Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
09-26-2012, 01:38 AM   #46
Veteran Member




Join Date: Oct 2007
Location: France
Photos: Albums
Posts: 804
QuoteOriginally posted by Zafar Iqbal Quote

Look at these for comparison: Nikon D800 vs D800E Digital SLR Review

[
Surprizingly the comparison shots between D800 and D800E are made at such aperture that diffraction becomes noticeable (f7.1, f11 and f16)
A test at f4 or f5.6 would be more convincing.
Note also that the D800 has a weak AA filter (see moiré patterns on the test made on printed document)
As for Pentax, I have understand on photozone web site that Klaus, for the lens test, use a K5. He found the AA filter too strong and tried to remove (or make it removed by a specialized company) but did notmanage do have the AA filter removed
Let's see how the new K52s will perform.

09-26-2012, 01:40 AM   #47
Pentaxian
gazonk's Avatar

Join Date: Feb 2009
Location: Oslo area, Norway
Photos: Gallery | Albums
Posts: 3,512
QuoteOriginally posted by Ishpuini Quote
Actually I heard the opposite. High ISO would have marginally less noise without an AA filter. I'm not sure how this would work though... I've looked around for a technical explanation of this, but found none...

Wim
Just guessing: The AA filter means that more sharpening is required. The sharpening will amplify noise.
09-26-2012, 01:40 AM   #48
Veteran Member




Join Date: Oct 2007
Location: France
Photos: Albums
Posts: 804
QuoteOriginally posted by mirelay Quote
Will the removal of the AA filter lower the high ISO performance or it has nothing to do with that?
If raw image is sharper due to AA removal there is much room left to use stronger noise reduction process that reduce image resolution
09-26-2012, 03:37 AM   #49
Loyal Site Supporter
baro-nite's Avatar

Join Date: Aug 2011
Location: North Carolina, USA
Photos: Gallery | Albums
Posts: 4,367
QuoteOriginally posted by goubejp Quote
Surprizingly the comparison shots between D800 and D800E are made at such aperture that diffraction becomes noticeable (f7.1, f11 and f16)
A test at f4 or f5.6 would be more convincing.
dpreview.com did such testing:

Nikon D800 Review: Digital Photography Review

Sample quote:

QuoteQuote:
In side by side print comparisons, we found that that discerning an improvement in D800E output over the stock D800 required examining 20 x 30 inch prints at very close viewing distances. In fact, when holding those same prints at arm's length, the differences were all but impossible to detect. Safe to say that if you're deciding between these two camera versions with an eye towards print output, you'd need to be producing very large prints or do significant cropping to gain much practical benefit to the D800E's increased resolving capabilities.


09-26-2012, 03:41 AM   #50
Pentaxian
Zafar Iqbal's Avatar

Join Date: Oct 2010
Photos: Gallery
Posts: 1,167
goubejp, the point I've been getting at was that the size depicted was too small for AA vs no AA to show differences. Same should go for any/most typical pixel level anomalies.

Following aren't exactly great examples nor taken for the purpose of proving my point but on which photo did I use the sharpest lens - yes, I believe the differences are large enough to make my point. One could cheat and read the exif to get a better idea but the answer isn't the key - the original photos are even if these photos aren't depicted at same sizes nor are they photos of prints, in which case it should be even harder to tell (original digital photo vs. fully framed photo of world class print - I'd say the digital photo would win in clarity, no?). I'll let these simmer for a bit and post links to high-res photos later.





Edit: baro-nite, ah - glad to see I'm not alone on this one.

Last edited by Zafar Iqbal; 08-19-2013 at 09:10 AM.
09-26-2012, 04:20 AM   #51
Pentaxian




Join Date: Jun 2009
Location: GMT +10
Photos: Albums
Posts: 10,851
QuoteOriginally posted by mirelay Quote
Will the removal of the AA filter lower the high ISO performance or it has nothing to do with that?
According to DXOMark, the D800E (@2979) actually has a slightly better low-light ISO score than the D800 (@2853).

Which makes sense. The more stuff you put in front of the sensor (like AA filters) the more light you block.
09-26-2012, 05:09 AM   #52
Veteran Member




Join Date: Oct 2007
Location: France
Photos: Albums
Posts: 804
QuoteOriginally posted by baro-nite Quote
dpreview.com did such testing:

Nikon D800 Review: Digital Photography Review

Sample quote:
Correct; I was thinking of these tests I had read a couple of month ago but was not able to remember where they were; thanks for posting.
I illustrate some points I mentioned :
the D800 AA filter is weak - moiré is visible on both test comparison between D800 and D800 E from f2.8 to f8. My believe is that the K5 filter is stronger thus the difference between K5 and K5 2 s may be greater.
The difference in resolution is visible from f4 to f5.6
From f8 onward, the diffraction causes blur that make the AA filter not useful; thus if there is risk of moiré stopping down to f11 should naturally remove it
(I assume an equvalence in pixel density between the D800 and the K5)
Regards
09-26-2012, 05:28 AM   #53
Pentaxian
normhead's Avatar

Join Date: Jun 2007
Location: Near Algonquin Park
Photos: Gallery | Albums
Posts: 25,869
QuoteQuote:
In side by side print comparisons, we found that that discerning an improvement in D800E output over the stock D800 required examining 20 x 30 inch prints at very close viewing distances. In fact, when holding those same prints at arm's length, the differences were all but impossible to detect. Safe to say that if you're deciding between these two camera versions with an eye towards print output, you'd need to be producing very large prints or do significant cropping to gain much practical benefit to the D800E's increased resolving capabilities.
To notice a difference on a D800 at 36 MP and to notice a difference on a 16 MP K-5 are two different things. T0 be the same size, say 20x30 you need to magnify a K-5 image 22 MP, larger than the original file. Any lack of sharpness will be magnified more than 2x compared to a D800. SO if you can see it at 20x30 on a D800 I'd expect to see it at 16x 20 on a K-5 image, and I would expect to see the difference on a 13x19 which I is my standard size for non-canvas printing. Couple that with the fact that the K-5 filter is fairly strong, and there's likely to be more difference between a K-5 and a K-5 IIs than between a D800 and a D800e and I'm expecting great things from this camera. WHile my K-5 will still have a home, the K-x and K20D are going to find new homes. ALthough, I may have to retire the K20D to the display case... I've had it 4 years now, we have a history.

09-26-2012, 07:22 AM   #54
Loyal Site Supporter




Join Date: Apr 2010
Location: Tennessee
Posts: 5,803
The removal of the AA filter on an APS-C will indeed be more noticeable because of the sensor size.

A3+ is 158,907mm^2
D800E sensor size is 864mm^2
K-5IIs sensor size is 370mm^2

To print at A3+ the image from a D800E has to be enlarged 183 times.
To print at A3+ the image from a K-5IIs has to be enlarged 430 times.

If DPR says that the difference in prints does not show up until 20x30 on the D800E then it would start to show up on prints from the K-5IIs at around 11x14.

Of course lens selection/aperture/subject matter all play a roll in how much difference you will see. Most lenses wide open will be soft enough to negate the added resolution and diffraction will do the same at really small apertures. An image taken with a 31mm LTD at F/5.6 should show a significant difference is detail.
09-26-2012, 07:23 AM   #55
Pentaxian
Transit's Avatar

Join Date: Aug 2007
Location: Whanganui NZ
Photos: Gallery | Albums
Posts: 4,077
QuoteOriginally posted by rawr Quote
..... The more stuff you put in front of the sensor (like AA filters) the more light you block.
My reading of the Nikon setup indicated there was just as much 'stuff' in front of the sensor,
just a different arrangement that eliminated the AA effect.
might be wrong, it has happened
09-26-2012, 05:21 PM   #56
Loyal Site Supporter
baro-nite's Avatar

Join Date: Aug 2011
Location: North Carolina, USA
Photos: Gallery | Albums
Posts: 4,367
QuoteOriginally posted by normhead Quote
To notice a difference on a D800 at 36 MP and to notice a difference on a 16 MP K-5 are two different things.
Good points here and above, and I for one don't need much convincing. I'm already planning to buy one (landscape and macro are two of my main subjects).
09-27-2012, 12:47 AM   #57
Site Supporter




Join Date: Jul 2011
Posts: 1,041
Obvisouly there are MANY cameras around that do not have a AA filter - the Kodak DCS cameras from 15 years ago did not have in internal AA filter, medium format backs don't have one and Leica M cameras also work without AA filter. The K5IIs will be one of them and honestly I am not sure that Pentax already has a clue what they just presented. It will all become much clearer in the next couple months. In theory this camera should be a great leap forward.
09-27-2012, 12:55 AM   #58
Site Supporter




Join Date: Jul 2011
Posts: 1,041
QuoteOriginally posted by Winder Quote
The removal of the AA filter on an APS-C will indeed be more noticeable because of the sensor size.

A3+ is 158,907mm^2
D800E sensor size is 864mm^2
K-5IIs sensor size is 370mm^2

To print at A3+ the image from a D800E has to be enlarged 183 times.
To print at A3+ the image from a K-5IIs has to be enlarged 430 times.

If DPR says that the difference in prints does not show up until 20x30 on the D800E then it would start to show up on prints from the K-5IIs at around 11x14.

Of course lens selection/aperture/subject matter all play a roll in how much difference you will see. Most lenses wide open will be soft enough to negate the added resolution and diffraction will do the same at really small apertures. An image taken with a 31mm LTD at F/5.6 should show a significant difference is detail.
The linear enlarging factor is not calculated by comparing the area (a squared value). APS-C to A3 is a linear enlargment of approximately 16x. The area increases by a factor of 16^2 or 256...

Lenses like the DA* 4/300 will probably benefit a lot wide open from removing the AA filter. You can actually feel that some potential is hold back by the AA filter.

You can always remove artefacts in PP - keeping the original resolution for most of the image - or place a hot mirror filter infront of the lens. No AA filter is an addition not a limitation.
09-27-2012, 04:40 PM   #59
Loyal Site Supporter




Join Date: Apr 2010
Location: Tennessee
Posts: 5,803
QuoteOriginally posted by zapp Quote
The linear enlarging factor is not calculated by comparing the area (a squared value). APS-C to A3 is a linear enlargment of approximately 16x. The area increases by a factor of 16^2 or 256...
I'm not trying to show the enlargement ratio. A 2x enlargement of a 4x5 negative yields and 8x10 print, but obviously this is 4 times the size of the original negative. The linear enlargement factor (ratio) is a hold over from the darkroom days. More and more small professional printers are working with and charging by the area of materials and ink consumed. If you have a coating applied over a print they will charge you by the size of the area covered.... not the enlargement ratio.

The enlargement factor for an A3+ print (13x19) from an Pentax APS-C is approximately 20.7x. The area of the print is 429 times greater than the area of the sensor. Compared to a true FF where the area of the print is 183 times greater than the area of the sensor. Using the area gives you a much more simplified view of what you are doing. The linear enlargement factor is misleading because when you say 2x you mean 4 times the size. When you say 4x you mean 16 times the original size. Let's drop the ratio/factor and say what we actually mean.
09-27-2012, 04:44 PM   #60
Pentaxian
Zafar Iqbal's Avatar

Join Date: Oct 2010
Photos: Gallery
Posts: 1,167
Links to original res versions of my previous photos:
Viggo's Mercedes - Zafar Iqbal's Photos - 85mm f/2
2011.05 May - Zafar Iqbal's Photos - 110mm f/5.6

I'm fully aware these are lame examples and these are obviously processed with added sharpness - still, the differences doesn't show in the smaller sizes.

I think I read in some other thread that Leica M9 does not have AA filter, so I looked for a RAW file and found this page: LEICA M9 Sample Images and DNG files

Open it up in PS - duplicate the layer and apply Gaussion blur with a value of 2. You now have a copy of the photo that looks like it was taken using an utterly crap lens. Scale way down to something similar to the K-II samples shot and toggle top layer on off - how big a difference do you see? Would you notice whatever difference you see if you saw them side by side as clearly as on the K-II samples shots?

I noticed this NASA Curiosity pic on Facebook - caption mentioned the photo was taken without AA filter: Catalog Page for PIA16134 I can no longer find that caption, but I'm 99,9% sure I read it correct. Does look sharper than M9 pics, though even though the tiff was likely created from highly compressed jpegs - anyhow, do the same thing with the scaling and blur. Even at a blur value at 1, it will be softer than the sharpness I can get from most of my lenses on my K-30 - and I don't have to pay attention to aperture size to get that sharpness. Increased (severe cases of) moire is what should really distinguish AA from non AA at smaller sizes.

In case anyone is in doubt - I'm not saying AA vs. No-AA filter is a load of crap. I'm saying the K-II samples posted by OP are doctored/fake.

Edit: who know, perhaps the camera used to take that K-II samples shot has greater sharpness towards the right - that's a possibility too. The one on left just looks too soft to be a AA thing.
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!
brochure, k5iis, pentax news, pentax rumors
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
10 things to know about the K5II/K5IIs jpzk Pentax K-5 16 09-24-2012 05:33 AM
DA*55 sample pics please axl Pentax SLR Lens Discussion 26 04-22-2011 04:39 AM
DA Limited, varying build quality from sample to sample. the swede Pentax SLR Lens Discussion 20 04-10-2011 01:52 PM
I like this camera, sample... FHPhotographer Pentax K-r 8 01-31-2011 11:04 PM
K-5: another sample schimi76 Pentax K-5 12 11-15-2010 06:49 AM



All times are GMT -7. The time now is 03:49 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