[jeremy_c]

I have the standard kit zoom 18-55mm 3.5-5.6 AL II pentax lens and also the Pentax FA 50mm f/1.4 prime. It seems that to get the same exposure as the prime on the zoom, I have to over expose the zoom by +1.0ev. The exposure looks good. If I leave it at +/- 0.0ev, then the zoom always looks under exposed.

Is this normal for the K20D and the kit zoom?

Jeremy

[Marc Sabatella]

Is this normal for the K20D and the kit zoom?
Jeremy

A whole EV? No. Maybe a fraction, especially when the zoom is wide open - most lenses overstate their maximum aperture. Could you post some samples?

[jeremy_c]

Here is an example:

Prime: http://jeremy.cowgar.com/photos/JNC10211.JPG
Prime: http://jeremy.cowgar.com/photos/JNC10211.PEF

Zoom: http://jeremy.cowgar.com/photos/JNC10212.JPG
Zoom: http://jeremy.cowgar.com/photos/JNC10212.PEF

Jeremy

[jeremy_c]

I'd also be interested in what you guys think about the differences between those photos. They were taken back to back, only a lens change was made (and the +1ev). I would have expected a little more from the prime vs the kit zoom or maybe I am missing something? I am new to DSLR.

Thanks!

Jeremy

[RiceHigh]

I have the standard kit zoom 18-55mm 3.5-5.6 AL II pentax lens and also the Pentax FA 50mm f/1.4 prime. It seems that to get the same exposure as the prime on the zoom, I have to over expose the zoom by +1.0ev. The exposure looks good. If I leave it at +/- 0.0ev, then the zoom always looks under exposed.

Is this normal for the K20D and the kit zoom?

Jeremy

Many Pentax digital lenses tend to underexpose by -0.5 to -1EV and particularly digital zooms like to be darker. The FA 50 is just a lens which tends to be just brighter (but not overexpose IMO) and thus you see the difference.

For more info about the relative brightness and darkness of Pentax lenses, you can have a look on my this summary:-

RiceHigh's (Pentax) DSLR and Lens Measurbation Page on Exposure Accuracy and More..

And, the K20D is not really famous for exposure accuracy and consistency, so just the body could introduce more errors than it should be, see my following simple and factual analysis:-

RiceHigh's Pentax Blog: K20D Exposure Inaccuracy and Inconsistency

But since you have already got your new K20D, I think you have to live with it. The only way to compensate is to apply a different suitable +/-EV compensation whenever you change a different lens.

[DAZ]

Looks like the deterrents between T-stop and F-stop. Put simply an F-stop is a mathematical relationship between focal length and aperture. Each “full” stop is half or twice as much light but is not telling how much light is transmitted though the lens. How much light is transmitted and not absorbed is a T-stop. A lens with less glass, better glass, less transitions and better coatings (a fast 50) will usually transmitted more light then a lens with more glass, lower quality glass, more transitions and not as high-quality coatings (slow zoom). There is also a mechanical part, as the camera has to move the aperture blades at the time the photo is taken. This means there can be a small change even on the same lens from photo to photo. Most of the time as we use TTL metering it is not important. For some who do exaction studio work and stop motion it can be critical. These people will use mechanical F-stop lenses and possible determine the T-stop of their lenses. See F-number - Wikipedia, the free encyclopedia for some more info.

DAZ

[dlacouture]

Well, I think this is exactly the same problem as with the k10 focus screens...

As the k20 focus screen must be the same, you'll end up with a severe underexposure for lenses above f/3.5, and an overexposure with lenses below f/4.5 (approx.).

As there is nearly no lens with a wide-open aperture below f/4.5, you won't have cases of severe overexposure in everyday use...

But going from the kit lens to the 50 f/1.4, you'll see the difference...

Using a LL-60 focus screen solves this problem...

[jeremy_c]

Using a LL-60 focus screen solves this problem...

Just so I learn, how does the focus screen affect the exposure? I thought the focus screen was simply a preference of how you want your eye piece to appear?

Jeremy

[dlacouture]

Well, the light meter is located on the pentaprism, and thus meters the image formed on the focus screen... So, changing the focus screen have (fortunately in the Kx0 case!) an impact on the metering process...
And before you ask, the AF module is located beneath the mirror, and so is not impacted whatsoever by a focus screen change.

[Marc Sabatella]

I think what happened here is tha the shot from kit lens shows more of the (light colored) background, causing the camera to want to choose a lower exposure. I note that the actual exposure was identical - ISO 200, f/16, 6". So the two lenses *do* let in the same amount of the light - the camera just chose a lower exposure for the kit lens in this particular case, and I suspect it was because it was looking at a slightly different scene. I'd still be surprised if there were a *consistent* 1 EV difference. But I'd also observe that settings that require a 6" shutter speed are maybe not the best test - you're kind of pushing the limits there. More to the point would be exposures in "normal" shutter speed ranges.

As for what differences you expected to see, I'm not sure what you mean. Are you saying you expected a bigger difference in overall image quality? Not at f/16 - both lenses are past their prime (so to speak) at that aperture. Shoot both at f/5.6 and the FA50 will eat the kit lens's lunch.

[Duplo]

Looks like the deterrents between T-stop and F-stop. Put simply an F-stop is a mathematical relationship between focal length and aperture. Each “full” stop is half or twice as much light but is not telling how much light is transmitted though the lens. How much light is transmitted and not absorbed is a T-stop. A lens with less glass, better glass, less transitions and better coatings (a fast 50) will usually transmitted more light then a lens with more glass, lower quality glass, more transitions and not as high-quality coatings (slow zoom). There is also a mechanical part, as the camera has to move the aperture blades at the time the photo is taken. This means there can be a small change even on the same lens from photo to photo. Most of the time as we use TTL metering it is not important. For some who do exaction studio work and stop motion it can be critical. These people will use mechanical F-stop lenses and possible determine the T-stop of their lenses. See F-number - Wikipedia, the free encyclopedia for some more info.

DAZ

I think you are spot on here.
As I always shoots in manual mode, I see these differences between lenses all the time.
In general my primes has around half a T-stop difference to my fast zooms.

[DAZ]

Like a lot of things Duplo just because we can measure something doesn’t mean that it is significant. Cameras and lenses are built to such high standards we sometimes think they are scientific instruments. They are not. They are machines that are built better then they needs to be most of the time. I to can see it on my lenses but just don’t need to worry about it. A long time ago lens revues use to measure this but with TTL metering they dropped it as not important.

Measuring some things are pretty much useless. Take multi point metering and this is not to say don’t use it as it is a useful tool. With multi point it is basically not possible to determine what the camera is metering. Move the camera just a little and the metering point fall on different things. Even if they don’t move the data is put into an algorithm that some engineer come up with to try and make a good exposure. Change engineers and you change exposures. The only thing you can say is that it deferent then some photographer thinks it should be exposured. That is why cameras have exposure compensation controls so the photographer can make it what they want. Most of the time it is just an excuse to bash one camera brand over another.

DAZ

[Duplo]

Like a lot of things Duplo just because we can measure something doesn’t mean that it is significant. Cameras and lenses are built to such high standards we sometimes think they are scientific instruments. They are not. They are machines that are built better then they needs to be most of the time. I to can see it on my lenses but just don’t need to worry about it. A long time ago lens revues use to measure this but with TTL metering they dropped it as not important.

Well in my case it actually is important knowledge... say when i switch from a 70-200 to a fast 50 during an event, I will need to make a correction to my exposure settings (as I prefer shooting manual mode), yes the metering info in the viewfinder will tell me the same thing, but knowing that I can drop the ISO half a stop when shifting to a fast prime and keep my other exposure settings unchanged actually is important at times.

I have not made a solid measurement of it, it is just something I have noticed while working and made a mental note of...

Measuring some things are pretty much useless. Take multi point metering and this is not to say don’t use it as it is a useful tool. With multi point it is basically not possible to determine what the camera is metering. Move the camera just a little and the metering point fall on different things. Even if they don’t move the data is put into an algorithm that some engineer come up with to try and make a good exposure. Change engineers and you change exposures. The only thing you can say is that it deferent then some photographer thinks it should be exposured. That is why cameras have exposure compensation controls so the photographer can make it what they want. Most of the time it is just an excuse to bash one camera brand over another.

DAZ

I am not sure I can follow all of what you write here... stuff like using to bash camera brands etc... why would anyone do that... I do not think I did that in my post above at least...
Not sure I got the link to multisegment metering either...

I guess it all depends on how you work, to me kknowing what I talked about in my first post is important for the way I work when shooting events.

I see knowing as much as possible about the gear one uses as important, it allows you to react to certain cituations by reflex and allows me to dedicate my mind to my surroundings.

[DAZ]

Sorry Duplo, the part about the metering was just a general rant not pointed at you. A common theme is “I took a bad photo because the camera metering is bad”. You are obviously not one of those. Knowing the limits of your equipment is always a good thing and limits are not always equipment fault. Knowing things like your camera metering is not as liner in dim orange light can be important to shooting nightclubs but of little use to someone who shoots car races. To some that do catalog shooting a lot they may take the time to determine the T-stop of there lenses so when they change a lens they know how much to change the shutter speed. For them anything that helps get more photos out help. For some they may never notice any of this. To you these thing are important.

Again sorry Duplo if it look like I was pointing something your way.

DAZ