Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
05-19-2014, 05:43 AM   #661
Veteran Member




Join Date: Apr 2011
Location: Chicago Area
Photos: Gallery
Posts: 366
QuoteOriginally posted by tromboads Quote
Now the exposure accuracy is always what I expected it to be, but never was with the katseye. I quickly tried a Tak 55 f1,8, Tak 50 f1.4, an A 50 f1.7, and a Tak 200 f4 and they all expose correctly when stopped down, all the way to f22. FREAKING AMAZING ! Heck I should have taken the punt years ago :P I just palmed it off as if one was going to use a non-stock different screen you'd forgo the right to correct computed exposures. How silly!.
Sounds like a substantial improvement over the Katzeye, and its sounds like your quick test did not show any exposure problems. Since you got the screen with NO markings could you please test the following:

In low light, with your Tak f4 (or another slow lens) point at the wall or another evenly light object, and - covering the eye piece - check the suggested exposure using normal metering, and then check the suggested exposure from using live view. If your camera does not have live view switch to one of the fast lenses and see if you get the same suggested exposure.

I - and I am sure others - would really like to know if it the non-etched screen avoids the exposure difference we are experiencing...

Thank you so much...

05-19-2014, 06:21 AM   #662
Pentaxian
tromboads's Avatar

Join Date: Apr 2012
Location: Melbs
Photos: Gallery | Albums
Posts: 1,130
This K10 of mine ain't fancy enough to have "live view" :P Sorry.

Switch to one of the fast lenses I have? f4 is f4 no matter what lens it is. but to satisfy the curiosity in some of you.. Stand bye...
05-19-2014, 06:32 AM   #663
Veteran Member




Join Date: Apr 2011
Location: Chicago Area
Photos: Gallery
Posts: 366
QuoteOriginally posted by tromboads Quote
This K10 of mine ain't fancy enough to have "live view" :P Sorry.

Switch to one of the fast lenses I have? f4 is f4 no matter what lens it is. but to satisfy the curiosity in some of you.. Stand bye...
You are absolutely right. F4 is f4. However, using aperture priority with my 31mm ltd at F4 suggests a shutter speed of 1/8 - my 18-135mm set at f4 gets a suggested shutter speed 1.5 stop slower (I.e. 1/20 or so). Switching the latter to live view (and not going though the regular light meter) I get a suggested shutter speed close to expected (I.e. 1/8)

Thanks for checking,,,
05-19-2014, 06:59 AM   #664
Pentaxian
tromboads's Avatar

Join Date: Apr 2012
Location: Melbs
Photos: Gallery | Albums
Posts: 1,130
I worry now we are dealing with actual efficacy, or T value of the lens. without first knowing actually if f4 is accurate on the 18-135, or indeed the 31mm ltd. (in regards to sharing a similar transmission value)

Having said that. Here is the result of my Super Tak 50mm f.14, a good 40yr at least I suppose, on the K10 with the EE-S.. all iso100

Index of /media/ees

Download these and go nuts;

I'm putting the variation down to the efficient of the aperture on the lens itself, oh.. as half stops are not printed on the barrel, I used these values over the lens to guess the middle values (I actually added the two on each end and / by 2 to get the middle.."shrugs" anyway,

f1.4 = s1/750
f2 = s1/500
f2.8 = s1/350
f3.4 = s1/250
f4 = s1/250
f5.6 = s1/125
f6.8 = s1/90
f8 = s1/60
f9.5 = s1/45
f11 = s1/30
f16 = s1/20


Having done all this I realize I have set me camera to only meter in 1/2 stops, not 1/3 stops.. .. This actually explains the discrepancy

Well there ya go. Long story short, I'm chuffed at how accruate it is next to a) the stock screen (useless) and the katseye (just as useless)

I mean those histograms are showing 1/3 to 1/4 most of exposure discrepancy, that's f**king nothing in the scheme of outputted work.

05-19-2014, 07:36 AM   #665
Veteran Member




Join Date: Apr 2011
Location: Chicago Area
Photos: Gallery
Posts: 366
QuoteOriginally posted by tromboads Quote
I worry now we are dealing with actual efficacy, or T value of the lens. without first knowing actually if f4 is accurate on the 18-135, or indeed the 31mm ltd. (in regards to sharing a similar transmission value)

Having said that. Here is the result of my Super Tak 50mm f.14, a good 40yr at least I suppose, on the K10 with the EE-S.. all iso100

Index of /media/ees

Download these and go nuts;

I'm putting the variation down to the efficient of the aperture on the lens itself, oh.. as half stops are not printed on the barrel, I used these values over the lens to guess the middle values (I actually added the two on each end and / by 2 to get the middle.."shrugs" anyway,

f1.4 = s1/750
f2 = s1/500
f2.8 = s1/350
f3.4 = s1/250
f4 = s1/250
f5.6 = s1/125
f6.8 = s1/90
f8 = s1/60
f9.5 = s1/45
f11 = s1/30
f16 = s1/20


Having done all this I realize I have set me camera to only meter in 1/2 stops, not 1/3 stops.. .. This actually explains the discrepancy

Well there ya go. Long story short, I'm chuffed at how accruate it is next to a) the stock screen (useless) and the katseye (just as useless)

I mean those histograms are showing 1/3 to 1/4 most of exposure discrepancy, that's f**king nothing in the scheme of outputted work.
I'm sorry you went through all that work - but the issue is different than what you are eluding too.

The FAST lens set at F4 causes the camera to suggest a shutter speed of 1/8 - and the resulting picture appears to be correctly exposed. Swiching out the FAST lens with a SLOW lens, again at F4, causes the camera to suggest a shutter speed of 1/20 - and the resulting picture is now 1.5 stop underexposed. The used aperture does not matter (ie.. you can use F8 instead). What matter is the slower lens combined with the S Screen messes up the light metter.

My THEORY is that the light AF markings etched in the focus screen throws off the camera meter, causing it to underexpose.Those AF markings are a lot more visible with the SLOW lens... However, it could be something else.. If YOUR camera does not show this difference for fast and slow lenses, then it supports the theory, since you don't have the etchings.

BTW - due to the differnet focal lenths of your lenses, a blank wall probably works better. In my case it is IMMEDIATELY apparently that the slow lens causes significant underexposure..

Thanks for your help...

Last edited by HenrikDK; 05-19-2014 at 07:44 AM.
05-19-2014, 07:58 AM   #666
Pentaxian
tromboads's Avatar

Join Date: Apr 2012
Location: Melbs
Photos: Gallery | Albums
Posts: 1,130
So you concern is an underexposure from a "slower" lens? And not from a "faster" lens that is stopped down, Which you feel is caused by the grid lines getting in the way...

Right well I'll throw the 200mm f4 on the front and see what the histogram looks like.
05-19-2014, 09:31 AM   #667
Veteran Member




Join Date: Apr 2011
Location: Chicago Area
Photos: Gallery
Posts: 366
QuoteOriginally posted by tromboads Quote
So you concern is an underexposure from a "slower" lens? And not from a "faster" lens that is stopped down, Which you feel is caused by the grid lines getting in the way...

Right well I'll throw the 200mm f4 on the front and see what the histogram looks like.
Exactly. I have seen some variation in suggested exposure in bright light, but nothing like the low light variation.

In practice I can live with it, since I would tend to use faster lenses in low light situations, and now I know I should look out for it. If I can fix it with a non-etched screen I may return my unopened k-5 s screen....
05-19-2014, 10:33 AM   #668
Site Supporter




Join Date: Feb 2011
Location: United Kingdom
Photos: Albums
Posts: 424
Hendrik,

I tried a bit more testing, comparing LV and viewfinder auto exposure of the same scenes. Low-ish light, just playing about indoors. Using matrix, centre weighted, and spot metering on a variety of scenes using a few different lenses.

This wasn't very scientific! But I wanted to have a very quick look if there was much of a problem. In some cases (though not always) I found repeatable differences in exposure between VF and LV mode. Worst case - approx 1.7 stops, noticed on a slow lens combo used together with matrix metering on a particular scene. Other usage profiles (especially switching away from matrix metering) actually gave the opposite result with some scene and lens combinations (i.e. the viewfinder mode exposed more than LV in some spot metering situations).

None of it was really too bad, for me, it was generally well within the normal requirement to apply exposure compensation to really nail an exposure. I was going to put the stock screen back in to see if that made any difference. But I decided the variability is just simply not that bad. I can certainly live with the results I am getting and I am not going back to the stock screen on the off-chance of more consistent behaviour at this point. And I decided to concentrate on doing some new photos rather than bottoming this one fully out at this time!

I retain the suspicion that slow lens + ee screen with brackets + matrix metering + low-ish light may be a combination to watch out for, because it may require a little compensation (or switching to LV). But I would usually take a test shot and adjust afterwards anyway.

I think the way to really prove this one way or the other is to use an artificially lit scene, and to try a number of different combinations, preferably with a stock, ee (brackets) and ee (unbracketed) screen. But too much for me - I'm going to get back to taking some (non-test) shots for now!

05-19-2014, 10:42 AM   #669
Site Supporter
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 26,158
QuoteOriginally posted by tromboads Quote
This K10 of mine ain't fancy enough to have "live view" :P Sorry.

Switch to one of the fast lenses I have? f4 is f4 no matter what lens it is. but to satisfy the curiosity in some of you.. Stand bye...
  • Take your Super Tak 50/1.4 and get the exposure at f/1.4 in Av mode, use a blank wall
  • Note the shutter speed and calculate down to what it would be at f/8
  • Mount your 18-55 kit and get the exposure for f/8 at 50mm (leave aperture open), also in Av mode
  • Compare the two, they should be very close
When I do the same test with the K-3, I get two stops underexposure with the kit lens. Differences between you and me:
  • K-3 vs. K10D (vastly different metering systems)
  • I have the Type-S screen with the AF zone markings

I suspect that what many of us are seeing is related to the K-3 metering system. It has a ton of sensor sites and is probably giving the bright lines more credibility than they deserve. I also see less of an effect when the subject is complex (the grid lines are not uniformly bright in that case).

Anyway, firing a message off to focusingscreens.com is on my list of things to do today.


Steve

Last edited by stevebrot; 05-19-2014 at 10:59 AM.
05-19-2014, 11:00 AM   #670
Veteran Member




Join Date: Apr 2011
Location: Chicago Area
Photos: Gallery
Posts: 366
QuoteOriginally posted by stevebrot Quote
  • Take your Super Tak 50/1.4 and get the exposure at f/1.4 in Av mode
  • Note the shutter speed and calculate down to what it would be at f/8
  • Mount your 18-55 kit and get the exposure for f/8 (leave aperture open), also in Av mode
  • Compare the two, they should be very close
I didn't think about the fact that the Tak is using stop down metering?.. So if the Tak is set at say f5.6 that would in essence be the same as metering with a slower lens. If so then it appears the non-etched screen may not affected...
05-19-2014, 11:11 AM   #671
Site Supporter
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 26,158
QuoteOriginally posted by tromboads Quote
f1.4 = s1/750
f2 = s1/500
f2.8 = s1/350
I just did a closer look at your numbers. Based on your f/1.4 exposure, you should be seeing something closer to 1/185s at f/2.8 (2 stops less light than at f/1.4) and your example photos support the numbers. For your chosen subject, you are not getting a linear response from the meter. That being said, the results are still a lot better than I have ever gotten with my K10D. In the future, I suggest using a well-lit exposure wedge or a blank wall for a test. The dynamic range in your subject is beyond that of the meter and may have factored into the results.


Steve
05-19-2014, 04:21 PM   #672
Pentaxian
tromboads's Avatar

Join Date: Apr 2012
Location: Melbs
Photos: Gallery | Albums
Posts: 1,130
Again we are assuming

1) The super tak's reported apertures are perfectly accurate
2) That aperture size (in regards to light transmission) from one lens to another is accurate enough to compare too
3) That two very different average weighted metering systems will work similarly
4) That multi glass elements and lens construction wont effect the transmission value.

This is even before we look at what the focus screen may do :P

You all see the madness in this right?

I'll do another test, (using a newer A series) and we'll comparing the manually computed values to what the light meter suggests. I'll work in whole stops to in order to keep things as simple as possible.

But in regards to how My K10 is now assessing the light coming into it, well it has never been this accurate.

Having said that, I fail to believe tho that Pentax with their own etched screen would be silly enough to put the etching's in a place to effect the meter... (i.e the etchings you have mimic the Pentax ones right?)
05-19-2014, 05:25 PM   #673
Veteran Member




Join Date: Apr 2011
Location: Chicago Area
Photos: Gallery
Posts: 366
QuoteOriginally posted by tromboads Quote
Again we are assuming

1) The super tak's reported apertures are perfectly accurate
2) That aperture size (in regards to light transmission) from one lens to another is accurate enough to compare too
3) That two very different average weighted metering systems will work similarly
4) That multi glass elements and lens construction wont effect the transmission value.

This is even before we look at what the focus screen may do :P

You all see the madness in this right?
My K-5, k-01, and k-3 with a variety of lenses will produce close to same exposure. Throwing the ee-S screen in suddenly - consistently - cause 1.5 stop under exposure. Yes - +/- half a stop is easily expected; this exceeds "normal".

QuoteOriginally posted by tromboads Quote
Having said that, I fail to believe tho that Pentax with their own etched screen would be silly enough to put the etching's in a place to effect the meter... (i.e the etchings you have mimic the Pentax ones right?)
The pentax screens have black markings/lines. These screens have what appear an etched line that reflects light, causing them to appear very bright on slower lenses - see stevebrots pictures a dozen posts back...
05-19-2014, 05:45 PM   #674
Pentaxian
tromboads's Avatar

Join Date: Apr 2012
Location: Melbs
Photos: Gallery | Albums
Posts: 1,130
Ah, well that's sound enough for me to have a point of difference (must have missed those shots of that focus screen of the K5 / K3) the Black vs Clear could very well be doing all this too!

Anyway more shots; Index of /media/ees/test 2

I couldn't bring myself to photograph a grey wall, consider this a compromise. This was with an A 50mm f1.7 (newest lens I own...)

The results on both are (in stops)

f2.8 = 1/10
f4 = 1/6
f5.6 = 1/3
f8 = 1s
f11 = 2s
f16 = 3s

That looks pretty good to me, There was a discrepancy at f1.7 however (in the histogram at least) The rest of the in the middle numbers look like this;

f1.7 = 1/30
f2.4 = 1/15
f3.5 = 1/8
f4.5 = 1/4
f6.7 = 1/2
f9.5 = 1s
f13 = 2s

Again.. that's pretty neat... I haven't got anything a full stop out yet.

But again, the histograms give more of the story away. And still, difference to the outputed picture whilst in work mode is hardly worth concern

The results at the small end, with both f11 and f13 reporting 2s, I'm going to palm that off again to an older lens (A series are 20yrs now?) and the fact that I didn't take my camera out of 1/2 stop exposure mode! DoH!
05-19-2014, 07:49 PM   #675
Veteran Member




Join Date: Apr 2011
Location: Chicago Area
Photos: Gallery
Posts: 366
The issue is still not how the individual lens behaves


Both slow and fast lenses will measure consistently, but metering with slow lenses will be 1 to 2 stops darker than with fast lenses. Using stop down metering appears to work well for fast apertures, but then starts underexposing when you go to slower exposures. Looking at your Tak/1.4 results it appears you may be seeing a little bit of this (about a stop lower metering at f4 vs f2. Hard to say...
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, dof, dslr, exposure, f1.4, f3.5, fa50, focus, folks, k-01, k-3, k-5, k-5 ii, k-5 iis, k5, katzeye, lenses, love, pentax k-5, photos, screen, shot, split
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Canon ee-S focus screen, anyone? dlacouture Pentax Camera and Field Accessories 2 04-30-2011 01:22 PM
Some Bird Shots - Manual focus with Canon 'Frankinlens' on K10D gscara Post Your Photos! 10 04-09-2011 05:16 PM
K-x manual focus screen? nixcamic Pentax DSLR Discussion 1 12-07-2009 10:45 AM
Manual Focus on Pentax DSLRs - I want my split prism screen PentHassyKon Pentax DSLR Discussion 17 02-22-2009 04:42 PM
Do we require split screen for manual focus? madhurvyas Pentax DSLR Discussion 32 07-23-2008 03:51 PM



All times are GMT -7. The time now is 06:28 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