K20D metering with older lenses (again)

Wayno · 06-21-2009, 06:26 PM

I know that there are several old threads about this topic, but I have a question for which I couldn't find an answer.

The K20D (actually GX20) I'm using seems to meter accurately in Av mode for M42 lenses using manual stop down. In M mode, metering for older K mount lenses is inaccurate (using the green button for stop down) and I understand that a focussing screen swap (to the LL-60) should largely solve this issue. My question is, will the screen swap affect the metering of M42 lenses in aperture priority? I will probably be using more M42 lenses than anything else, so I don't want to mess that up if it's working fine at the moment. I would, however, also like to use old K mount lenses from time to time as well. If the LL-60 screen allows accurate metering with both then I will give it a try.

By the way, I don't really understand why metering should be accurate when manually stopping down an M42, but not accurate with the K mount lenses using the green button. Perhaps the lens isn't stopped down long enough to get an accurate reading? Thanks in advance for any advice.

stevebrot · 06-21-2009, 09:28 PM - **1 Like**

Originally posted by Wayno

...By the way, I don't really understand why metering should be accurate when manually stopping down an M42, but not accurate with the K mount lenses using the green button. Perhaps the lens isn't stopped down long enough to get an accurate reading? Thanks in advance for any advice.

There is a general problem with meter accuracy for all non-A lenses on the K10D and K20D and to a lesser extent on the KxxxD (mirror prism) bodies. Strangely, the meter behavior is somewhat lens-specific, is aperture dependent, and is worse (by my observation) for lenses with a maximum aperture wider than about f/4 regardless of mount. My Jupiter-9 85/2 is by far the worst offender with about two stops underexposure wide open.

The general pattern is:

Underexposure wide open
Essentially correct exposure at f/3.5-f/8
Mild overexposure at f/11 and smaller assuming adequate light for shooting at those apertures

A short note regarding stop-down metering at narrower apertures in moderate or low light. The published sensitivity for the meter on the K10D is 1 EV (ISO 100). My experience has been that sensitivity is somewhat less with linearity falling off at about 2 EV (ISO 100). Now you might think that 1 or 2 EV is pretty dark and it is when metering wide open. However, when metering stopped down in moderately low light (say f/2 @ 1/30s, ISO 100), at f/11 using stop-down metering, the amount of light actually striking the sensor is only EV 1 (ISO 100) and is at the limit of sensitivity for the meter.

The result is that the meter gives the same reading for all apertures below a certain point at a given light intensity resulting in severe underexposure. This loss of meter linearity in low light and narrow apertures is true for all stop-down metering cameras:

Many M42 film cameras
All K-mount cameras (film and digital) when mounted to M42 lenses
All crippled KA mount cameras (film and digital) when mounted to non-A lenses

Steve

(Much more than a simple answer I know...I guess I am in a verbose mood...)

Wayno · 06-21-2009, 10:21 PM

Thanks for the detailed response.

What puzzles me is that metering seems accurate for M42 lenses (compared with the kit lens) in Av mode, and the reading drops in a linear fashion as I stop down the lens. However, in M mode with an old K lens, it's obviously off.

I've only done limited testing, so I'll have to do some more to confirm those observations.

stevebrot · 06-21-2009, 10:37 PM

Originally posted by Wayno

Thanks for the detailed response.

What puzzles me is that metering seems accurate for M42 lenses (compared with the kit lens) in Av mode, and the reading drops in a linear fashion as I stop down the lens. However, in M mode with an old K lens, it's obviously off.

I've only done limited testing, so I'll have to do some more to confirm those observations.

Depends on the lens. My Tamron 70-150/3.5 has essentially accurate metering with the PK adapter while my K-mount Zenitar 16/2.8 is inaccurate at all apertures. There is also some variability depending on focus. My Pentax-M 50/1.7 meters 1/2 stop different between close-focus and infinity for the same blank-wall target. Go figure... I have done some detailed testing with multiple lenses using both the stock and KatzEye screens, but have not taken the time to finish collating the data.

My general approach has been to determine the appropriate exposure for a given scene using a combination of meter reading and "chimping" the histogram to determine how much compensation to dial in. It is highly tempting to invest in a hand-held meter and just be done with it all

.

Steve

(Not happy with the stopped down metering situation...but knows that the Nikon and Canon camps have the same issues...)

Last edited by stevebrot; 06-21-2009 at 10:42 PM.

**Lowell Goudge** · 06-22-2009, 06:57 AM

check trhis link

https://www.pentaxforums.com/forums/241716-post69.html

**Canada_Rockies** · 06-22-2009, 06:33 PM

Originally posted by Lowell Goudge

check trhis link

https://www.pentaxforums.com/forums/241716-post69.html

Matches my not so scientific results. Well done, LG.

Wayno · 06-22-2009, 11:23 PM

Thanks for the replies.

Well I guess I spoke too soon. After some further testing the screwmount metering in Av mode isn't as accurate as I thought... It looks like I'll be trying to get an LL-60 screen some time soon. Mind you, coming from a Spotmatic it's really quite easy with the digi to check the exposure after the shot & correct if necessary. However, it'd be nice to not have to worry about that, so I'll try swapping out the screen.

**Canada_Rockies** · 06-23-2009, 06:41 PM

It probably does the same thing as it does with K and M lenses - overexposed at one end of the aperture range, underexposed at the other. PITA - and I am not referring to bread. There are a stop or three in the middle where the metering is close enough, and once you find them, you can work around it for a while. "Let me see, it meters ok at f/8 so it metered 1/500 at f/8, but I want f/16 for depth of field, so that's two stops less light, I need 1/125 second exposure." Absolutely brilliantly accurate. But it is still PITA.

stevebrot · 06-23-2009, 08:57 PM

Originally posted by Canada_Rockies

It probably does the same thing as it does with K and M lenses - overexposed at one end of the aperture range, underexposed at the other. PITA - and I am not referring to bread. There are a stop or three in the middle where the metering is close enough, and once you find them, you can work around it for a while. "Let me see, it meters ok at f/8 so it metered 1/500 at f/8, but I want f/16 for depth of field, so that's two stops less light, I need 1/125 second exposure." Absolutely brilliantly accurate. But it is still PITA.

You discovered my approach. Meter at f/5.6 - f/8 and calculate the shutter speed for the aperture you really want.

Steve

(Either that or just shamelessly chimp, adjust, and reshoot...)

06-21-2009, 06:26 PM	#1
Wayno Inactive Account Join Date: Jun 2009 Location: Australia Posts: 5	K20D metering with older lenses (again) I know that there are several old threads about this topic, but I have a question for which I couldn't find an answer. The K20D (actually GX20) I'm using seems to meter accurately in Av mode for M42 lenses using manual stop down. In M mode, metering for older K mount lenses is inaccurate (using the green button for stop down) and I understand that a focussing screen swap (to the LL-60) should largely solve this issue. My question is, will the screen swap affect the metering of M42 lenses in aperture priority? I will probably be using more M42 lenses than anything else, so I don't want to mess that up if it's working fine at the moment. I would, however, also like to use old K mount lenses from time to time as well. If the LL-60 screen allows accurate metering with both then I will give it a try. By the way, I don't really understand why metering should be accurate when manually stopping down an M42, but not accurate with the K mount lenses using the green button. Perhaps the lens isn't stopped down long enough to get an accurate reading? Thanks in advance for any advice.

06-21-2009, 09:28 PM - 1 Like	#2
stevebrot Otis Memorial Pentaxian Join Date: Mar 2007 Location: Vancouver (USA) Photos: Gallery \| Albums Posts: 42,007	Originally posted by Wayno ...By the way, I don't really understand why metering should be accurate when manually stopping down an M42, but not accurate with the K mount lenses using the green button. Perhaps the lens isn't stopped down long enough to get an accurate reading? Thanks in advance for any advice. There is a general problem with meter accuracy for all non-A lenses on the K10D and K20D and to a lesser extent on the KxxxD (mirror prism) bodies. Strangely, the meter behavior is somewhat lens-specific, is aperture dependent, and is worse (by my observation) for lenses with a maximum aperture wider than about f/4 regardless of mount. My Jupiter-9 85/2 is by far the worst offender with about two stops underexposure wide open. The general pattern is: Underexposure wide open Essentially correct exposure at f/3.5-f/8 Mild overexposure at f/11 and smaller assuming adequate light for shooting at those apertures A short note regarding stop-down metering at narrower apertures in moderate or low light. The published sensitivity for the meter on the K10D is 1 EV (ISO 100). My experience has been that sensitivity is somewhat less with linearity falling off at about 2 EV (ISO 100). Now you might think that 1 or 2 EV is pretty dark and it is when metering wide open. However, when metering stopped down in moderately low light (say f/2 @ 1/30s, ISO 100), at f/11 using stop-down metering, the amount of light actually striking the sensor is only EV 1 (ISO 100) and is at the limit of sensitivity for the meter. The result is that the meter gives the same reading for all apertures below a certain point at a given light intensity resulting in severe underexposure. This loss of meter linearity in low light and narrow apertures is true for all stop-down metering cameras: Many M42 film cameras All K-mount cameras (film and digital) when mounted to M42 lenses All crippled KA mount cameras (film and digital) when mounted to non-A lenses Steve (Much more than a simple answer I know...I guess I am in a verbose mood...) Last edited by stevebrot; 06-21-2009 at 09:53 PM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
GEneral question about K20D and older lenses	jeryst	Pentax SLR Lens Discussion	6	11-25-2009 11:38 PM
Older lenses for K20D?	poyntnshuteguy	Troubleshooting and Beginner Help	5	02-23-2009 09:52 PM
Metering Question On Older Lenses	metroeloise	Pentax SLR Lens Discussion	3	02-21-2009 08:34 PM
K20D and older flashes	ljlayne	Flashes, Lighting, and Studio	9	07-16-2008 12:00 AM
Stop Down Metering with K20D and Tamron SP Lenses	petez	Pentax DSLR Discussion	39	04-05-2008 08:53 PM

06-21-2009, 10:21 PM	#3
Wayno Inactive Account Join Date: Jun 2009 Location: Australia Posts: 5 Original Poster	Thanks for the detailed response. What puzzles me is that metering seems accurate for M42 lenses (compared with the kit lens) in Av mode, and the reading drops in a linear fashion as I stop down the lens. However, in M mode with an old K lens, it's obviously off. I've only done limited testing, so I'll have to do some more to confirm those observations.

06-21-2009, 10:37 PM	#4
stevebrot Otis Memorial Pentaxian Join Date: Mar 2007 Location: Vancouver (USA) Photos: Gallery \| Albums Posts: 42,007	Originally posted by Wayno Thanks for the detailed response. What puzzles me is that metering seems accurate for M42 lenses (compared with the kit lens) in Av mode, and the reading drops in a linear fashion as I stop down the lens. However, in M mode with an old K lens, it's obviously off. I've only done limited testing, so I'll have to do some more to confirm those observations. Depends on the lens. My Tamron 70-150/3.5 has essentially accurate metering with the PK adapter while my K-mount Zenitar 16/2.8 is inaccurate at all apertures. There is also some variability depending on focus. My Pentax-M 50/1.7 meters 1/2 stop different between close-focus and infinity for the same blank-wall target. Go figure... I have done some detailed testing with multiple lenses using both the stock and KatzEye screens, but have not taken the time to finish collating the data. My general approach has been to determine the appropriate exposure for a given scene using a combination of meter reading and "chimping" the histogram to determine how much compensation to dial in. It is highly tempting to invest in a hand-held meter and just be done with it all . Steve (Not happy with the stopped down metering situation...but knows that the Nikon and Canon camps have the same issues...) Last edited by stevebrot; 06-21-2009 at 10:42 PM.

06-22-2009, 06:57 AM	#5
Lowell Goudge Site Supporter Join Date: Jan 2007 Location: Toronto Photos: Gallery \| Albums Posts: 17,886	check trhis link https://www.pentaxforums.com/forums/241716-post69.html

06-22-2009, 06:33 PM	#6
Canada_Rockies Loyal Site Supporter Join Date: Nov 2007 Location: Sparwood, BC, Canada Photos: Gallery \| Albums Posts: 12,385	Originally posted by Lowell Goudge check trhis link https://www.pentaxforums.com/forums/241716-post69.html Matches my not so scientific results. Well done, LG.

06-22-2009, 11:23 PM	#7
Wayno Inactive Account Join Date: Jun 2009 Location: Australia Posts: 5 Original Poster	Thanks for the replies. Well I guess I spoke too soon. After some further testing the screwmount metering in Av mode isn't as accurate as I thought... It looks like I'll be trying to get an LL-60 screen some time soon. Mind you, coming from a Spotmatic it's really quite easy with the digi to check the exposure after the shot & correct if necessary. However, it'd be nice to not have to worry about that, so I'll try swapping out the screen.

06-23-2009, 06:41 PM	#8
Canada_Rockies Loyal Site Supporter Join Date: Nov 2007 Location: Sparwood, BC, Canada Photos: Gallery \| Albums Posts: 12,385	It probably does the same thing as it does with K and M lenses - overexposed at one end of the aperture range, underexposed at the other. PITA - and I am not referring to bread. There are a stop or three in the middle where the metering is close enough, and once you find them, you can work around it for a while. "Let me see, it meters ok at f/8 so it metered 1/500 at f/8, but I want f/16 for depth of field, so that's two stops less light, I need 1/125 second exposure." Absolutely brilliantly accurate. But it is still PITA.

06-23-2009, 08:57 PM	#9
stevebrot Otis Memorial Pentaxian Join Date: Mar 2007 Location: Vancouver (USA) Photos: Gallery \| Albums Posts: 42,007	Originally posted by Canada_Rockies It probably does the same thing as it does with K and M lenses - overexposed at one end of the aperture range, underexposed at the other. PITA - and I am not referring to bread. There are a stop or three in the middle where the metering is close enough, and once you find them, you can work around it for a while. "Let me see, it meters ok at f/8 so it metered 1/500 at f/8, but I want f/16 for depth of field, so that's two stops less light, I need 1/125 second exposure." Absolutely brilliantly accurate. But it is still PITA. You discovered my approach. Meter at f/5.6 - f/8 and calculate the shutter speed for the aperture you really want. Steve (Either that or just shamelessly chimp, adjust, and reshoot...)