K10D dynamic range

drwho9437 · 07-06-2008, 10:05 AM

Let me give you the scoop.

I had a Canon 300D and just shot the kit for a long time. Then I got a 50mm, then the Sigma 10-20, then the 35mm f2, sold the 50mm. Then I got sick of the lag of the 300D on start up and got a black (hated the silver 300D's looks) 400D. One week after I shelled out for the 400D Pentax released the K10D. Purchased a 100mm macro.

Now that there is a K20D, people are willing to part with the K10D for about 500 dollars it seems.

Now I only really shoot RAW, I use lightroom. I looked over the review of the K10D again at dpreview and many lens tests. The item that concerns me is the dynamic range test. I find I don't have enough dynamic range with my 400D and according to their result there is strange clipping of highlights and no real roll off for the K10D while the K20D has a normal roll off.

Is this true, that the DR is much better on the K20D? I guess landscape people would notice most dark shadows and brights skies.

I'm attracted to pentax for the limited lenses and the in body IS. I don't like how heavy the bodies have gotten though.

I'm mostly just interested in the DR issue I am asking about. I also love that Pentax shooters are a community.

Wheatfield · 07-06-2008, 02:33 PM

Originally posted by drwho9437

Let me give you the scoop.

I had a Canon 300D and just shot the kit for a long time. Then I got a 50mm, then the Sigma 10-20, then the 35mm f2, sold the 50mm. Then I got sick of the lag of the 300D on start up and got a black (hated the silver 300D's looks) 400D. One week after I shelled out for the 400D Pentax released the K10D. Purchased a 100mm macro.

Now that there is a K20D, people are willing to part with the K10D for about 500 dollars it seems.

Now I only really shoot RAW, I use lightroom. I looked over the review of the K10D again at dpreview and many lens tests. The item that concerns me is the dynamic range test. I find I don't have enough dynamic range with my 400D and according to their result there is strange clipping of highlights and no real roll off for the K10D while the K20D has a normal roll off.

Is this true, that the DR is much better on the K20D? I guess landscape people would notice most dark shadows and brights skies.

I'm attracted to pentax for the limited lenses and the in body IS. I don't like how heavy the bodies have gotten though.

I'm mostly just interested in the DR issue I am asking about. I also love that Pentax shooters are a community.

First, the best dynamic range on the K10 is had at ISO 100, and is ~11 stops. The DR progressively shortens as you raise the ISO, until at ISO 1600 you are down to ~9 stops.
This is still a couple of stops more DR than the longest range colour print films ever were.
While I haven't actually tested the K20, my impression is that the DR is similar to that of the K10, perhaps a little longer. The K20 does have a dynamic range extension in the Fn menu (in ISO adjust) that seems to increase the DR by about a stop, at the cost of losing 100 ISO.

Jodokast96 · 07-06-2008, 03:13 PM

According to Dpreview, the K10D has a DR of 7.3 stops (-4.5 to 2.8), the XTi 8.4 stops (-4.9 to 3.5), the XSi 8.7 stops (-5.1 to 3.6) and the K20D 9 stops (-5.8 to 3.2). Not sure where you got 11 stops from, or which is right for that matter, but that's a pretty big difference. At any rate, it seems pretty clear that Pentax has opted to favor shadow detail as opposed to highlight detail.

drwho9437 · 07-06-2008, 03:21 PM

Originally posted by Jodokast96

According to Dpreview the K10D has a DR of 7.3 stops (-4.5 to 2.8), the XTi has 8.4 stops (-4.9 to 3.5) and the K20D 9 stops (-5.8 T

Correct those numbers and charts are my reference above.

arbutusq · 07-06-2008, 06:21 PM

Those charts are for Jpegs only. Imaging Resource has a chart that ranks the raw performance of all these cameras. If you check the K10d review on that site you'll find it.

drwho9437 · 07-06-2008, 06:29 PM

Originally posted by arbutusq

Those charts are for Jpegs only. Imaging Resource has a chart that ranks the raw performance of all these cameras. If you check the K10d review on that site you'll find it.

I didn't think they were because they talk about headroom in ACR as well, but I will look at IR as well...

Edit: you are correct, the main part is about the JPEG engine which I don't care about really..

Also looked at the IR data, but it doesn't have as good a commentary. In the raw comparison the K10D fairs better, and probably it would be OK. Thanks for pointing it out to me.

Wheatfield · 07-06-2008, 07:32 PM

Originally posted by Jodokast96

According to Dpreview, the K10D has a DR of 7.3 stops (-4.5 to 2.8), the XTi 8.4 stops (-4.9 to 3.5), the XSi 8.7 stops (-5.1 to 3.6) and the K20D 9 stops (-5.8 to 3.2). Not sure where you got 11 stops from, or which is right for that matter, but that's a pretty big difference. At any rate, it seems pretty clear that Pentax has opted to favor shadow detail as opposed to highlight detail.

I got my data from a fellow on the Pentax Discuss Mail List who, among other things, has worked for Adobe and Nasa.
He wouldn't have been testing jpegs, which do have a rather short DR.

Zebooka · 07-07-2008, 02:46 AM

DPreview has a wrong method for measuring.
They measure DR in JPEGs. Numbers they measure cannot be called Dynamic Range.
And topic starter mentioned here that he shoots in raw.

Jodokast96 · 07-07-2008, 02:53 AM

Hey, easy guys.

I never said Dpreveiw was right.

breischl · 07-15-2008, 11:29 AM

Slightly off topic here, but I'm wondering how all this relates to the histogram the camera shows. IIRC it shows a 4 or 5 stop range in the histogram. But even the most conservative number cited here says the camera records a 7.5 stop range.

So what gives? Is the histogram showing me a smaller range than what the camera actually records? If so, which parts of the true range are getting clipped out of the histogram?

Gooshin · 07-15-2008, 11:33 AM

the histogram you see represents the histogram of the jpeg output at the settings that you set it at.

usualy a histogram that clips highlights on your screen means you still got some headroom in ACR

breischl · 07-15-2008, 11:44 AM

Originally posted by Gooshin

the histogram you see represents the histogram of the jpeg output at the settings that you set it at.

usualy a histogram that clips highlights on your screen means you still got some headroom in ACR

Interesting. Seems like it would be handy to get the true range, or at least something closer to it. I'll have to look at the available settings when I get home to see if that's possible.

I don't suppose you would know off the top of your head?

Gooshin · 07-15-2008, 11:46 AM

Originally posted by breischl

Interesting. Seems like it would be handy to get the true range, or at least something closer to it. I'll have to look at the available settings when I get home to see if that's possible.

I don't suppose you would know off the top of your head?

you "get used to it"

meaning, you get used to looking at your LCD screen and having a good idea of how its going to end up on your monitor.

**Canada_Rockies** · 07-15-2008, 01:37 PM

Originally posted by breischl

Interesting. Seems like it would be handy to get the true range, or at least something closer to it. I'll have to look at the available settings when I get home to see if that's possible.

I don't suppose you would know off the top of your head?

The setting I use is the highlight blinkies. I expose to have one or two very small specular highlights that are blown out (blinking red). Then, I look for shadows that are beyond recovery (blinking yellow). I find that normally, I don't have the shadow blinkies. The blinkies are rated at the jpeg level, and with one or two tiny ones, they are not usually blown out on my RAW images when I open the image in DxO or Adobe.

I shoot only RAW - for the same reasons that I used to shoot Fujichrome Velvia - the better base quality.

breischl · 07-15-2008, 02:35 PM

Thanks guys. I had pretty much arrived at the same conclusions myself. I was just hoping there was a way to make the histogram accurate, instead of needing to work around it's inaccuracies.

The workarounds & guessing are usually fine, but occasionally I get it wrong and screw up a shot. It seems silly that we have to guess at this stuff when the camera already knows the answer.

I suppose it would be hard to show a 9-stop range on the LCD, but does anyone even look at the midtones on the in-camera histogram? I'd be happy if it showed me the bottom two stops and the top two stops, with nothing in between.

07-06-2008, 10:05 AM	#1
drwho9437 Junior Member Join Date: Jul 2008 Posts: 31	K10D dynamic range Let me give you the scoop. I had a Canon 300D and just shot the kit for a long time. Then I got a 50mm, then the Sigma 10-20, then the 35mm f2, sold the 50mm. Then I got sick of the lag of the 300D on start up and got a black (hated the silver 300D's looks) 400D. One week after I shelled out for the 400D Pentax released the K10D. Purchased a 100mm macro. Now that there is a K20D, people are willing to part with the K10D for about 500 dollars it seems. Now I only really shoot RAW, I use lightroom. I looked over the review of the K10D again at dpreview and many lens tests. The item that concerns me is the dynamic range test. I find I don't have enough dynamic range with my 400D and according to their result there is strange clipping of highlights and no real roll off for the K10D while the K20D has a normal roll off. Is this true, that the DR is much better on the K20D? I guess landscape people would notice most dark shadows and brights skies. I'm attracted to pentax for the limited lenses and the in body IS. I don't like how heavy the bodies have gotten though. I'm mostly just interested in the DR issue I am asking about. I also love that Pentax shooters are a community.

07-06-2008, 03:13 PM	#3
Jodokast96 Veteran Member Join Date: Oct 2007 Location: Erial, NJ USA Photos: Gallery \| Albums Posts: 2,134	According to Dpreview, the K10D has a DR of 7.3 stops (-4.5 to 2.8), the XTi 8.4 stops (-4.9 to 3.5), the XSi 8.7 stops (-5.1 to 3.6) and the K20D 9 stops (-5.8 to 3.2). Not sure where you got 11 stops from, or which is right for that matter, but that's a pretty big difference. At any rate, it seems pretty clear that Pentax has opted to favor shadow detail as opposed to highlight detail. Last edited by Jodokast96; 07-06-2008 at 03:23 PM.

07-06-2008, 06:29 PM	#6
drwho9437 Junior Member Join Date: Jul 2008 Posts: 31 Original Poster	Originally posted by arbutusq Those charts are for Jpegs only. Imaging Resource has a chart that ranks the raw performance of all these cameras. If you check the K10d review on that site you'll find it. I didn't think they were because they talk about headroom in ACR as well, but I will look at IR as well... Edit: you are correct, the main part is about the JPEG engine which I don't care about really.. Also looked at the IR data, but it doesn't have as good a commentary. In the raw comparison the K10D fairs better, and probably it would be OK. Thanks for pointing it out to me. Last edited by drwho9437; 07-06-2008 at 06:56 PM.

07-15-2008, 11:29 AM	#10
breischl Forum Member Join Date: Aug 2007 Location: Denver Photos: Gallery Posts: 92	Histogram Slightly off topic here, but I'm wondering how all this relates to the histogram the camera shows. IIRC it shows a 4 or 5 stop range in the histogram. But even the most conservative number cited here says the camera records a 7.5 stop range. So what gives? Is the histogram showing me a smaller range than what the camera actually records? If so, which parts of the true range are getting clipped out of the histogram?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
k20d or k10d - which has better dynamic range	Michael Barker	Pentax DSLR Discussion	10	12-18-2008 10:47 AM
Dynamic Range	Rene`	Pentax DSLR Discussion	6	09-25-2008 07:46 AM
Dynamic Range	Vlad	Photographic Technique	5	06-01-2008 10:40 PM
K10D dynamic range...ISO100, single exposure HDR	d.bradley	Pentax DSLR Discussion	15	02-10-2008 06:11 PM
Impressed by Dynamic Range of K10D	Avantphoto	Pentax DSLR Discussion	11	05-25-2007 06:37 AM

07-06-2008, 02:33 PM	#2
Wheatfield Moderator Join Date: Apr 2008 Location: The wheatfields of Canada Posts: 15,986	Originally posted by drwho9437 Let me give you the scoop. I had a Canon 300D and just shot the kit for a long time. Then I got a 50mm, then the Sigma 10-20, then the 35mm f2, sold the 50mm. Then I got sick of the lag of the 300D on start up and got a black (hated the silver 300D's looks) 400D. One week after I shelled out for the 400D Pentax released the K10D. Purchased a 100mm macro. Now that there is a K20D, people are willing to part with the K10D for about 500 dollars it seems. Now I only really shoot RAW, I use lightroom. I looked over the review of the K10D again at dpreview and many lens tests. The item that concerns me is the dynamic range test. I find I don't have enough dynamic range with my 400D and according to their result there is strange clipping of highlights and no real roll off for the K10D while the K20D has a normal roll off. Is this true, that the DR is much better on the K20D? I guess landscape people would notice most dark shadows and brights skies. I'm attracted to pentax for the limited lenses and the in body IS. I don't like how heavy the bodies have gotten though. I'm mostly just interested in the DR issue I am asking about. I also love that Pentax shooters are a community. First, the best dynamic range on the K10 is had at ISO 100, and is ~11 stops. The DR progressively shortens as you raise the ISO, until at ISO 1600 you are down to ~9 stops. This is still a couple of stops more DR than the longest range colour print films ever were. While I haven't actually tested the K20, my impression is that the DR is similar to that of the K10, perhaps a little longer. The K20 does have a dynamic range extension in the Fn menu (in ISO adjust) that seems to increase the DR by about a stop, at the cost of losing 100 ISO.

07-06-2008, 03:21 PM	#4
drwho9437 Junior Member Join Date: Jul 2008 Posts: 31 Original Poster	Originally posted by Jodokast96 According to Dpreview the K10D has a DR of 7.3 stops (-4.5 to 2.8), the XTi has 8.4 stops (-4.9 to 3.5) and the K20D 9 stops (-5.8 T Correct those numbers and charts are my reference above.

07-06-2008, 06:21 PM	#5
arbutusq Veteran Member Join Date: Sep 2006 Location: Vancouver BC canada Photos: Gallery Posts: 380	Those charts are for Jpegs only. Imaging Resource has a chart that ranks the raw performance of all these cameras. If you check the K10d review on that site you'll find it.

07-06-2008, 07:32 PM	#7
Wheatfield Moderator Join Date: Apr 2008 Location: The wheatfields of Canada Posts: 15,986	Originally posted by Jodokast96 According to Dpreview, the K10D has a DR of 7.3 stops (-4.5 to 2.8), the XTi 8.4 stops (-4.9 to 3.5), the XSi 8.7 stops (-5.1 to 3.6) and the K20D 9 stops (-5.8 to 3.2). Not sure where you got 11 stops from, or which is right for that matter, but that's a pretty big difference. At any rate, it seems pretty clear that Pentax has opted to favor shadow detail as opposed to highlight detail. I got my data from a fellow on the Pentax Discuss Mail List who, among other things, has worked for Adobe and Nasa. He wouldn't have been testing jpegs, which do have a rather short DR.

07-07-2008, 02:46 AM	#8
Zebooka Veteran Member Join Date: May 2008 Location: Russia, Siberia, Novosibirsk Posts: 323	DPreview has a wrong method for measuring. They measure DR in JPEGs. Numbers they measure cannot be called Dynamic Range. And topic starter mentioned here that he shoots in raw.

07-07-2008, 02:53 AM	#9
Jodokast96 Veteran Member Join Date: Oct 2007 Location: Erial, NJ USA Photos: Gallery \| Albums Posts: 2,134	Hey, easy guys. I never said Dpreveiw was right.

07-15-2008, 11:33 AM	#11
Gooshin Veteran Member Join Date: Sep 2007 Location: Toronto, the one in Canada. Posts: 5,610	the histogram you see represents the histogram of the jpeg output at the settings that you set it at. usualy a histogram that clips highlights on your screen means you still got some headroom in ACR

07-15-2008, 11:44 AM	#12
breischl Forum Member Join Date: Aug 2007 Location: Denver Photos: Gallery Posts: 92	Originally posted by Gooshin the histogram you see represents the histogram of the jpeg output at the settings that you set it at. usualy a histogram that clips highlights on your screen means you still got some headroom in ACR Interesting. Seems like it would be handy to get the true range, or at least something closer to it. I'll have to look at the available settings when I get home to see if that's possible. I don't suppose you would know off the top of your head?

07-15-2008, 11:46 AM	#13
Gooshin Veteran Member Join Date: Sep 2007 Location: Toronto, the one in Canada. Posts: 5,610	Originally posted by breischl Interesting. Seems like it would be handy to get the true range, or at least something closer to it. I'll have to look at the available settings when I get home to see if that's possible. I don't suppose you would know off the top of your head? you "get used to it" meaning, you get used to looking at your LCD screen and having a good idea of how its going to end up on your monitor.

07-15-2008, 01:37 PM	#14
Canada_Rockies Loyal Site Supporter Join Date: Nov 2007 Location: Sparwood, BC, Canada Photos: Gallery \| Albums Posts: 12,385	Originally posted by breischl Interesting. Seems like it would be handy to get the true range, or at least something closer to it. I'll have to look at the available settings when I get home to see if that's possible. I don't suppose you would know off the top of your head? The setting I use is the highlight blinkies. I expose to have one or two very small specular highlights that are blown out (blinking red). Then, I look for shadows that are beyond recovery (blinking yellow). I find that normally, I don't have the shadow blinkies. The blinkies are rated at the jpeg level, and with one or two tiny ones, they are not usually blown out on my RAW images when I open the image in DxO or Adobe. I shoot only RAW - for the same reasons that I used to shoot Fujichrome Velvia - the better base quality.

07-15-2008, 02:35 PM	#15
breischl Forum Member Join Date: Aug 2007 Location: Denver Photos: Gallery Posts: 92	Thanks guys. I had pretty much arrived at the same conclusions myself. I was just hoping there was a way to make the histogram accurate, instead of needing to work around it's inaccuracies. The workarounds & guessing are usually fine, but occasionally I get it wrong and screw up a shot. It seems silly that we have to guess at this stuff when the camera already knows the answer. I suppose it would be hard to show a 9-stop range on the LCD, but does anyone even look at the midtones on the in-camera histogram? I'd be happy if it showed me the bottom two stops and the top two stops, with nothing in between.