Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
05-10-2008, 05:36 PM   #61
Junior Member




Join Date: Apr 2008
Location: Prague
Posts: 42
So far it seems to me that there are actually two firmware problems with incorrect exposure using lenses in non-A position - one is probably incorrect timing/synchronization between closing lens' apreture and running shutter, one is incorrect algorithm when evaluating measured values.
Two examples (using e.g. Pentax-A 50/1.7):

* set Av mode, and take two pictures - one with apertrure ring set to wide open, second with aperture stopped down few stops. Because the camera does not (can't) know what aperture is set on the lens, it will use identical shutter speeds for both shots, so one would expect that the second picture will be underexposed. But surprisingly both pictures will look identically! The only explanation I can think of is that the shutter is fired to early (or too late), when the lens' aperture is not yet in the chosen position.

* set M mode, set lens to wide open, press Green to measure, take picture.
Set aperture to one stop smaller, press Green, take picture.
Set aperture to two stops smaller, press Green, take picture. etc.
You will notice that shutter speeds will vary faster - in 2EV steps instead of 1EV, and the resulting pictures will confirm it (they will be brighter and brighter).
One possible explanation is that the camera makes two measurements - one with lens wide open, one with lens "closed" - and the difference will (incorrectly) apply to "closed measurement".

(I made these tests with two different lenses, and also I looked into the lens to verify that the aperture is really closing/opening during shooting.)

05-10-2008, 06:10 PM   #62
Pentaxian
Lowell Goudge's Avatar

Join Date: Jan 2007
Location: Toronto
Photos: Gallery | Albums
Posts: 15,346
QuoteOriginally posted by ivanp Quote
So far it seems to me that there are actually two firmware problems with incorrect exposure using lenses in non-A position - one is probably incorrect timing/synchronization between closing lens' apreture and running shutter, one is incorrect algorithm when evaluating measured values.
Two examples (using e.g. Pentax-A 50/1.7):

* set Av mode, and take two pictures - one with apertrure ring set to wide open, second with aperture stopped down few stops. Because the camera does not (can't) know what aperture is set on the lens, it will use identical shutter speeds for both shots, so one would expect that the second picture will be underexposed. But surprisingly both pictures will look identically! The only explanation I can think of is that the shutter is fired to early (or too late), when the lens' aperture is not yet in the chosen position.
no, in AV mode the camera does not activate the apature at all, it leaves it wide open regardless of apature setting, all shots are wide open, but metered for wide open. the only error is any error in the metering, which is not consistant as a function of f-stop as discussed earlier.
QuoteQuote:
* set M mode, set lens to wide open, press Green to measure, take picture.
Set aperture to one stop smaller, press Green, take picture.
Set aperture to two stops smaller, press Green, take picture. etc.
You will notice that shutter speeds will vary faster - in 2EV steps instead of 1EV, and the resulting pictures will confirm it (they will be brighter and brighter).
One possible explanation is that the camera makes two measurements - one with lens wide open, one with lens "closed" - and the difference will (incorrectly) apply to "closed measurement".

(I made these tests with two different lenses, and also I looked into the lens to verify that the aperture is really closing/opening during shooting.)
this is correct, in manual mode, pressing the green button should give correct exposure (and does so on the *istD) but on the K10D there is simething funny with the metering and viewing screen, where the meter is inacurate. This is compensated in software so that when a lens is in AE mode, the camera knows (by the contact encoding) the maximum apature, and corrects the reading, but if you take that lens out of AE mode, the exposure is not correct. The general trend is to uinder expose at F1.4, by 1 stop to over expose by 2 stops between F8 and F11 fallingback to over expose by 1 1/2 stops at F22/32.

There is a follow on effect with this metering problem on the K10,

Add a teleconverter. All teleconverters pass through not only focal length (which has been a great debate on this thread already), but it also passes through the maximum apature. But by adding a TC you change the apature, and the metering follows it's same inacurate reading as with non AE lenses. As a result, you get a constant exposure error also with adding a teleconverter, the error is constant for each lens but will change as a function of the resultant maximum apature and the origonal apature.

As I have stated before, none of these problems are present on the *istD. it is something specific in my experience, with the K10D (I don't own any of the other models) but people have reported it also on the K20D.

People have also reported that different viewing screens can fix the metering issue on K mount lenses, at the expense of correct metering on AE lenses. It seems there is no good single fix.

I have yet to try changing screens because for the sake of this experiment, I don't want to spend a hundred dollars. I was hoping people would post results using a uniform exposed field to evaluate the metering, but as you have seen earlier in this post, people want to post only pictures where scene and metering evaluation are not reliably repordused.

This is not for the same of "measurebating as has been implied, but only to better define the problem, so that reliable exposure can be made, and perhaps to help persuade pentax to come to the table and offer some fix.
05-11-2008, 08:38 AM   #63
Pentaxian
Wheatfield's Avatar

Join Date: Apr 2008
Location: The wheatfields of Canada
Posts: 10,058
My initial test indicated that there is an exposure irregulaity problem with the K20 and pre A lenses.
You came up with the idea that the focussing screen was the problem, so i changed my K20 screen out for an original istD screen (it was the one that was in my istD when I bought it). I repeated the exposure series, and found that the screen made no difference.
Sorry bub, your wrong on that one.
I then repeated the test in very bright light, and confirmed that the metering problem relates to poor linearity of the meter when the light dims.
This confirms what one of the electical engineers on PDML says about the metering system.

This confirms that there is a problem, it just doesn't confirm what you have decided the problem is.
To me it looks like you are letting your arrogance make you blind.

Last edited by Wheatfield; 05-11-2008 at 09:22 AM.
05-11-2008, 10:23 AM   #64
Pentaxian
Lowell Goudge's Avatar

Join Date: Jan 2007
Location: Toronto
Photos: Gallery | Albums
Posts: 15,346
QuoteOriginally posted by Wheatfield Quote
My initial test indicated that there is an exposure irregulaity problem with the K20 and pre A lenses.
You came up with the idea that the focussing screen was the problem, so i changed my K20 screen out for an original istD screen (it was the one that was in my istD when I bought it). I repeated the exposure series, and found that the screen made no difference.
Sorry bub, your wrong on that one.
I then repeated the test in very bright light, and confirmed that the metering problem relates to poor linearity of the meter when the light dims.
This confirms what one of the electical engineers on PDML says about the metering system.

This confirms that there is a problem, it just doesn't confirm what you have decided the problem is.
To me it looks like you are letting your arrogance make you blind.
Not a question of arrogance at all.

There are many posters on this forum that have stated changing out the focusing screen has improved or corrected the metering with manual lenses. At the same time they also post that using AE lenses the metering becomes erratic when they change out the screens. Pentax has actually confirmed this to me in writing, that changing the screens will impact metering.

Additionally, and I can attest to this witht he K10D, it is not a question of linearity with high vs low ambient light, because I can do the same test and have done so, with bright suunlight, overcase sky, shade, and dim incandascent light. While I might need to change ISO settings between the different lighting situations, the characteristic does not change. This to me suggests the behavior is much more a question of the F_Stop only at the time of metering. Clearly there may be more to it than this, but from what can be measured using the camera in normal lighting conditions this is what is apparent.

If you want to continue in personal attacks, go ahead, I am finished with this thread. but that will not stop me from posting observations that I can and have measured.

05-11-2008, 10:41 AM   #65
Pentaxian
Wheatfield's Avatar

Join Date: Apr 2008
Location: The wheatfields of Canada
Posts: 10,058
As soon as you change ISO, you are changing the EV range, with the potential to introduce errors in your test.
Sorry, you've nullified your test by creating a variable which will throw the results sideways.
My informal test, which agrees with the shooting I have done with both K10 and K20 cameras is that the meter loses linearity as the light levels drop below a threshhold which seems to be fairly high, unfortunately.

It's a well known fact that if you put in a screen that has different transmissive properties from what the meter is calibrated for, you will affect metering. My point was specifically addressing your assertation that the screen was causing the problem (apparently wrong).

Note, I'm not saying there isn't some loss of metering accuracy, it;s just that your reasons for it don't agree with mine and others who I know are educated in the field reasons.

It's likely that your testing procedure is faulty, based on your description, and I expect that anyone who is using your faulty testing procedure (introducing variables during the test) will share your results.
05-11-2008, 12:06 PM   #66
Pentaxian
Wheatfield's Avatar

Join Date: Apr 2008
Location: The wheatfields of Canada
Posts: 10,058
QuoteOriginally posted by falconeye Quote
Why are you resorting to Dilbert cartoon like arguments here?

Following your logic, Pentax should remove support for SR with all pre-SR lenses from Sigma and Tamron. Immediately!

Now see this: Pentax even allows you to dial in a focus length for Sigma or Tamron lenses made before even AF existed! Even for Zeiss lenses!


It is just a bug in their firmware which prevents this from working with a TC, and the fix is the removal of one line of code (the 'if'-statement which grays out the menu entry).

Would Pentax run a bug-entry system like most software companies do, we woudn't even have to discuss this here.

It is this simple. Please do not complicate things.
So what you are saying is that Pentax is responsible to reverse engineer their equipment so that it will support faultily reverse engineered third party equipment.
And you call my arguments Dilbert like.
Put the blame where it belongs, on the manufacture that apparently said their equipment was compatible when it obviously isn't.
At least buddy with his exposure issues has a valid concern, no matter whether he is wrong about the actual cause or not.
05-11-2008, 03:47 PM   #67
Pentaxian
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,863
QuoteOriginally posted by Wheatfield Quote
So what you are saying is that Pentax is responsible to reverse engineer their equipment so that it will support faultily reverse engineered third party equipment.
And you call my arguments Dilbert like.
I said what I said, not what you claim I said.

Maybe, repetition helps in your case... The Pentax firmware has a bug (it greys out a menu item while it must never grey out) and everything we demand is they listen to their customers and fix it.

And I will not reply to you on the matter anymore.
05-11-2008, 03:58 PM   #68
Pentaxian
Wheatfield's Avatar

Join Date: Apr 2008
Location: The wheatfields of Canada
Posts: 10,058
Bottom line, you want them to engineer a fix for someone else's equipment compatability problem.
Perhaps if you offered to pay them to rewrite the firmware......

05-13-2008, 11:30 AM - 1 Like   #69
Pentaxian
Lowell Goudge's Avatar

Join Date: Jan 2007
Location: Toronto
Photos: Gallery | Albums
Posts: 15,346
I'm an arrogant, measurebating, pixel peeping SOB, but

I'm also right.

I put my *istD screen into my K10D and as my title implies, then I pxiel peeped and measurebated.

the result is below, when compared to the *istD and K10D each with their own screens.

the bottom line is that the K10D focusing screen causes a huge linearity problem with exposure, compared to the *istD screen. It has nothing at all to do with different sensors, because as you can see, the results of the *istD screen in both the *istD and K10D are virtually identical.

Note that in PSPX2 where measuremens were made, greyscale of 45 is roughly 1 stop (JPEG with nominal contrast)

Note that the measurement for the K10D was done today, the *istD over a year ago.

All measurements were done using the same method, using a uniformly lit concrete surface, with ISO held constant over the entire test, using green button for setting exposure.

All measurements use the same 50mm F1.4 K mount lens.

ALthough I promised I would not add any additional comments to this post, the facts speak for themselves.

It should be noted that the SMC Pentax 50mm F1.4 seems to have a slight and increasing error in apature as you stop down, but that is a different issue, and perhaps someone else can comment.

Last edited by Lowell Goudge; 09-24-2010 at 06:22 AM.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
camera, dslr, fl, k20d, lens, photography, teleconverter
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
2X teleconverters gawan Pentax SLR Lens Discussion 8 05-10-2009 11:27 AM
teleconverters... again cwood Pentax SLR Lens Discussion 4 08-24-2008 11:29 AM
K20D & Teleconverters HawaiianOnline Pentax News and Rumors 1 01-24-2008 02:14 AM
Teleconverters FotoPete Pentax SLR Lens Discussion 0 10-12-2007 04:08 PM
Teleconverters garedagmad Pentax DSLR Discussion 0 11-13-2006 05:44 AM



All times are GMT -7. The time now is 09:46 AM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top