Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
08-10-2015, 02:16 PM   #46
Pentaxian
thibs's Avatar

Join Date: Jun 2007
Location: Belgium
Photos: Albums
Posts: 5,292
QuoteOriginally posted by Edgar_in_Indy Quote
I've been very tempted by the GR in the past, and could see myself owning one if the right deal comes along. If I could just get the LX100's lens on the GR's sensor, without it being any larger...
Yep, I get that but that lens is soo sharp . Actually a good optical Tele complement would be tempting to me. But VF will come first.

That thing really is designed with photog on mind. Just look at the menus content is revealing. A lovely cam really.

08-13-2015, 12:04 AM   #47
Senior Member




Join Date: Nov 2009
Location: Newcastle, AU!
Posts: 259
QuoteOriginally posted by falconeye Quote

Pixelshift is a variant of image stacking with computable effect on sensor scores. It effectively multiplies base iso by 1/#frames and therefore, pushes DR and Dxomarks.

However, it is no sensor score and must be excluded from the official list of scores. Dxo somewhat break their own rule for their own DxO ONE camera which is sad.
I'd be fine with this if they awarded results to specific sensors but they dont, they award it to the camera. The camera has these options, so regardless of how clunky/specific its use is, should be included. It's like when dpreview used to weight image quality based on default jpg processing. You can get better image quality, do it!
08-13-2015, 04:37 AM   #48
Pentaxian
Edgar_in_Indy's Avatar

Join Date: May 2010
Location: Indiana, USA
Photos: Gallery
Posts: 2,637
QuoteOriginally posted by bibz Quote
I'd be fine with this if they awarded results to specific sensors but they dont, they award it to the camera. The camera has these options, so regardless of how clunky/specific its use is, should be included. It's like when dpreview used to weight image quality based on default jpg processing. You can get better image quality, do it!
But wouldn't that kind of be like DxO basing image quality on a camera's ability to shoot a bracketed HDR image? A lot of cameras can do that, but that doesn't mean the sensor actually performs any better. Using the HDR mode as the basis for the sensor rating would skew the results.
08-13-2015, 05:05 AM   #49
Veteran Member
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,863
QuoteOriginally posted by bibz Quote
I'd be fine with this if they awarded results to specific sensors but they dont, they award it to the camera.
You are misled.
DxOMark is a sensor-only benchmark figure. It doesn't even weight resolution at all.

08-13-2015, 05:53 AM   #50
Site Supporter
Site Supporter
Zygonyx's Avatar

Join Date: Jun 2011
Location: Ile de France
Posts: 3,101
It does, but only when testing lenses
08-13-2015, 06:33 AM   #51
Pentaxian




Join Date: Nov 2013
Posts: 4,761
QuoteOriginally posted by Edgar_in_Indy Quote
But wouldn't that kind of be like DxO basing image quality on a camera's ability to shoot a bracketed HDR image? A lot of cameras can do that, but that doesn't mean the sensor actually performs any better. Using the HDR mode as the basis for the sensor rating would skew the results.
+1 I can take any camera and shoot many pictures and merge then and get better quality. I can also use longer exposure time and lower the isos. With theses strategies I can increase the dynamic range and color deph to go beyond what of the best FF even with a old APSC body.

I can also use a pano head and stitch many shoot together and get picture with resolution of hundred mega pixels or even gigapixels.

What the sensor can do on one side and what you can do with some tricks on very specific shooting conditions is not the same matter at all.
08-13-2015, 06:34 AM   #52
Site Supporter
Site Supporter




Join Date: Mar 2009
Location: Gladys, Virginia
Photos: Gallery
Posts: 16,437
QuoteOriginally posted by Edgar_in_Indy Quote
But wouldn't that kind of be like DxO basing image quality on a camera's ability to shoot a bracketed HDR image? A lot of cameras can do that, but that doesn't mean the sensor actually performs any better. Using the HDR mode as the basis for the sensor rating would skew the results.
Well. DXO Mark did test the "super raw plus" on their own DXO Mark One and gave it a score of 85 (normal raw score of 70). I guess when you are selling something, you can test it in such a way as to maximize your own scores. Just not other folk's gear.
08-13-2015, 09:22 AM   #53
Veteran Member
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,863
QuoteOriginally posted by Zygonyx Quote
It does, but only when testing lenses
Yeah, I checked and you're right, they dropped their P-MPix and use some "DxOMark lens" now. What I meant then is "DxOmark sensor" which isn't called "DxOmark camera" btw ...

QuoteOriginally posted by Rondec Quote
Well. DXO Mark did test the "super raw plus" on their own DXO Mark One and gave it a score of 85 (normal raw score of 70). I guess when you are selling something, you can test it in such a way as to maximize your own scores. Just not other folk's gear.
This is what I meant when saying above "Dxo somewhat break their own rule for their own DxO ONE camera which is sad."

08-13-2015, 02:50 PM   #54
Site Supporter
Site Supporter
Zygonyx's Avatar

Join Date: Jun 2011
Location: Ile de France
Posts: 3,101
Not even granted this will help them selling more Ones btw.... I am not convinced at all by this new and so expensive "geekery".

Last edited by Zygonyx; 08-13-2015 at 02:57 PM.
08-13-2015, 04:30 PM   #55
Site Supporter
Site Supporter




Join Date: Mar 2009
Location: Gladys, Virginia
Photos: Gallery
Posts: 16,437
QuoteOriginally posted by Zygonyx Quote
Not even granted this will help them selling more Ones btw.... I am not convinced at all by this new and so expensive "geekery".
There are a lot of cameras that have multiple exposures to reduce noise or HDR modes. They have utility in certain situations and they are nice to know about, but I don't really think that testing them tells you much.

The K3 II sensor shift image enhancement definitely is known about and I don't really think that DXO Mark needs to test it for photographers to know that it is present on the camera and whether or not it will be useful for them.
08-15-2015, 07:07 PM   #56
Senior Member




Join Date: Nov 2009
Location: Newcastle, AU!
Posts: 259
They aren't just measuring the sensor, they're measuring the file output, so theres a TONNE of stuff inbetween what the sensor sees and what gets written to the card. The pixel shift technology isn't just taking 4 shots, I thought it was taking data readings for each sub-pixel. From what I can see with that output theres a significant increase in image quality using much more data instead of guessing or averaging. It gets written to the card as one file. HDR modes are using different exposures and guessing further, leading to worse iq imo.

I guess I just don't think dxo's 'technical' facade and basic result interpretation benefits as much as it could. We need Cicala-esque sensor geekery!
08-16-2015, 12:44 AM   #57
Veteran Member
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,863
QuoteOriginally posted by bibz Quote
The pixel shift technology isn't just taking 4 shots
It is, with shifting the sensor by 1 pixel in between shots. It stores 4 separate full images concatenatwd into a single file, like what a zip archive does.

BTW, HDR from multiple shots is the same exactly, provided you skip the second step of tone mapping. The result is deterministic and a raw file with added dynamic range, depending on separation of EV during bracketing. This is best seen by Lightroom 6's new capability to process an HDR bracket into a single raw with increased bit depth.

Moreover, modern sensors feature a full digital read out and there is much less happening in between reading the sensor and writing the file than you may think. And DxO's methodology includes to protect against processed sensor data, like averaging, tone mapping or black level subtraction. What they test is indeed the sensor output basically.

You can easily do the same with a Leica Monochrome: Have a revolving filter wheel in front of the lens with RGGB filters rotating during a burst of 4 shots. While such an accessory is easily doable, it wouldn't qualify the Leica Monochrome for a better DxO sensor score. Maybe, you can now accept why the pixelshift mode is a nice camera capability but must remain outside the domain of what is tested by DxO mark sensor.

Last edited by falconeye; 08-16-2015 at 01:00 AM.
08-16-2015, 02:02 AM   #58
Pentaxian
thibs's Avatar

Join Date: Jun 2007
Location: Belgium
Photos: Albums
Posts: 5,292
It might be interesting to add such a score but only as a secondary not used to categorize etc which in the end would be a lot of work for what it would bring to the table.

Testing lenses are a lot more useful IMO.
08-16-2015, 08:34 AM   #59
Veteran Member
MJSfoto1956's Avatar

Join Date: Dec 2013
Location: Boston
Posts: 1,287
QuoteOriginally posted by falconeye Quote
You can easily do the same with a Leica Monochrome: Have a revolving filter wheel in front of the lens with RGGB filters rotating during a burst of 4 shots. While such an accessory is easily doable, it wouldn't qualify the Leica Monochrome for a better DxO sensor score. Maybe, you can now accept why the pixelshift mode is a nice camera capability but must remain outside the domain of what is tested by DxO mark sensor.
Actually, merging 3 or 4 images together will indeed increase DR significantly -- this has been demonstrated over and over.

Michael
08-16-2015, 12:23 PM   #60
Pentaxian




Join Date: Nov 2013
Posts: 4,761
QuoteOriginally posted by bibz Quote
They aren't just measuring the sensor, they're measuring the file output, so theres a TONNE of stuff inbetween what the sensor sees and what gets written to the card. The pixel shift technology isn't just taking 4 shots, I thought it was taking data readings for each sub-pixel. From what I can see with that output theres a significant increase in image quality using much more data instead of guessing or averaging. It gets written to the card as one file. HDR modes are using different exposures and guessing further, leading to worse iq imo.

I guess I just don't think dxo's 'technical' facade and basic result interpretation benefits as much as it could. We need Cicala-esque sensor geekery!
If you want more resolution the most practical solution are either shooting large/medium format film and scan, take a MF digital or latest Canon 50MP FF. Then less practical but still applicable in many situations, even with some moving sbujects is panos shoots. This work with everything from smartphones to latest FF.

Pixel shift might be practical in a very few occasion, that's true. In some cases it might be more practical than pano shoots but the situation where it apply are more limited and the gain in quality are more limited because your pano can always be made of more shoots as necessary !

So the overall impact is very small.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
aps-c, body, cameras, canon, crop, dxo, dxomark, ff, ii, k-3, k-3ii, lr, marks, mask, motion, nikon, pentax, pentax news, pentax rumors, pixel, results, score, sensor, sensors, shift, tests, thread
Thread Tools Search this Thread
Search this Thread:

Advanced Search


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 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 11 07-05-2015 10:36 AM
New K-3ii vict Pentax K-3 8 05-22-2015 01:05 AM



All times are GMT -7. The time now is 08:12 AM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top