Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
12-05-2009, 08:58 AM   #1
New Member




Join Date: Nov 2009
Posts: 14
KX Noise

I tried out the KX in a store, and was surprised that contrary to all the excitement about its low noise, my photos were quite noisy even at ISO400. I thought that ISO200 was also alot noisier than my K100D! It was like comparing a compact with an SLR.

Also, when using the digital hdr for boosting shadows, they were very grainy and the whole thing was unusable. I thought a super low noise camera would be awesome for post improving dynamic range.

I later realised though that I was mostly using live view and seeing what settings worked in video, and read that the sensor can get hot and produce noise. I also had the shake reduction turned on for video.

So my question is if you want to take some videos while out then your photos will turn out noisy as a result?

Does anyone know if the SR will also make the sensor hot?

Can video be shot without live view?

Does anyone have any tricks for reducing the amount of noise on video? I remember one person saying for the nikon d90 they closed the aperture pointed at a bright light then opened the aperture again or something...

I know that it's not really a video camera but hey come on, it can take sweet looking video that's way better than HD camcorders IMO and that is worth playing with


Last edited by fan22; 12-05-2009 at 09:03 AM.
12-05-2009, 11:01 AM   #2
Veteran Member
Pentaxor's Avatar

Join Date: May 2009
Location: Vancouver, B.C.
Photos: Gallery
Posts: 6,513
did you make sure that you set exposure correctly? the same was even mentioned about the K-7, but it was more of a user-error that is causing the problem.
12-05-2009, 11:21 AM   #3
Veteran Member




Join Date: Sep 2007
Location: USA
Posts: 1,812
QuoteOriginally posted by fan22 Quote
I tried out the KX in a store, and was surprised that contrary to all the excitement about its low noise, my photos were quite noisy even at ISO400. I thought that ISO200 was also alot noisier than my K100D! It was like comparing a compact with an SLR.
I have recently got the Kx and like you I also have the K100D.

The beauty of the K100D is the relatively large pixel sites since it is only 6Mp occupying the same size sensor.

Having said that the Kx is in just about every way better in terms of image quality to the K100D, and the K100D I consider really good up to about ISO1600. Kx definitely shows its superiority at ISO1600 and above.

However you shouldn't have to rely on anyone else's opinion -
basically conduct your own tests to see for yourself -

eg: I did that - please see

Kx (jpg) ISO Performance

- it took no more than a few minutes shooting
and maybe up to about 1/2 hour doing the cropping
which you would not have to do, as you should be able to just view the images on your computer screen at 100% for yourself -
12-05-2009, 12:23 PM   #4
New Member




Join Date: Nov 2009
Posts: 14
Original Poster
Thanks for your replies.

I just used P mode. Can you explain how the shutter or aperture can contribute to noise if the ISO is say set to 200?

The k100d is super smooth at ISO200, I wonder if setting the kx to 6mp will help in terms of noise? I realised pixel cramming would contribute to deterioration of the image and wish megapixels wasn't such a hype.

UnknownVT, have you compared the kx and k100d side by side with the same manual settings, both set to 6mp and of the same scene? Does the larger pixel sites of the k100d show its strength by producing a slightly bright image in dimmer lighting?

The k100d was my first dslr and I remember being disappointed that my compact (Canon A510) would always produce much brighter images indoors with the same manual settings. I'm still confused about allocating set numbers to rating iso etc when the results can be so different camera to camera. Can anyone help my understand why?

By the way according to the manual live preview will reduce photo quality but I haven't heard the same for SR, and if video can be shot without live preview.

12-05-2009, 01:35 PM   #5
Veteran Member




Join Date: Jun 2009
Posts: 11,913
Assessing the image quality of a camera in-store, at the pixel peeping level, must be close to impossible, I would have thought.

You can compare images from out of the K100D and from the K-x here against a more or less standardised test, with RAW samples available too:

K100D:
Pentax K100D Digital Camera Samples - Full Review - The Imaging Resource!

K-x:
Pentax K-x Digital Camera Samples - Hands-On Preview - The Imaging Resource!

I notice a difference in sensitivity, as well as IQ, between my K200D and my K-x. I'd be surprised if there wasn't any visible difference between the K100D and the K-x.

Video can't be shot without LiveView being active because video on the K-x is just essentially a recording of LiveView, as it is with all DSLR's that do video.

I think you are mistaken about the K-x's video noise too. It is not an issue. Have a look at the many K-x video samples here:

https://www.pentaxforums.com/forums/pentax-video-forum/77199-k-x-video-samples.html

including the batch at the end of the thread. I've also personally taken some video's at 6400 and even 12800 ISO and the noise didn't bother me.

The 'tricks' for producing low noise video are the same as any 'tricks' for producing low noise stills out of any camera: provide enough light or get a faster lens, set exposure correctly etc. And practice, practice, practice.

SR shouldn't make the sensor hot since it is a mechanism that sits outside the sensor.
12-05-2009, 02:44 PM   #6
Veteran Member
Marc Sabatella's Avatar

Join Date: Sep 2006
Location: Denver, CO
Photos: Gallery
Posts: 10,685
QuoteOriginally posted by fan22 Quote
The k100d is super smooth at ISO200, I wonder if setting the kx to 6mp will help in terms of noise? I realised pixel cramming would contribute to deterioration of the image and wish megapixels wasn't such a hype.
Do you have test photos that show the difference you noticed? My guess is, you made the mistake of comparing at whatever the maximum zoom was on the rear LCD in the store - which certainly was *far* greater magnification on the K-x than the K100D. Had you compared images at the same same size, it's extremely unlikely you'd have noticed a difference (or if you did, it would probably have been in favor of the the K-x).

Don't cripple the K-x by not shooting with all its pixels - just don't place unreasonable expectations on it by comparing its images to the k100D with the k-x images blown up four times as big.

QuoteQuote:
The k100d was my first dslr and I remember being disappointed that my compact (Canon A510) would always produce much brighter images indoors with the same manual settings.
P&S cameras may be applying auto exposure corrections during their JPEG conversion; a DSLR generally won't. If you're seeing a difference of anything more than a fraction of a stop, then that is probably what is going on. Either that or the A510 is just completly wrong about its ISO settings.
12-05-2009, 02:55 PM   #7
wll
Veteran Member




Join Date: Nov 2006
Location: Mission Hills, CA
Posts: 773
Hands down !

The K-x has the best high ISO of all my Pentax Cameras ...my K20D's, the K200D's, and it blows the doors off the K10D's that I sold.

My DS I thought was a good high ISO camera for its day....well the K-x is light years ahead in terms of high ISO performance and DR.

Man alive, I wish the Sony 12MP sensor was used in the K-7


wll

12-05-2009, 11:06 PM   #8
Veteran Member




Join Date: Sep 2007
Location: USA
Posts: 1,812
QuoteOriginally posted by fan22 Quote
have you compared the kx and k100d side by side with the same manual settings, both set to 6mp and of the same scene? Does the larger pixel sites of the k100d show its strength by producing a slightly bright image in dimmer lighting?
Here are 100% crops from the standardized studio images from Imaging-Resource -
I also downsized the Kx 100% crop to be the same size as the K100D crop (ie: equivalent of 6Mp image)

ISO200 (shadow) Noise -



ISO details -



I think the images are comparable - the shadow noise crops seem to show that the Kx crop gained when downsized - seems somewhat more detailed?

However in the detail crops the K100D seems better - but then I think this due to the different treatment of their default "Bright" settings - the K100D tends to over-saturate its colors so the brush hairs looks "better", but if one looks at the edge of the ferrule then perhaps the Kx version looks sharper? however there is a slight difference in orientation so that may not be so clear.

In all practical usage there's probably no difference between the Kx and K100D at ISO200 - other than the obvious fact the Kx has an image that's twice the size of the K100D.

If we look at high ISO - in particular at ISO3200 (K100D's weak spot)

ISO3200 (shadow) noise -



ISO3200 detail -



The Kx images show a definite improvement even in the full crops -
and when downsized to the same size as the K100D -
then the improvements are quite substantial.
12-06-2009, 10:42 AM   #9
New Member




Join Date: Nov 2009
Posts: 14
Original Poster
This is the (cropped) image I took in a store, showing the very grainy shadow detail at iso400 with shadow correction done in camera. With such a low noise camera you'd think you could get way better shadow correction especially at that iso. The original noise was finer with more color noise.




And a noisy video
IMGP5314.avi (24 MB)


I think this is all because the sensor got hot? I wasn't using it for long though.
12-06-2009, 12:04 PM   #10
Veteran Member




Join Date: Mar 2007
Location: U.K.
Posts: 685
I've seen much the same level of dark noise in Canon 5D iso 100 shots that were only brought up about 1.5 stops in the shadows to get the black suits to show a bit of detail, and that was outdoor wedding shots (in RAW) where you wouldn't expect any noise problems. My K100D can also look rough at iso 400 in the dark shadows although a bit of NR in the RAW converter can make it very clean without losing much detail. So, basically there may not be much wrong with it, especially if it's jpg as you're working in the last couple of bits and noise gains artefacts that don't normally show but when correction is used, who knows.
12-06-2009, 10:04 PM   #11
Veteran Member
Marc Sabatella's Avatar

Join Date: Sep 2006
Location: Denver, CO
Photos: Gallery
Posts: 10,685
It's normal to see noise in the very deep shadows eve at medium ISO values like that, and by using the shadow correction feature, you basically asked the camera to double the normal amount of noise. I don't think there's any camera on the market that would do significantly better in those conditions.
12-07-2009, 06:03 AM   #12
New Member




Join Date: Nov 2009
Posts: 14
Original Poster
I never saw that on my k100d. In RAW i've taken photos that were very dark from underexposure and processed it to reveal a very useable photo. I preferred that than use higher iso because i could expose the image later with lower iso with better noise. On occasion if it was too severely underexposed i'd get lines in the image so couldn't use those. I boosted shadows many times and they were perfect. I could do a bit of NR but in the example I showed it looks beyond saving with NR. It must be the way it applies the digital filter to try and add contrast to the shadows.

I'm assuming this is from pixel cramming then from 6mp to 12mp on the same size sensor, along with the heat caused by live view?

I still bought the kx anyway but it hasn't arrived yet. I think i'll use raw and try use iso100 as much as possible.

Has anyone who owns one noticed a severe deterioration of image quality from using LV? After how long? I guess it won't be a problem in winter in the northern hemisphere but I'll be using it in Australia mostly...
12-07-2009, 08:04 AM   #13
Veteran Member
er1kksen's Avatar

Join Date: Oct 2008
Location: Forestville, NY
Photos: Albums
Posts: 1,801
QuoteOriginally posted by fan22 Quote

I'm assuming this is from pixel cramming then from 6mp to 12mp on the same size sensor, along with the heat caused by live view?
The "pixel-cramming causes noise" concept is, at the DSLR level, mostly a myth. Noise is a function of sensor size and chip design. Smaller chips (with higher pixel density) are enlarged more for the same output size, so noise is magnified. Larger chips (with lower pixel density) are enlarged less, as is the noise in their images. Many conclude from this that the higher and lower pixel densities are in fact the factor to blame, but its a case of correlation rather than cause.

On two equally-sized chips, one with high pixel density and one with low pixel density, you are again magnifying the higher-pixel-density sensor (and its noise) to a greater degree when you look at noise at the pixel level. Actual image noise (over the whole image area) may be about the same, but at pixel level the high-density sensor will show more noise, essentially because it can simply "resolve" that noise better (being a higher-resolution sensor).

What you're seeing here is that the Sony 6mp sensor is a very low-noise chip design, but also that 6mp simply isn't enough to resolve most of the sensor's noise. It gets averaged out between the pixels. The 12mp sensor, on the other hand, is an even lower-noise design (at least at high ISOs) but the higher pixel density is more revealing of the noise there is at the pixel level. At equal print sizes the difference is negligible.
12-07-2009, 09:20 AM   #14
Veteran Member
Marc Sabatella's Avatar

Join Date: Sep 2006
Location: Denver, CO
Photos: Gallery
Posts: 10,685
QuoteOriginally posted by fan22 Quote
I never saw that on my k100d.
As I said before, show me comparison images made with the same settings and reproduced at the same size. As it is, you are seeing the effect of using the "noise increasing" option (you called it "shadow correction", but same difference) and then blowing up and image over four times larger than you ever did with your K100D. Of course you never saw this on your K100D - because you never *did* this with your K100D. You never turned on shadow correction because your camera doesn't have it, and you never viewed your images four times larger than 1:1 because there'd have been no point.

Perform an apples-to-apples to comparison, and you;ll see either no difference worth mentioning, or else a win for the K-x.

QuoteQuote:
I'm assuming this is from pixel cramming then from 6mp to 12mp on the same size sensor
Not at all. It's like I said; you used an option known to increase noise, and then you blew your image up four times larger than you've viewed an image before.
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!
aperture, camera, dslr, kx, noise, photography, photos, sensor, video, view
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
SR noise in K-x Hemi345 Video Recording and Processing 3 02-17-2011 09:17 PM
Noise Nubi Pentax DSLR Discussion 30 06-28-2010 02:17 PM
K7 and noise ... again. jpzk Pentax DSLR Discussion 128 11-24-2009 09:55 AM
Another.....K-7 noise..... the swede Pentax DSLR Discussion 11 10-17-2009 02:57 AM
cs3 noise filter vs. noise ninja vs. ??? reknelb Pentax DSLR Discussion 0 03-04-2008 04:55 PM



All times are GMT -7. The time now is 03:56 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