Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
02-26-2011, 02:27 AM   #151
Veteran Member
jolepp's Avatar

Join Date: Oct 2010
Location: Finland
Photos: Albums
Posts: 3,196
QuoteOriginally posted by mgoblue Quote
Just a quick fix idea: 80 series tungsten light filter? Used to use them all the time with film. I know it's not ideal, but if it works it could save a lot of trouble and let people keep shooting in low tungsten light. I don't have a K5 (though I'd like one) so I can't test it. Anyone else? In theory it should work.
I tried a Hoya 80A briefly: it would seem to help some, but then it seems to take quite a bit of light (~1.5 stops) away too.

02-26-2011, 03:14 AM   #152
Veteran Member




Join Date: Aug 2009
Posts: 2,311
I just got a temporarily raise in IQ (not......)

But could it be that its all in the sensor??? The D7000 obviously has problems to, as the K-5 has. But they dont share the same AF systems.

Maybe the sensor was rushed and neither Nikon or Pentax had the time to make things 100% right with it.

Just a thought, i have no understanding about these technical stuff so...
02-26-2011, 04:00 AM   #153
Pentaxian
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,862
Original Poster
QuoteOriginally posted by JeffJS Quote
I Will acknowledge that this is probably greater than the typical misfocus EV (ISO 2000, 1/125, f2.0).. This is however a typical shooting condition for me.
4.6 EV and the writing is high contrast. Should work ok according to the study and it does in your case.

QuoteOriginally posted by mtngal Quote
On the AF assist lamp - did you happen to notice where it was illuminating, compared to the focus point you were using?
Good point.
I don't think the AF assist light beam is decentered. Rather I think the problem is that it can be easily overcasted by a lens hood. It wasn't in my FA31 test.

But to be sure, I just confirmed again:

I took a free-hand photo at -8 EV (the K-5 is such a wonderful beast, 1s exposure time).
(at -8 EV, my non-adapted eyes could only guess the presence of a large dark object in front of a white wall -- which isn't bad in itself though )

And oh wonder, the AF assist light came on and engaged multiple times to eventually deliver an in-focus image (which implies it covered the AF sensor area).


So, thanks to your question, we learned this:

At really low EV (probably below -4 EV with the FA31), the AF assist light works as expected.

At higher enough EV (above 0 EV or so), the AF assist light works as expected too (stays off).

At an intermediate level (around -4 EV to 0 EV or so), the AF assist light engages only once for the initial measurement but not for consecutive corrections, resulting in misfocus.

This in turn implies this:

The AF system decides itself if it needs the AF assist light.

Let me define two terms:

1. The AF "sees": it receices light and isn't in the dark.
2. The AF "reads": it coumputes a signal from the shifted cross-correlation of the two partial AF ray bundels, and reads an optimum shift (phase) p.

This means:

If the AF can read clear enough (not just see well enough) then the assist beam stays off. Otherwise, it is engaged.

This implies that the AF module is really sure it can read and that it wouldn't need the assist beam.

This leaves two possibilities:

1. It only thinks to be able to "read" well enough but when it must, can't.
2. It can read correctly but the value is messed up with down the line.

In either case, it would point to a malfunctioning.

So, I evebtually think that Christine's camera wasn't different but that she just didn't hit the intermediate region where the AF assist beam engages only once.

QuoteOriginally posted by mgoblue Quote
Just a quick fix idea: 80 series tungsten light filter?
My work points towards a lack of light problem, amplified by non-white light which simply is less light at the same exposure (as determined by channel saturation).

Then, the filter would make it worse be eating even more light.

Last edited by falconeye; 02-26-2011 at 04:11 AM.
02-26-2011, 04:35 AM   #154
Pentaxian
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,862
Original Poster
QuoteOriginally posted by the swede Quote
But could it be that its all in the sensor??? The D7000 obviously has problems to, as the K-5 has.
It's not the sensor and I don't think the D7000 problems are alike (maybe as severe but not alike).

In my "Understanding Image Sharpness" paper I have a section about the correlation of focus accuracy and sensor size. I even go as far as saying that APS-C cameras with phase detect AF and 5Ám pixels run into a conceptual limitation.

Between an APS-C camera with an f/5.6 AF center sensor and an f/1.8 lens on the one hand, and an FF camera with a f/2.8 double AF center sensor (more expensive camera) and an equivalent f/2.8 lens, there is a factor 4.5 in defocus blur after print. IMHO, that's a reason why 645D images in the studio are all razor sharp.

No wonder why more and more people start complaining about soft results with AF and the current generation of >16MP APS-C cameras. You really have to stop down to exploit the resolution. Just because of AF. And then comes motion blur.

Personally, I consider AF accuracy to be a much stronger reason for FF than the famous one stop light advantage (which needs more expensive (non equivalent) lenses to be exploited). 1st AF accuracy, 2nd lens performance, 3rd DR, 4th the VF, 5th low light (but nobody please discuss other meaning of FF than front focus in this thread ).

02-26-2011, 04:56 AM   #155
Pentaxian
gazonk's Avatar

Join Date: Feb 2009
Location: Oslo area, Norway
Photos: Gallery | Albums
Posts: 3,509
QuoteOriginally posted by falconeye Quote
In my "Understanding Image Sharpness" paper I have a section about the correlation of focus accuracy and sensor size. I even go as far as saying that APS-C cameras with phase detect AF and 5Ám pixels run into a conceptual limitation.
I guess at some point contrast detect AF will be superior for these cameras, and at that time, EVFs will also be superior to OVFs. So I don't think the Sony SLT solution (or other attempts at PDAF in cameras without optical viewfinders) will have a long life.
Pentax seems to take CDAF very serious now - K-5 CDAF is vastly improved, isn't it?
02-26-2011, 05:33 AM   #156
Veteran Member
jolepp's Avatar

Join Date: Oct 2010
Location: Finland
Photos: Albums
Posts: 3,196
I suppose one way to go about it would be dropping the AF sensor and its mirror in which case light would end up on the image sensor which could then be used for contrast AF (this is essentially what Sony has been up to?). This would seem to eliminate AF / image sensor misalignment and the lens specific FF/BF plague with some cost and complexity .
02-26-2011, 05:51 AM   #157
Veteran Member




Join Date: Jul 2008
Location: Var, South of France
Photos: Albums
Posts: 1,071
Fuji has some interesting things going on with their sensor's half-masked pixels used for a kind of phase AF...
02-26-2011, 06:32 AM   #158
Veteran Member
Smeggypants's Avatar

Join Date: Sep 2010
Posts: 1,536
QuoteOriginally posted by the swede Quote
I just got a temporarily raise in IQ (not......)

But could it be that its all in the sensor??? The D7000 obviously has problems to, as the K-5 has. But they dont share the same AF systems.

Maybe the sensor was rushed and neither Nikon or Pentax had the time to make things 100% right with it.

Just a thought, i have no understanding about these technical stuff so...
The sensor isn't used in the regular AF system. The mirror is down and a secondry mirror behind it directs it down in the AF module at the botto of the camera.

Autofocus: phase detection

02-26-2011, 06:35 AM   #159
Veteran Member
Smeggypants's Avatar

Join Date: Sep 2010
Posts: 1,536
QuoteOriginally posted by jolepp Quote
I suppose one way to go about it would be dropping the AF sensor and its mirror in which case light would end up on the image sensor which could then be used for contrast AF (this is essentially what Sony has been up to?). This would seem to eliminate AF / image sensor misalignment and the lens specific FF/BF plague with some cost and complexity .
Then the AF would be incredibly slow, and the inherent hunting would make Pentax a laughing stock amongst DSLRs

Autofocus: contrast detection
02-26-2011, 10:25 AM   #160
Pentaxian




Join Date: Apr 2007
Location: Sweden
Posts: 1,887
QuoteOriginally posted by falconeye Quote
This in turn implies this:

The AF system decides itself if it needs the AF assist light.

Let me define two terms:

1. The AF "sees": it receices light and isn't in the dark.
2. The AF "reads": it coumputes a signal from the shifted cross-correlation of the two partial AF ray bundels, and reads an optimum shift (phase) p.

This means:

If the AF can read clear enough (not just see well enough) then the assist beam stays off. Otherwise, it is engaged.

This implies that the AF module is really sure it can read and that it wouldn't need the assist beam.
Sort of what I hinted at back in post nr 58 in this thread.
https://www.pentaxforums.com/forums/pentax-news-rumors/134215-lumolabs-pentax...ml#post1399069
02-26-2011, 11:21 AM   #161
Pentaxian
jsherman999's Avatar

Join Date: Dec 2007
Photos: Gallery
Posts: 8,228
QuoteOriginally posted by falconeye Quote

Personally, I consider AF accuracy to be a much stronger reason for FF than the famous one stop light advantage (which needs more expensive (non equivalent) lenses to be exploited). 1st AF accuracy, 2nd lens performance, 3rd DR, 4th the VF, 5th low light (but nobody please discuss other meaning of FF than front focus in this thread ).
Agreed, it shall not be mentioned!


.

Last edited by jsherman999; 02-26-2011 at 11:29 AM.
02-26-2011, 11:35 AM   #162
Loyal Site Supporter
dadipentak's Avatar

Join Date: Jul 2008
Location: Baltimore, Maryland
Photos: Gallery
Posts: 10,073
QuoteOriginally posted by falconeye Quote
other meaning of FF
french fries?
02-26-2011, 12:32 PM   #163
Veteran Member




Join Date: Jan 2007
Location: Hawaii
Posts: 593
One question: This problem has no effect on those who normally shoot in full daylight or use full spectrum bulbs for their modeling lights, correct?
02-26-2011, 12:39 PM   #164
Banned




Join Date: Jan 2011
Posts: 242
Yes, it's correct.
02-26-2011, 01:33 PM   #165
Pentaxian
gazonk's Avatar

Join Date: Feb 2009
Location: Oslo area, Norway
Photos: Gallery | Albums
Posts: 3,509
QuoteOriginally posted by dadipentak Quote
french fries?
They're french again - "freedom fries" didn't last?
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!
copy, ev, focus, front, issue, k-5, light, pentax, pentax news, pentax rumors, plane, study
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Any tips for low light focus with K-5 designinme_1976 Pentax K-5 5 11-21-2010 08:38 PM
focus hunting in low light sorin Pentax DSLR Discussion 12 07-17-2010 02:20 PM
Low Light auto focus JohnKSA Troubleshooting and Beginner Help 6 03-10-2010 04:19 AM
Pentax Low Light Focus indy1984 Pentax DSLR Discussion 15 02-17-2010 09:42 AM
EV low light focus question tarsus Photographic Technique 3 06-26-2008 08:22 AM



All times are GMT -7. The time now is 06:59 PM. | 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