K10D light reading location?

**ebooks4pentax** · 06-29-2007, 11:35 PM

I can't remember which thread(s) were talking about how replacing the original focusing screen on K10D was giving false metering, (over or under exposure). That would indicate that the metering is taking place passed the focusing screen. In the 35mm camera days, I believe the metering was done at the mirror, before the screen. I don't know if anyone has the technical "know how" but here is my question. Where is the light measured on the K10D? Is it at the mirror/lens side of the screen or at viewfinder side of the screen? I can't believe that one would design a metering system with anything between the lens and the meter., and if so, why?

and · 06-29-2007, 11:53 PM

Deffinetley behind the screen. I have a split prism screen myself and with slow lenses it will totally mess up the metering! using a lens with max aperature of f5.6 needs +3 ev to come out ok, even then its a bit underexposed. Probably due to the prism blacking out and making the scene appear dark. My split prism screen also has darker, more grainer appearence in general and so the viewfinder is more dim.

With f2.8 or faster lenses the meter is not affected in any way though, not even spot.

pschlute · 06-30-2007, 02:08 AM

I was under the impression that the meter was behind the mirror. That the mirror actually allows some light to pass through. But i might have been dreaming.

blende8 · 06-30-2007, 03:11 AM

Behind the mirror is the AF-system.
The metering system is slightly above the view finder, so behind the screen. This was always so, also in film days.

pschlute · 06-30-2007, 04:26 AM

Thanks for clarifying that. This explains why the focus screen will have an effect on exposure.

matix · 06-30-2007, 05:30 PM

Originally posted by pschlute

Thanks for clarifying that. This explains why the focus screen will have an effect on exposure.

I received my split prism screen this week, easy to install but not so easy to keep all the dust specks off!!, will remove and clean it again.

It works a treat on my K10D, I thought I would miss the lines, but guess I ignored them anyway, it is a lot brighter than the stock K10D screen, so that makes me happy.

Does not seem to have had any affect on auto focus accuracy, but I have noted and this may be a coincidence, that it seems to focus quicker in lower light than before?? I would now say that the unassisted focus is as good as any DSLR I have tested.

Has anyone else noted this change in low light focusing, to the better?

Really happy with it overall, the split prism is at approx 45deg which surprised me, but that makes sense as it allows many angles in the frame to be used, ie vertical, horizontal and most in between, as focal points to align the split images.

I got mine from; jinfinance for USD $24.95 plus $4.95 postage, good communications and fast shipping.

Phil

jeffkrol · 06-30-2007, 06:20 PM

Originally posted by matix

Does not seem to have had any affect on auto focus accuracy, but I have noted and this may be a coincidence, that it seems to focus quicker in lower light than before?? I would now say that the unassisted focus is as good as any DSLR I have tested.

Has anyone else noted this change in low light focusing, to the better?

Phil

Shouldn't make any difference. The AF cell is in the bottom of the light box and is fed light through the semi silvered part of the instant return mirror. Behind this mirror is another front surface mirror that directs light down into the AF cells. I believe the AF chamber has at least one other mirror in it as well .
Light meter/p-ttl meter is behind a face of the porro-prism.

**ebooks4pentax** · 06-30-2007, 08:03 PM

Originally posted by jeffkrol

Light meter/p-ttl meter is behind a face of the porro-prism.

If so, that would mean that the focusing screen has a direct effect on the light being metered. If a certain focusing screen is brighter, the TTL metering system will be fooled in thinking that there is more light than there really is. The mirror is equidistant to the CCD surface as it is to the top face of the focusing screen. The screen then acts like a filter to the light. Denser or darker screen will change the exposure? I always thought that the light entering the camera was metered either on the mirror or somewhere before it goes through the focusing screen.

In other word, different focusing screens will make the camera over or under expose! Is that correct?

Jonas B · 06-30-2007, 08:29 PM

Originally posted by k10dbook

-snip-
In other word, different focusing screens will make the camera over or under expose! Is that correct?

Yes, that's correct.
Then it is not only about having a "light" or "dark" screen. The viewfinder screens are etched with the pattern on them in different angles.

Above is the, now many times posted, chart I get when comparing lenses and screens. The DS screen is a bit darker but the plot indicates that the meter reads more light...

This way Canon, as a current example, can offer the Ee-S screen for us people fond of manually focusing. The screen is good for manual focusing as the etched pattern is different resulting in a more shallow DOF than the standard screen. It also is a bit darker. Canon offers a firmware setting shifting the meters baseline a bit compensating for this difference.

As you understand I wish Pentax made something similar.

In the case of the errors we see with the stopped down metering with the K10D it is even more complicated. We don't get many elaborated reports but as I understand it some have different results with some tele lenses. I wonder if there is another factor in play as well with the original K10D screen. It may be that light from certain angles is read by the camera. The D/DS screens handles the situation better and give correct or nearly correct metering also with lenses having shorter focal lengths.

2 cents,

**jfdavis58** · 06-30-2007, 08:48 PM

Just curious Jonas; been read the older threads with exposure information. How did you get these data points: what was the experimental set-up and procedure, what sort of data reduction/manipulation to put this graph together? How many times did you run the experiment?

You could just point me at the original thread if that info has been shared. Thanks!

**barondla** · 07-01-2007, 07:06 AM

Not all cameras have meter cell in the prism. My Olympus Om4t has a simi transparent mirror with a sub mirror behind it to direct the light to the bottom of mirror box metering cell. It is far superior to what we have now. It is very accurate - beats my hand held Pentax analog spot meter, reads much lower light (around 2 min +), and not bothered by light entering viewfinder. Hope we get something similar some day.
thanks
barondla

Jonas B · 07-01-2007, 05:35 PM

Originally posted by jfdavis58

Just curious Jonas; been read the older threads with exposure information. How did you get these data points: what was the experimental set-up and procedure, what sort of data reduction/manipulation to put this graph together? How many times did you run the experiment?

You could just point me at the original thread if that info has been shared. Thanks!

Hi jfdavis58,

Hmm... I don't know... last time you asked me about my color management setup and I really described it you never replied and turned me down...

But OK then, here we go:

The chart has many sources, most of them based on experience and reading good books on exposure, digital picture technology, raw converters and zone systems but I also take a few Internet opinions into account.

The first thing to mention is that the baselines for EV0, EV1 and so on in both directions are taken from my own experience and preferences. The EV0 line represents an RGB-value of 97 in the chart. Some prefer 110, some even higher figures. I am slowly moving towards 110. Knowing this you may move the lines upwards or downwards to your own taste. The values for the other lines are the results from my own experiments using a lot of cameras to see what happens when you take the double or the half exposure time with the same target.

Then we have the curves: In this case they come from one single camera only (my first K10D). They are the average result from three exposures with each lens and screen combo without changing target.

The target was a Kodak grey card (but could have been anything neutral) more or less filling the viewfinder. The background behind the card was a grey fabric and the room has grey tapestries. The light source were a couple of daylight type bulbs.

Exposure method was Av mode or M mode, spot metering against the center of the grey card and all pictures were shot raw.

Then the pictures were developed. I used ACR 3.7 and as usual I had all settings set to default, straight (curve), zero (sharpening, noise reduction). Only the WB was manipulated.

Finally we have the figures. They are all taken from ACR by setting a sample point in the center of the picture while having all pictures selected. By using the WB tool at the same point the RGB values all became very much the same.

Well, that's about it. Doing it didn't take long.

(I can add that I, having the flimsy *ist DS metering in mind, also moved out in the overcast daylight and took one series of pictures with one of the lenses allready tested while my friend held a white sheet giving some neutral and soft light to the grey card which was placed on a grey sheet. The results were to my surprise the same.)

Lucky me not being into statistics. That saved me a lot of money as I didn't have to test houndreds of cameras. and time, accordingly. The weak point is that I missed to try my Jupiter-9 and other M42 lenses. I was to focused on having my daily equipment "calibrated".

Now let me down.

regards,

Jonas B · 07-01-2007, 05:38 PM

Originally posted by barondla

Not all cameras have meter cell in the prism. My Olympus Om4t has a simi transparent mirror with a sub mirror behind it to direct the light to the bottom of mirror box metering cell. It is far superior to what we have now. It is very accurate - beats my hand held Pentax analog spot meter, reads much lower light (around 2 min +), and not bothered by light entering viewfinder. Hope we get something similar some day.
thanks
barondla

I agree (except of that I don't have any Pentax analog spot meter). I also have an old OM-4Ti and since I got it I haven't used any other camera better at metering the light.

regards,

jeffkrol · 07-01-2007, 08:22 PM

Originally posted by Jonas B

I agree (except of that I don't have any Pentax analog spot meter). I also have an old OM-4Ti and since I got it I haven't used any other camera better at metering the light.

regards,

Pretty sure the Olympus is an OTF type meter, similar to the Pentax LX (which was also said to have one of the best metering systems ever

Jonas B · 07-01-2007, 10:52 PM

Originally posted by jeffkrol

Pretty sure the Olympus is an OTF type meter, similar to the Pentax LX (which was also said to have one of the best metering systems ever

OTF... Yes and no. Above all it is a reliable tool with a great exposure system including multispot and much more. Reading the light from the film comes to use during long exposures and perhaps also when using flash (not sure about the later).

06-29-2007, 11:35 PM	#1
ebooks4pentax Site Supporter Join Date: Apr 2007 Location: California, USA Photos: Albums Posts: 464	K10D light reading location? I can't remember which thread(s) were talking about how replacing the original focusing screen on K10D was giving false metering, (over or under exposure). That would indicate that the metering is taking place passed the focusing screen. In the 35mm camera days, I believe the metering was done at the mirror, before the screen. I don't know if anyone has the technical "know how" but here is my question. Where is the light measured on the K10D? Is it at the mirror/lens side of the screen or at viewfinder side of the screen? I can't believe that one would design a metering system with anything between the lens and the meter., and if so, why?

06-30-2007, 02:08 AM	#3
pschlute Moderator Loyal Site Supporter Join Date: Mar 2007 Location: Surrey, UK Photos: Gallery Posts: 8,193	Meter I was under the impression that the meter was behind the mirror. That the mirror actually allows some light to pass through. But i might have been dreaming.

06-30-2007, 04:26 AM	#5
pschlute Moderator Loyal Site Supporter Join Date: Mar 2007 Location: Surrey, UK Photos: Gallery Posts: 8,193	Thanks Thanks for clarifying that. This explains why the focus screen will have an effect on exposure.

06-30-2007, 05:30 PM	#6
matix Senior Member Join Date: Jan 2007 Location: Brisbane, Australia Posts: 239	Low light focusing now seems quicker with new split screen. Originally posted by pschlute Thanks for clarifying that. This explains why the focus screen will have an effect on exposure. I received my split prism screen this week, easy to install but not so easy to keep all the dust specks off!!, will remove and clean it again. It works a treat on my K10D, I thought I would miss the lines, but guess I ignored them anyway, it is a lot brighter than the stock K10D screen, so that makes me happy. Does not seem to have had any affect on auto focus accuracy, but I have noted and this may be a coincidence, that it seems to focus quicker in lower light than before?? I would now say that the unassisted focus is as good as any DSLR I have tested. Has anyone else noted this change in low light focusing, to the better? Really happy with it overall, the split prism is at approx 45deg which surprised me, but that makes sense as it allows many angles in the frame to be used, ie vertical, horizontal and most in between, as focal points to align the split images. I got mine from; jinfinance for USD $24.95 plus $4.95 postage, good communications and fast shipping. Phil

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
No light (Pentax K10d)	Larsenio	Troubleshooting and Beginner Help	7	07-22-2010 05:49 AM
light meter in k10d and k20d	platinum	Pentax DSLR Discussion	9	11-13-2008 12:38 PM
K10D low light metering	mc5	Pentax DSLR Discussion	15	01-19-2008 02:30 AM
Picture/pixel location to sensor location mapping	eman	Pentax DSLR Discussion	5	08-06-2007 10:03 PM

06-29-2007, 11:53 PM	#2
and Veteran Member Join Date: Mar 2007 Photos: Gallery Posts: 2,476	Deffinetley behind the screen. I have a split prism screen myself and with slow lenses it will totally mess up the metering! using a lens with max aperature of f5.6 needs +3 ev to come out ok, even then its a bit underexposed. Probably due to the prism blacking out and making the scene appear dark. My split prism screen also has darker, more grainer appearence in general and so the viewfinder is more dim. With f2.8 or faster lenses the meter is not affected in any way though, not even spot.

06-30-2007, 03:11 AM	#4
blende8 Veteran Member Join Date: Feb 2007 Location: Bremen, Germany Photos: Albums Posts: 1,521	Behind the mirror is the AF-system. The metering system is slightly above the view finder, so behind the screen. This was always so, also in film days.

06-30-2007, 06:20 PM	#7
jeffkrol Veteran Member Join Date: Sep 2006 Location: Wisconsin USA Photos: Gallery Posts: 8,434	Originally posted by matix Does not seem to have had any affect on auto focus accuracy, but I have noted and this may be a coincidence, that it seems to focus quicker in lower light than before?? I would now say that the unassisted focus is as good as any DSLR I have tested. Has anyone else noted this change in low light focusing, to the better? Phil Shouldn't make any difference. The AF cell is in the bottom of the light box and is fed light through the semi silvered part of the instant return mirror. Behind this mirror is another front surface mirror that directs light down into the AF cells. I believe the AF chamber has at least one other mirror in it as well . Light meter/p-ttl meter is behind a face of the porro-prism.

06-30-2007, 08:03 PM	#8
ebooks4pentax Site Supporter Join Date: Apr 2007 Location: California, USA Photos: Albums Posts: 464 Original Poster	Originally posted by jeffkrol Light meter/p-ttl meter is behind a face of the porro-prism. If so, that would mean that the focusing screen has a direct effect on the light being metered. If a certain focusing screen is brighter, the TTL metering system will be fooled in thinking that there is more light than there really is. The mirror is equidistant to the CCD surface as it is to the top face of the focusing screen. The screen then acts like a filter to the light. Denser or darker screen will change the exposure? I always thought that the light entering the camera was metered either on the mirror or somewhere before it goes through the focusing screen. In other word, different focusing screens will make the camera over or under expose! Is that correct?

06-30-2007, 08:29 PM	#9
Jonas B Inactive Account Join Date: Sep 2006 Location: Göteborg, Sweden Posts: 810	Originally posted by k10dbook -snip- In other word, different focusing screens will make the camera over or under expose! Is that correct? Yes, that's correct. Then it is not only about having a "light" or "dark" screen. The viewfinder screens are etched with the pattern on them in different angles. Above is the, now many times posted, chart I get when comparing lenses and screens. The DS screen is a bit darker but the plot indicates that the meter reads more light... This way Canon, as a current example, can offer the Ee-S screen for us people fond of manually focusing. The screen is good for manual focusing as the etched pattern is different resulting in a more shallow DOF than the standard screen. It also is a bit darker. Canon offers a firmware setting shifting the meters baseline a bit compensating for this difference. As you understand I wish Pentax made something similar. In the case of the errors we see with the stopped down metering with the K10D it is even more complicated. We don't get many elaborated reports but as I understand it some have different results with some tele lenses. I wonder if there is another factor in play as well with the original K10D screen. It may be that light from certain angles is read by the camera. The D/DS screens handles the situation better and give correct or nearly correct metering also with lenses having shorter focal lengths. 2 cents,

06-30-2007, 08:48 PM	#10
jfdavis58 Site Supporter Join Date: Nov 2006 Location: 13 S 0357397-3884316 Photos: Gallery Posts: 887	Just curious Jonas; been read the older threads with exposure information. How did you get these data points: what was the experimental set-up and procedure, what sort of data reduction/manipulation to put this graph together? How many times did you run the experiment? You could just point me at the original thread if that info has been shared. Thanks!

07-01-2007, 07:06 AM	#11
barondla Site Supporter Join Date: Sep 2006 Location: midwest, United States Photos: Gallery \| Albums Posts: 5,714	Not all cameras have meter cell in the prism. My Olympus Om4t has a simi transparent mirror with a sub mirror behind it to direct the light to the bottom of mirror box metering cell. It is far superior to what we have now. It is very accurate - beats my hand held Pentax analog spot meter, reads much lower light (around 2 min +), and not bothered by light entering viewfinder. Hope we get something similar some day. thanks barondla

07-01-2007, 05:35 PM	#12
Jonas B Inactive Account Join Date: Sep 2006 Location: Göteborg, Sweden Posts: 810	Originally posted by jfdavis58 Just curious Jonas; been read the older threads with exposure information. How did you get these data points: what was the experimental set-up and procedure, what sort of data reduction/manipulation to put this graph together? How many times did you run the experiment? You could just point me at the original thread if that info has been shared. Thanks! Hi jfdavis58, Hmm... I don't know... last time you asked me about my color management setup and I really described it you never replied and turned me down... But OK then, here we go: The chart has many sources, most of them based on experience and reading good books on exposure, digital picture technology, raw converters and zone systems but I also take a few Internet opinions into account. The first thing to mention is that the baselines for EV0, EV1 and so on in both directions are taken from my own experience and preferences. The EV0 line represents an RGB-value of 97 in the chart. Some prefer 110, some even higher figures. I am slowly moving towards 110. Knowing this you may move the lines upwards or downwards to your own taste. The values for the other lines are the results from my own experiments using a lot of cameras to see what happens when you take the double or the half exposure time with the same target. Then we have the curves: In this case they come from one single camera only (my first K10D). They are the average result from three exposures with each lens and screen combo without changing target. The target was a Kodak grey card (but could have been anything neutral) more or less filling the viewfinder. The background behind the card was a grey fabric and the room has grey tapestries. The light source were a couple of daylight type bulbs. Exposure method was Av mode or M mode, spot metering against the center of the grey card and all pictures were shot raw. Then the pictures were developed. I used ACR 3.7 and as usual I had all settings set to default, straight (curve), zero (sharpening, noise reduction). Only the WB was manipulated. Finally we have the figures. They are all taken from ACR by setting a sample point in the center of the picture while having all pictures selected. By using the WB tool at the same point the RGB values all became very much the same. Well, that's about it. Doing it didn't take long. (I can add that I, having the flimsy *ist DS metering in mind, also moved out in the overcast daylight and took one series of pictures with one of the lenses allready tested while my friend held a white sheet giving some neutral and soft light to the grey card which was placed on a grey sheet. The results were to my surprise the same.) Lucky me not being into statistics. That saved me a lot of money as I didn't have to test houndreds of cameras. and time, accordingly. The weak point is that I missed to try my Jupiter-9 and other M42 lenses. I was to focused on having my daily equipment "calibrated". Now let me down. regards,

07-01-2007, 05:38 PM	#13
Jonas B Inactive Account Join Date: Sep 2006 Location: Göteborg, Sweden Posts: 810	Originally posted by barondla Not all cameras have meter cell in the prism. My Olympus Om4t has a simi transparent mirror with a sub mirror behind it to direct the light to the bottom of mirror box metering cell. It is far superior to what we have now. It is very accurate - beats my hand held Pentax analog spot meter, reads much lower light (around 2 min +), and not bothered by light entering viewfinder. Hope we get something similar some day. thanks barondla I agree (except of that I don't have any Pentax analog spot meter). I also have an old OM-4Ti and since I got it I haven't used any other camera better at metering the light. regards,

07-01-2007, 08:22 PM	#14
jeffkrol Veteran Member Join Date: Sep 2006 Location: Wisconsin USA Photos: Gallery Posts: 8,434	Originally posted by Jonas B I agree (except of that I don't have any Pentax analog spot meter). I also have an old OM-4Ti and since I got it I haven't used any other camera better at metering the light. regards, Pretty sure the Olympus is an OTF type meter, similar to the Pentax LX (which was also said to have one of the best metering systems ever

Thread Tools	Search this Thread
	Search this Thread: Advanced Search

07-01-2007, 10:52 PM	#15
Jonas B Inactive Account Join Date: Sep 2006 Location: Göteborg, Sweden Posts: 810	Originally posted by jeffkrol Pretty sure the Olympus is an OTF type meter, similar to the Pentax LX (which was also said to have one of the best metering systems ever OTF... Yes and no. Above all it is a reliable tool with a great exposure system including multispot and much more. Reading the light from the film comes to use during long exposures and perhaps also when using flash (not sure about the later).