Go Back   Steve's Digicams Forums > Digital SLR and Interchangeable Lens Cameras > Pentax / Samsung dSLR, K Mount Mirrorless

Reply
 
Thread Tools Search this Thread
Old Feb 19, 2010, 1:41 AM   #11
Senior Member
 
Eetu's Avatar
 
Join Date: Jan 2010
Location: Jyväskylä, Finland
Posts: 2,212
Default

Quote:
Originally Posted by mtngal View Post
I think I gave up putting too much store into sensor figures and charts a long time ago ... So when you look at any charts, you need to understand what they are trying to tell you, then decide if that's anything important to you.
Many thanks. Ever since I started looking closely into cameras, checking them out, there is a lot of bias and things to sort through. Not always pleasant. There is no perfect camera.

Many thanks.

Ned
Eetu is offline   Reply With Quote
Old Feb 25, 2010, 11:26 PM   #12
Member
 
Join Date: Feb 2010
Posts: 87
Default

Intersting,

D90 and D5000 are between pro cameras, and with very near results. They are higher than any cameras in their tier and level of price, if I did not miss anyone...

But the D5000 had a worse result in Lo-Light ISO, what is strange, becase AFAIK thei have the same sensor, no? Maybe due NR differences?

Last edited by MarceloRSC; Feb 25, 2010 at 11:34 PM.
MarceloRSC is offline   Reply With Quote
Old Feb 25, 2010, 11:30 PM   #13
Senior Member
 
bigdawg's Avatar
 
Join Date: Feb 2005
Location: Thach Alabama
Posts: 14,981
Default

Quote:
Originally Posted by Hards80 View Post
yea, i have shot canon for going on 15 years now. currently have a 50d.

but, i stay abreast with all major manufacturers to help give the most unbiased opinions i can.
And both you and Jim do it well.
__________________
Big Dawg
bigdawg is offline   Reply With Quote
Old Feb 25, 2010, 11:33 PM   #14
Senior Member
 
bigdawg's Avatar
 
Join Date: Feb 2005
Location: Thach Alabama
Posts: 14,981
Default

Quote:
Originally Posted by mtngal View Post
I think I gave up putting too much store into sensor figures and charts a long time ago - I tend to look at sample pictures (which aren't always an accurate indicator either) and go by ergonomics mostly. A long time ago there was a person who hung out around here (until he got banned) who delights with telling everyone who'll listen just how horrible Pentax cameras are and his charts always show them as being worse than any other camera. At least DxO is far more objective than this individual is.

Not everyone has the same tastes, either. For instance, Pentax has traditionally had more detail/resolution and more noise in their higher ISO pictures than some of their competition (including ones using the same sensor). So someone looking strictly at noise would say the camera lags behind the one with less noise. On the other hand, the person looking at lines and detail would prefer it, wanting to decide for themselves how much detail they want to sacrifice to noise reduction.

So when you look at any charts, you need to understand what they are trying to tell you, then decide if that's anything important to you.
I agree Hariet. I turn off the in camera noise reduction myself then use Neatimage http://www.neatimage.com/download.html if I find I need the noise reduction.
__________________
Big Dawg
bigdawg is offline   Reply With Quote
Old Feb 25, 2010, 11:34 PM   #15
Senior Member
 
bigdawg's Avatar
 
Join Date: Feb 2005
Location: Thach Alabama
Posts: 14,981
Default

Quote:
Originally Posted by Eetu View Post
Well, I must say, yours and some others have been very even-steven, unbiased and helpful... which I thinks makes for a really good forum.

Many thanks.

Ned
One of the most unbiased forums out there Ned.
__________________
Big Dawg
bigdawg is offline   Reply With Quote
Old Feb 26, 2010, 6:23 AM   #16
Senior Member
 
Join Date: Nov 2004
Location: Mount Shasta, California
Posts: 1,525
Default

I consider DXO useless in evaluating a camera. It analyizes a very narrow range of factors for a very complex subject. For more on how DXO works and its shortcomings, see here www.luminous-landscape.com/reviews/dxo-explained.shtml
pboerger is offline   Reply With Quote
Old Feb 26, 2010, 8:30 AM   #17
Senior Member
 
VTphotog's Avatar
 
Join Date: Mar 2005
Location: Extreme Northeastern Vermont, USA
Posts: 4,232
Default

There are those more concerned with taking measurements than taking pictures, and that is fine as far as it goes.
My opinion is that for any practical purpose, all of the DSLR cameras are more than adequate. User interface, and functionality mean more to me than noise figures and resolution charts. The things that are easiest to measure, are those things which have the least meaning as far as being out using the camera.

brian
VTphotog is offline   Reply With Quote
Old Feb 26, 2010, 9:17 AM   #18
Administrator
 
Join Date: Jun 2003
Location: Savannah, GA (USA)
Posts: 22,378
Default

Again, I'd take some of those tests with a "grain of salt" (especially the S/N ratio tests).

One problem with some of those tests is that they don't take retained detail into consideration. For example, you can have a camera with a great Signal to Noise ratio and have soft images from it at higher ISO speeds. ;-)

Manufacturers are increasingly using Noise Reduction at the raw level. Sony does it with some models. Pentax appears to do it with their current dSLR lineup (for example, if you look at the S/N ratio graphs for Pentax models at DxOMark.com, you'll often see the color of the graph change at higher ISO speed settings, indicating that they've detected a camera model is using Noise Reduction at the raw level, even though it's turned off in the camera menus. In those cases (since Noise Reduction can make one camera look better than it really is on most Signal to Noise ratio tests), they use extrapolated results (hence the color of their graph lines changing).

I've also seen very strong evidence that Nikon is using Noise Reduction at the raw level with some of it's models (for example, the D90). Yet, they haven't detected it with that model yet (so they're not using extrapolated results for it's S/N ratio tests). It also wouldn't surprise me if Canon is doing the same thing (in a way that's harder to detect).

Another issue impacting results is the sensitivity of the sensor for different colors (that the color filter array design and other characteristics of a sensor's design can impact) and what kind of RGB multipliers and tone curves need to be used in different lighting conditions for correct white balance and to achieve the desired Dynamic Range and contrast in an image; and if you need to increase levels in one color channel more than another, you can increase noise levels in real world conditions and cause detail in one channel over another to be clipped in some image areas (impacting Dynamic Range) depending on the lighting you're shooting in because of differences in the way a sensor responds in that lighting.

So, I'd look at images (not just things like S/N ratio graphs from DxOMark) to get a better idea of how camera models compare in that area (preferably with the same types of subjects in the same type of lighting you plan on shooting in at higher ISO speeds).

Note that some noise measurements are more sophisticated now, and give you an idea of things like grain pattern, how individual channels are impacted and more. For example, Imatest has some pretty sophisticated tests available. But, I'd still take them with a "grain of salt" when comparing cameras, rather than assuming those tests are "gospel", especially since subject color and lighting will impact results.
JimC is offline   Reply With Quote
Old Feb 28, 2010, 10:46 AM   #19
Member
 
Motorman's Avatar
 
Join Date: Dec 2007
Posts: 58
Default

Interesting thread as I have viewed this site before and this is my simple take on the Pentax evaluation on the sensor/camera rating at DXO..

If you eliminate the MF and FF cameras and do APS-C cams the Pentaxes do better,

K10D.....8th

K20D....12th

K200...18th

KM....23rd

K7......33rd...hmmm,really

Now Samsung GX 20 got 4th on the APS-C list and shares pretty much the same sensor as K20 and K7 and it,GX 20, gets 4th,hmmm not that I have anything bad to say about the camera but the numbers if the sensor data that it uses should be closer and many have tested and sited very similar images and much closer than DXO's ranking shows or would have you believe.
And where is the KX ranking?Not tested but but this Canon is????Well it is pretty much accepted that the sensor in the KX is probably the same one used in the Nikon D90, number 1 in the DXO ranking for APS-C cameras and(maybe starting a conspiracy theory here)that they don't want (for what ever reason)the entry level,under 500 bucks Pentax KX to be number 1 or 2(really a possibility if ya think about it if infact they do share the same sensor) and that would be a ground shaker would it not...Not tested a KX,sure they haven't,guess they couldn't find one......
Motorman is offline   Reply With Quote
 
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



All times are GMT -5. The time now is 1:41 AM.