Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
03-06-2014, 07:01 PM   #16
Veteran Member
fgaudet's Avatar

Join Date: Oct 2013
Location: Ontario
Posts: 726
QuoteOriginally posted by Ex Finn. Quote
Could be the high ISO NR and camera settings.
Could also be the fact that if you want to highlight how good the camera is, you'll pick your best pictures. If you want to highlight how bad it is, you'll pick your worst shots...

As the OP said, this was in the worst possible conditions.

QuoteOriginally posted by j2photos Quote
which makes me wonder why I see so many high noise shots when mine doesnt seem to produce much noise at all.
Hence my theory on the inconsistent sensor quality. You don't see many people complaining about bad noise handling from the K3 (very few), but those who are complaining are dealing with very bad High ISO images. Like I experienced. I'm not a Pro but I have been into photography for over 20 years and I'm all about the numbers, the tests, the ratings and such. The K3 I got, made me feel as if I was back in 2004. Very harsh noise, ugly coloured noise. Nothing remotely looking like grain.

03-06-2014, 07:07 PM   #17
Veteran Member




Join Date: Jun 2010
Location: Manteca, CA
Photos: Gallery | Albums
Posts: 1,653
QuoteOriginally posted by fgaudet Quote
Could also be the fact that if you want to highlight how good the camera is, you'll pick your best pictures. If you want to highlight how bad it is, you'll pick your worst shots...
There are just as many reviews on how great this camera handles noise as there are ones that say it doesnt. I will post some high iso shots from mine when I have a moment as I am still at the office.

QuoteOriginally posted by fgaudet Quote
Hence my theory on the inconsistent sensor quality. You don't see many people complaining about bad noise handling from the K3 (very few), but those who are complaining are dealing with very bad High ISO images. Like I experienced. I'm not a Pro but I have been into photography for over 20 years and I'm all about the numbers, the tests, the ratings and such. The K3 I got, made me feel as if I was back in 2004. Very harsh noise, ugly coloured noise. Nothing remotely looking like grain.
I can see some possibility there.
03-06-2014, 07:23 PM   #18
New Member




Join Date: Mar 2014
Posts: 4
Have you tryed turning iso noise reduction off, was talk on you tube of it making images not sharp
03-06-2014, 07:47 PM   #19
Veteran Member
fgaudet's Avatar

Join Date: Oct 2013
Location: Ontario
Posts: 726
QuoteOriginally posted by TimC Quote
making images not sharp
Any NR basically reduces noise by applying a small blur to the image (or section of), taking some sharpness away, in body NR or a plugin in post. I've never been impressed with the results from the camera built-in NR I always do my NR in post if required.

03-06-2014, 07:52 PM   #20
Emperor and Senpai
Loyal Site Supporter
VoiceOfReason's Avatar

Join Date: Jun 2010
Location: Mishawaka IN area
Photos: Albums
Posts: 6,124
Wow, that is REALLY bad noise.
03-06-2014, 08:43 PM - 4 Likes   #21
Pentaxian
dosdan's Avatar

Join Date: Dec 2007
Location: Brisbane, Australia
Photos: Gallery | Albums
Posts: 1,741
Stagnant, the AR of your test image is 2.16:1. This indicates you've already cropped the original image. Can you please indicate whether this is a 100% crop, or what percentage of the original frame you cropped off the H & W of the original frame? When discussing the visibility of noise, the proportion of the frame used and any resizing is important.

See what I mean here: https://www.pentaxforums.com/forums/172-pentax-k-3/242969-sports-images-k-3-a-3.html#post2710908




BTW, any APS-C at ISO12800, unless resized to a quite small size, is going to look poor without drastic invention. ISO12800 is due purely to digital gain. (Analogue gain is not used at high ISOs because image shot noise at this level is, by far, the dominant noise source.) Compared to ISO100, you are using 128x gain. This means individual sensels have collected only 1/128th of the photo-electrons for the same rendering brightness level (e.g. the exposure was 7 stops lower due to the weak scene luminance level, requiring 7 stops more digital gain to produce the same image rendering brightness level) . So if a sensel holds 30,000 photo-electrons at FWC (usually defined as the 3% non-linearity point - November ) at ISO100, at ISO12800 that sensel will only be able to hold approx. 238 photo-electrons before the ADC clips.

Photonic shot noise (an intrinsic property of light and other forms of radiation) is the square-root of the number of photo-electrons. And the wanted-signal-to-shot-noise ratio is also the sq-root of the number of photo-electons. This means that, compared to ISO100, ISO12800 will have an 11.3x worse SNR (sq-root of 1:128), e.g. 173:1 (at ISO100; sq-root of 30,000) vs 15.4:1 (at ISO12800; sq-root of 238).
03-06-2014, 09:35 PM   #22
Pentaxian
calsan's Avatar

Join Date: Nov 2008
Location: Perth, Australia
Photos: Gallery
Posts: 1,543
Tangent open:

I need to step in here to redeem the honour of the K-7 which gets held up as the "gold standard" of bad noise.

K-7 somehow is regarded as the world's noisiest camera - and it is noisy by the standards of newer cameras, particularly K-5, - but it has no where near the high noise of earlier cameras.
My K10d can't be used beyond iso 800 and at 1600, forget being able to identify things in your photo even at 4x6inch print size. K-7 gets me to 1600, which is the difference being able to shoot at night, or not. I've even got good results at 2000 in daylight (birding - that's why..).
DXO confusingly rate the K10d higher than the K7, but I think this may be because the K10d is exceptional at iso 100 and 200. Because it certainly tanks after iso 400…

Tangent closed.

03-06-2014, 10:04 PM   #23
Veteran Member




Join Date: Jun 2009
Posts: 11,913
I've not seen those noise levels with my K-3. I can make it look that noisy, if I wanted to though. Also remember - 100% crops of a 24MP image...

QuoteOriginally posted by Stagnant Quote
The images were developed from RAW in DxO 9
For these sort of comparisons, I suggest you run the RAW's through PDCU 4 to generate the output instead of DxO (or anything else). It is, after all, the 'official' image processor, and one that every K-3 user will have. So it gives consistency.

QuoteOriginally posted by Adam Quote
DxO 9 must love color noise
DxO isn't the best for NR.
It has other strengths, but NR (even using PRIME) isn't one of them.
03-06-2014, 10:38 PM - 1 Like   #24
Pentaxian
dosdan's Avatar

Join Date: Dec 2007
Location: Brisbane, Australia
Photos: Gallery | Albums
Posts: 1,741
Another point. There is a scientific paper that argues that, in an ideal imaging system, 1000 is the minimum number of photons per sensel required to get an image without shot noise becoming visible:

Psychophysical thresholds and digital camera sensitivity: the thousand photon limit
Quoting: In summary, an ideal single capture camera must be designed to capture 10^3 photons in the dark part of an image to avoid visible photon noise. The pixel must be able to capture 10^6 photons to encode the dynamic range of natural images. These are the basic constraints for an ideal camera that can render the vast majority of natural images with no visible noise and no saturated pixels. For real cameras, the requirement can be even higher due to the addition of electronic noise and color process functions (such as color correction).
So if the sensel in the K-3 has a 30Ke- FWC (I haven't seen the actual value reported by anyone yet, but I suspect it's around 30Ke-), then 1000e-, which would needed for a 33:1 (30dB) SNR, would be 2 stops below ISO12800 i.e. at ISO3200.

BTW, my K-5 calculations indicate that its sensels have a 34:1 (30.5dB) SNR at ISO3200.

Of course, real-life sensors also have read noise, made up of sensor read noise, PGA noise and ADC noise, and images will start looking noisy at lower ISOs.

Dan.

Last edited by dosdan; 03-07-2014 at 12:15 AM.
03-07-2014, 03:11 AM   #25
Veteran Member




Join Date: Dec 2013
Photos: Gallery
Posts: 603
Original Poster
QuoteOriginally posted by dosdan Quote
Stagnant, the AR of your test image is 2.16:1. This indicates you've already cropped the original image. Can you please indicate whether this is a 100% crop, or what percentage of the original frame you cropped off the H & W of the original frame? When discussing the visibility of noise, the proportion of the frame used and any resizing is important.
The "thumbnails" you see on the main page are not full size. If you click the thumbnail in the Flickr enviroment and chose all sizes, you will get the option to see the original size. With the vertical resolution of 1294, they are highly unlikely to fit most of the screens. So it would be incorrect to call these images 100 % crop.
- 100 % Crop processed in SilkyPix
- 100 % Crop processed in Lightroom
- 100 % Crop processed in PDCU

Full scene (uploaded in its original size and downloadable) :
- SilkyPix
- Lightroom
- PDCU

In SilkyPix I used Standard contrast and zeroed sharpness, noise reduction, demosaic sharpness. In Lightroom i turned off all the possible adjustments. PDCU was tricky. Assuming that it is based on SilkyPix I set the contrast and sharpness to minimum and turned off all the other adjustments.
03-07-2014, 03:29 AM   #26
Veteran Member




Join Date: Jun 2009
Posts: 11,913
ISO 12800, 100% crop (or close enough), hazy night scene, essentially un-processed images ...
let's just remind outselves of what we are seeing. All things considered, the results are quite OK.

The PDCU image looks best, interestingly enough.
03-07-2014, 04:08 AM   #27
Loyal Site Supporter
Loyal Site Supporter




Join Date: Mar 2009
Location: Gladys, Virginia
Photos: Gallery
Posts: 27,603
I don't think the whole scene looks too bad for being iso 12,800. I don't crop at all when I am shooting that high iso and don't shoot that high unless in extremis. Just to say as well that I think that the K5 and K3 are equivalent for high iso shooting. However, comparing a heavy crop to a K5 photo is probably not totally fair and must remember that K5s don't look great with heavy crops at iso 12800 either.

This is iso 6400 on my K3 -- obviously not a photo with deep shadows, but there wasn't a lot of light either. At 100 percent there is some noise (I didn't apply noise reduction), but a fair amount of detail as well.


03-07-2014, 04:11 AM   #28
Veteran Member




Join Date: Dec 2013
Photos: Gallery
Posts: 603
Original Poster
Rondec, would you mind, uploading the original image size, or at least a 100% crop ?
03-07-2014, 05:26 AM   #29
Pentaxian
dosdan's Avatar

Join Date: Dec 2007
Location: Brisbane, Australia
Photos: Gallery | Albums
Posts: 1,741
Stagnant, the cropped images, as presented here, are 800px high. You want to present them so we can easily compare them by scrolling up and down. Clicking through to Flicker and selecting the original size is not convenient for this purpose.

The linear dimension compression is 800/1294 = 61.8%. The area compression is (800/1294)^2 = 37.5% of the original area, or 290,000px (370x800) / 773,812px (598x1294) = 37.5%

The improvement in SNR in dB with the area reduction is 20log(sq-root(area1/area2)). You can also substitute the ratio of MPs for the ratio of areas. (See: Detailed computation of DxOMark Sensor normalization - DxOMark)

It's easier to work just with the linear dimension reduction: 20log(height1/height2)
= 20log(1294/800)
= 4.2dB higher SNR .

The full-scene "thumbnails" you presented are 333px high. Since the K-3 frame height is 4,000px, the improvement in SNR is:

20log(4000/333)
= 21.6dB higher SNR.

This improvement is equivalent to shooting at ISO1055 instead of at ISO128000. That's why the small full-scene versions hardly show any noise compared to the 100% crops, and are suitable for web presentation.

So knowing both the original size & the presentation size are critical when assessing noise performance. Using higher ISO means that an image is suitable for a smaller viewing size, or a larger image size, but viewed from a greater distance.

Dan.

Last edited by dosdan; 03-07-2014 at 05:41 AM.
03-07-2014, 05:42 AM   #30
Unregistered User
Guest




The Exif on Flickr says all pics are taken @ ISO12800 which is more like I would expect when I look at them. I wonder what Stagnant is trying to pull...
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, color, dslr, image, images, iso, iso100, iso12800, iso3200, k-3, k-3 high iso, k-3 high iso dxo jpg raw, k3, noise, pentax, pentax k-3, photon, pp, scene, settings, shot, shots, size, snr, sports, wind
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
K-3 backyard ISO test PALADIN85020 Pentax K-3 & K-3 II 15 03-19-2014 09:00 AM
Pentax K-3 vs K-30 on AF & high ISO samuelyuen3 Pentax K-3 & K-3 II 8 12-05-2013 10:13 PM
Full size high ISO K-3 JPEGS Ayoh Pentax K-3 & K-3 II 34 10-19-2013 02:35 AM
Q test with images and high iso test pictures StigVidar Pentax Q 9 10-05-2011 08:23 PM
K-x High ISO Test Shots :- ) wll Pentax DSLR Discussion 9 11-09-2009 12:43 AM



All times are GMT -7. The time now is 04:40 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