[Peter Zack]

Ok I know that most if not all viewfinders don't show the full image. The K10D is at 95% and I find in many shots I have to account for that if I'm looking for tight framing to eliminate something distracting.

So does anyone know why the viewfinders can't show more of the image seen by the lens/sensor? Is it that there's not enough room for a large enough mirror? Just curious...

[Lowell Goudge]

I think the reason is largely historical, in that film cameras used to account for framing and mounting of slides.

While this does not apply to digital obviously, what about shake reduction? If the sensor can move, then if you had 100% sensor area you might omit something you wanted.

In myu view, if you are that critical on the crop of the shot, it is best to waste a few pixles and PP the image. You can always cut away, but you can never add

[stewart_photo]

Ok I know that most if not all viewfinders don't show the full image. The K10D is at 95% and I find in many shots I have to account for that if I'm looking for tight framing to eliminate something distracting. So does anyone know why the viewfinders can't show more of the image seen by the lens/sensor? Is it that there's not enough room for a large enough mirror? Just curious...

It's the same principle behind why a lens only throws at certain image area onto the image sensor, Peter. However, in this case, the image sensor is your viewfinder display, the lens is the glass prism, and the distance from the back of the lens to the image sensor is the distance from the mirror to the glass prism. To increase the image area on the viewfinder display without introducing distortions, they would have to increase the distance through the prism or the distance from the mirror to the prism. Either of those would increase the physical size of the camera. This is one reason why the K10D is smaller than those few mammoth Nikon and Canon cameras which display 100% of the image in the viewfinder.

And, yes, it's mildly irritating to account for this when shooting, but you'll get used to it as time passes. I use a couple of tactics for this, which are entirely second nature for me after all these years. With a fixed focal length lens, I do a small circle with the camera to quickly account for anything along the edges before pressing the shutter. And, with a zoom lens, I quickly zoom out a tiny bit and then back in again. Of course, you'll eventually develop your own tactics.

stewart

[Peter Zack]

Thanks Stewart. I was more curious than anything. As I've been shooting Pentax's for over 25 years and basically do the same when shooting as you suggested. I was more curious about the physics and wondered what the limitations might be in giving a 100% viewfinder image. That explains it pretty well.

[stewart_photo]

Sorry, I was on auto-pilot by the time I got to that last paragraph (dumping all I was thinking) and forgot once again that you've been involved with photography so long. When I started years ago, I was too naive to know the viewfinder wasn't displaying the full image. I noticed it over time and developed solutions over time, so it never was a big issue. And it wasn't a big issue until I used a medium format camera (100% viewfinder) almost exclusively for several years and then switched back to 35mm. Those first few rolls of 35mm film afterwards were an embarrassment of strange objects along the edges of the images. At that point, it became an big issue to deal with.

Anyway, it's all a matter of camera size and how much distortion the manufacturer is willing to accept. It's obviously possible to get a 100% view in a smaller camera, but the effect in the viewfinder would be about what a wide angle lens would add to an image. Most manufactuers are not willing to go there (certainly not with a serious camera) and would rather crop the viewfinder image instead.

stewart

[Miserere]

Thanks for the explanation, Stewart!

[GaryML]

It's the same principle behind why a lens only throws at certain image area onto the image sensor, Peter. However, in this case, the image sensor is your viewfinder display, the lens is the glass prism, and the distance from the back of the lens to the image sensor is the distance from the mirror to the glass prism. To increase the image area on the viewfinder display without introducing distortions, they would have to increase the distance through the prism or the distance from the mirror to the prism. Either of those would increase the physical size of the camera. This is one reason why the K10D is smaller than those few mammoth Nikon and Canon cameras which display 100% of the image in the viewfinder.

Umm, no. Not true at all.

First of all, we are dealing with the focusing screen and not the image sensor or the film when discussing the viewfinder. The focusing screen must show 100% of the image on the sensor plane, and the view system (prism + optical lenses) must show 100% of the focusing screen.

Second, there is no issue with size or distortion with a 100% viewfinder. My Pentax LX has a 100% viewfinder and it is quite a small compact camera. [Edit: Actually, the specs show 98% vertical and 95% horizontal.] The Nikon D300 has a 100% finder and it is no bigger than the D200 with 95% coverage. There is no engineering reason at all to not show 100% of the image. No distortion problem at all. The reason is economic (the cost of more precise engineering and manufacturing to get 100% coverage) and historical. In the past, consumers who shot reversal film and had prints made would get prints that cropped off the edges of the negative. And slide mounts may crop the image slightly. Camera makers routinely reduced the image size on consumer grade cameras to reflect what most consumers would get on their prints. However, professional grade cameras (like the Nikon F series) would have 100% finders for more control for professionals working with unmounted positive film or making custom prints from negatives.

Third, the finder in modern DSLRS is crowded with all sorts of extra info (camera settings, exposure info, flash status, etc.) In order to squeeze all that stuff in, it is tempting to cut a few percent off the image size. Most people won't notice the difference.

Fourth, the other issue with the viewfinder is magnification. With the smaller focusing screen in cameras with an 18x24mm sensor, the apparent image will be smaller in the viewfinder. You can modify the optical system to increase magnification, but you may also affect the eyepoint of the finder (the eye relief necessary to see the entire finder image). If you have any experience with telescopes, you know that a higher magnification eyepiece has less eye relief, and a better quality eyepiece with better eye relief is much more expensive. A viewing system with higher magnification and good eye relief will be more expensive, but it will be more comfortable to use, especially for eyeglass wearers. Again, a smaller image that is few percent off 100% will same some money and most people won't notice.

---

Last edited by GaryML; 11-30-2007 at 09:40 AM. Reason: Corrected error on Pentax LX viewfinder

[OniFactor]

Umm, no. Not true at all.

First of all, we are dealing with the focusing screen and not the image sensor or the film when discussing the viewfinder. The focusing screen must show 100% of the image on the sensor plane, and the view system (prism + optical lenses) must show 100% of the focusing screen.

Second, there is no issue with size or distortion with a 100% viewfinder. My Pentax LX has a 100% viewfinder and it is quite a small compact camera. The Nikon D300 has a 100% finder and it is no bigger than the D200 with 95% coverage. There is no engineering reason at all to not show 100% of the image. No distortion problem at all. The reason is economic (the cost of more precise engineering and manufacturing to get 100% coverage) and historical. In the past, consumers who shot reversal film and had prints made would get prints that cropped off the edges of the negative. And slide mounts may crop the image slightly. Camera makers routinely reduced the image size on consumer grade cameras to reflect what most consumers would get on their prints. However, professional grade cameras (like the Nikon F series, Pentax LX, medium format cameras, etc.) would have 100% finders for more control for professionals working with unmounted positive film or making custom prints from negatives.

Third, the finder in modern DSLRS is crowded with all sorts of extra info (camera settings, exposure info, flash status, etc.) In order to squeeze all that stuff in, it is tempting to cut a few percent off the image size. Most people won't notice the difference.

Fourth, the other issue with the viewfinder is magnification. With the smaller focusing screen in cameras with an 18x24mm sensor, the apparent image will be smaller in the viewfinder. You can modify the optical system to increase magnification, but you may also affect the eyepoint of the finder (the eye relief necessary to see the entire finder image). If you have any experience with telescopes, you know that a higher magnification eyepiece has less eye relief, and a better quality eyepiece with better eye relief is much more expensive. A viewing system with higher magnification and good eye relief will be more expensive, but it will be more comfortable to use, especially for eyeglass wearers. Again, a smaller image that is few percent off 100% will same some money and most people won't notice.

what's the magnification of your LX, Gary? is there a correlation between magnification and per cent view?

[mattdm]

I think the reason is largely historical, in that film cameras used to account for framing and mounting of slides.

As I understand it, the film cameras that didn't have 100% viewfinders claimed that as a justification, but the truth was that it was just cheaper.

[GaryML]

what's the magnification of your LX, Gary? is there a correlation between magnification and per cent view?

I checked the specs that I could find:

My Pentax LX instruction manual states that the viewfinder shows 98% vertical coverage and 95% horizontal coverage, so that is somewhat less than the 100% I claimed earlier. No mention of the finder magnification. With the LX, there are interchangable finders with different magnifications. I can't find the instruction sheet for the standard FA-1 finder.

On the B. Dimitrov Web site he simply states that the LX viewfinder has "98% Magnification [Coverage]" which confuses two different things:

LX*(late*version)

But on the page with the listing for the finders, he states that the standard FA-1 finder for the LX has 90% magnification with a 15.8mm eyepoint:

Pentax Interchangeable Viewfinders

A few more data points for comparison: The Nikon D300 has 100% horizontal and vertical coverage, 94% magnification with a 19.5mm eyepoint. The full-frame D3 has 100% horizontal and vertical coverage, 70% magnification with an 18mm eyepoint. The Nikon F6 film camera has 100% horizontal and vertical coverage, 74% magnification with an 18mm eyepoint. The FC-1 "action" finder for the Pentax LX (with the FB-1 base) has just 55% magnification, but an eyepoint of 60mm!

The issue about magnification is more critical for the smaller focusing screens on the digital bodies. This is especially an issue for cameras with smaller sensors like the lower end Canons and especially the 4/3 system cameras from Olympus and Panasonic. I read that the finder on the new Olympus E-3 had something like 115% magnification. Obviously, with the smaller focusing screen a higher magnification is needed for a comparable image size, but I don't know the mathematical formula.

I hope that gives you some ideas about the design choices for SLR finders.

[stewart_photo]

Umm, no. Not true at all. (snip)

Of course, the explination given was greatly simplified. However, viewfinder design is indeed a balance between camera size, magnification, distortion, and so on. If you keep the distances equal and add a greater view to the viewfinder without losing magnification, there is going to be some distortion in the same manner as a wide angle lens on a projection system adds distortion to the display screen. If you don't want that distortion, you use a normal lens and increase distance between the projection system and display screen to keep the same image size. You can argue against that if you want, but it is a well established fact of optics design.

Your Pentax LX and the Nikon D300 gain a 100% viewfinder by sacrificing magnification. The K10D, and other Pentax models, can have a 100% viewfinder if users are willing to accept a smaller image (lose magnification) in the viewfinder. Of course, if that were to happen, there would probably be individuals here in this forum shortly afterwards complaining about the small image in the viewfinder, especially since there is already some interest in magnifying viewfinder eyepieces as it is now.

And I don't subscribe to the notion that manufacturers, in this day and age, are saving big money by throwing in a viewfinder prism that generates an image size a one or two millimeters smaller than another prism. The cost difference between those two pieces of glass would be miniscule, if not outright nonexistent. Instead, manufacturers seek to offer a big, bright, viewfinder with nice magnification, and are willing to sacrifice a very small area along the edges of the viewfinder image in the process. Since that has been the norn for many decades, in many cameras, without any major impact on photography, most find that compromise quite acceptable.

stewart

[noblepa]

It seems to me that the distance from the nodal point of the lens to the sensor and the distance from the nodal point to the focusing screen MUST be identical. Otherwise, the focus would not be the same. IOW, an image could be in focus on the screen, but not on the sensor, or vice versa. This is obviously an unacceptable situation. For that reason, moving the screen one way or the other is simply not an option.

The "cropping" of the image to 95%, therefore, must be happening between the screen and the eye; that is, somewhere in the pentaprism/pentamirror assembly. Either that, or the screen is not the same size as the sensor. But it must be the same distance as the sensor is from the nodal point of the lens.

I, too, don't understand why the camera makers do this. I find it very hard to believe that it saves much money. I also find it hard to believe that showing 100% would make the camera that much bigger.

It is very annoying. When I'm trying to do a very tight framing of my shot, I am forced to guess at how much extra image will be included in the final shot.

Just my $0.02 worth.

Paul Noble

[mattdm]

It is very annoying. When I'm trying to do a very tight framing of my shot, I am forced to guess at how much extra image will be included in the final shot.

Or, you can not worry about it and trim it off in post-processing.

[GaryML]

It seems to me that the distance from the nodal point of the lens to the sensor and the distance from the nodal point to the focusing screen MUST be identical. Otherwise, the focus would not be the same. IOW, an image could be in focus on the screen, but not on the sensor, or vice versa. This is obviously an unacceptable situation. For that reason, moving the screen one way or the other is simply not an option.

Exactly right. Otherwise, the image will be out of focus on the film/sensor when it is in focus on the focusing screen.

The "cropping" of the image to 95%, therefore, must be happening between the screen and the eye; that is, somewhere in the pentaprism/pentamirror assembly. Either that, or the screen is not the same size as the sensor. But it must be the same distance as the sensor is from the nodal point of the lens.

I, too, don't understand why the camera makers do this. I find it very hard to believe that it saves much money. I also find it hard to believe that showing 100% would make the camera that much bigger.

It is very annoying. When I'm trying to do a very tight framing of my shot, I am forced to guess at how much extra image will be included in the final shot.

Yes, most likely the camera viewfinder/eyepiece shows slightly less than 100% of the full focus screen, so you don't see the edge/mounting of the screen. This way, you don't have to exactly align the thing, making it cheaper and easier to manufacture. Also the camera viewfinder must super-impose the other viewfinder date (camera settings) over the edge of the focusing screen image.

Yes, it is annoying when extra things show up at the edge of the image frame. But imagine the annoyance if you include something right at the edge of the viewfinder and it doesn't show up in your final print or slide due to the routine cropping of the film processor! Which scenerio is more likely to generate a costly return to the manufacturer for a supposed defect in the camera by an unsophisticated user?

[noblepa]

Or, you can not worry about it and trim it off in post-processing.

I realize that I could trim it in pp, but I'm not terribly excited about post processing anyway. For one thing, I am a computer systems programmer all day. I enjoy photography because it takes me away from that. I don't relish spending all day fighting with a recalcitrant computer at work and then coming home to spend all evening at my computer, doing pp.

Its only a minor nit, that I probably shouldn't be picking. In no way does it spoil my enjoyment of the K10D. Its just that I don't see any reason, whatsoever for not showing the full frame. I don't believe that it makes any real difference in camera size or manufacturing cost.

At the risk of bringing up speculation about the next cameras to come from Pentax, if I were making the list of desired enhancements, a 100% viewfinder would be at least twenty places higher on my list of 100 new features, than live view. Neither would be in the top 50. The viewfinder can be done for next to nothing, while live view will probably add at least $50 to the price.

As always, YMMV.

Paul Noble