DxOMark measures K-3II

bibz · 08-08-2015, 06:02 PM

Pixelshift results would have been interesting. Most interesting. I think the D7200 difference is entirely in software. They've been working with 24mp sensors for quite a while now and I'm sure have tweaked it for that kind of output. Pentax's approach will always be different (hopefully that still applies to FF), and I prefer what they do.

Also Pentax shouldn't have given them a 645z earlier as they were sold out. I'd rather it go to a user then whatever dxo is.

Edgar_in_Indy · 08-08-2015, 07:01 PM

Originally posted by Winder

I check out DxO just like most of the people here, but I really don't but a whole lot importance on what they report.

I hear a lot of people scoff at DxO sensor scores, but I've come to respect them since they match what I've seen with my own eyes.

I've progressed from early DSLR's starting with the Olympus e-410 and then e-520, and then Canon t2i, Pentax K-x, and finally Pentax K-30. I also owned an Olympus PEN E-P1 m43 mirrorless along the way. And my compacts have included the Olympus XZ-1, Sony RX-100-II, Sony RX-1, and now the Panasonic LX100. And I have always found that DxO scores track very closely with the relative real-world image quality from each of my cameras.

**hjoseph7** · 08-08-2015, 09:44 PM

Originally posted by Edgar_in_Indy

I hear a lot of people scoff at DxO sensor scores, but I've come to respect them since they match what I've seen with my own eyes.

I've progressed from early DSLR's starting with the Olympus e-410 and then e-520, and then Canon t2i, Pentax K-x, and finally Pentax K-30. I also owned an Olympus PEN E-P1 m43 mirrorless along the way. And my compacts have included the Olympus XZ-1, Sony RX-100-II, Sony RX-1, and now the Panasonic LX100. And I have always found that DxO scores track very closely with the relative real-world image quality from each of my cameras.

Seems like the K-5IIs still beats the K3 when it comes to sensors according to DxO...

gorme · 08-09-2015, 01:57 AM

Originally posted by bibz

Pixelshift results would have been interesting. Most interesting. I think the D7200 difference is entirely in software. They've been working with 24mp sensors for quite a while now and I'm sure have tweaked it for that kind of output. Pentax's approach will always be different (hopefully that still applies to FF), and I prefer what they do.

Also Pentax shouldn't have given them a 645z earlier as they were sold out. I'd rather it go to a user then whatever dxo is.

From what I know the maximal difference because of software is around 3 points. So the sensor technology is also involved in the difference between K3-II and D7200.

Simen1 · 08-09-2015, 02:45 AM - **1 Like**

The DR are highly affected by how low the lowest ISO setting are. The ISO 80 that really measured to about ISO 70 makes the K-5 series shine in DR measurement. The ISO range is programmable via camera firmware.

DxOMark forum admin, Sophie, told me the 3. of june:

Quote:

I've just got confirmation that the sensor review for the Pentax 645Z will be published this summer. Unfortunately, I cannot provide you with any specific date, since our priorities in terms of test production is subject to variations.

I asked for an ETA on Friday, but i guess she was on vacation or just busy that day. Anyway, i'm happy it gets tested.

**Rondec** · 08-09-2015, 02:50 AM - **1 Like**

Originally posted by hjoseph7

Seems like the K-5IIs still beats the K3 when it comes to sensors according to DxO...

If you look at the SNR and dynamic range curves, they lie right on top of each other. The K5 II does have noise reduction applied to RAW files per DXO Mark above iso 1600.

The only difference is that the base iso on the K5 II is 80 and the K3 II is iso 100. This produces the .5 EV difference in measured maximal dynamic range (14.1 versus 13.6). The final scores for the K5 II and K3 II are 82 and 80 respectively. Not a visible difference. DXO Mark says that there needs to be a difference of 5 points to see a difference in sensors. On the other hand, resolution is not included in their score and the K3 does have better resolution with 24 megapixels than the K5 model cameras.

I own a K3 and a K5 II and there is minimal difference in ability to process these cameras, but the K3 does have better resolution at lower isos.

gorme · 08-09-2015, 05:14 AM

Originally posted by Simen1

The DR are highly affected by how low the lowest ISO setting are. The ISO 80 that really measured to about ISO 70 makes the K-5 series shine in DR measurement. The ISO range is programmable via camera firmware.

DxOMark forum admin, Sophie, told me the 3. of june:

I asked for an ETA on Friday, but i guess she was on vacation or just busy that day. Anyway, i'm happy it gets tested.

I can try to get a precise date for you...

clockworkrat · 08-09-2015, 06:25 AM

Originally posted by Edgar_in_Indy

I hear a lot of people scoff at DxO sensor scores, but I've come to respect them since they match what I've seen with my own eyes.

I've progressed from early DSLR's starting with the Olympus e-410 and then e-520, and then Canon t2i, Pentax K-x, and finally Pentax K-30. I also owned an Olympus PEN E-P1 m43 mirrorless along the way. And my compacts have included the Olympus XZ-1, Sony RX-100-II, Sony RX-1, and now the Panasonic LX100. And I have always found that DxO scores track very closely with the relative real-world image quality from each of my cameras.

How does the LX100 compare to the RX100? Is there a noticeable improvement in image quality?

Nicolas06 · 08-09-2015, 06:49 AM - **1 Like**

Originally posted by pathdoc

Subtext: we did not test in areas that would show Pentax cameras to be superior?

If we think that DxO algorithms are based on standard bayer matrix and do not support X-trans sensor for example, it seems only logical that they can't analyse pixel shift performance.

DxO is a small company, they need to make money to stay in business and their DxOMark is just a by product of their bundled calibration/optical/sensor correction applied automatically to each camera they support in DxO Optics Pro.

**Mikesul** · 08-09-2015, 07:02 AM

+1 Nicolas06 for a good rational explanation. Forums are too quick to criticize companies without realizing the constraints those companies have. DXOMark is a gift to us and a marketing tool for them. We should be grateful for the utility it does offer.

totsmuyco · 08-09-2015, 07:12 AM

Originally posted by hjoseph7

Seems like the K-5IIs still beats the K3 when it comes to sensors according to DxO...

Though the difference is negligibIe, I still hope that the next APS-C flagship will be better than the K-5 series. Otherwise, there will b le NP reason for some of us to upgrade.

Just kidding.

rawr · 08-09-2015, 07:19 AM

I like DxOMark. They are useful. I do have a gripe with their lens testing [in]consistency, but not their sensor scoring.

As others have noted, while the individual scores are a useful shorthand, the real utility of the DxOMark process is the measurement charts and curves. It's very useful, for example, to follow the dynamic range profile sometimes between cameras - where it starts from, where it looks good, where it falls apart.

pathdoc · 08-09-2015, 07:24 AM

Nicolas06: In that case they need to admit their inability to perform the test. This is what honest scientists would do.

Nicolas06 · 08-09-2015, 07:47 AM

Originally posted by pathdoc

Nicolas06: In that case they need to admit their inability to perform the test. This is what honest scientists would do.

Man, give them a rest... I know we all love Pentax but they said they didn't test it explicitely as a warning so we are all aware. To me that quite honest.

And I'am quite sure they have the hability to do it... They design raw processing software, they design hardware to process photos in phone and cameras andt they recently made a small camera... I'd say their set of skill for anything related to photo processing is quite respectable. But making the dedicated software for such things might take years and cost millions euros... If that just to satisfy a few fanboys that will always find something more to argue anway, that's not very productive.

gorme · 08-10-2015, 02:45 AM

Originally posted by Simen1

The DR are highly affected by how low the lowest ISO setting are. The ISO 80 that really measured to about ISO 70 makes the K-5 series shine in DR measurement. The ISO range is programmable via camera firmware.

DxOMark forum admin, Sophie, told me the 3. of june:

I asked for an ETA on Friday, but i guess she was on vacation or just busy that day. Anyway, i'm happy it gets tested.

You should have more info about it soon, the sensor really shines

(cannot say more, sorry guys)

08-09-2015, 02:45 AM - 1 Like	#20
Simen1 Veteran Member Join Date: Jul 2009 Location: Tromsø, Norway Photos: Albums Posts: 1,031	The DR are highly affected by how low the lowest ISO setting are. The ISO 80 that really measured to about ISO 70 makes the K-5 series shine in DR measurement. The ISO range is programmable via camera firmware. DxOMark forum admin, Sophie, told me the 3. of june: Quote: I've just got confirmation that the sensor review for the Pentax 645Z will be published this summer. Unfortunately, I cannot provide you with any specific date, since our priorities in terms of test production is subject to variations. I asked for an ETA on Friday, but i guess she was on vacation or just busy that day. Anyway, i'm happy it gets tested. Last edited by Simen1; 08-09-2015 at 02:52 AM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
How do you see the K-3ii vs K-5iis?	Conqueror	Pentax DSLR Discussion	11	01-27-2016 07:12 AM
K-3II First Impressions	cali92rs	Pentax K-3 & K-3 II	9	07-17-2015 05:05 AM
Firmware the same for K-3 & K-3II?	glee46	Pentax DSLR Discussion	5	07-05-2015 06:02 PM
K-3 (not K-3II) future availability & price guesses	Lititz	Pentax K-3 & K-3 II	11	07-05-2015 10:36 AM
New K-3ii	vict	Pentax K-3 & K-3 II	8	05-22-2015 01:05 AM

08-08-2015, 06:02 PM	#16
bibz Senior Member Join Date: Nov 2009 Location: Newcastle, AU! Posts: 276	Pixelshift results would have been interesting. Most interesting. I think the D7200 difference is entirely in software. They've been working with 24mp sensors for quite a while now and I'm sure have tweaked it for that kind of output. Pentax's approach will always be different (hopefully that still applies to FF), and I prefer what they do. Also Pentax shouldn't have given them a 645z earlier as they were sold out. I'd rather it go to a user then whatever dxo is.

08-08-2015, 07:01 PM	#17
Edgar_in_Indy Veteran Member Join Date: May 2010 Location: Indiana, USA Photos: Gallery Posts: 2,685	Originally posted by Winder I check out DxO just like most of the people here, but I really don't but a whole lot importance on what they report. I hear a lot of people scoff at DxO sensor scores, but I've come to respect them since they match what I've seen with my own eyes. I've progressed from early DSLR's starting with the Olympus e-410 and then e-520, and then Canon t2i, Pentax K-x, and finally Pentax K-30. I also owned an Olympus PEN E-P1 m43 mirrorless along the way. And my compacts have included the Olympus XZ-1, Sony RX-100-II, Sony RX-1, and now the Panasonic LX100. And I have always found that DxO scores track very closely with the relative real-world image quality from each of my cameras.

08-08-2015, 09:44 PM	#18
hjoseph7 Site Supporter Join Date: May 2014 Location: Minnesota Photos: Gallery \| Albums Posts: 2,047	Originally posted by Edgar_in_Indy I hear a lot of people scoff at DxO sensor scores, but I've come to respect them since they match what I've seen with my own eyes. I've progressed from early DSLR's starting with the Olympus e-410 and then e-520, and then Canon t2i, Pentax K-x, and finally Pentax K-30. I also owned an Olympus PEN E-P1 m43 mirrorless along the way. And my compacts have included the Olympus XZ-1, Sony RX-100-II, Sony RX-1, and now the Panasonic LX100. And I have always found that DxO scores track very closely with the relative real-world image quality from each of my cameras. Seems like the K-5IIs still beats the K3 when it comes to sensors according to DxO...

08-09-2015, 01:57 AM	#19
gorme Senior Member Join Date: May 2014 Location: Galway - Ireland Posts: 213	Originally posted by bibz Pixelshift results would have been interesting. Most interesting. I think the D7200 difference is entirely in software. They've been working with 24mp sensors for quite a while now and I'm sure have tweaked it for that kind of output. Pentax's approach will always be different (hopefully that still applies to FF), and I prefer what they do. Also Pentax shouldn't have given them a 645z earlier as they were sold out. I'd rather it go to a user then whatever dxo is. From what I know the maximal difference because of software is around 3 points. So the sensor technology is also involved in the difference between K3-II and D7200.

08-09-2015, 02:50 AM - 1 Like	#21
Rondec Loyal Site Supporter Join Date: Mar 2009 Location: Gladys, Virginia Photos: Gallery Posts: 27,653	Originally posted by hjoseph7 Seems like the K-5IIs still beats the K3 when it comes to sensors according to DxO... If you look at the SNR and dynamic range curves, they lie right on top of each other. The K5 II does have noise reduction applied to RAW files per DXO Mark above iso 1600. The only difference is that the base iso on the K5 II is 80 and the K3 II is iso 100. This produces the .5 EV difference in measured maximal dynamic range (14.1 versus 13.6). The final scores for the K5 II and K3 II are 82 and 80 respectively. Not a visible difference. DXO Mark says that there needs to be a difference of 5 points to see a difference in sensors. On the other hand, resolution is not included in their score and the K3 does have better resolution with 24 megapixels than the K5 model cameras. I own a K3 and a K5 II and there is minimal difference in ability to process these cameras, but the K3 does have better resolution at lower isos.

08-09-2015, 05:14 AM	#22
gorme Senior Member Join Date: May 2014 Location: Galway - Ireland Posts: 213	Originally posted by Simen1 The DR are highly affected by how low the lowest ISO setting are. The ISO 80 that really measured to about ISO 70 makes the K-5 series shine in DR measurement. The ISO range is programmable via camera firmware. DxOMark forum admin, Sophie, told me the 3. of june: I asked for an ETA on Friday, but i guess she was on vacation or just busy that day. Anyway, i'm happy it gets tested. I can try to get a precise date for you...

08-09-2015, 06:25 AM	#23
clockworkrat Veteran Member Join Date: May 2015 Location: Black Isle, Scotland Posts: 405	Originally posted by Edgar_in_Indy I hear a lot of people scoff at DxO sensor scores, but I've come to respect them since they match what I've seen with my own eyes. I've progressed from early DSLR's starting with the Olympus e-410 and then e-520, and then Canon t2i, Pentax K-x, and finally Pentax K-30. I also owned an Olympus PEN E-P1 m43 mirrorless along the way. And my compacts have included the Olympus XZ-1, Sony RX-100-II, Sony RX-1, and now the Panasonic LX100. And I have always found that DxO scores track very closely with the relative real-world image quality from each of my cameras. How does the LX100 compare to the RX100? Is there a noticeable improvement in image quality?

08-09-2015, 06:49 AM - 1 Like	#24
Nicolas06 Veteran Member Join Date: Nov 2013 Posts: 4,854	Originally posted by pathdoc Subtext: we did not test in areas that would show Pentax cameras to be superior? If we think that DxO algorithms are based on standard bayer matrix and do not support X-trans sensor for example, it seems only logical that they can't analyse pixel shift performance. DxO is a small company, they need to make money to stay in business and their DxOMark is just a by product of their bundled calibration/optical/sensor correction applied automatically to each camera they support in DxO Optics Pro.

08-09-2015, 07:02 AM	#25
Mikesul Loyal Site Supporter Join Date: Sep 2009 Location: Phoenix, AZ Posts: 7,594	+1 Nicolas06 for a good rational explanation. Forums are too quick to criticize companies without realizing the constraints those companies have. DXOMark is a gift to us and a marketing tool for them. We should be grateful for the utility it does offer.

08-09-2015, 07:12 AM	#26
totsmuyco Pentaxian Join Date: Aug 2013 Location: Iloilo City Photos: Gallery Posts: 2,276	Originally posted by hjoseph7 Seems like the K-5IIs still beats the K3 when it comes to sensors according to DxO... Though the difference is negligibIe, I still hope that the next APS-C flagship will be better than the K-5 series. Otherwise, there will b le NP reason for some of us to upgrade. Just kidding.

08-09-2015, 07:19 AM	#27
rawr Veteran Member Join Date: Jun 2009 Posts: 11,913 Original Poster	I like DxOMark. They are useful. I do have a gripe with their lens testing [in]consistency, but not their sensor scoring. As others have noted, while the individual scores are a useful shorthand, the real utility of the DxOMark process is the measurement charts and curves. It's very useful, for example, to follow the dynamic range profile sometimes between cameras - where it starts from, where it looks good, where it falls apart.

08-09-2015, 07:24 AM	#28
pathdoc Pentaxian Join Date: Mar 2015 Photos: Gallery \| Albums Posts: 6,381	Nicolas06: In that case they need to admit their inability to perform the test. This is what honest scientists would do.

08-09-2015, 07:47 AM	#29
Nicolas06 Veteran Member Join Date: Nov 2013 Posts: 4,854	Originally posted by pathdoc Nicolas06: In that case they need to admit their inability to perform the test. This is what honest scientists would do. Man, give them a rest... I know we all love Pentax but they said they didn't test it explicitely as a warning so we are all aware. To me that quite honest. And I'am quite sure they have the hability to do it... They design raw processing software, they design hardware to process photos in phone and cameras andt they recently made a small camera... I'd say their set of skill for anything related to photo processing is quite respectable. But making the dedicated software for such things might take years and cost millions euros... If that just to satisfy a few fanboys that will always find something more to argue anway, that's not very productive.

08-10-2015, 02:45 AM	#30
gorme Senior Member Join Date: May 2014 Location: Galway - Ireland Posts: 213	Originally posted by Simen1 The DR are highly affected by how low the lowest ISO setting are. The ISO 80 that really measured to about ISO 70 makes the K-5 series shine in DR measurement. The ISO range is programmable via camera firmware. DxOMark forum admin, Sophie, told me the 3. of june: I asked for an ETA on Friday, but i guess she was on vacation or just busy that day. Anyway, i'm happy it gets tested. You should have more info about it soon, the sensor really shines (cannot say more, sorry guys)