'm sure half-way through I'll realise I should have asked in a different way or something, lol.
Anyway, I've been thinking about 35mm half-frame cameras and their lenses and what focal lengths constitute "wide angle" and "normal", and whether these are actually wide-angle and actually "normal".
I've read that in 35mm full frame usage, you find the "normal" focal length (which gives roughly the angle of view of the human eye) by measuring a length of the 35mm negative, and depending on if you measure the diagonal or the long side, you get a different focal length suggested.
And the same goes for 35mm half-frame camera.
I've read that 28mm on a half-frame camera is the equivalent to a 40mm on a full frame camera, but I've looked at the DoF tables for the 28mm lens on my Petri compact and for my tamron 35-70mm 17a lens at 35mm.
at the same distances and apertures, the DoF for the 28mm half frame is wider (deeper?) than for the 17a at 35mm.
This suggests to me that on 35mm half frame cameras, a 28mm lens is NOT actually equivalent to a 40mm lens on 35mm full frame camera.
So, is a 28mm lens on a half-frame camera STILL a 28mm wide-angle lens, like it would be if it could be fitted to a full frame camera?
Thanks for any help anyone can give.
Anyway, I've been thinking about 35mm half-frame cameras and their lenses and what focal lengths constitute "wide angle" and "normal", and whether these are actually wide-angle and actually "normal".
I've read that in 35mm full frame usage, you find the "normal" focal length (which gives roughly the angle of view of the human eye) by measuring a length of the 35mm negative, and depending on if you measure the diagonal or the long side, you get a different focal length suggested.
And the same goes for 35mm half-frame camera.
I've read that 28mm on a half-frame camera is the equivalent to a 40mm on a full frame camera, but I've looked at the DoF tables for the 28mm lens on my Petri compact and for my tamron 35-70mm 17a lens at 35mm.
at the same distances and apertures, the DoF for the 28mm half frame is wider (deeper?) than for the 17a at 35mm.
This suggests to me that on 35mm half frame cameras, a 28mm lens is NOT actually equivalent to a 40mm lens on 35mm full frame camera.
So, is a 28mm lens on a half-frame camera STILL a 28mm wide-angle lens, like it would be if it could be fitted to a full frame camera?
Thanks for any help anyone can give.
Oren Grad
Well-known
The "normal" focal length is the format diagonal. For full-frame 35 mm (24x36 mm), this is 43.3 mm. For half-frame 35 mm (nominally 18x24 mm, some cameras depart from that a bit), this is 30 mm.
You have to be careful about comparing DOF tables or DOF markings on different lenses. Depth-of-field is defined in terms of a particular circle of confusion - a parameter defining how much blur you can have before an image is visibly unsharp. This in turn depends on how big you enlarge and how closely and critically you view the prints. Remember also that for any given print size, a half-frame negative will need to be enlarged quite a bit more than a full-frame negative.
To compare DOF numbers from different sources you need to know what circle of confusion was assumed in the calculations for each.
A 28 mm lens on half frame will indeed give you approximately the same field of view as a 40 mm lens on full frame. So it is "equivalent" in that very specific sense.
In general, a 28 mm lens *designed for half-frame* cannot be used on a full-frame camera, for two reasons. It will not project an image circle large enough to cover a full frame negative, and in most cases it will be designed for a flange-to-film distance shorter than that of any full-frame camera. However, with a suitable mount adapter, you can use a 28 mm lens designed for full-frame cameras on half-frame cameras. On the full-frame camera it will give you a wide view, while on the half-frame camera it will give you a view just barely on the wide side of normal, because the film records a much smaller slice of the image circle projected by the lens. But the lens remains a 28 mm lens either way, regardless of which camera it is mounted on - focal length is a property of the lens itself.
You have to be careful about comparing DOF tables or DOF markings on different lenses. Depth-of-field is defined in terms of a particular circle of confusion - a parameter defining how much blur you can have before an image is visibly unsharp. This in turn depends on how big you enlarge and how closely and critically you view the prints. Remember also that for any given print size, a half-frame negative will need to be enlarged quite a bit more than a full-frame negative.
To compare DOF numbers from different sources you need to know what circle of confusion was assumed in the calculations for each.
A 28 mm lens on half frame will indeed give you approximately the same field of view as a 40 mm lens on full frame. So it is "equivalent" in that very specific sense.
In general, a 28 mm lens *designed for half-frame* cannot be used on a full-frame camera, for two reasons. It will not project an image circle large enough to cover a full frame negative, and in most cases it will be designed for a flange-to-film distance shorter than that of any full-frame camera. However, with a suitable mount adapter, you can use a 28 mm lens designed for full-frame cameras on half-frame cameras. On the full-frame camera it will give you a wide view, while on the half-frame camera it will give you a view just barely on the wide side of normal, because the film records a much smaller slice of the image circle projected by the lens. But the lens remains a 28 mm lens either way, regardless of which camera it is mounted on - focal length is a property of the lens itself.
David Hughes
David Hughes
What we call normal in 35mm cameras is based on a convention that started with a magazine article about Leica lenses in the 30's. You can't really compare it with the human eye as the film is flat the eye is curved and the eye has a focal length that is nothing like 50mm but more like 22 or 23mm. Also we can see things out of the corner of our eye which suggest more than 180° vision.
Others use a rough rule that says it should be the diameter of the film but that is based on the entire width of the frame and not what is used. Look at a slide holder or enlargers masking frame for that one.
And if 42mm, the diagonal, is standard then ask yourself why so many cameras use 50 or 53mm.
And many photographers with compact cameras will tell you that normal is 35mm and a few will say 28mm.
The best way to judge it is by looking at the angle of the lens coverage, the field of view (FoV). That's about 47° for a 50mm lens. Or you could look at the horizontal and vertical angles, which would work better for say 4:3 aspect ratios as in a half frame...
I think the problem is that we like hard and fast rules and this is a vague measurement we are discussing; worse still, we are confused by all the conflicting rules on the internet. It has a lot to do with what you like in a picture.
Barnack prototype used several lenses* wider and narrower than 50mm that he got off the shelf; I often wonder if they picked 50mm as a compromise between them.
Regards, David
*Search ebay etc for 'mikro-summar' and you'll see the microscope lenses for sale from time to time. I guess he used one -or some - of them as they were to hand.
Others use a rough rule that says it should be the diameter of the film but that is based on the entire width of the frame and not what is used. Look at a slide holder or enlargers masking frame for that one.
And if 42mm, the diagonal, is standard then ask yourself why so many cameras use 50 or 53mm.
And many photographers with compact cameras will tell you that normal is 35mm and a few will say 28mm.
The best way to judge it is by looking at the angle of the lens coverage, the field of view (FoV). That's about 47° for a 50mm lens. Or you could look at the horizontal and vertical angles, which would work better for say 4:3 aspect ratios as in a half frame...
I think the problem is that we like hard and fast rules and this is a vague measurement we are discussing; worse still, we are confused by all the conflicting rules on the internet. It has a lot to do with what you like in a picture.
Barnack prototype used several lenses* wider and narrower than 50mm that he got off the shelf; I often wonder if they picked 50mm as a compromise between them.
Regards, David
*Search ebay etc for 'mikro-summar' and you'll see the microscope lenses for sale from time to time. I guess he used one -or some - of them as they were to hand.
Rob-F
Likes Leicas
Here's a simple way to figure it out:
Let's make the comparison by thinking about the horizontal, or long side of the frame. For full frame, that is 36mm. For half frame, the width of the film becomes the long side, and that is 24mm. 24/36 = 2/3, so the lens for the half frame camera, to cover the same horizontal view, would be 2/3 the focal length of the full frame camera's lens, which is often 50mm. 50mm x 2/3 = 34mm (approximately).
Some fixed-lens full frame cameras come with a 45mm lens. 2/3 x 45mm = 30mm. So the half-frame equivalent would be 30mm.
Or if we are thinking about a Rollei 35 with its 40mm lens then 2/3 x 40mm = 26mm (approximately).
In the last two case, a 28mm seems the closest equivalent.
Let's make the comparison by thinking about the horizontal, or long side of the frame. For full frame, that is 36mm. For half frame, the width of the film becomes the long side, and that is 24mm. 24/36 = 2/3, so the lens for the half frame camera, to cover the same horizontal view, would be 2/3 the focal length of the full frame camera's lens, which is often 50mm. 50mm x 2/3 = 34mm (approximately).
Some fixed-lens full frame cameras come with a 45mm lens. 2/3 x 45mm = 30mm. So the half-frame equivalent would be 30mm.
Or if we are thinking about a Rollei 35 with its 40mm lens then 2/3 x 40mm = 26mm (approximately).
In the last two case, a 28mm seems the closest equivalent.
'Normal' is really a meaningless term, it doesn't refer to any specific technical measurement or calculation. Perhaps 'traditional' works better.
When a 24x36 frame is cut in half to 18x24, the aspect ratio changes, too; ergo, field of view comparisons are not directly comparable between the two formats using the diagonal. But near enough, I suppose.
When a 24x36 frame is cut in half to 18x24, the aspect ratio changes, too; ergo, field of view comparisons are not directly comparable between the two formats using the diagonal. But near enough, I suppose.
Rob-F
Likes Leicas
'Normal' is really a meaningless term, it doesn't refer to any specific technical measurement or calculation. Perhaps 'traditional' works better.
When a 24x36 frame is cut in half to 18x36, the aspect ratio changes, too; ergo, field of view comparisons are not directly comparable between the two formats using the diagonal. But near enough, I suppose.
Yep, right on! That's why I base it on horizontal size, which doesn't involve aspect ratio.
zuiko85
Mentor
One factor in choosing a focal length is how much area the negative has. Miniature negatives, such as the 24x36mm 35mm frame can only stuff so much information into that postage stamp area. With 18x24mm of course, only half that can be recorded.
So….a wider angle on a really small frame may include too much information, trying to record too much detail that an expansive scene possibly contains.
I mess around with 16mm cameras and even smaller the 8x11 Minox. Always looking for bold, simple scenes, with few elements. That is easier with a somewhat more narrow angle.
So….a wider angle on a really small frame may include too much information, trying to record too much detail that an expansive scene possibly contains.
I mess around with 16mm cameras and even smaller the 8x11 Minox. Always looking for bold, simple scenes, with few elements. That is easier with a somewhat more narrow angle.
David Hughes
David Hughes
Part of the problem is that they measure the FoV across the diagonal; I can see that as important from the point of view of the lens designer as the frame must be covered and more to avoid fuzzy edges but for the rest of us the horizontal FoV would make a lot more sense.
Some might say that the diagonal on the film, monitor or TV screen is the largest measurement they can use without lying but I won't...
And then there's APS and digital cameras with 3 or 4 aspect ratios available and they all have the same focal length and that makes it meaningless.
Regards, David
Some might say that the diagonal on the film, monitor or TV screen is the largest measurement they can use without lying but I won't...
And then there's APS and digital cameras with 3 or 4 aspect ratios available and they all have the same focal length and that makes it meaningless.
Regards, David
seany65
Well-known
Thanks to everyone for all the replies and info.
So it seems 28mm on half-frame is "equivalent" to 40mm on 35m full-frame. I can work with that, as I've got a 40mm on a ricoh 500gx.
I wonder what the size of the circle of confusion Petri were using for their Orrikor 28mm lens on my compact/Junior/Half/dejur petri compact is? (I hope the novelty of this camera having 4 names wears off soon, lol).
The eye is a bit like a 22mm lens? Is that just one eye? What size is the "sensor" in the eye?
Anyway, it's 2:45am and I think my quota of attempting to "think" for the day has been reached.
So it seems 28mm on half-frame is "equivalent" to 40mm on 35m full-frame. I can work with that, as I've got a 40mm on a ricoh 500gx.
I wonder what the size of the circle of confusion Petri were using for their Orrikor 28mm lens on my compact/Junior/Half/dejur petri compact is? (I hope the novelty of this camera having 4 names wears off soon, lol).
The eye is a bit like a 22mm lens? Is that just one eye? What size is the "sensor" in the eye?
Anyway, it's 2:45am and I think my quota of attempting to "think" for the day has been reached.
David Hughes
David Hughes
You can create DoF tables here:-
http://www.dofmaster.com/doftable.html
The default is a Canon 7D but scroll down the choices and you'll find several CoC's and then can make comparisons.
Have fun.
Regards, David
http://www.dofmaster.com/doftable.html
The default is a Canon 7D but scroll down the choices and you'll find several CoC's and then can make comparisons.
Have fun.
Regards, David
seany65
Well-known
Thanks for the link David.
Rob-F
Likes Leicas
I've seen several estimates of the focal length of our eyes. Of course there must be some individual variation; some of us have larger eyeballs, some smaller. No point in my throwing in any more estimates to add to the internet "facts" already out there, but this may be of interest: The area of sharp vision occurs in a part of the retina called the fovea, which is very small. At any given moment the area we can see in sharp focus is about the size of a dime held at arm's length. We think we can see a much larger area at a time, but that's because the eye scans around in a series of little jumps--saccades--which the brain assembles into a larger image. Still, much of our FOV at a given moment is out of focus and blurry, like an OOF background.
So I think that while the figure of 22mm is very plausible, since it sounds consistent with the size of the eyeball, I imagine that the field that we are conscious of at any given time is smaller, equivalent to a rather longer lens. Many of us seem to think that a focal length of around 35mm is the most natural. Allowing a confidence interval around that figure, could it be that the "experiential focal length" of the eye is in a range of 28 to 40mm?
zuiko85
Mentor
I’m getting cataract surgery next week and they have already done all the measurements on my eyes to get the right lens size. I'll ask them what my focal length measurements are and post what the answers are.
Rob-F
Likes Leicas
Great idea, let us know! You are going to love the results!
seany65
Well-known
zuiko85, I'd like to know what they say as well.
seany65
Well-known
I've read on the 'web that even though a 28mm lens on 35mm half-frame is "equivalent" to a 40mm lens on 35mm full-frame, the lens is still a 28mm wide-angle.
With this in mind I've been fiddling with stuff on the link David gives. It's a little confusing as the DoF given in the tables isn't quite what the manual for the Petri junior gives, even when trying different CoC's for 28mm and 40mm, though at first glance it seems the DoF for 28mm is closer to the DoF table given in the manual. Here's a pic of the table from the manual:
With this in mind I've been fiddling with stuff on the link David gives. It's a little confusing as the DoF given in the tables isn't quite what the manual for the Petri junior gives, even when trying different CoC's for 28mm and 40mm, though at first glance it seems the DoF for 28mm is closer to the DoF table given in the manual. Here's a pic of the table from the manual:
David Hughes
David Hughes
The point about DoF tables is that they are based on maths and some of the figures in the sum are variable according to what you want out of it or what the lens makers think they can get away with.
In reality/theory the lens focuses everything to a point and only at that point is it in focus but we use our eyes and how small a point we can see is a variable.
And the enlargement of the film etc comes into it.
What they/we do with a DoF calculation is guess how small or large a point we can get away with, without the image looking out of focus.
So we/they might decide that a point of 1/100th inch is the smallest thing we can see. And then you think that a 1" high negative enlarged to 10" high means 1/100" on the print but /1000th on the film.
OTOH, the 1" negative at 4" means 1/100th on the print is only 1/400th on the film.
A lot of makers assume you'll enlarge to (say) the 1/1000 version and a few think otherwise and assume you'll never go beyond 4x6 and so can quote a bigger DoF.
In the days of box cameras they assumed you did same size prints and so could tell you that every thing from 10 ft to infinity would be in focus and they weren't exactly lying.
That brings infinity into it. A lot of them assume infinity is a lot nearer than we'd think. In the Orikkor chart infinity seems to start at over 95.47 ft ...
BTW, the size you can get away with on the negative is called the circle of confusion, because it can be confused with a sharply focussed point.
I hope that makes sense. The website I suggested gives you a choice of the CoC if you explore it. Needless to say, it's metric but converting 1/100th of an inch to 0.254mm and so on isn't difficult.
Regards, David
In reality/theory the lens focuses everything to a point and only at that point is it in focus but we use our eyes and how small a point we can see is a variable.
And the enlargement of the film etc comes into it.
What they/we do with a DoF calculation is guess how small or large a point we can get away with, without the image looking out of focus.
So we/they might decide that a point of 1/100th inch is the smallest thing we can see. And then you think that a 1" high negative enlarged to 10" high means 1/100" on the print but /1000th on the film.
OTOH, the 1" negative at 4" means 1/100th on the print is only 1/400th on the film.
A lot of makers assume you'll enlarge to (say) the 1/1000 version and a few think otherwise and assume you'll never go beyond 4x6 and so can quote a bigger DoF.
In the days of box cameras they assumed you did same size prints and so could tell you that every thing from 10 ft to infinity would be in focus and they weren't exactly lying.
That brings infinity into it. A lot of them assume infinity is a lot nearer than we'd think. In the Orikkor chart infinity seems to start at over 95.47 ft ...
BTW, the size you can get away with on the negative is called the circle of confusion, because it can be confused with a sharply focussed point.
I hope that makes sense. The website I suggested gives you a choice of the CoC if you explore it. Needless to say, it's metric but converting 1/100th of an inch to 0.254mm and so on isn't difficult.
Regards, David
seany65
Well-known
David, I know that only the plane of the point of focus is actually IN focus and that DoF is the "acceptable" zone of "not so blurred as to look out of focus, even though it's not sharp".
I also know that the smaller the negative the more magnification is needed for a decent sized print, so the more critical it is to have an actual sharp image with no camera shake.
I also know that the more DoF an image has, the less likely you are to see the actual point of focus of the lens, and so could have trouble seeing the focal point of the image if other things don't point to it.
The "sums" you quote are a bit above the level of my wo braincells, but I understand that DoF also depends on the CoC the maker has decided to use with any particular lens, partly realised as I played with the different CoC's and looked at the different DoF's for the same focal lentgh, f No. and focused distance. I started off with "0.001" and took it from there. It seems that CoC=0.024 is quite near to what Petri were thinking of. f11 at 10ft for 28mm is fairly close: the site gives (4' 10.7'' - ∞), the manual gives (4.98 - ∞), but other focused distances on the site and in the manual give slightly different numbers, although the manual gives the feet in decimals so for me it's a bit more difficult to equate them. I shall have to faff about with the site and manual table sin metres.
The Orikkor infinity being over 95.57 feet away, seems consistant with the idea that I'd been working with that infinty would be about 128ft. doing the old "half-way between the last marked distance and infinity would be twice that distance, repeat for the new guessed distance and infginity until the gap gets too small". The last stated distance is 16ft, so guessed 32ft, 64ft and so infinity distance 128ft.
I also know that the smaller the negative the more magnification is needed for a decent sized print, so the more critical it is to have an actual sharp image with no camera shake.
I also know that the more DoF an image has, the less likely you are to see the actual point of focus of the lens, and so could have trouble seeing the focal point of the image if other things don't point to it.
The "sums" you quote are a bit above the level of my wo braincells, but I understand that DoF also depends on the CoC the maker has decided to use with any particular lens, partly realised as I played with the different CoC's and looked at the different DoF's for the same focal lentgh, f No. and focused distance. I started off with "0.001" and took it from there. It seems that CoC=0.024 is quite near to what Petri were thinking of. f11 at 10ft for 28mm is fairly close: the site gives (4' 10.7'' - ∞), the manual gives (4.98 - ∞), but other focused distances on the site and in the manual give slightly different numbers, although the manual gives the feet in decimals so for me it's a bit more difficult to equate them. I shall have to faff about with the site and manual table sin metres.
The Orikkor infinity being over 95.57 feet away, seems consistant with the idea that I'd been working with that infinty would be about 128ft. doing the old "half-way between the last marked distance and infinity would be twice that distance, repeat for the new guessed distance and infginity until the gap gets too small". The last stated distance is 16ft, so guessed 32ft, 64ft and so infinity distance 128ft.
phofseth
Established
since this is a forum section on optics and a question about lenses I might add that the Zeiss Tenax right before WWII had 24x24mm frames and a 40mm Sonnar lens as standard plus longer and shorter alternatives. The non-interchangable-lens version had a Tessar 40mm.
p.
p.
raid
Dad Photographer
I have such a Zeiss Tenax with the 40mm Sonnar lens. It is a very nice looking camera and it has a very good lens.