Reading mode: Light Dark

K20D full test - NEW REVIEW

Go to last post
M

me_tarzan

Well-known member
Messages
238
Reaction score
36
Location
Queensland, AU
Here's a full test on the K20D. It's interesting - the tester compares a shot (at 14.6MP and 3200 with the K20D -v- a K10D shot at 800asa (and underexposed 2 stops - ie "effectively" at 3200 asa). He makes the point that there isn't much difference.
Points to note

1) The images are incorrectly designated as both "K20D" - it'll be obvious which is which from the file properties - one is 10MP and the other 14.6.

Enlarge the serial number on the lens, so that it fills the entire screen and compare the difference between the two cameras. - Not necessarily talking about noise here (although this is part of the equation). You'll see that the K20D shows a hugely improved image. More interestingly, though, is that at the periphery of the image, the resolution is much worse! I guess this could easily be due to a different aperture but there is no EXIF data on either picture. My thoughts are that the K20D picture seems a great improvement (noise wise) and even if it didn't, it's impressive that Pentax have managed to squeeze 50% more pixels without a hint of extra noise.

Like I say - very interesting....

http://www.digitalcamerareview.com/default.asp?newsID=3374&review=pentax+k20
 
Thanks for posting this. I had not seen it before.

I cannot comment on the absolute levels of noise in the ISO 3200 shots from the K20D and the K10D as I have not found a way to have both images open on my monitor simultaneously. But I think that it is mistaken to consider only absolute noise levels, when what is more important is the signal to noise ratio. In this measure, the K20D trounces the K10D, as the former's image contains so much more detail. For this reason, any sentient being who looked at the two sample images would immediately prefer the K20D's. I mention this, because I think that many readers of the review will infer that there is no difference in the final IQ at ISO 3200.

Rob
 
Not new at all, i saw this review before i bought my K20D, i was astouned with the detail in the pictures with the cards.

--
Enrique
 
I'll be fascinated to see DPRIEW's full test of the K20D - if they ever get round to doing one...... Similarly, I wonder when Steve's Digicams will do a full test. In the sample photos, there's a standard shot of "Nicholson Street" (brick building). In the foreground, there is usually a dumpster and this has a serial number on it. Some cameras (very few) show a clear image of this and other cameras can't resolve the digits. I'll be fascinated to see how the K20D compares. (But I guess that will depend on what lens they use. The old kit lens wasn't such a good performer.
 
Robogo2 said: "cannot comment on the absolute levels of noise in the ISO 3200 shots from the K20D and the K10D as I have not found a way to have both images open on my monitor simultaneously."

Here is a free program that will let you view up to four images at a time. Wonderful program, I really like it, and use it all the time for comparing images.

It can do a lot!

http://www.faststone.org/FSViewerDetail.htm
--
Don
 
This review was fine, but I have to totally disagree when it comes to the issue of high ISO shooting. The tester may be well qualified, but he is dead wrong when he says it may not be worth the upgrade from the K10D based on ISO performance. I have spent a week shooting mostly at ISO 1600 on my K20D and find it excellent. K10D owners, very honestly, have you ever spent, or even considered shooting a week at ISO 1600? Many times on my K10D I regretted even using 800. The noise is very different on the K20D, very even and predictable, unlike the often bunched-up clusters you find withthe K10D. In short, there is no reasonable comparison between the cameras past ISO 400 IMHO. I was not even going to buy the K20D, had already pretty firmly decided against it, until I shot in a low light store with one and was overwhelmed by the difference. To pass off the difference as marginal is an injustice to those that might want/need high ISO abilities and take this review to mean that they won't get much improvement with the K20D. That is just plain Wrong!! I mean Very wrong! And how many want to push process your K10D images to get 3200, when you just set the K20D to 3200 and shoot, without much worry about the results?

Do you ever wonder where things come from, like in a month from now when you are talking with someone and they tell you they heard the high ISO was no better in the K20D than it was in the K10D? Well, it comes from places like this review, which is just totally wrong to a real world shooter. Maybe on his paper he calculated some fancy figures, but I have shot real shots, and printed the results, I know what I see.

I was out visiting a friend today and was showing her the K20D, shot this for her (e-mailed her the original) at 6400. Ever done this with your K10D? I know I never did. Another one I shot at lunch at 1100 of a friend checking out, again, it beats my K10D by a longshot. If any K10D or K20D owners disagree with my findings, let me know, I am no scientist but have near perfect vision according to a recent exam. I know what I see, and I have seen a lot of 800+ ISO in the last week that were fully acceptable to a non-pixel peeping shooter in the real world, not pushing some shot in post-processing, just turning the ISO dial in an upward position and shooting real world shots.
6400



1100



BTW-The reviewer can kiss my ISO 6400 foot.



A couple of cat shots I took for my friend IO 400-Might as well post them here! LOL





--
'This is more serious than I thought.....but it is still fun!
http://www.pbase.com/rupertdog Take a look- It's Free!
 
The reviewer also reported no AF improvement, something I find to be incorrect also. I have a K10D (Love it BTW) but find a definate improvement with my K20D in side by side useage of the same lens. I mostly prefer MF, my eyes are usually as good as or better than the camera, but still, there is an improvement of noticeable difference. Others have reported this too. Maybe we just got lucky, maybe the reviewer just got unlucky, I have no idea, but if he couldn't see an ISO advantage, I have serious doubts that he could see anything else. Although his report was basically positive, it hit my trash bin based on his gross miscalculations on ISO. I guarantee you I would not have paid .50 cents for a K20D if the high ISO had not been overwhelmingly better than my K10D. It is not about the money, it is about the IQ, and the K20D performs.
--
'This is more serious than I thought.....but it is still fun!
http://www.pbase.com/rupertdog Take a look- It's Free!
 
chrswggl said:
I could never pull up an ISO 800 shot to 3200 on my k10d. I would
see intense horizontal banding. Maybe my copy is just defective, but
I see it if I fool with high ISO shots at all.
Click to expand...
You've got horizontal patterns (in landscape orientation) using your K10D? You wouldn't happen to always use it with the grip, would you? Horizontal patterns have often been reported at higher ISO's using the grip for the K10D, but almost never otherwise unless the camera was proved defective in some way.

Regards, GordonBGood
 
has HPN a bit in the upper right corner's bokeh.

I've shot my K10D at ISO1600 in bright conditions outdoors to get shutter speed up w/o problems. Indoors, I lock auto ISO down to the 100-400 range...
 
Hi, indeed, that has been posted here before.

And someone has noted, that the focus points of the two images are different. The K20D can is way much sharper at the upper part (look at the half-tone dots and the tiny engraving behind the lever), the K10D slightly on the bottom of the can (also visible on the dots and also at the upper part, where the dots in front are gone but the engraving behind the lever is sharp).

So judge for yourself how serious the conclusions based on this pictures can be taken...

I don't know if people doing a review don't notice things like that or ... what?
me_tarzan said:
Here's a full test on the K20D. It's interesting - the tester
compares a shot (at 14.6MP and 3200 with the K20D -v- a K10D shot at
800asa (and underexposed 2 stops - ie "effectively" at 3200 asa). He
makes the point that there isn't much difference.
Points to note
1) The images are incorrectly designated as both "K20D" - it'll be
obvious which is which from the file properties - one is 10MP and the
other 14.6.
Enlarge the serial number on the lens, so that it fills the entire
screen and compare the difference between the two cameras. - Not
necessarily talking about noise here (although this is part of the
equation). You'll see that the K20D shows a hugely improved image.
More interestingly, though, is that at the periphery of the image,
the resolution is much worse! I guess this could easily be due to a
different aperture but there is no EXIF data on either picture. My
thoughts are that the K20D picture seems a great improvement (noise
wise) and even if it didn't, it's impressive that Pentax have
managed to squeeze 50% more pixels without a hint of extra noise.

Like I say - very interesting....

http://www.digitalcamerareview.com/default.asp?newsID=3374&review=pentax+k20
Click to expand...
 
me_tarzan said:
Here's a full test on the K20D. It's interesting - the tester
compares a shot (at 14.6MP and 3200 with the K20D -v- a K10D shot at
800asa (and underexposed 2 stops - ie "effectively" at 3200 asa). He
makes the point that there isn't much difference.
Click to expand...
I am the only one? this seems completely counterintuitive (read: stupid) to me. If he can push the image he made 2 stops from the K10D you should be able to do the same with the K20D. So the only acceptable comparison would be both pushed by 2 steps. Only if you loose more shadow detail with the K20D than this would be worth considering.
 
Thinking about this a little more: Doesn't this show that the noise level at ISO800 of the K10D is the same as the noise level at ISO3200 with the K20D?
 
I cannot say you are wrong. Nor can say so to those who say the opposite about the AF in another thread here. But could it be that your k10d AF may have been below average ?

lock
 
Then you should send it in. Vertical ? Yes... and Pentax probably won't do anything for you. But horizontal ? That should not happen without the grip, like Gordon stated.

lock
 
chrswggl said:
No. You know that ISO's are always a signal boost from base. Doing
it in pp isn't all that much different. What they're saying is the
noise level of the k10d at an equivalent of ISO 3200 is the same as
the K20d at ISO 3200.
Click to expand...
I believe there is a big difference: when you choose a higher ISO at you camera more current is sent through the sensor, making it more sensitive but also increasing the noise. If you push in software you use the resources of the raw file, but you also loose lots of shadow detail and do not gain all the brightness detail, that is why the image in the test has more contrast. Otherwise you could just shoot raw at base iso and than choose the sensitivity on your computer.

Some cameras have a 'high' mode which is the same as pushing, most do this only for one step.
 
Where do you get this!!! Comments as follows:
cseiler said:
cseiler said:
No. You know that ISO's are always a signal boost from base. Doing
it in pp isn't all that much different. What they're saying is the
noise level of the k10d at an equivalent of ISO 3200 is the same as
the K20d at ISO 3200.
Click to expand...
I believe there is a big difference: when you choose a higher ISO at
you camera more current is sent through the sensor, making it more
sensitive but also increasing the noise. If you push in software you
use the resources of the raw file, but you also loose lots of shadow
detail and do not gain all the brightness detail, that is why the
image in the test has more contrast. Otherwise you could just shoot
raw at base iso and than choose the sensitivity on your computer.
Some cameras have a 'high' mode which is the same as pushing, most do
this only for one step.
Click to expand...
What "more current is sent through the sensor"? The photosites themselves don't do anything different for high ISO's then for low ISO's other than see much less light. All compensation for the higher ISO is in then increased gain in the signal after the photosite, as in analog gain prior to the Analog to Digital Convertor (ADC) for most cameras and/or digital multiplication of the converted raw readings as in push processing.

The reason that push processed low ISO images from raw data sometimes have different noise than the native high ISO images using raw from the camera is that the amplifier and ADC sometimes introduce proportionally much more noise at low ISO's than dark readout noise from the sensor; when the signal is allowed to be amplified in the camera before digital conversion, this ADC noise becomes much less of a proportion. The K10D has very low dark readout noise right across the ISO range due to its 22-bit digital conversion system, thus this ADC noise does not impact the comparison and it is valid.

As lock says, this is a valid comparison of the noise from the K20D at ISO 3200 compared to the equivalent noise from the K10D in which the noise is more or less the same, with a slight advantage to the K20D in this example due to processing differences (different raw convertors?, different processing of raw in the camera as the K20D applies a slight amount of NR to raw data for ISO's of 1600 and higher).

The comparison was never intended to show the differences in focus or detail resolution between the two cameras; thus the author did not pay much attention to where the cameras focussed (unfortunately).

Regards, GordonBGood
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top