Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
05-31-2008, 10:52 PM   #1
Inactive Account




Join Date: Apr 2008
Posts: 464
K200D Noise Performance Chart - Pixel Peepers Rejoice!

I got a desire today to make a noise chart, as I wanted to see the various levels of noise reduction that the K200D would give.


K200D with 18-55 ALII kit lens:
-------

Methodology:


1) I placed various items on a small table with a lamp shining on them.

2) I placed my camera on a stable surface with most items in full frame.

3) Camera set to DNG files, long shutter noise reduction always off. Av mode with aperature at f/8, 2sec timer, custom white balance.

4) Shots were taken at ISO 100-1600 for a baseline, then ISO 400-1600 "Weak High ISO" and ISO 400-1600 "Strong High ISO" settings.

5) I then ran the baseline ISO1600 image through NoiseNinja plugin-- default filter settings but with the USM set to around 90 I believe, 1.0 pixels.

6) After all these images were taken, I then lined them up into layers in Adobe Photoshop and cropped 4 various positions (this ensured near-pixel-accurate crops of the same area). I exported these crops as files, then imported them in a new canvas and named them.

--------------------

The Results (Click on any image for a full size version to see differences!)

First, I found right off the bat I would discard the weak ISO noise reduction settings -- it was virtually indistinguishable from the strong.

Trust me, I pixel peeped a heck of a lot during this procedure -- You could give me A/B examples of Weak/Strong and I wouldn't be able to tell the difference. I spent 10 minutes in Lightroom in "comparison" mode looking at strong/weak versions...they're basically clones.


Without further ado, the pictures:


First we have the Baseline ISO100 no NR picture


And now we have the Baseline ISO1600 picture


Here we have ISO1600 Strong NR


And here we have ISO1600 baseline through NoiseNinja


Now we have the first chart. This one focuses on large blocks of same color


Finally we have the second chart. This focuses on details




-------------------------------------------------


What did I learn today?

1) K200D "weak" vs "Strong" high ISO noise reduction is virtually the same.

2) In a non-crop, non-zoom situation, it's difficult to tell differences between some ISO100 and ISO1600 images.

3) In crop situations, high ISO might come back to bite you in the butt, but in-camera noise reduction does tend to help without washing out details.

4) NoiseNinja does a great job at noise reduction, but to maintain detail you really have to play around with the parameters. I'm still hesitant to run it through a batch mode because each picture needs tweaking for maximum performance.

5) I spend way too much time doing crazy things like this. And it's tiring!

06-01-2008, 04:20 AM   #2
Veteran Member




Join Date: Jan 2008
Location: Ste-Anne des Plaines, Qc., Canada
Posts: 2,014
If you do the same test in poor lighting, the noise would likely be more obvious. Longer exposures generate more noise.
06-01-2008, 04:49 AM   #3
Inactive Account




Join Date: Apr 2008
Posts: 464
Original Poster
QuoteOriginally posted by flyer Quote
If you do the same test in poor lighting, the noise would likely be more obvious. Longer exposures generate more noise.

Oh yes, of course

To be fair, this was pretty poor lighting, as this single lamp is only something like a 100W equivelant flourescent bulb.

If you look at the origial files at 100% magnification, the noise is quite evident, though yes looking at the image fitting to a screen you don't notice very much at all.
06-01-2008, 06:51 AM   #4
Veteran Member
arbutusq's Avatar

Join Date: Sep 2006
Location: Vancouver BC canada
Photos: Gallery
Posts: 380
Generally if you shoot RAW or DNG the in camera NR settings should have no effect on noise reduction in Lightroom. Still your results look pretty good.

Would it be possible to repeat the experiment with lower light? I'm so demanding aren't I....

06-01-2008, 08:55 AM   #5
Inactive Account




Join Date: Apr 2008
Posts: 464
Original Poster
QuoteOriginally posted by arbutusq Quote
Generally if you shoot RAW or DNG the in camera NR settings should have no effect on noise reduction in Lightroom. Still your results look pretty good.

Would it be possible to repeat the experiment with lower light? I'm so demanding aren't I....
Yes true.

The whole point of the experiment was to look at various methods of noise reduction and their effectiveness. I only used Photoshop CS3 in this test, though I may have been browsing with Lightroom at the time also (I have no idea, it was late!)

I wanted to see:

1) An image with no noise reduction
2) Whether any difference really existed between "weak" and "strong" NR in camera (in these cases, none, so I discarded "weak" images)
3) How much better/worse Noise Ninja would in noise reduction on a baseline 1600.


But, perhaps I am misunderstanding -- Is in-camera noise reduction applied to both RAW and JPEG or just JPEG?

Truth be told, I cannot see any difference in my images between 1600STRONG and 1600NONE - they look exactly the same to me. Perhaps I should run the test again in JPEG?

Also I will consider very low light situations -- This type of test takes several hours to properly finish, though.
06-01-2008, 10:47 AM   #6
Site Supporter
ChipB's Avatar

Join Date: Nov 2007
Location: Austin, TX area
Photos: Gallery | Albums
Posts: 1,518
Thanks for going the "extra mile"!!

cputeq, you obviously spent a LOT of time and effort on this endeavor - very interesting to (actually) not see a whole bunch of difference. Truth be told, I'm not too much of a pixel peeper, but I am interested in seeing tests like this as they tend to show what the camera is capable of.

I'm an old film guy, so have been very reluctant to go over ISO 400 - but tests like these give me hope for low light/high ISO performance.

Thanks again for your efforts!!
06-01-2008, 11:09 AM   #7
Loyal Site Supporter
Canada_Rockies's Avatar

Join Date: Nov 2007
Location: Sparwood, BC, Canada
Photos: Gallery | Albums
Posts: 9,147
QuoteOriginally posted by ChipB Quote
cputeq, you obviously spent a LOT of time and effort on this endeavor - very interesting to (actually) not see a whole bunch of difference. Truth be told, I'm not too much of a pixel peeper, but I am interested in seeing tests like this as they tend to show what the camera is capable of.

I'm an old film guy, so have been very reluctant to go over ISO 400 - but tests like these give me hope for low light/high ISO performance.

Thanks again for your efforts!!
Chip, I have found that even at 1600 ISO on the K10D, the noise beats any film grain hands down. There is a sample on my Flickr site taken in miserable lighting conditions. Check the detail in the corners, and the noise.
06-01-2008, 03:06 PM   #8
Inactive Account




Join Date: Mar 2008
Location: Dallas Texas
Posts: 319
QuoteOriginally posted by Canada_Rockies Quote
Chip, I have found that even at 1600 ISO on the K10D, the noise beats any film grain hands down. There is a sample on my Flickr site taken in miserable lighting conditions. Check the detail in the corners, and the noise.
Canada, ahh those days when I use to use Kodak Tri-X 400 ASA B/W film and would cringe everytime I'd make an 8x10 print in my makeshift darkroom aka bedroom growing up hoping it would look like my 32 ASA Panatomic-X prints, we have come a long ways haven't we and it's all good and none of that nasty spoiled soup smell either. LOL

Barry

06-01-2008, 07:04 PM   #9
Veteran Member
Jewelltrail's Avatar

Join Date: May 2008
Location: Rhode Island
Photos: Albums
Posts: 4,180
cputeq

Perhaps you should quit your day job and send your resume over to DPReview--LOL! Seriously, this is an ambitious & diligent undertaking--I only wish you had a K20 so I could pick your brain on that camera's high ISO capabilities.

Efforts and contributions like this are very much appreciated!

Best Regards,

Ernest



"Humanity subdues inhumanity as water subdues fire."

Mencius 6A:18.
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, chart, dslr, iso, iso1600, k200d, noise, noiseninja, photography, picture, reduction, settings
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Sharp lens for pixel-peepers?? EliotK Pentax SLR Lens Discussion 45 02-12-2010 12:10 PM
K-7 video mode and sharpening for pixel-peepers richtrav Pentax DSLR Discussion 0 08-19-2009 12:28 AM
For you pixel peepers... wildman Pentax SLR Lens Discussion 12 01-27-2009 10:33 PM
Wanted! Pixel Peepers - new Sigma 17-70mm superfuzzy Pentax SLR Lens Discussion 9 01-15-2008 07:39 PM
Hay pixel peepers-Help regken Photographic Technique 1 12-02-2007 11:30 AM



All times are GMT -7. The time now is 02:10 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