[wolfier]

Actually, I get a faster shutter speed for the same DOF on the APSC camera. The FF camera is only "better" if I want "less" DOF, not if I want more, and only if I have a lens of equivalent speed at the longer focal length. If I have a 200mm F2.8 lens on the APSC camera and a 300mm F4 lens on the FF camera (about the same size, weight and cost) there is no real advantage to the FF camera. I would need a 300mm F2.8 lens and they are a lot bigger and more expensive.

Hey, I thought the context here is to keep the DOF the same, so it's understood i this case people would have to turn up the ISO by 1.3 stops on the 135 body to achieve the same shutter speed.

And don't take me wrong. I don't belong to the camp that believes FF is better in all regards. FF and APS-C are better than each other at different things.

As to comparing WA lenses, I'd use the 16-35 II instead of the 17-40 (http://photozone.de/canon_eos_ff/435-canon_1635_28_5d?start=1), which, in terms of resolution and CA, seem to tell a much rosier story for FF, especially after a slight stop-down to match the DOF of the APS-C body. (You're welcome to pick a better lens on APS-C for comparison, if there is any - I took a look at the Tokina 11-16 and it doesn't seem to be much better than the Canon 10-22 - they're all great lenses and honestly they're all better lenses than I can make use of them, but the 16-35 II seems to have the edge near the edge).

I have to make my preference clear, though. I don't consider vignetting and a distortion to be serious issues on WA lenses, in my mind they're both minor when compared to resolution and CA, especially both are correctable in software with a relative minor quality loss - and in most situations for my shooting, the vignetting and especially the distortion don't need to be corrected at all.

---

Last edited by wolfier; 09-07-2009 at 03:17 PM.

[*isteve]

Hey, I thought the context here is to keep the DOF the same, so it's understood i this case people would have to turn up the ISO by 1.3 stops on the 135 body to achieve the same shutter speed.

And don't take me wrong. I don't belong to the camp that believes FF is better in all regards. FF and APS-C are better than each other at different things.

I was keeping DOF the same, just not necessarily at max aperture.

As to comparing WA lenses, I'd use the 16-35 II instead of the 17-40 (Canon EF 16-35mm f/2.8 USM L II (full format) -Test Report / Review), which, in terms of resolution and CA, seem to tell a much rosier story for FF, especially after a slight stop-down to match the DOF of the APS-C body. (You're welcome to pick a better lens on APS-C for comparison, if there is any - I took a look at the Tokina 11-16 and it doesn't seem to be much better than the Canon 10-22 - they're all great lenses and honestly they're all better lenses than I can make use of them, but the 16-35 II seems to have the edge near the edge).

Hmm - just compare the prices though. I agree the 16-35II is better than the 17-40 (which is close to unusable). Note also, the original version of the 16-35 was awful. Using a legacy WA lens on FF does not always work. Canon and Nikon have already had to bring out updates to many of their wide to mid range zooms so that they work well on FF.

I have to make my preference clear, though. I don't consider vignetting and a distortion to be serious issues on WA lenses, in my mind they're both minor when compared to resolution and CA, especially both are correctable in software with a relative minor quality loss - and in most situations for my shooting, the vignetting and especially the distortion don't need to be corrected at all.

The major issues are corner softness and field curvature. I really dont think FF is much better in this respect.

[arbib]

I also was somewhat perplexed from the result of the excercise. But when I thought of it it did make some sense. I was too used of looking at low noise performance by comparing noise at different ISO levels. But when you think of it, ISO is a not a parameter you set when you construct on image. You want to control the FOV, the DOF and Shutter Speed to achieve your wanted "look". Myself I mostly use aperture priority to get my wanted DOF, then I set the longest time I can use depending on if the object moves, or if I use handheld. The ISO I have to use is then a consequence of those parameters. And it so happens that the relation between FOV, DOF and sensor size is such that the gain I have in better performance at a given ISO for the FF (roughly one stop) is exactly opposite to the loss I have due to having to go to a smaller aperture to achieve the same DOF.

Of course the relation assumes that you are within the limits set by the lenses, but this is likley to be the case when looking at longer focal lenses where max aperture is typically getting smaller as focal length increases. For wide lenses the FF will give you the possibility to go to a smaller DOF than what is typically possible with current set of lenses. But for me the problem with having to go to high ISO is most often occuring on the slightly longer focal lengths (sports, nature/animals etc.) and then the benefit with FF in terms of noise performance seems more or less gone.

Best regards,
Haakan

I'm with you here. I use Av mode mostly to set DOF. I use the ISO to achieve the shutter speed in where I need to do this too. I am very conformable with higher ISOs of 500/800 in daytime photos where I may be under tons of trees or other things that affect the brightness of the light quality.

Myself, with todays tech., FF cameras for me are not such a big goal for IQ. But maybe for Fast primes in candid photography for the limited DOF. But Really, I like a slightly deep DOF in head shots..I want from the nose to just in front of the ear to sharp. And f/1.4 or f/1.8 may not do it at that close of a distance. Depending on the FL of the prime. Now the 31mm Limited may be perfect wide open at f/2.4 for a head shoot for what I want anyway

[Photomy]

Hi wolfier,
I agree that FF in some cases have an advantage (and as you pointed out there are also the opposite cases). But you first example is where it actually in real life does not have a noise advantage.

If you can increase the shutter time for FF, you can of course to the same for the APS-C, again giving the same noise performance. As discussed in the thread, when aperture is stepped down in order to gain DOF, you will not have any noise advantage with FF.

Best regards,
Haakan

Yea, but it is easier to design and make a lens that goes to a smaller aperture than to a larger one. If they do not make the lens with one more larger aperture, unfortunately you can't use it . So does this change the debate at all? The fact that a higher ISO is available to the FF, but there is a limit to how big the APS-C aperture is? I wonder why Pentax did not make the 50-135 at f2.0? Could it be the doubling in cost and weight in the lens? Now do this for all the lenses in the system. The cost of the system is now 3 to 4 times.

There seems to be a point where it is less expensive and more practical to get more light gathering capability for reduced noise or faster shutter speed (if desired by the photographer) with a single investment in a larger photosite sensor (FF) than make several expensive faster lenses (Analogous to in body SR?). And actually less weight on system basis if this is a goal. Olympus did not make their top zoom f1.4 even though the photosite size is a limiting factor in ISO performance for the E single digit line pro camera. With the much smaller mount size (and the smart 4:3 ratio) they were able to come up with a fantastic, and very expensive, f2.0 14-35 (28-70) however. I guess you could make the same analysis between the E-3 and the K7.

A critical balancing act appears to be in play with the lens designs, performance, and cost as well as for the sensors. Each company tries to get the best on-printed-page quality within a given system budget (market price point). Is this why Leica is jumping ahead to an even larger sensor system for the large budget buyer to achieve another system performance/cost benefit? Not sure.

By the way, is the ISO noise advantage above ISO 800 one or two stops for full frame? It seems to be much closer to two stops in most cases these days. Two stops is a lot.

[falconeye]

By the way, is the ISO noise advantage above ISO 800 one or two stops for full frame? It seems to be much closer to two stops in most cases these days. Two stops is a lot.

It isn't above ISO800. It is at every ISO if ISO is kept constant, i.e., if lens size is allowed to change.

Then, the noise advantage is 2log(1.53)/log(2) or 1.23 stops. Independent of the day

[Photomy]

It isn't above ISO800. It is at every ISO if ISO is kept constant, i.e., if lens size is allowed to change.

Then, the noise advantage is 2log(1.53)/log(2) or 1.23 stops. Independent of the day

Getting back to Pentax system specifically, it would help us if Pentax came out with some lower priced, wide angle, fast, quality, primes as Nikon did recently with the 35mm f1.8. SLRgear shows it to be decent lens for $235. Maybe not quite top pro level, but decent.

Is it not feasible to make a APS-C version (33.3mm) of the 50mm 1.4 at about the same cost? I assume not or they would have done it. I guess it would actually have to be f1.0 to have the same capability on APS-C as the 50 1.4 does for a FF camera. There seems to be a great challenge to completely scale-down lenses to the APS-C size sensor as to cost and quality. Is this because the K-Mount is actually a former 35mm film size mount? Is it because of the lack of commitment (uncertainty) long term to the sensor format? Lens physics? Maybe this would be a whole other discussion.

[*isteve]

Getting back to Pentax system specifically, it would help us if Pentax came out with some lower priced, wide angle, fast, quality, primes as Nikon did recently with the 35mm f1.8. SLRgear shows it to be decent lens for $235. Maybe not quite top pro level, but decent.

Is it not feasible to make a APS-C version (33.3mm) of the 50mm 1.4 at about the same cost? I assume not or they would have done it. I guess it would actually have to be f1.0 to have the same capability on APS-C as the 50 1.4 does for a FF camera. There seems to be a great challenge to completely scale-down lenses to the APS-C size sensor as to cost and quality. Is this because the K-Mount is actually a former 35mm film size mount? Is it because of the lack of commitment (uncertainty) long term to the sensor format? Lens physics? Maybe this would be a whole other discussion.

The problem is the legacy of the registration distance required for SLR cameras. If the focal length of the lens is > the registration distance (about 45mm) then no problem. The lens' front element is the objective lens and the dimensions of the lens can be kept quite compact (vis 40mm pancake) as the objective lens can be at the correct distance from the sensor. In this case the length of the lens is dictated purely by focal length (distance from objective lens to focal point) and width is dictated by F number (diameter = focal length/Fmax).
.
However lenses with a focal length < the registration distance require an additional divergent lens/lens group in FRONT of the objective lens to increase the angle of view because you cannot move the objective lens closer to the sensor. This is why WA lenses are so wide and bulbous despite the small focal length. The shorter the FL and faster the lens the larger this divergent element becomes and the more aspherical correction it requires, which means the cost and size goes up exponentially as F stop gets wider and focal length gets shorter.

So on a Pentax SLR, the smallest possible lens will be around 45mm (although you can push this a bit because the final lens element actually projects slightly into the camera). With APSC you also need less aspherical correction for a given focal length because the angle of view is smaller.

However, the cost of a lens is directly proportional to weight of glass and number of aspherical and special glass elements. As a result, costs will increase as you get wider or longer than the reg distance.

Of course if you buy a rangefinder or micro4/3 camera the reg distance is much smaller, and small cheap WA lenses with minimal correction are a possibility.

[falconeye]

The problem is the legacy of the registration distance required for SLR cameras.

Great explaination, thanks!

Add to it the legacy of the registration (or mount hole) diameter. As studied in another thread, the PK mount is limited to f/1.2, some other SLR mounts to f/1.1 and some range finder mounts to f/0.95.

So, a PK mount 50/1.4 is possible while a 30mm/1.0 isn't. The excellent PK mount 30/1.4 from Sigma may be as good as it gets.

[Photomy]

Well, I guess it is also true that it is much easier with current digital technology to add bokeh (reduce DOF) to APS-C than to add DOF to FF. They have face detect now for AF. Maybe in the future we will have in-camera subject select, and digital bokeh added in the camera after the shot.

See this new product.

Alien Skin Software: Bokeh Depth of Field Examples

[Photomy]

The problem is the legacy of the registration distance required for SLR cameras. If the focal length of the lens is > the registration distance (about 45mm) then no problem. The lens' front element is the objective lens and the dimensions of the lens can be kept quite compact (vis 40mm pancake) as the objective lens can be at the correct distance from the sensor. In this case the length of the lens is dictated purely by focal length (distance from objective lens to focal point) and width is dictated by F number (diameter = focal length/Fmax).
.
However lenses with a focal length < the registration distance require an additional divergent lens/lens group in FRONT of the objective lens to increase the angle of view because you cannot move the objective lens closer to the sensor. This is why WA lenses are so wide and bulbous despite the small focal length. The shorter the FL and faster the lens the larger this divergent element becomes and the more aspherical correction it requires, which means the cost and size goes up exponentially as F stop gets wider and focal length gets shorter.

So on a Pentax SLR, the smallest possible lens will be around 45mm (although you can push this a bit because the final lens element actually projects slightly into the camera). With APSC you also need less aspherical correction for a given focal length because the angle of view is smaller.

However, the cost of a lens is directly proportional to weight of glass and number of aspherical and special glass elements. As a result, costs will increase as you get wider or longer than the reg distance.

Of course if you buy a rangefinder or micro4/3 camera the reg distance is much smaller, and small cheap WA lenses with minimal correction are a possibility.

Despite people getting all bent out of shape at the notion of no optical viewfinder, I believe there is a bright future for high quality mirror-less cameras. Samsung NX or whatever else that may out there on the drawing board. But, new lens designs with the option of much shorter registration distance will need to be taken advantage of. Maybe this is one of the delays of the Samsung NX - the need for new lens designs and adapters for Pentax and other lens brands. And probably faster, accurate contrast detect AF is still a challenge.

I started out using large format where speed and clarity of image preview is not important. Final image quality is the important thing.

[falconeye]

Well, I guess it is also true that it is much easier with current digital technology to add bokeh (reduce DOF) to APS-C than to add DOF to FF.

This comment is not adding to the discussion

BTW, to have APS-C DoF in FF, just stop down

Despite people getting all bent out of shape at the notion of no optical viewfinder, I believe there is a bright future for high quality mirror-less cameras.

Mirror-less cameras are an old thing and the best implementation to date was just announced yesterday: The Leica M9.

To really benefit from the concept, you need excellent lens quality and super accurate focus for wide angle as well. Something Leica is famous for. Not sure about NX though

[Marc Sabatella]

I've always assumed that a shorter registration distance is "good" for wide angle. But the M9 announcement and dpreview preview mentioned the flip side disadvantage that hadn't occured to me: a "simple" (not sure of the technical term, but not retrofocus or reverse/inverted telephoto or whatever one calls a typical DSLR wide angle) wide angle lens will have the light striking the sensor at pretty steep angles as you move away from the center, which complicates the design of the sensor. Wondering how that will play out in practice.

[DAZ]

I've always assumed that a shorter registration distance is "good" for wide angle. But the M9 announcement and dpreview preview mentioned the flip side disadvantage that hadn't occured to me: a "simple" (not sure of the technical term, but not retrofocus or reverse/inverted telephoto or whatever one calls a typical DSLR wide angle) wide angle lens will have the light striking the sensor at pretty steep angles as you move away from the center, which complicates the design of the sensor. Wondering how that will play out in practice.

I think you are looking for image-space telecentric (not object-space telecentric or maybe double telecentric). These lenses can be big $$$.

DAZ

[falconeye]

a "simple" (not sure of the technical term, but not retrofocus or reverse/inverted telephoto or whatever one calls a typical DSLR wide angle) wide angle lens will have the light striking the sensor at pretty steep angles as you move away from the center, which complicates the design of the sensor. Wondering how that will play out in practice.

Don't know the term either. Symmetrical lenses are a possibility

Yes, the steeper angle was of concern for early digital Leicas. I think they now have it pretty much under control, with clever microlens design and stealth firmware fixes (e.g., Adobe was forced to include vignetting correction as a mandatory and non visible step in its raw converters).

Even for film and with a perfect lens, the genuine cos^4 correction was always there. cos^4(45°) = 0.25 or 2 stops ... 45° is the corner angle for a 22mm lens on 135mm FF (14mm APS-C).

[24X36NOW]

Sorry for the perhaps controverisial heading, but I thought that it might catch the attention, as well as it does convey something that, at least to me, came somewhat as a surprise.

When considering FF vs APS-C I always took as absolute fact that if all things equal, the FF will have about one stop better low light performance. This based on that with double the area, it will have one stop better ISO performance.

But then I started to think what parameters I play with when constructing an image. The parameters I control are FOV, DOF (aperture) and shutter time. The ISO setting comes as a consequence of these parameters (and sometimes as a limitation if noise becomes too high).

So then I thought that in order to have more of an "apple to apple" comparison, I should compare the noise performance under the same image parameters, i.e. under the same FOV, DOF and shutter time.

For a given FOV and DOF, the required f-number is proportional to the format size. Thus with FOV, DOF and same shutter time, the FF will require a one stop higher ISO compared to the APS-C.

In order to see how a SNR curve between APS-C and FF would look like if comparing images taken under the same image conditions (same FOV, same DOF and same shutter speed), I took the DxOMark numbers for Nikon D90 and D700 (I used these cameras since they have the same pixel count and was released roughly at the same time, so they should be similar in technology evolution status, but I do not want to trigger a brand vs brand discussion).

I have attached the curve I ended up with. Of course, the comparison holds only within the limits that the aperture of the lens can be changed.

A very practical situation would be e.g. in sports or in nature shooting using long focal length lenses. Using a lens of 400mm f/2.8 for APS-C and 600mm f/4.0 would actually result in images having the same noise performance. But the 600mm lens would set one back an additional $2000 dollars (using Nikon prices)

I am sorry if I have stated something that is old news to you all, but too me the result came somewhat as a surprise and was not what I had intuitively thought before. If I have made any logical sommersaults in my thinking, I will be happy to be corrected. As for now, it made me realize that for the way I shoot images, a FF would actually not give me any benefit in terms of low noise performance.

Best regards,
Haakan

Sorry, gotta call BS on this. Your mistaken assumption is that exposure requires a higher ISO on the FF camera, thus negating the noise advantage in your rather contrived scenario. Unfortunately, an APS-C dSLR doesn't get equivalent exposure to a FF at the same aperture, since the smaller sensor loses a stop of light gathering ability. So, exposure on the APS-C at an aperture of f 2.8 is equivalent to f 4 on FF. Ergo, you must ALSO increase the ISO by a stop on APS-C to take your equivalent DOF/AOV/Shutter Time picture, given the same amount of light on subject.

This is easy enough to test; take some "Sunny 16" images with your APS-C dSLR of a front-lit subject on a bright sunny day at the appropriate aperture/shutter/ISO combination (e.g., ISO 200, 1/250 @ f 16 or 1/500 @ f 11 or 1/1000 @ f 8; ISO 400, 1/500 @ f 16, 1/1000 @ f 11, 1/2000 @ f 8). Your pictures will be dark. Increase exposure by one stop, and they'll be exposed spot-on as they would be on slide film (or, assuming accurate ISO values, on a FF dSLR). I discovered this (much to my annoyance) when shooting as I prefer to, the old fashioned way, using known exposure values on "manual" and/or with a hand held meter.