Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
10-30-2013, 09:18 AM   #16
New Member




Join Date: Aug 2012
Location: Kerala
Photos: Albums
Posts: 16
Original Poster
Anyone have any idea whether turning on the AA filter will affect IS or not as both rely on the same hardware and mechanics?

10-30-2013, 10:36 AM   #17
Pentaxian




Join Date: Jan 2013
Posts: 428
Thismshould be written in the user guide, but I expect that this kind of AA filter does not influence IS because of using higher frequencies for this purpose
10-30-2013, 12:16 PM   #18
Senior Member




Join Date: Jun 2012
Photos: Gallery
Posts: 139
A small digression : those DNG are quite big (35~40MB)... I was expecting more like 30MB per DNG...
10-30-2013, 01:49 PM   #19
Pentaxian
PePe's Avatar

Join Date: Nov 2007
Location: Helsinki, Finland
Photos: Gallery | Albums
Posts: 596
QuoteOriginally posted by hkiran1 Quote
Anyone have any idea whether turning on the AA filter will affect IS or not as both rely on the same hardware and mechanics?
I have been wondering about this too.
There is nothing about this in the information released. Both functions being intended for continuous use, I quess they would have to openly announce it if the systems would interfere with each others.
These systems do use the same hardware. However, I would imagine the functions are different enough to be run in separate modes parallel to each others.
Would definitely like to see this confirmed by Pentax.

10-30-2013, 02:57 PM   #20
Pentaxian
dosdan's Avatar

Join Date: Dec 2007
Location: Brisbane, Australia
Photos: Gallery | Albums
Posts: 1,741
I've created two comparison images from IR's K-3 AA examples:

1. Downloaded AA0 (no AA), AA1 (AA simulator strength 1) & AA2 (AA simulator strength 2) DNGs and converted these to TIFs in SilkyPix Developer Studio v4, using the Natural sharpening preset. No USM output sharpening applied. Also used the Demosaic Sharp setting of 100, rather than the default of 80. This increases the resolution, but takes longer, and is not suitable for noisy images.

2. In PXP X4, cropped and combined the 3 spatial frequency strips in AA0, AA1, AA2 top-to-bottom order.

3. Saved this 100% crop without sharpening. Used PNG so that no JPEG artefacts or softening would be present. 1252 x 241 pixels.

https://www.dropbox.com/s/oaewbq8wa356u9k/AA%20comparison%20100%20percent.png

4. Resized this to 200% (Bicubic, 75% sharpening) and cropped further to only include the right half of the image. 1253x476 pixels.

https://www.dropbox.com/s/8sgjasv11eety7b/AA%20comparison%20200%20percent.png

I've made these availble as download links rather than embedded images, so the website will not resize or convert them when displayed. When viewing make sure you view these at 100%.

It is obvious that the mazing is almost completely gone by AA2. The false colour aliasing is still present, but fainter, and it would probably be difficult to see in real life images.


Dan.
10-30-2013, 03:07 PM   #21
Veteran Member
traderdrew's Avatar

Join Date: Jun 2010
Location: Florida
Photos: Gallery | Albums
Posts: 639
ISO 1600 looks very good. I don't like to go above ISO 1250 on my K-5 II unless I'm not going to crop the photo. There is this thinking the more megapixels you have then you will have more noise. That might be true in some cases but I haven't bought into that thinking for the past year or two
10-30-2013, 03:31 PM   #22
Veteran Member




Join Date: May 2012
Posts: 733
QuoteOriginally posted by traderdrew Quote
ISO 1600 looks very good. I don't like to go above ISO 1250 on my K-5 II unless I'm not going to crop the photo. There is this thinking the more megapixels you have then you will have more noise. That might be true in some cases but I haven't bought into that thinking for the past year or two
That is only true if you look at 100% crops. But in reality the high megapixel 100% crop shows more noise because it is magnified more. At the same level of magnification the high megapixel picture will have no more and no less noise than the low megapixel one.

Put another way, the difference in noise (increase) of the high megapixel sensor will only be perceptible at a magnification beyond the point where you start seeing the individual pixels of the lower megapixel sensor.

All the above is assuming same sensor size, same level of technology (that is not comparing an old high MP sensor to a modern low MP sensor)and that the pixel density is not taken to the extreme that interpixel gaps start to be a significant fraction of the pixel sensitive area.

10-30-2013, 03:32 PM - 1 Like   #23
Veteran Member




Join Date: Mar 2007
Location: U.K.
Posts: 685
QuoteOriginally posted by PePe Quote
I have been wondering about this too.
There is nothing about this in the information released. Both functions being intended for continuous use, I quess they would have to openly announce it if the systems would interfere with each others.
These systems do use the same hardware. However, I would imagine the functions are different enough to be run in separate modes parallel to each others.
Would definitely like to see this confirmed by Pentax.
It's much the same as a loudspeaker playing all the different notes of music at the same time. It WILL work.
10-30-2013, 03:38 PM - 1 Like   #24
Pentaxian
dosdan's Avatar

Join Date: Dec 2007
Location: Brisbane, Australia
Photos: Gallery | Albums
Posts: 1,741
QuoteOriginally posted by traderdrew Quote
There is this thinking the more megapixels you have then you will have more noise.
Please read this post which demonstrates that this is no longer true: https://www.pentaxforums.com/forums/16-pentax-news-rumors/238970-pentax-k-3-h...ml#post2534630

Dan.
10-30-2013, 08:19 PM   #25
Senior Member
scole's Avatar

Join Date: Jan 2009
Location: Everett, WA
Posts: 216
Out of curiousity, which setting (1 or 2) would be the equivalent of what a K5's normal operation?
10-30-2013, 09:40 PM - 1 Like   #26
Pentaxian
dosdan's Avatar

Join Date: Dec 2007
Location: Brisbane, Australia
Photos: Gallery | Albums
Posts: 1,741
QuoteOriginally posted by scole Quote
Out of curiousity, which setting (1 or 2) would be the equivalent of what a K5's normal operation?
Neither. AA2 has the most filtering, but with the higher res of the K-3 vs K5 (24MP vs 16.2MP), there is a difference.

I've gone back to IR's K-5 review and taken the Optical/Electronic viewfinder accuracy JPG down the bottom of the samples as it uses the same test image as used in the K-3 AA examples. Unfortunately, it's only there as a JPG rather than a DNG. It has the same shutter speed, f-stop & ISO as what was used in the K-3 sample. I've enlarged it to be the same pixel size as represented in the higher-res K-3 shots, using Bicubic resizing with a sharpening of 100 in PSPX4, and then added it down the bottom i.e. AA0, AA1, AA2, K5.

https://www.dropbox.com/s/2qj5jgvvy4jpzng/AA%20comparison%20inc%20K5.jpg

Dan.

Last edited by dosdan; 10-30-2013 at 09:49 PM.
10-31-2013, 10:24 AM   #27
Veteran Member
devorama's Avatar

Join Date: Apr 2009
Location: Chicago, IL
Photos: Gallery | Albums
Posts: 638
If the intermediate AA1 setting is really a linear vs circular shake of the sensor (as in AA2), this would mean it would only reduce the moire in one direction. I'd love to see these same tests repeated with the target rotated 90 and 45 degrees.
10-31-2013, 03:10 PM   #28
Veteran Member




Join Date: May 2012
Posts: 733
From where does the linear vs circular thing come? I've seen it mentioned several times in this forum but not a source for it.

I would find it quite odd if this were so. IMHO circular motion is the only reasonable way to go otherwise two things go wrong. First it will only reduce moire in one direction and secondly it could also cause the moire to be coloured. If the sensor is shaken horizontally or vertically then the de-moire effect will only 'mix' the green with red and the green with blue but there will be no mixing between the blue and red pixels. This will probably end you up with a less distinct but more colourful moire than if the mechanical AA was switched off.

I think it much more likely that it is just a matter of changing the radius of the oscillation with the lower mode giving only partial overlap between the pixels while the higher one would give full overlap. You would perhaps have the lower mode oscillating in a radius of half a pixel while the higher mode would be a radius of one full pixel so that all four colour pixels are fully mixed.

@Dosdan, the AA2 mode seems not to fully eliminate Moire but retains a much higher resolution than the K-5II. The increased resolution looks to be considerably more than one would expect from merely the 16->24 MP sensor resolution. I'm wondering whether there might be a focus issue in the K-5II shot.
10-31-2013, 04:14 PM   #29
Pentaxian
dosdan's Avatar

Join Date: Dec 2007
Location: Brisbane, Australia
Photos: Gallery | Albums
Posts: 1,741
QuoteOriginally posted by lister6520 Quote
he increased resolution looks to be considerably more than one would expect from merely the 16->24 MP sensor resolution. I'm wondering whether there might be a focus issue in the K-5II shot.
Could be.

Theoretically you would expect approx 20% difference in the extinction frequency. Other factors: raw processing instead of JPEG; the strength of the AA filter in the K-5.

Dan.
11-13-2014, 08:27 AM - 1 Like   #30
Inactive Account




Join Date: Aug 2014
Posts: 76
QuoteOriginally posted by dosdan Quote
Could be.

Theoretically you would expect approx 20% difference in the extinction frequency. Other factors: raw processing instead of JPEG; the strength of the AA filter in the K-5.

Dan.
Ricoh's white paper on the AA Filter:
http://c2b6d376b97bcc466063-5420c200a1f030d1394a9548df6eadbd.r5.cf2.rackcdn....aper_16Dec.pdf
Thought it might prove helpful to some, it isn't easy to locate on the website.

---------- Post added 11-13-14 at 09:31 AM ----------

QuoteOriginally posted by PePe Quote
I have been wondering about this too.
There is nothing about this in the information released. Both functions being intended for continuous use, I quess they would have to openly announce it if the systems would interfere with each others.
These systems do use the same hardware. However, I would imagine the functions are different enough to be run in separate modes parallel to each others.
Would definitely like to see this confirmed by Pentax.
Ricoh's white paper on the AA Filter:
http://c2b6d376b97bcc466063-5420c200a1f030d1394a9548df6eadbd.r5.cf2.rackcdn....aper_16Dec.pdf
Thought it might prove helpful to some, it isn't easy to locate on the website.
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!
dslr, filter, functions, k-3, k3, paper, pentax, pentax k-3, systems, website
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
K-5 IIs -- The Pros and Cons of Omitting an AA-Filter Class A Pentax DSLR Discussion 114 12-18-2012 10:22 PM
Pentax K-30 Test Images (Finally!) Welfl Pentax K-30 & K-50 24 06-24-2012 09:54 PM
Q test with images and high iso test pictures StigVidar Pentax Q 9 10-05-2011 08:23 PM
Some K-5s shipped with stains on the sensor/AA filter? Peter Fang Pentax K-5 & K-5 II 292 03-06-2011 01:57 PM
For Sale - Sold: Pentax MZ-3 AutoFocus Film SLR Camera with AA Battery Pack FG and Cable Releas frank Sold Items 3 07-20-2009 03:18 AM



All times are GMT -7. The time now is 04:27 PM. | 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