Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
04-24-2012, 02:06 AM   #31
Site Supporter




Join Date: Apr 2011
Location: Copenhagen
Photos: Gallery | Albums
Posts: 488
Original Poster
QuoteOriginally posted by Digitalis Quote
because in live view mode the metering is done off the sensor itself - the optical exposure meters that are used when the mirror is flipped down are behind the focusing screen - and if you are using an after-market focusing screen with focusing aids on it then that is most likely causing the erratic AE behaviour.
Here is what KatzEye write about the issue:

Metering Information The KatzEye™ focusing screen for the Pentax K-5 and K-7 shows only a very slight effect on light metering. In our testing, the KatzEye™ screen tested within factory tolerance in nearly all situations. The only exception is when using spot metering with lenses having apertures of f2.8 or smaller. This combination may produce overexposure, so it is advisable to use either center-weighted average metering or multi-segment metering to preserve accurate meter response when using small aperture lenses. In the event that spot metering mode must be used with a small aperture lens, it is advisable to set an appropriate exposure compensation by reading the histogram of a test shot. Approximate exposure compensation values are as follows*:

Aperture Multi-Segment Center-Weighted Spot
1.4 0 0 0
2.0 0 0 0
2.8 0 0 -0.6EV
4.0 0 0 -1.5 EV
5.6 0 0 -2.5 EV

Note: These exposure compensation suggestions are a recommended starting point only and actual results may vary depending on conditions. Users are advised to take a test photo and check the histogram before beginning any critical work. Apertures indicated in the above chart are the lens maximum aperture as indicated on the lens body and are not affected by the actual aperture chosen for a given photo.
* Metering test results presented are based on testing of the K-7 model. While results are expected to be similar for the K-5, users of this model are strongly encouraged to take a test shot and check the histogram before beginning any critical shooting.

04-24-2012, 03:08 AM   #32
Site Supporter




Join Date: Apr 2011
Location: Copenhagen
Photos: Gallery | Albums
Posts: 488
Original Poster
Here is a test I made showing the light meter perfomance w/KatzEye Plus prism on
Pentax K7 - Exposure mode: Manual - Metering mode: Centerweighted-Average - Custom Image EV+/- set to: -4 w/S-M-C Takumar 1.8/55

With Optical View Finder


With Optical View Finder


With Live View


Conclusions:
- Errors occur from 4 - up
- Errors occur in ‘M‘ mode even w/ Cent.weighted-average metering
- Errors donīt occur when in Live View
- Metering is brighter with optical view finder from 5.6 - down


Note: After a comment I want to add: This is not in any formal way a scientific test. Merely some practical findings of mine.

Last edited by jt_cph_dk; 04-25-2012 at 03:05 AM. Reason: edit text
04-24-2012, 10:53 AM - 1 Like   #33
Loyal Site Supporter
pacerr's Avatar

Join Date: Jan 2008
Location: Henry, TN
Photos: Gallery | Albums
Posts: 2,848
*This is NOT intended as Personal Criticism, but rather as thoughtful critique and commentary.*

While I appreciate the spirit and intent of this comparison and mostly agree with the implications (and often indulge in such exercises for personal reasons and calibration of my own gear), I have to point out that there are too many un-addressed (or at least un-annotated) variables to consider this an evaluation of the Katz-eye screen and its potential exposure biases.

Subtle variations, such as in framing (eg, 2nd frame in Live View), and a documentation of lighting source calibration and consistency, preclude this from being taken as a de facto documentary of the Katz-eye screen's effect on EV -- even though it may seem "right" at first glance to many experienced users.

While this presentation may be reasonably accurate and without bias, there's also no reason to assume it is so and offer a caution that it isn't a universal result.

We owe any manufacturer the respect of reading the definition of the capabilities of the device and understanding what the benefits and limitations of the product are. All too often we, as consumers, are presumptuous of the actual capabilities of a device.

Personally, I have no expectation that metering through a split-screen/micro-prism will conform to the OEM standard of the matt screen as designed. Fortunately, I have instant review with histogram and 'blinkies' ("chimping") to compensate for any extra-ordinary variations of my own choice of modifications.

The utility of ANY tool is equally dependent on the skill and knowledge of the craftsman.

[ Disclosure? -- I do use four split-prisms, including both Katz-eye and others', screens on seven Pentax SLR/DSLR bodies and use them selectively -- with appropriate adjustments -- for specific purposes. ]

H2

(AS a nonsequitur I'll suggest that those dissatisfied with their results with split-prism metering use the advantage of the exposure bracketing mode to establish a personal preference for EV adjustments with their own gear.)
04-24-2012, 11:06 AM - 1 Like   #34
Pentaxian
Lowell Goudge's Avatar

Join Date: Jan 2007
Location: Toronto
Photos: Gallery | Albums
Posts: 15,312
QuoteOriginally posted by jt_cph_dk Quote
Hello Lowell, Thankīs for posting your chart.

I donīt get this meter? With any given lens and aperture, it is supposed to measure the amount of light that comes through to the sensor calculating the time and/or ISO needed for a correct exposure. Then from that and for my reasons, I can then choose to over- or under-expose the whole or parts of the scene. However, and as your chart shows, This meter does not deliver consistent results regardless of what ever aperture setting I choose (I have made the test shots going through all the steps) but instead comes up with a fluctuating (or inconsistent) result as described earlier. Then Iīm told itīs the split prism. And I can understand that. This needs calibration. But I still donīt get why there is a kind of algorithmic disorder. Why is it not the same for every step of aperture, if the problem derives from the split prism being added to the flow of light?

I know how to work around it and get the results that I want, but I find it inconvenient to have to do so every time I change the aperture. Itīs supposed to be enough to push the green button when in M mode and should be manual (by switch) in Av mode with the Takumars.
I cant answer fully as I did not design the system, and have not discected the cameras to understand what is happening, BUT, what I can tell you, and which I learned by a simple test of placing an *istD focusing screen in the K10D, is that I believe part of the issue is the finsih on the focusing screen. For the K10D they changed from the LL60 screen to the LL80 screen, and the screen was brighter, but I think, and here is where the assumption comes in, the LL80 screen scatters light differently, and if there is light coming in at low angles (which would be the case with wide apertures, that light hits the sensors from other than where they should be looking.

When I put the *istD screen in the K10D the metering settled down and matched the *istD performance, which until the K5 came along, was arguably the best metering camera pentax made for legacy lenses.


I believe even with the *istD they knew there was some non linearity and with A lenses calibrated it out, (although the non linearity is quite well behaved from my chart) the K10D was much worse, and drew a ton of complaints about metering. the K20D was the same, the K7 slightly better, and the K5 very good (like the *istD )

Knowing how your camera behaves is part of your job. I have charted many lenses on each of my bodies, it takes about 10 minutes per lens per body. Where I fall behind, is when I add a new body, as this necessitates testing 46 lenses at present. I have become somewhat lazy in keeping up, testing only a few lenses to see how the trend holds.

04-25-2012, 12:57 AM   #35
Site Supporter




Join Date: Apr 2011
Location: Copenhagen
Photos: Gallery | Albums
Posts: 488
Original Poster
QuoteOriginally posted by pacerr Quote
Subtle variations, such as in framing (eg, 2nd frame in Live View), and a documentation of lighting source calibration and consistency, preclude this from being taken as a de facto documentary of the Katz-eye screen's effect on EV -- even though it may seem "right" at first glance to many experienced users. While this presentation may be reasonably accurate and without bias, there's also no reason to assume it is so and offer a caution that it isn't a universal result. We owe any manufacturer the respect of reading the definition of the capabilities of the device and understanding what the benefits and limitations of the product are. All too often we, as consumers, are presumptuous of the actual capabilities of a device.
Hello Pacerr,

I agree. I could have written ‘Suggested conclusions‘. And we do owe the manufactures to be thorough and precise about our criticism and comments to their products. And they owe it to us to be honest about their products limitations. I quoted KatzEye in their own evaluation of the Plus prism screen and show some results that suggest other conclusions to the abilities of their product. Thatīs it. My test was never intended to be scientific and I make no such claims, they are just examples of my practical findings. I find the DSLR world difficult to comprehend and need to be able to ask basic, informal questions and discuss whatever small progress I make with other users in forums and threads like this one.

Thankīs for your thoughtful critique and commentary.

Regards
04-25-2012, 01:44 AM   #36
Site Supporter




Join Date: Apr 2011
Location: Copenhagen
Photos: Gallery | Albums
Posts: 488
Original Poster
QuoteOriginally posted by Lowell Goudge Quote
what I can tell you, and which I learned by a simple test of placing an *istD focusing screen in the K10D, is that I believe part of the issue is the finsih on the focusing screen. For the K10D they changed from the LL60 screen to the LL80 screen, and the screen was brighter, but I think, and here is where the assumption comes in, the LL80 screen scatters light differently, and if there is light coming in at low angles (which would be the case with wide apertures, that light hits the sensors from other than where they should be looking.
Thatīs an interesting assumption you make Lowell. It seems reasonable that the different angles of light and screen scattering of it, could be the cause to the problems/errors/inconsistencies/need for calibrations. Is it at all possible, to fine-tune the meters behavior, the algorithms in the K7s firmware, with any of the many ‘handles‘ available in the camera menus? I havenīt found it yet.
Maybe I need to start looking for a K5 to substitute my K7 instead of trying to find an answer to this calibration issue. Or just accept the K7 (w/KatzEye Plus split prism) for what it is: In many ways a great photographic tool, that Iīm enjoying almost every day. Thankīs. Regards
04-25-2012, 03:59 AM   #37
Pentaxian
Lowell Goudge's Avatar

Join Date: Jan 2007
Location: Toronto
Photos: Gallery | Albums
Posts: 15,312
QuoteOriginally posted by jt_cph_dk Quote
Thatīs an interesting assumption you make Lowell. It seems reasonable that the different angles of light and screen scattering of it, could be the cause to the problems/errors/inconsistencies/need for calibrations. Is it at all possible, to fine-tune the meters behavior, the algorithms in the K7s firmware, with any of the many ‘handles‘ available in the camera menus? I havenīt found it yet.
Maybe I need to start looking for a K5 to substitute my K7 instead of trying to find an answer to this calibration issue. Or just accept the K7 (w/KatzEye Plus split prism) for what it is: In many ways a great photographic tool, that Iīm enjoying almost every day. Thankīs. Regards
The k7 is a fine camera, and was excellent at the time it was released, but it was rapidly surpassed with respect to sensor technology. I actually don't use the K7 with manual lenses, or for that matter the K5. I have never traded any of my bodies, so at present, and until they fail, I will use my *istD and K10D for manual lenses. Both have split image finders installed. My. k5 and. K7 are my main every day bodies, used with (almost exclusively) a sigma 10-20/4-5.6, a Tamron 28-75/2.8 and a sigma 70-200/2.8 plus sigma 2x TC. My various M42 and K mount MF lenses do get used but only, for the moment on the older bodies. Note this is partially because my K10D has had the lens locking pin mechanism fail, and this keeps the locking pin and screw drive recessed, so I have an M42 adaptor semi perminantly installed. The *istD gets my K series lenses because it can do TTL flash
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, k7, meter, photography
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
How to meter if you don't have a light meter? LFLee Pentax Medium Format 25 02-09-2012 11:45 AM
Do I really now need a light meter? atomic Troubleshooting and Beginner Help 15 11-15-2011 02:46 PM
Help with Light Meter Debra Heschl Pentax Camera and Field Accessories 3 11-03-2011 11:35 AM
K7 Light Meter eccs19 Pentax DSLR Discussion 2 09-27-2010 05:01 PM
Light Meter sam.p Pentax Film SLR Discussion 10 08-05-2010 03:51 PM



All times are GMT -7. The time now is 06:16 AM. | See also: NikonForums.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