Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
03-24-2014, 02:37 AM - 1 Like   #1
Site Supporter
Stone G.'s Avatar

Join Date: Jun 2010
Location: North Zealand, Denmark
Photos: Gallery | Albums
Posts: 1,511
My Life with Pentax DSLR Sensors

I have had (have) the Pentax *ist DL (6Mp - 0.0078mm pixel pitch), the Pentax K-200D (10Mp - 0.0061mm) and the Pentax K-5 (16 Mp - 0.0048mm). All have served me very well and I have been (am) very pleased with them all. Now, that I have taken a deep breath and bought the Pentax K-3 (24Mp and 0.0039mm pixel pitch), I thought that I had better investigate the benefits of these upgrades, should my Chief Financial Contoller some day ask.

So, the question is: Has sensor- and associated technology evolved in such a way, that the ever higer spatial resolutions of the sensors have not been lost in the inevitable photon noise and smaller well capacity dictated by the laws of physics - and the camera manufacturers' tricks in firmware to smoothen image data and hide that noise???

To test that, I set up a 200mm f/2.8 lens on a tripod and pointed it towards an antenna on the roof of an office building 230 meters away:



I then exchanged cameras and shot a short series of images at ISO 200 and ISO 800. Next, I made 400 x 400 pixel crops around the antenna and here is the result:



Click on image to see the full, non-resized 400 pixel crops.

Unfortunately, the ligtening conditions changed rather much during the test, but I think the result is still clear enough to speak for itself.

Anyway, to me it appears obvious that the meaningfull pixel race didn't stop with the K-5 16 Mp sensor - at least not when used with a quality lens with excellent optical resolution. Even at ISO 800, there is finer detail and better image quality with the K-3 sensor than ever before.

Is this also / to a high degree attributable to the disabling of the AA filter simulator? Further tests with the three levels of AA filtering will show. I have the Moon with its incredibly delicate detail (when seeing permits) in mind as my next test target......


Last edited by Stone G.; 03-24-2014 at 02:45 AM.
03-24-2014, 02:52 AM   #2
Site Supporter
geomez's Avatar

Join Date: Mar 2013
Location: Preskit Arizona
Photos: Gallery
Posts: 1,561
Neither images are linked.
03-24-2014, 02:55 AM   #3
Site Supporter




Join Date: Apr 2011
Photos: Albums
Posts: 2,893
Interesting set of comparisons.

I recently got a K3 and have been using it with the Super Takumars and SMC Takumars exclusively. It is a step up on my prior K100DS which seems to perform like your *istD and K2000. I was attributing much of the improvement to the much greater ease of focusing with the K3 screen. A couple of my colleagues were impressed by what I showed today.
03-24-2014, 03:36 AM   #4
Pentaxian
calsan's Avatar

Join Date: Nov 2008
Location: Perth, Australia
Photos: Gallery
Posts: 1,367
Terrible that there are Chief Financial Officers in Denmark also.
Quite amazing difference. Funnily, the best image seems to be at ISO 800 with the K-3, or at least there's no difference between 100 and 800 in resolution.

03-24-2014, 04:36 AM   #5
Site Supporter
Stone G.'s Avatar

Join Date: Jun 2010
Location: North Zealand, Denmark
Photos: Gallery | Albums
Posts: 1,511
Original Poster
QuoteOriginally posted by geomez Quote
Neither images are linked.
Sorry, I had grave troubles in inserting the link. Don't know why, but it works now!

QuoteOriginally posted by calsan Quote
Funnily, the best image seems to be at ISO 800 with the K-3, or at least there's no difference between 100 and 800 in resolution.
Well, I think that's attributable to the ever-changing light. As you can see, light was somewat better with the K-3 800 ISO image and, hence, the contrast. My conclusion is that the 200 and 800 ISO images are quite identical and that - generally speaking - under good ligthening conditions, ISO 800 isn't really a problem with K-3.
03-24-2014, 05:11 AM   #6
Pentaxian
bdery's Avatar

Join Date: Mar 2008
Location: Quebec city, Canada
Photos: Gallery
Posts: 5,661
One perk of the K-3 is clearly resolution when using a sharp lens. another is high ISO performances. Shoot at ISO 1600 (not sure your first camera goes above that) and you'll see that the K-3 (and K-5) perform really well there too.

Incidentally, to test pure resolution, you'd be better at base ISO.
03-24-2014, 06:10 AM   #7
Site Supporter
Stone G.'s Avatar

Join Date: Jun 2010
Location: North Zealand, Denmark
Photos: Gallery | Albums
Posts: 1,511
Original Poster
QuoteOriginally posted by bdery Quote
One perk of the K-3 is clearly resolution when using a sharp lens. another is high ISO performances. Shoot at ISO 1600 (not sure your first camera goes above that) and you'll see that the K-3 (and K-5) perform really well there too.
Well, the *ist DL goes to ISO 3200 (very noisy - at least in low light) and the K-200D goes to ISO 1600 (so-so, noise wise), but exposure times are limited to 1/4000 s and thus, these two cameras constrain what is realistic/feasible for a comparative test at both high and low ISO in broad daylight.

QuoteOriginally posted by bdery Quote
Incidentally, to test pure resolution, you'd be better at base ISO.
Your are of course right, but the base ISO of the *ist DL is 200 and in an "open-air" test - shooting over 230 meters with a fast telephoto lens in turbulent air, you will want to keep exposure times reasonably identical - if not, seeing conditions will add to the uncertainty in the interpretation.

Anyway, with my K-200D, K-5 and K-3 I really never see much of a difference.
03-24-2014, 09:45 AM   #8
Veteran Member




Join Date: May 2008
Location: SF Bay Area
Posts: 417
THanks for posting. This test confirms what I have been observing with my K3. ISO 10,000 on K3 = ISO 1250/ISO 1600 on K10D. There simply is no comparison. I very regularly shoot up to ISO 5000 with very little to no noise. Combine that with the smaller,more sensitive AF points and K3 is simply impossible to beat especially in low light... Call me 'amazed'....

03-24-2014, 08:18 PM   #9
Site Supporter
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 26,233
QuoteOriginally posted by Stone G. Quote
shooting over 230 meters with a fast telephoto lens in turbulent air, you will want to keep exposure times reasonably identical
Forgive me, but I just gotta ask...Why were you shooting at distance? You can test the same thing with a shorter focal length in still air inside.


Steve
03-24-2014, 08:36 PM   #10
Pentaxian
carrrlangas's Avatar

Join Date: Mar 2012
Location: Cottbus (Berlin)
Photos: Gallery
Posts: 1,738
QuoteOriginally posted by calsan Quote
Funnily, the best image seems to be at ISO 800 with the K-3, or at least there's no difference between 100 and 800 in resolution.
Actually, I think this is due to apparent sharpness due to higher noise in the ISO 800 shot. Take a look here: Tutorials – Sharpness

Thanks for the useful comparison. I am still reluctant to higher megapixel count but sensor technology is quite complex nowadays, far form well capacity and SNR..
03-24-2014, 09:01 PM   #11
Pentaxian




Join Date: Jan 2011
Location: New York
Posts: 3,018
QuoteOriginally posted by Stone G. Quote
SNIP ... Is this also / to a high degree attributable to the disabling of the AA filter simulator? Further tests with the three levels of AA filtering will show. I have the Moon with its incredibly delicate detail (when seeing permits) in mind as my next test target......
Glad I read all the way through to the end before posting. I was about to comment that no AA filter rather than more pixels might be why the K-3 is sharper than the K-5. Thanks for the comparison.
03-25-2014, 02:29 AM   #12
Site Supporter
Stone G.'s Avatar

Join Date: Jun 2010
Location: North Zealand, Denmark
Photos: Gallery | Albums
Posts: 1,511
Original Poster
QuoteOriginally posted by stevebrot Quote
Forgive me, but I just gotta ask...Why were you shooting at distance? You can test the same thing with a shorter focal length in still air inside.


Steve
That's a perfectly legitimate question. The brief answer is: I wanted a "real-life" test.

While I do not dicard tests under laboratory-like, strictly controlled conditions as useless, I personally - as a happy amateur - have come to realize that testing my gear under realistic conditions tells me more about what I have and what I can use it for / hope for.

And the good thing about a comparative test is that one can indeed examinate the trends/features one is looking for under less than ideal conditions, for as long as these conditions do not change drastically during the test.
03-25-2014, 02:35 AM   #13
Site Supporter




Join Date: Apr 2011
Photos: Albums
Posts: 2,893
QuoteOriginally posted by Stone G. Quote
That's a perfectly legitimate question. The brief answer is: I wanted a "real-life" test.

While I do not dicard tests under laboratory-like, strictly controlled conditions as useless, I personally - as a happy amateur - have come to realize that testing my gear under realistic conditions tells me more about what I have and what I can use it for / hope for.

And the good thing about a comparative test is that one can indeed examinate the trends/features one is looking for under less than ideal conditions, for as long as these conditions do not change drastically during the test.
I think I might say that differently. As I use my stuff I learn what it does under various circumstances. Eventually I will be able to intuitively reach for the right bit of kit to achieve the result I want. Then the technical kit becomes a means of doing creative work, even 'art'.
03-26-2014, 04:44 AM   #14
New Member




Join Date: Feb 2011
Location: Seattle
Photos: Gallery
Posts: 6
I think your test may not be perfect but it gives a good indication of the improvement you can expect. What I find interesting is that the resolution of the K-3 is just about as good as the resolution of the very best lenses available. In other words the next generation of sensor (36Mp?) may have better resolution than any lens available. This leads me to think that maybe Pentax Ricoh should stop at 24Mp and concentrate on improving dynamic range?
03-31-2014, 03:51 AM   #15
Site Supporter
Stone G.'s Avatar

Join Date: Jun 2010
Location: North Zealand, Denmark
Photos: Gallery | Albums
Posts: 1,511
Original Poster
QuoteOriginally posted by stevebrot Quote
You can test the same thing with a shorter focal length in still air inside.

QuoteOriginally posted by Halchemy Quote
What I find interesting is that the resolution of the K-3 is just about as good as the resolution of the very best lenses available. In other words the next generation of sensor (36Mp?) may have better resolution than any lens available.

QuoteOriginally posted by DeadJohn Quote
I was about to comment that no AA filter rather than more pixels might be why the K-3 is sharper than the K-5.

As also suggested by Halchemy, I belive that we are approaching sensor (spatial) resolution where optical resolutions matters. So, 'any lens' will not do for this kind of comparison tests. But here then, are some still air tests taken with my DA*200 lens at about closests focusing distance (1400mm). As my "amateur test chart", I use a stamp because, as much as with a real test-chart, the incredible fine details in the gravure usually makes it easier for me to see differences in resolution.



And first let's take a look at the differences between "Full" (Level 2), "Medium" (Level 1) and no (off) AA-filter simulation:



The difference between level 2 and level 1/off is quite obvious: The AA-filter simulator does have the expected effect on resolution. The difference between level 1 and off appears much more subtle to my eye.

Next, let's have a look at the differences between the K-5 and the A-3 sensor (with no AA-simulation) under these controlled conditions. I took series of 5 images at f/5.6 and f/8 and picked two images from each series - one for the left column and one for the right column crops - at random. And here is the result (K-5 images on top and K-3 images at the bottom):


K-5 and K-3 with DA*200mm lens at f/5.6:





K-5 and K-3 with DA*200 at f/8:


I think it is fair and obvious to conclude that a high-resolution lens like the DA*200 does benefit from the finer resolution of the K-3 sensor. However, when one compares with the AA-filter simulation image above, the difference appears to me to be as much due to the lack of AA-filter as to the smaller pixel pitch of the K-3 sensor.

To take this one step further, I made yet another test with my dedicated macro lens, Tamron Adaptall-2 35-80mm f/2.8-3.8 (Model 01A). This lens is renowned for it fine resolution characteristics but should, on paper at least, have a smaller resolving power than the DA*200. On the other hand, it allows for a larger viewing angle at a closest focusing distance of only 270mm. Again, the K-5 images are on top and K-3 images at the bottom:



So, what does one conclude here? I am somewhat in doubt. The larger scale of the K-3 images makes it easier to see the tiniest details and i THINK that I can still see a K-3 resolution advantge over the K-5 here. But I belive that halchemy may well be right: We are approaching a situation where only the lenses with the highest optical resolution will benefit from more pixels.........

Last edited by Stone G.; 03-31-2014 at 04:06 AM.
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!
*ist dl, camera, detail, dslr, image, iso, k-200d, k-3, k-5, noise, pentax, pentax dslr, pentax dslr sensors, photography, pixel, quality, resolution, result, sensor, test
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Simple DSLR with my old Takumar lens Hosperantasa Pentax SLR Lens Discussion 6 07-07-2013 12:00 AM
DSLR vs SLR camera life Gary Pentax DSLR Discussion 27 02-23-2013 05:11 PM
I hope Pentax keeps going with Crop Sensors coreyhkh Pentax DSLR Discussion 76 01-02-2013 08:52 AM
Having a blast with my first DSLR DLCrow Welcomes and Introductions 4 12-07-2012 11:04 AM
Greatest Threat to DSLR Sensors mithrandir Photographic Technique 8 07-29-2010 07:17 PM



All times are GMT -7. The time now is 10:24 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