Dpreview and K-3s ISO noise

rrstuff · 04-01-2014, 06:11 PM

I looked through the Exif data of their JPGs... and they expose K-3 with ~50% shorter shutter speeds than Nikon and Canon at the highest ISO settings.

Next thing you know they will write that 'high ISO noise is not quite up to par with competition'... facepalm.

Pentax K-3 First Impressions Review: Digital Photography Review

According to DXO data, K-3 does not have inflated ISO figures, unlike most other cameras. Perhaps that is the reason why so many people in user reviews complain about high iso noise...

**kiwi_jono** · 04-01-2014, 07:51 PM

I noticed this on DXO. Pentax does a great job with ISO accuracy which does make for a bit of disadvantage on noise test reporting (if the review reporting against set ISO not actual ISO). I think it should be pointed out in reviews like this because it is relevant.

MD Optofonik · 04-01-2014, 08:19 PM

Please explain.

**kiwi_jono** · 04-01-2014, 08:59 PM - **2 Likes**

Have a look at this example ISO Sensitivity of K3 versus Nikon 7100: Pentax K-3 versus Nikon D7100 - Side by side camera comparison - DxOMark

When the Nikon is set for ISO 400 its actual sensitivity is measured as ISO 279 (significantly short of 400). But if you look at the K-3 the ISO reading are pretty close to spot on (ISO 400 is actually ISO 381).
When you set the cameras to ISO 3200, the K-3 is actually ISO 2933 (around 3000) and 7100 is ISO 2192 (around 2000).
So if DPR provide comparative test images at say ISO 400 then the K-3 is achieving a higher ISO and therefore can be expected to have a higher noise component than the Nikon 7100 but people viewing the test images won't generally make an allowance for the greater error in the Nikon (which is really operating at ISO 279 not 400).

Compare other Canon / Nikon cameras and you will find that most don't achieve the stated ISO but Pentax is normally about right.

MD Optofonik · 04-01-2014, 09:11 PM

Thanks!

causey · 04-02-2014, 04:45 AM

Honesty vs. marketing.

audiobomber · 04-02-2014, 10:37 AM - **1 Like**

ISO fudging has become a pet peeve for me. From DPR's XE-2 review:

"However the X-E2's need for unusually long exposures explains much of the difference we see between it and the Sony NEX-6; in fact it's probably better compared to the performance of its peers set 1EV lower (i.e. ISO 400 on the Fujifilm compared to ISO 200 on its rivals)."

Fujifilm X-E2 Review: Digital Photography Review

DPR is aware of this issue, yet they continue to use ISO as the basis for performance comparison, allowing manufacturers to distort public perception by manipulating ISO. The only way to level the playing field would be to shoot test photos under specific standard lighting conditions, using similar shutter speeds and apertures (or better yet, light transmission). ISO could then be ignored and noise performance could be fairly evaluated.

I'm not sure why they gave Sony a pass in the quote above. I believe Sony and Olympus are fudging ISO just like Fuji does.

jens.w · 04-02-2014, 12:51 PM

wow, this ISO-tweaking was something I've never thought about. One more argument to put in the bag when friends wonder why I prefer my K-3 over a canikon alternative

**rbefly** · 04-02-2014, 12:56 PM

Yes, DPR somehow manages to find ways to boost their favs and slag the rest. You would almost think it's intentional, but of course it's not.
Simon says.
Ron

**Rondec** · 04-02-2014, 01:11 PM

Originally posted by audiobomber

ISO fudging has become a pet peeve for me. From DPR's XE-2 review:

"However the X-E2's need for unusually long exposures explains much of the difference we see between it and the Sony NEX-6; in fact it's probably better compared to the performance of its peers set 1EV lower (i.e. ISO 400 on the Fujifilm compared to ISO 200 on its rivals)."

Fujifilm X-E2 Review: Digital Photography Review

DPR is aware of this issue, yet they continue to use ISO as the basis for performance comparison, allowing manufacturers to distort public perception by manipulating ISO. The only way to level the playing field would be to shoot test photos under specific standard lighting conditions, using similar shutter speeds and apertures (or better yet, light transmission). ISO could then be ignored and noise performance could be fairly evaluated.

I'm not sure why they gave Sony a pass in the quote above. I believe Sony and Olympus are fudging ISO just like Fuji does.

I agree. If you are testing, you definitely should shoot manual and then comment after the fact that shooting at iso 1600 causes under exposure by 1/2 stop or, whatever.

JinDesu · 04-02-2014, 01:18 PM

Originally posted by Rondec

I agree. If you are testing, you definitely should shoot manual and then comment after the fact that shooting at iso 1600 causes under exposure by 1/2 stop or, whatever.

I would rather they do that as a secondary. The primary images that show up for comparo should be shot with focal length, aperture, and shutter speed set the same, and then ISO adjusted to make the exposure equal. This should give an accurate interpretation of ISO performance. Then a second set of images can be shot at the same aperture, shutter speed, and ISO to show that one of the cameras underexposes or overexposes, etc.

**Rondec** · 04-02-2014, 01:27 PM

Originally posted by JinDesu

I would rather they do that as a secondary. The primary images that show up for comparo should be shot with focal length, aperture, and shutter speed set the same, and then ISO adjusted to make the exposure equal. This should give an accurate interpretation of ISO performance. Then a second set of images can be shot at the same aperture, shutter speed, and ISO to show that one of the cameras underexposes or overexposes, etc.

I doubt the cameras under expose in real world situations. The issue is that they may require higher iso or longer shutter speeds to get that decent exposure.

audiobomber · 04-02-2014, 01:56 PM

Originally posted by Rondec

I doubt the cameras under expose in real world situations. The issue is that they may require higher iso or longer shutter speeds to get that decent exposure.

That's correct. They overexpose with a longer shutter speed, then lower the exposure with tone mapping. To compare a Pentax exposure fairly to Fuji or Olympus, you would need to activate Highlight correction on the Pentax.

Cannikin · 04-02-2014, 02:16 PM

Originally posted by audiobomber

That's correct. They underexpose with a longer shutter speed, then boost the exposure with tone mapping. To compare a Pentax exposure fairly to Fuji or Olympus, you would need to activate Highlight correction on the Pentax.

Er... no, using a slower shutter speed compensates for the lower ISO, giving you the same exposure.

The whole point of higher ISO is to give you a higher shutter speed when you are aperture constrained. These manufacturers are lying by not giving you the high shutter speed you expect at their reported ISO.

audiobomber · 04-02-2014, 02:22 PM

Originally posted by Cannikin

Er... no, using a slower shutter speed compensates for the lower ISO, giving you the same exposure.

The whole point of higher ISO is to give you a higher shutter speed when you are aperture constrained. These manufacturers are lying by not giving you the high shutter speed you expect at their reported ISO.

Two photos, same exposure:
Fuji @ 1/50s, f/4, ISO 100
Pentax @ 1/100s, f/4, ISO 100

04-01-2014, 06:11 PM	#1
rrstuff Veteran Member Join Date: Sep 2013 Location: Pasadena, CA Photos: Albums Posts: 1,126	Dpreview and K-3s ISO noise I looked through the Exif data of their JPGs... and they expose K-3 with ~50% shorter shutter speeds than Nikon and Canon at the highest ISO settings. Next thing you know they will write that 'high ISO noise is not quite up to par with competition'... facepalm. Pentax K-3 First Impressions Review: Digital Photography Review According to DXO data, K-3 does not have inflated ISO figures, unlike most other cameras. Perhaps that is the reason why so many people in user reviews complain about high iso noise...

04-02-2014, 12:56 PM	#9
rbefly Site Supporter Join Date: Oct 2009 Location: Denver, Colorado Photos: Albums Posts: 2,030	More Coincidences? Yes, DPR somehow manages to find ways to boost their favs and slag the rest. You would almost think it's intentional, but of course it's not. Simon says. Ron

04-02-2014, 01:56 PM	#13
audiobomber Veteran Member Join Date: May 2008 Location: Sudbury, Ontario Photos: Albums Posts: 6,806	Originally posted by Rondec I doubt the cameras under expose in real world situations. The issue is that they may require higher iso or longer shutter speeds to get that decent exposure. That's correct. They overexpose with a longer shutter speed, then lower the exposure with tone mapping. To compare a Pentax exposure fairly to Fuji or Olympus, you would need to activate Highlight correction on the Pentax. Last edited by audiobomber; 04-02-2014 at 02:35 PM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
K-5 noise even with ISO 200 (was: K-5 IIs noise)	lunany	Pentax K-5 & K-5 II	9	12-28-2013 06:54 PM
Pentax K-3 and 3s scheduled to appear in April	JohnBee	Pentax News and Rumors	296	03-23-2013 01:57 PM
K-30 banding noise, aka vertical lines at ISO 1600 and up	qwertyazerty	Troubleshooting and Beginner Help	10	08-13-2012 03:48 AM
DPreview K-01 RAW high ISO samples botched!	JohnBee	Pentax K-01	8	05-04-2012 05:08 AM
k-5 and k-r ISO noise higher than 3200?	metungnasi	Pentax K-5 & K-5 II	14	12-11-2010 11:57 AM

04-01-2014, 07:51 PM	#2
kiwi_jono Loyal Site Supporter Join Date: Dec 2009 Location: Christchurch, New Zealand Photos: Gallery \| Albums Posts: 2,437	I noticed this on DXO. Pentax does a great job with ISO accuracy which does make for a bit of disadvantage on noise test reporting (if the review reporting against set ISO not actual ISO). I think it should be pointed out in reviews like this because it is relevant.

04-01-2014, 08:19 PM	#3
MD Optofonik Veteran Member Join Date: Feb 2010 Location: California Photos: Gallery \| Albums Posts: 962	Please explain.

04-01-2014, 08:59 PM - 2 Likes	#4
kiwi_jono Loyal Site Supporter Join Date: Dec 2009 Location: Christchurch, New Zealand Photos: Gallery \| Albums Posts: 2,437	Have a look at this example ISO Sensitivity of K3 versus Nikon 7100: Pentax K-3 versus Nikon D7100 - Side by side camera comparison - DxOMark When the Nikon is set for ISO 400 its actual sensitivity is measured as ISO 279 (significantly short of 400). But if you look at the K-3 the ISO reading are pretty close to spot on (ISO 400 is actually ISO 381). When you set the cameras to ISO 3200, the K-3 is actually ISO 2933 (around 3000) and 7100 is ISO 2192 (around 2000). So if DPR provide comparative test images at say ISO 400 then the K-3 is achieving a higher ISO and therefore can be expected to have a higher noise component than the Nikon 7100 but people viewing the test images won't generally make an allowance for the greater error in the Nikon (which is really operating at ISO 279 not 400). Compare other Canon / Nikon cameras and you will find that most don't achieve the stated ISO but Pentax is normally about right.

04-01-2014, 09:11 PM	#5
MD Optofonik Veteran Member Join Date: Feb 2010 Location: California Photos: Gallery \| Albums Posts: 962	Thanks!

04-02-2014, 04:45 AM	#6
causey Veteran Member Join Date: Feb 2009 Location: Arlington, VA Photos: Albums Posts: 3,757	Honesty vs. marketing.

04-02-2014, 10:37 AM - 1 Like	#7
audiobomber Veteran Member Join Date: May 2008 Location: Sudbury, Ontario Photos: Albums Posts: 6,806	ISO fudging has become a pet peeve for me. From DPR's XE-2 review: "However the X-E2's need for unusually long exposures explains much of the difference we see between it and the Sony NEX-6; in fact it's probably better compared to the performance of its peers set 1EV lower (i.e. ISO 400 on the Fujifilm compared to ISO 200 on its rivals)." Fujifilm X-E2 Review: Digital Photography Review DPR is aware of this issue, yet they continue to use ISO as the basis for performance comparison, allowing manufacturers to distort public perception by manipulating ISO. The only way to level the playing field would be to shoot test photos under specific standard lighting conditions, using similar shutter speeds and apertures (or better yet, light transmission). ISO could then be ignored and noise performance could be fairly evaluated. I'm not sure why they gave Sony a pass in the quote above. I believe Sony and Olympus are fudging ISO just like Fuji does.

04-02-2014, 12:51 PM	#8
jens.w New Member Join Date: Dec 2013 Posts: 12	wow, this ISO-tweaking was something I've never thought about. One more argument to put in the bag when friends wonder why I prefer my K-3 over a canikon alternative

04-02-2014, 01:11 PM	#10
Rondec Loyal Site Supporter Join Date: Mar 2009 Location: Gladys, Virginia Photos: Gallery Posts: 27,652	Originally posted by audiobomber ISO fudging has become a pet peeve for me. From DPR's XE-2 review: "However the X-E2's need for unusually long exposures explains much of the difference we see between it and the Sony NEX-6; in fact it's probably better compared to the performance of its peers set 1EV lower (i.e. ISO 400 on the Fujifilm compared to ISO 200 on its rivals)." Fujifilm X-E2 Review: Digital Photography Review DPR is aware of this issue, yet they continue to use ISO as the basis for performance comparison, allowing manufacturers to distort public perception by manipulating ISO. The only way to level the playing field would be to shoot test photos under specific standard lighting conditions, using similar shutter speeds and apertures (or better yet, light transmission). ISO could then be ignored and noise performance could be fairly evaluated. I'm not sure why they gave Sony a pass in the quote above. I believe Sony and Olympus are fudging ISO just like Fuji does. I agree. If you are testing, you definitely should shoot manual and then comment after the fact that shooting at iso 1600 causes under exposure by 1/2 stop or, whatever.

04-02-2014, 01:18 PM	#11
JinDesu Veteran Member Join Date: Jun 2011 Location: New York City Photos: Gallery Posts: 5,638	Originally posted by Rondec I agree. If you are testing, you definitely should shoot manual and then comment after the fact that shooting at iso 1600 causes under exposure by 1/2 stop or, whatever. I would rather they do that as a secondary. The primary images that show up for comparo should be shot with focal length, aperture, and shutter speed set the same, and then ISO adjusted to make the exposure equal. This should give an accurate interpretation of ISO performance. Then a second set of images can be shot at the same aperture, shutter speed, and ISO to show that one of the cameras underexposes or overexposes, etc.

04-02-2014, 01:27 PM	#12
Rondec Loyal Site Supporter Join Date: Mar 2009 Location: Gladys, Virginia Photos: Gallery Posts: 27,652	Originally posted by JinDesu I would rather they do that as a secondary. The primary images that show up for comparo should be shot with focal length, aperture, and shutter speed set the same, and then ISO adjusted to make the exposure equal. This should give an accurate interpretation of ISO performance. Then a second set of images can be shot at the same aperture, shutter speed, and ISO to show that one of the cameras underexposes or overexposes, etc. I doubt the cameras under expose in real world situations. The issue is that they may require higher iso or longer shutter speeds to get that decent exposure.

04-02-2014, 02:16 PM	#14
Cannikin Veteran Member Join Date: Mar 2010 Location: SW Washington Posts: 833	Originally posted by audiobomber That's correct. They underexpose with a longer shutter speed, then boost the exposure with tone mapping. To compare a Pentax exposure fairly to Fuji or Olympus, you would need to activate Highlight correction on the Pentax. Er... no, using a slower shutter speed compensates for the lower ISO, giving you the same exposure. The whole point of higher ISO is to give you a higher shutter speed when you are aperture constrained. These manufacturers are lying by not giving you the high shutter speed you expect at their reported ISO.

04-02-2014, 02:22 PM	#15
audiobomber Veteran Member Join Date: May 2008 Location: Sudbury, Ontario Photos: Albums Posts: 6,806	Originally posted by Cannikin Er... no, using a slower shutter speed compensates for the lower ISO, giving you the same exposure. The whole point of higher ISO is to give you a higher shutter speed when you are aperture constrained. These manufacturers are lying by not giving you the high shutter speed you expect at their reported ISO. Two photos, same exposure: Fuji @ 1/50s, f/4, ISO 100 Pentax @ 1/100s, f/4, ISO 100