Upset with my new K20D

Marc Sabatella · 06-14-2009, 02:42 PM

Originally posted by pcfranchina

Thanks PBO. You understood what I meant.

Same Lens
Same F-Stop
Same ISO
Same Shutter Speed
In my opinion should produce the same or better result with the new lens.

You used two different lenses too? Or do you mean the new *camera*?

In any case, I think the source of confusion is that the picture from the K20D is indeed better to most of our eyes. Maybe your monitor is calibrated too dark, or maybe you just prefer unrealistically bright images when shooting darkish scenes. But as I said before, we're talking about a fraction of a stop difference. I fail to see what the big deal is. I'll bet if you lined 100 cameras, put in the exact same ISO, aperture, and shutter speed on all of them, you'd get 100 slightly different exposures. That's pretty normal, really.

If you're of the opinion than 1/15" is too slow a shutter speed, up the ISO as I said before. But I can definitely confirm that is right about what one should expect for the kind of lighting that seems to be involved. A camera that produced a good exposure at a significantly faster shutter speed at ISO 400 and f/2.8 would be defective - it would probably be lying *big time* about its actual ISO.

Last edited by Marc Sabatella; 06-15-2009 at 12:57 AM.

nostatic · 06-14-2009, 02:57 PM

I'm upset because my K20D makes terrible espresso.

dylansalt · 06-14-2009, 03:59 PM

To be quite honest - I find the Spot metering on my K10D generally to be inaccurate & inconsistent (sic?) especially compared to my Oly 5060.

I ran some tests of a red & black backpack in bright sun and when I spot metered off the black I got perfect histogram readings where really in theory I should have got overexposed results?????

I suggest you do the same test with CW & EVal metering and then compare your findings.

Dyl

pcfranchina · 06-14-2009, 04:45 PM

Typo.. It was the same lens. Sigma DG 24-60 2.8

I was a little more pleased with the results with my Pentax 50 1.4.

crispy0009 · 06-14-2009, 06:33 PM

Originally posted by nostatic

I'm upset because my K20D makes terrible espresso.

haha
nice.

defiant · 06-14-2009, 11:20 PM

Originally posted by pcfranchina

Ok here is an example. I really had to slow it down with the K20 to get what I wanted. Same settings on the K10 was blown out.

Obviously not a big deal for taking a photo of my messy kitchen but if people were in the room... I'd have blur.

Spot metering is probably not the best way to do an exposure comparison. Because spot metering meters a relatively small portion of the picture, you're basing the metering on only a small section, and from the photos, this section has extremely bright areas (the reflections) as well as darker areas (the wood), so the slightest angle difference can alter the metering significantly, and it also means the metering algorithms get depended upon much more which could also explain a difference in behaviour.

It appears that the K20D shot was taken at an angle slightly to the left of the angle of the K10D shot, with the very bright reflection being more in the centre of the spot metering area for the K20D. It looks like the reflection in the centre of the shot along with the wooden door in the middle of the shot were metered perfectly with the K20D, and thus you have a very nicely metered reflection and wooden frame that's not blown out. The rest of the photo is much darker compared to what is in the centre of the photo, but spot metering is meant to ignore the rest of the area for metering anyway.

The K10D could have metered more on the wooden frame instead of the reflection since the reflection near the centre is completely overexposed. Even then i wonder if the wooden frame in the centre is also overexposed, and thereby the spot metering of the K10D hasn't done its job properly in this shot? I haven't used a K10D and don't know how it compares, but is there any chance the K10D wasn't on spot metering for that shot? Because i agree with you that the kitchen is nicely exposed with the K10D photo, but the spot metering section itself seems to be overexposed.

little laker · 06-15-2009, 12:00 AM

pcfranchina,
At least with my main system (not a Pentax) just the white balance can alter the photo that much, making one look blown out and the next look under exposed.

So I am going to suggest that you use manual white balance, and try the test over again.

Your also almost better off using completely artificial lighting as well. If the sunlight passes through a light cloud you might not notice the difference, but the cameras sure will.

flyer · 06-15-2009, 02:11 AM

There should not be any (or hardly) difference in exposure between the two cameras. Having more pixel just means you can make larger prints, but it doesn't change the exposure between the cameras.

**Rondec** · 06-15-2009, 04:37 AM

I own both a K10 and a K20 and I have to say that I like the metering of the K20 a lot better than the K10. As to your initial question, I routinely shoot at ISO 800 to 1600 with my K20 inside, just to keep the shutter speed faster. Even though our eyes see plenty of light, the camera still needs to increase its sensitivity or decrease the shutter speed. The nice thing about the K20 is you are able to increase your ISO a lot more than with K10 without bad results.

slomojoe · 06-15-2009, 05:01 AM

I agree with defiant above that your subject does not lend itself to the test, because of the dark shadows and bright reflections that may dramatically change the metering with minor shifts in focus point. Basically, the picture you show does not allow to draw any conclusion (this does not mean your observation of consistent "underexposure" with the 20D is wrong, just that we can't tell from that picture).

Pick a uniformly lit subject in a constant light situation that you know gives you the observed probelm (e.g. low light) and repeat the test, both with spot and multi-segment metering. That will allow us to judge better what may be going on. Frankly, I wouldn't find it so surprising if there was some difference between the two cameras, given all the differences between them, starting from the sensor itself, though it shouldn't be more than one stop.

sentro · 06-15-2009, 08:52 AM

It does underexpose. I have a 100D super, a K10D and K20D. My K20D does underexpose by 1/2 stop. Ran many test shots and compared to my other cameras. All the talk about angles, and ISO setting and camera settings are kind'a pointless....personally I already checked ALL that already. Many die hard Pentax users are just trying to dismiss the possibility of something like this happening But It just Does happen.

Personally doesn't really bother me....That's why the compensation 'dial' is there. When I want to get picky I just grab my Sekonic...and THEN apply the compensation on the camera.

Sorry if I stepped on any toes.

PS:

My K100D overexposes by 1/3 and my K10 is right on with some lenses. (hooray to to the compensation dial again)

stevebrot · 06-15-2009, 09:02 AM

Ummm...I made a comment above that was confirmed by the OP...

In case you guys did not notice, the issue is not with the metering system. Both cameras were in M mode with the same shutter speed, aperture, and ISO. The lens was also the same for both shots.

The complaint is that the exposure itself, not the metering, is not comparable between the two bodies.

Steve

Marc Sabatella · 06-15-2009, 11:58 AM

Originally posted by sentro

It does underexpose. I have a 100D super, a K10D and K20D. My K20D does underexpose by 1/2 stop.

Compared to the other cameras, you mean? And do you mean that the meter *chooses* a lower exposure, or that if you use M mode to set the same exposure on all three cameras, the K20D comes out darker at the same exposure settings? That's the source of confusion here.

If you are talking about the camera meter choosing a darker exposure, then that may well be (I have no way of testing that idea), but would not seem to be relevant here, since the OP is talking about manually making the same exposure settings on both cameras.

If on the other hand you are confirming his observation that for the same exposure settings, the K20D comes out half a stop darker, then great - good information to know. My guess is that might be deliberate, having to do with differences in the dynamic range of the sensor or the response curves applied in the in-camera processing.

But anyhow, I agree with your bottom line - none of this should be a big deal. Exposure is not an exact science, small variations are absolutely normal, and easy enough to adjust for.

egordon99 · 06-15-2009, 12:34 PM

Originally posted by little laker

If the sunlight passes through a light cloud you might not notice the difference, but the cameras sure will.

+1

I'm willing to bet the light coming in through the window changed quite a bit for the two pictures.

Not sure why you're upset with the K20D when the sun seems to be the culprit (or some clouds

)

**Douglas_of_Sweden** · 06-15-2009, 12:58 PM

Originally posted by defiant

Spot metering is probably not the best way to do an exposure comparison. Because spot metering meters a relatively small portion of the picture, you're basing the metering on only a small section, and from the photos, this section has extremely bright areas (the reflections) as well as darker areas (the wood), so the slightest angle difference can alter the metering significantly, and it also means the metering algorithms get depended upon much more which could also explain a difference in behaviour.

It appears that the K20D shot was taken at an angle slightly to the left of the angle of the K10D shot, with the very bright reflection being more in the centre of the spot metering area for the K20D. It looks like the reflection in the centre of the shot along with the wooden door in the middle of the shot were metered perfectly with the K20D, and thus you have a very nicely metered reflection and wooden frame that's not blown out. The rest of the photo is much darker compared to what is in the centre of the photo, but spot metering is meant to ignore the rest of the area for metering anyway.

The K10D could have metered more on the wooden frame instead of the reflection since the reflection near the centre is completely overexposed. Even then i wonder if the wooden frame in the centre is also overexposed, and thereby the spot metering of the K10D hasn't done its job properly in this shot? I haven't used a K10D and don't know how it compares, but is there any chance the K10D wasn't on spot metering for that shot? Because i agree with you that the kitchen is nicely exposed with the K10D photo, but the spot metering section itself seems to be overexposed.

I'm with defiant on this. One can see that the two photos are not framed identically. There are differences on how much that are included of the kitchen on all edges. Hence the center is placed somewhat different, but within an area of very large differences between bright glass reflections and darker wood. This may be enough to explain the difference? Why on earth are you using spot meetering on such a motive? Spot metering is for situations when you know you want to expose according to a specific surface, e.g. a face. Meter on the face, lock exposure, recompose, shoot. But when you care about the average exposure (like you appear to do here), you want to exposure according to the whole picture or the whole center of it. There is simply no relevant answer to your questions because the question is wrong.

06-14-2009, 04:45 PM	#34
pcfranchina Forum Member Join Date: Nov 2007 Location: USA, NYC Posts: 57 Original Poster	Same lens Typo.. It was the same lens. Sigma DG 24-60 2.8 I was a little more pleased with the results with my Pentax 50 1.4.

06-15-2009, 08:52 AM	#41
sentro New Member Join Date: Mar 2009 Location: Chicago Posts: 2	K20D underexposing It does underexpose. I have a 100D super, a K10D and K20D. My K20D does underexpose by 1/2 stop. Ran many test shots and compared to my other cameras. All the talk about angles, and ISO setting and camera settings are kind'a pointless....personally I already checked ALL that already. Many die hard Pentax users are just trying to dismiss the possibility of something like this happening But It just Does happen. Personally doesn't really bother me....That's why the compensation 'dial' is there. When I want to get picky I just grab my Sekonic...and THEN apply the compensation on the camera. Sorry if I stepped on any toes. PS: My K100D overexposes by 1/3 and my K10 is right on with some lenses. (hooray to to the compensation dial again)

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Upset my K-7	Danny Delcambre	Pentax DSLR Discussion	10	11-18-2010 08:00 AM
In Canada: Summer Rebates for K20d or K20d plus lens	Pentaxtic	Pentax DSLR Discussion	1	07-17-2009 11:34 AM
Magic Lantern Guides: Pentax K20D and MasterWorks: Jumpstart Guide for the K20D.	Reportage	Troubleshooting and Beginner Help	10	02-12-2009 10:24 PM
Why "Scene Modes" Upset Some On The K200D?	shutterpuppy	Pentax DSLR Discussion	33	01-10-2009 05:28 PM

06-14-2009, 02:42 PM	#31
Marc Sabatella Veteran Member Join Date: Sep 2006 Location: Denver, CO Photos: Gallery Posts: 10,685	Originally posted by pcfranchina Thanks PBO. You understood what I meant. Same Lens Same F-Stop Same ISO Same Shutter Speed In my opinion should produce the same or better result with the new lens. You used two different lenses too? Or do you mean the new camera? In any case, I think the source of confusion is that the picture from the K20D is indeed better to most of our eyes. Maybe your monitor is calibrated too dark, or maybe you just prefer unrealistically bright images when shooting darkish scenes. But as I said before, we're talking about a fraction of a stop difference. I fail to see what the big deal is. I'll bet if you lined 100 cameras, put in the exact same ISO, aperture, and shutter speed on all of them, you'd get 100 slightly different exposures. That's pretty normal, really. If you're of the opinion than 1/15" is too slow a shutter speed, up the ISO as I said before. But I can definitely confirm that is right about what one should expect for the kind of lighting that seems to be involved. A camera that produced a good exposure at a significantly faster shutter speed at ISO 400 and f/2.8 would be defective - it would probably be lying big time about its actual ISO. Last edited by Marc Sabatella; 06-15-2009 at 12:57 AM.

06-14-2009, 02:57 PM	#32
nostatic Veteran Member Join Date: Feb 2008 Location: socal Photos: Gallery Posts: 1,575	I'm upset because my K20D makes terrible espresso.

06-14-2009, 03:59 PM	#33
dylansalt Veteran Member Join Date: Mar 2008 Location: Durban, South Africa Photos: Gallery Posts: 1,052	To be quite honest - I find the Spot metering on my K10D generally to be inaccurate & inconsistent (sic?) especially compared to my Oly 5060. I ran some tests of a red & black backpack in bright sun and when I spot metered off the black I got perfect histogram readings where really in theory I should have got overexposed results????? I suggest you do the same test with CW & EVal metering and then compare your findings. Dyl

06-14-2009, 06:33 PM	#35
crispy0009 Forum Member Join Date: Jan 2009 Location: Palm Beach Gardens, FLORIDA Posts: 62	Originally posted by nostatic I'm upset because my K20D makes terrible espresso. haha nice.

06-14-2009, 11:20 PM	#36
defiant New Member Join Date: Mar 2009 Location: Perth Australia Posts: 19	Originally posted by pcfranchina Ok here is an example. I really had to slow it down with the K20 to get what I wanted. Same settings on the K10 was blown out. Obviously not a big deal for taking a photo of my messy kitchen but if people were in the room... I'd have blur. Spot metering is probably not the best way to do an exposure comparison. Because spot metering meters a relatively small portion of the picture, you're basing the metering on only a small section, and from the photos, this section has extremely bright areas (the reflections) as well as darker areas (the wood), so the slightest angle difference can alter the metering significantly, and it also means the metering algorithms get depended upon much more which could also explain a difference in behaviour. It appears that the K20D shot was taken at an angle slightly to the left of the angle of the K10D shot, with the very bright reflection being more in the centre of the spot metering area for the K20D. It looks like the reflection in the centre of the shot along with the wooden door in the middle of the shot were metered perfectly with the K20D, and thus you have a very nicely metered reflection and wooden frame that's not blown out. The rest of the photo is much darker compared to what is in the centre of the photo, but spot metering is meant to ignore the rest of the area for metering anyway. The K10D could have metered more on the wooden frame instead of the reflection since the reflection near the centre is completely overexposed. Even then i wonder if the wooden frame in the centre is also overexposed, and thereby the spot metering of the K10D hasn't done its job properly in this shot? I haven't used a K10D and don't know how it compares, but is there any chance the K10D wasn't on spot metering for that shot? Because i agree with you that the kitchen is nicely exposed with the K10D photo, but the spot metering section itself seems to be overexposed.

06-15-2009, 12:00 AM	#37
little laker Veteran Member Join Date: Sep 2006 Photos: Gallery Posts: 5,697	pcfranchina, At least with my main system (not a Pentax) just the white balance can alter the photo that much, making one look blown out and the next look under exposed. So I am going to suggest that you use manual white balance, and try the test over again. Your also almost better off using completely artificial lighting as well. If the sunlight passes through a light cloud you might not notice the difference, but the cameras sure will.

06-15-2009, 02:11 AM	#38
flyer Veteran Member Join Date: Jan 2008 Location: Ste-Anne des Plaines, Qc., Canada Posts: 2,013	There should not be any (or hardly) difference in exposure between the two cameras. Having more pixel just means you can make larger prints, but it doesn't change the exposure between the cameras.

06-15-2009, 04:37 AM	#39
Rondec Loyal Site Supporter Join Date: Mar 2009 Location: Gladys, Virginia Photos: Gallery Posts: 27,650	I own both a K10 and a K20 and I have to say that I like the metering of the K20 a lot better than the K10. As to your initial question, I routinely shoot at ISO 800 to 1600 with my K20 inside, just to keep the shutter speed faster. Even though our eyes see plenty of light, the camera still needs to increase its sensitivity or decrease the shutter speed. The nice thing about the K20 is you are able to increase your ISO a lot more than with K10 without bad results.

06-15-2009, 05:01 AM	#40
slomojoe Veteran Member Join Date: Mar 2007 Location: Upstate NY Posts: 788	I agree with defiant above that your subject does not lend itself to the test, because of the dark shadows and bright reflections that may dramatically change the metering with minor shifts in focus point. Basically, the picture you show does not allow to draw any conclusion (this does not mean your observation of consistent "underexposure" with the 20D is wrong, just that we can't tell from that picture). Pick a uniformly lit subject in a constant light situation that you know gives you the observed probelm (e.g. low light) and repeat the test, both with spot and multi-segment metering. That will allow us to judge better what may be going on. Frankly, I wouldn't find it so surprising if there was some difference between the two cameras, given all the differences between them, starting from the sensor itself, though it shouldn't be more than one stop.

06-15-2009, 09:02 AM	#42
stevebrot Otis Memorial Pentaxian Join Date: Mar 2007 Location: Vancouver (USA) Photos: Gallery \| Albums Posts: 42,007	Ummm...I made a comment above that was confirmed by the OP... In case you guys did not notice, the issue is not with the metering system. Both cameras were in M mode with the same shutter speed, aperture, and ISO. The lens was also the same for both shots. The complaint is that the exposure itself, not the metering, is not comparable between the two bodies. Steve

Thread Tools	Search this Thread
	Search this Thread: Advanced Search

06-15-2009, 11:58 AM	#43
Marc Sabatella Veteran Member Join Date: Sep 2006 Location: Denver, CO Photos: Gallery Posts: 10,685	Originally posted by sentro It does underexpose. I have a 100D super, a K10D and K20D. My K20D does underexpose by 1/2 stop. Compared to the other cameras, you mean? And do you mean that the meter chooses a lower exposure, or that if you use M mode to set the same exposure on all three cameras, the K20D comes out darker at the same exposure settings? That's the source of confusion here. If you are talking about the camera meter choosing a darker exposure, then that may well be (I have no way of testing that idea), but would not seem to be relevant here, since the OP is talking about manually making the same exposure settings on both cameras. If on the other hand you are confirming his observation that for the same exposure settings, the K20D comes out half a stop darker, then great - good information to know. My guess is that might be deliberate, having to do with differences in the dynamic range of the sensor or the response curves applied in the in-camera processing. But anyhow, I agree with your bottom line - none of this should be a big deal. Exposure is not an exact science, small variations are absolutely normal, and easy enough to adjust for.

06-15-2009, 12:34 PM	#44
egordon99 Inactive Account Join Date: Dec 2006 Location: Outside of Philly Posts: 1,561	Originally posted by little laker If the sunlight passes through a light cloud you might not notice the difference, but the cameras sure will. +1 I'm willing to bet the light coming in through the window changed quite a bit for the two pictures. Not sure why you're upset with the K20D when the sun seems to be the culprit (or some clouds )

06-15-2009, 12:58 PM	#45
Douglas_of_Sweden Site Supporter Join Date: Apr 2008 Location: Stockholm Photos: Gallery Posts: 3,374	Originally posted by defiant Spot metering is probably not the best way to do an exposure comparison. Because spot metering meters a relatively small portion of the picture, you're basing the metering on only a small section, and from the photos, this section has extremely bright areas (the reflections) as well as darker areas (the wood), so the slightest angle difference can alter the metering significantly, and it also means the metering algorithms get depended upon much more which could also explain a difference in behaviour. It appears that the K20D shot was taken at an angle slightly to the left of the angle of the K10D shot, with the very bright reflection being more in the centre of the spot metering area for the K20D. It looks like the reflection in the centre of the shot along with the wooden door in the middle of the shot were metered perfectly with the K20D, and thus you have a very nicely metered reflection and wooden frame that's not blown out. The rest of the photo is much darker compared to what is in the centre of the photo, but spot metering is meant to ignore the rest of the area for metering anyway. The K10D could have metered more on the wooden frame instead of the reflection since the reflection near the centre is completely overexposed. Even then i wonder if the wooden frame in the centre is also overexposed, and thereby the spot metering of the K10D hasn't done its job properly in this shot? I haven't used a K10D and don't know how it compares, but is there any chance the K10D wasn't on spot metering for that shot? Because i agree with you that the kitchen is nicely exposed with the K10D photo, but the spot metering section itself seems to be overexposed. I'm with defiant on this. One can see that the two photos are not framed identically. There are differences on how much that are included of the kitchen on all edges. Hence the center is placed somewhat different, but within an area of very large differences between bright glass reflections and darker wood. This may be enough to explain the difference? Why on earth are you using spot meetering on such a motive? Spot metering is for situations when you know you want to expose according to a specific surface, e.g. a face. Meter on the face, lock exposure, recompose, shoot. But when you care about the average exposure (like you appear to do here), you want to exposure according to the whole picture or the whole center of it. There is simply no relevant answer to your questions because the question is wrong.