The DUMB question...

OK guys,

There have been a few things said on the forums lately, about the difference between 'older' full frame ( FF ) lenses, and the 'newer' cropped ( DX ) lenses. Some of these statements have left me with more questions than answers...and therefore I am fully prepared to ask the DUMB questions...

I will try to be as thorough as I can, and then everyone should be able to follow my stupidity...

When an older FF lens is put onto a digital SLR (DSLR) with an APS-C sensor, the lens is subject to a

"cropping factor" of either 1.5 Nikon or 1.6 Canon. Therefore if the lens is a 50mm focal length, it then becomes the equivalent of a 75mm focal length lens. I understand the theory behind this, so I won't 'muddy' the waters by trying to explain it, although it is fair to say that the resulting image will be the same as that of a 75mm lens on a FF camera...

Now if we use a newer DX lens on our DSLR, of lets say, 75mm focal length, is this newer DX lens and its resulting image the equivalent of the older 50mm lens on the older FF camera??

I think that I got that right...

So I'm thinking basically, that the newer DX lenses have been converted to supply the same images as the older FF lenses after cropping...is this right???

I think that I better lie down for a while...

Bruce...

Focal length is focal length is a constant - a lens can only have one focal length (ok, ok so zooms have many, but these remain the same regardless of the camera they are attached to).

What changes is the Field of View (FoV) which is reduced on a smaller than full frame camera (by the, poorly named, crop factor - should really be called something like the field of view reduction factor but its a bit of a mouthful). So a lens with a field of view of 60 degrees on full frame will only have a field of view of 40 degrees on DX.

However ..... images from different focal length lenses operating on different formats to give the same field of view, have slightly different characteristics - in essence the smaller formats appear to have a greater depth of field at any given aperture. So as an example, to produce a lens for DX format which had very similar characteristics when used wide open to a 24/1.4 Canon, you'd need to use something like a 16mm~ f/1ish on DX format (not exact but not far off).

If you want to read up a far more accurate and better explanation, then Michael Langford's "Basic Photography" and Advanced Photography" books are good primers for basic theory and have been updated for digital (Focal Press are the publishes).

As Paul says focal length is a constant so a 50mm FX and a 50mm DX lens will give the same field of view on a DX DSLR.

What is different is that a DX lens is designed for the smaller sensor and will not fully expose an FX sensor. You'd either get a circular image or at least an image with significant light fall off at the corners, vingnetting.

The other thing about DX lenses is that they are supposed to be optimised for use on digital cameras which need the light to fall on the sensor at right angles to the sensor for best results, unlike film when the angle the light fell on the film was less important.

Sorry, I should have commented on the covering power of lenses which Ken has pointed out!

FWIW lenses are designed with varying covering power - that is the circular image which they actually produce must be able to adequately cover the format for which it is designed. The DX and similar formats suffers from the legacy of 35mm - we assume that DX cameras are 35mm equivatents because they look similar to 35mm SLRs, but of course they are not. They are actually more akin to half frame SLRs (such as the old Olympus Pen range - which were very compact). So we tend to think in terms of 35mm lenses which were designed with much greater covering power than is needed by DX cameras. In effect what we have been doing is to use the central area of lenses designed for a large format - much like using a Hasselblad lens on a Nikon (possible but actually not generally as good a solution optically as using a Nikon lens on a Nikon!

As Ken points out, using a lens designed for a specific format means using one which has been designed with a smaller covering power (ie smaller circular image in the sensor plane) and which has been optimised to give higher resolution over this smaller area. This can be a better solution than using the central section of a larger format lens.

(The reason why I can use a 60mm EFS macro Canon lens on Full Frame with a 12mm extension tube is because by using this much extension, the cone of light which forms the small covering circle needed to cover the smaller Canon format will then also just cover Full Frame but not at infinity - and the results are extremely good on Full Frame from greater than 1:1 to about ~40cm where it just vignettes).

As Ken also says, lenses which project light so that it falls onto the sensor as close to 90 degrees as possible will also help produce better images BUT trying to do this with a conventional ultra-wide (weitwinkel) rectilinear design is difficult and introduces all sorts of optical problems - which might explain which there has been a resurgence in fisheye designs recently (given that they weren't designed for underwater photographers as we are too small a part of the market).

Does this all make sense? Is it helpful?

Hi Paul and Ken,

Your answers have helped me out fantastically. I understand what you have said

and want to thank you for helping me clear up the mess in my head...well one of the messes anyway.

As I said, I had a fair idea as to the field of view issue, BUT, it is when everyone else starts talking/writing about "cropping factors" and "equivalencies" that I start getting confused...

Paul, although it is long winded, I believe that your description of the 'cropping' process is the one that should be adopted, taught and discussed...IMHO

One last question, although slightly off topic...

If I am using a DX lens on a DSLR, and install a screw on diopter to the front of the lens, to shorten the minimum focusing distance, will I also shorten the working range of the lens?

e.g., if my lens has a minimum focusing distance of 40cm and a maximum focusing distance of infinity, and I then install the diopter to achieve a minimum focus distance of 4cm, will this also shorten the maximum focusing distance from infinity down to something sustantially shorter???

Thanks guys, you have been great...

Respect,regards,

Bruce...

One last question, although slightly off topic...

If I am using a DX lens on a DSLR, and install a screw on diopter to the front of the lens, to shorten the minimum focusing distance, will I also shorten the working range of the lens?

e.g., if my lens has a minimum focusing distance of 40cm and a maximum focusing distance of infinity, and I then install the diopter to achieve a minimum focus distance of 4cm, will this also shorten the maximum focusing distance from infinity down to something sustantially shorter???

Short answer - Yes!

Many manufacturers used to publish this data but a quick look through my box of info shows that Canon only publish magnification achievable with their close-up lenses and their own lenses. In essence when your lens is focussed to infinity with the diopter lens fitted, it will actually focus far closer - very easy to check - and this is how far your furthest subject can be. Minimum focus will be closer than the lens alone. Be aware that most 'cheap' close-up 'diopter' lenses cause either or both spherical aberration or curvature of field (soft corners) so need to be used a relatively small apertures. Better, more expensive (achromat) types reduce these. Generally speaking diopters are available in +1 to +4 although +5 are available as well as some specialist stronger versions. Nikon, Canon, Marumi, Leitz (Leica), Zeiss and others all make or made achromats and you can pick these up reasonably used if you look hard enough. Zeiss also build motion picture versions at truly horrendous prices!!!

Dioptres shorten the focal length of a lens therefore for the same back-focus (distance from back node of the lens to the sensor) the lens will focus closer - but not as far as before. You have effectively racked the lens out further. Shorter focal length does not necessarily mean wider angle.

If you use a lens with a larger image circle than you need to cover the sensor (or film) it can cause problems with internal reflections of the spare light within the camera body. For example, I found using a certain long lens on an APS-size sensor camera gave me a flare problem when shooting black subjects on a white background. The extra white light that was not forming part of the image I was using bounced around inside the camera. This is why I suspect you are better off using DX lenses on DX cameras.

Hi Guys,

I just wish to ask one more question about the FF vs DX lens issue.

If I use two identical DSLR cameras and attach two 70mm focal length lenses to these two cameras,

with the only difference between the two lenses being that one is 'older' FF lens and the other is a 'newer' DX lens, will the resulting images be the same? Nothing else done, just point , shoot and look at the images.

Will they be the same image, with the same field of view???

Thanks guys,

Bruce...

It will depend on the lens!

Some older designs are so good that by sampling their central image section they yield results just as good as a modern purpose designed smaller format lens - this is probably more true of macro lenses than others. However, lens designers and manufacturers have many constraints put on them and redesigning a lens for smaller format use may be because of manufacturing costs or for other reasons which we don't know about (Leica for example have had to redisign or cease manufacture of lenses due to unavailability of certain glass types which stopped the production of certain elements). If you were to compare high quality macro primes I doubt that you'd see a great deal of difference but if you are thinking about zooms then modern, purpose built designs small format designs should offer better performance.

As Paul says there are a lot of things that could affect the final image quality but the field of view should be the same.

As Paul says there are a lot of things that could affect the final image quality but the field of view should be the same.

Internal reflections using a lens with an image circle for FF can lead to flare on a DX camera. I have produced an example here (observe the centre of the picture) , purposely over exposing the image in the RAW converter to make the effect more obvious. It is most visible when using long focus lenses and small f/stops and if you have a dark subject surrounded by a light area that is not in the picture and probably won't effect you much underwater. The answer for this shot would have been to use a longer lenshood or reduce the size of the white background. In the event, I had to fiddle with it in Photoshop to hide my mistake!

Dioptres shorten the focal length of a lens therefore for the same back-focus (distance from back node of the lens to the sensor) the lens will focus closer - but not as far as before. You have effectively racked the lens out further. Shorter focal length does not necessarily mean wider angle.

Can you explain why this is? Based on another thread here recently, I placed a diopter on one of my wide angle lenses to check the FOV. I was expecting it to be negligibly wider based on the new calculated focal length but found it to be significantly narrower. It seems that my understanding is incomplete.

Cheers,

Mike

Can you explain why this is? Based on another thread here recently, I placed a diopter on one of my wide angle lenses to check the FOV. I was expecting it to be negligibly wider based on the new calculated focal length but found it to be significantly narrower. It seems that my understanding is incomplete.

 

Cheers,

 

Mike

Thinking in terms of straightforward lenses (not telephoto or retrofocus w/a) the angle-of-view is a characteristic of the distance from the back of the lens to the image. When the lens is focussed on infinity that equals the focal length. However, to focus on something closer, you need to move the lens further from the image plane, narrowing the focal length. This is very apparent when using 10x8 sheet-film cameras as the bellows extension in a studio environment is normally so consequential that the f/stops are no longer accurate and exposure compensation is made by measuring the bellows extension and using the inverse-square law to calculate. (F/stop is an expression of the focal length over the lens aperture diameter but really it should be back-focus over the diam of the lens opening.) The angle-of-view noticeably narrows (and the image circle gets bigger) the closer you focus. That's why people who used to use such cameras rarely used long-focus lenses. Typically I would use a 240mm lens as standard and had a 360mm (theoretically standard view at infinity on 10x8) as my longest lens.

So if your say 50mm lens is set at 50mm from the image plane when focussed on infinity and you add a dioptre to shorten its focal length, you still have the 50mm extension, ergo it is focussed closer.

Are you still awake at the back of the class?

Telephoto lenses are designed to have a shorter back-focus than standard lens of the same focal length and wide-angle retrofocus lenses do the converse to leave room for the camera mirror to operate.

Are you still awake at the back of the class?

Errrr.... Yup! After re-reading your explanation several times and a few nose bleeds, I finally understand. It's amazing how difficult it is to understand a simple concept when your mind expects it to be more complicated. Thanks for the explanation!

Cheers,

Mike