Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
03-28-2018, 08:48 AM   #31
Pentaxian
Class A's Avatar

Join Date: Aug 2008
Location: Wellington, New Zealand
Posts: 10,056
QuoteOriginally posted by photoptimist Quote
The RAWs from hand-held pixel shift cannot be processed in the same way that RAWs from K-1 pixel shift were.
True, but I expect these to be very different files, i.e., there won't be any way to confuse the two formats, even if the same camera info were present.

I find the term "hand-held pixel shift" very unfortunate. There is clearly a different technology at work that has nothing to do with a precise shifting of sensels in order to obviate demosaicing.

QuoteOriginally posted by photoptimist Quote
It's also possible that the accelerator chip affects the optimal demosaicing parameters of every RAW produced by the K-1ii.
I have no insider information, but I believe that to be highly unlikely. I can see that different noise reduction and sharpening levels may be called for, but the underlying demosaicing strategy should be unaffected.

03-28-2018, 08:52 AM   #32
Pentaxian
Class A's Avatar

Join Date: Aug 2008
Location: Wellington, New Zealand
Posts: 10,056
QuoteOriginally posted by peterh337 Quote
That method will 100% guarantee a loss of compatibility with every new camera, however, so I am sure most apps don't do it.
I believe you have this the wrong way round.

I'm pretty sure most software is finicky about camera model information.

The exceptions are some software applications that deal with DNG files. As the DNG format is fixed, it is possible to read in DNG files from new cameras even with old software. However, that also implies that one typically does not benefit from tailored camera profiles which can make quite a difference to how much one will have to tweak a RAW file.
03-28-2018, 09:31 AM   #33
Veteran Member




Join Date: Jan 2009
Location: London
Posts: 559
"The exceptions are some software applications that deal with DNG files. As the DNG format is fixed, it is possible to read in DNG files from new cameras even with old software. However, that also implies that one typically does not benefit from tailored camera profiles which can make quite a difference to how much one will have to tweak a RAW file"

That makes sense. AIUI, a DNG contains a default profile for the camera. The camera name is not important in order for the software to use this profile. And that is how the old software continues to work.
03-28-2018, 09:36 AM   #34
Pentaxian




Join Date: Apr 2007
Location: Romania
Posts: 11,621
You can open DNGs - should be able to open K-1 II's DNGs too; but perhaps not with optimal result. But check the release notes for e.g. Lightroom - you'll see, for each version, the list of cameras they're adding support for.
Cameras supported by Camera Raw
Just try opening a PEF with a previous version...

03-28-2018, 09:44 AM - 2 Likes   #35
On Hiatus. Back Later.
Loyal Site Supporter
monochrome's Avatar

Join Date: Mar 2008
Location: Flyover Country
Photos: Gallery | Albums
Posts: 24,447
Yikes! Real Resolution IS NOT ‘Hand Held PixelShift’ It is in-camera SuperResolution stacking using the motion sensor data from the IBIS system only to align images 2, 3 and 4 with image 1. There is no deterministic sensor movement associated with ‘creating’ the RAW, as there is with PixelShift.

I wish I could make this bold flashing red.
03-28-2018, 09:55 AM - 2 Likes   #36
Moderator
Not a Number's Avatar

Join Date: Mar 2012
Location: Venice, CA
Posts: 6,017
QuoteOriginally posted by monochrome Quote
Yikes! Real Resolution IS NOT ‘Hand Held PixelShift’ It is in-camera SuperResolution stacking using the motion sensor data from the IBIS system only to align images 2, 3 and 4 with image 1. There is no deterministic sensor movement associated with ‘creating’ the RAW, as there is with PixelShift.

I wish I could make this bold flashing red.
So it's like focus stacking done in camera. Instead of aligning image data it uses the sensor positioning data from IBIS which potentially would be more accurate. Neat!
03-28-2018, 10:06 AM   #37
Pentaxian
photoptimist's Avatar

Join Date: Jul 2016
Photos: Albums
Posts: 3,372
QuoteOriginally posted by Class A Quote
True, but I expect these to be very different files, i.e., there won't be any way to confuse the two formats, even if the same camera info were present.

I find the term "hand-held pixel shift" very unfortunate. There is clearly a different technology at work that has nothing to do with a precise shifting of sensels in order to obviate demosaicing.
Yes, the term may be unfortunate but reflects the tough choices that feature namers face. Should the name reflect the results it creates or how it works?

For most photographers, how it works is less important than the results. And the results of this feature seems to be pixel-shift-like-high-resolution output under hand-held conditions. That it accomplishes this outcome using different internal tricks than the ones used for original pixel-shift may be irrelevant to most photographers. It's only the geekiest photographers (like you and me!) who care about the magic inside the literal black box of the camera.

QuoteOriginally posted by Class A Quote
I have no insider information, but I believe that to be highly unlikely. I can see that different noise reduction and sharpening levels may be called for, but the underlying demosaicing strategy should be unaffected.
I have no insider information, either, so you may right. But when I think about how the accelerator chip might be working it's magic I can imagine Pentax engineers using their knowledge of the camera's CFA color spectra to reduce chromatic noise in the image. Yet a consequence of that process might be a corrected estimate of the R, G, and B pixel values that changes the effective spectra of the three channels. That would affect the demosaicing process almost as if the CFA materials were changed between the K-1 and K-1ii.

We shall see!

---------- Post added 03-28-18 at 11:13 AM ----------

QuoteOriginally posted by monochrome Quote
Yikes! Real Resolution IS NOT ‘Hand Held PixelShift’ It is in-camera SuperResolution stacking using the motion sensor data from the IBIS system only to align images 2, 3 and 4 with image 1. There is no deterministic sensor movement associated with ‘creating’ the RAW, as there is with PixelShift.

I wish I could make this bold flashing red.
But the term "Pixel Shift" is entirely ambiguous on this. It could equal apply to a system that carefully induces a controlled shift in the pixels or a system that carefully measures the shift in the pixels.

One is an IBIS-induced shift in the pixels and the other is an IBIS-measured shift in the pixels. Both systems use the fact that the pixels have shifted even if the first might be considered the superior technology albeit only usable under more restrictive conditions.

Thus, both involve shifted pixels and both could be called "pixel shift."
03-28-2018, 10:24 AM - 1 Like   #38
Veteran Member




Join Date: Jan 2009
Location: London
Posts: 559
There is absolutely nothing which a "chip" can do (in terms of image processing) which the camera could not do in software

In this case it probably does it faster, because fast processors draw a lot of power and a camera like the K1 is very limited in that department.


Last edited by peterh337; 03-28-2018 at 10:29 AM.
03-28-2018, 10:32 AM - 1 Like   #39
On Hiatus. Back Later.
Loyal Site Supporter
monochrome's Avatar

Join Date: Mar 2008
Location: Flyover Country
Photos: Gallery | Albums
Posts: 24,447
QuoteOriginally posted by photoptimist Quote

Thus, both involve shifted pixels and both could be called "pixel shift."
Except they aren’t. One is called PixelShift. The other is called Real Resolution. One moves the sensor intentionally. One captures information about random sensor movement and uses it. The pixels aren’t actually shifted by the process.

There are LCD product shots out there with the Real Resolution menu choice highlighted. This is a perfect example of how Ricoh’s lack of Pentax brand marketing is abysmal. It is such a clever use of their tech. They should be brand-managing all over this. Real Resolution, AstroTracer, PixelShift, IBIS (SR) - nobody knows what they are. And they all ‘shift pixels’ or photosites (ambiguously).

Last edited by monochrome; 03-28-2018 at 10:40 AM.
03-28-2018, 10:42 AM   #40
Pentaxian
photoptimist's Avatar

Join Date: Jul 2016
Photos: Albums
Posts: 3,372
QuoteOriginally posted by peterh337 Quote
There is absolutely nothing which a "chip" can do (in terms of image processing) which the camera could not do in software

In this case it probably does it faster, because fast processors draw a lot of power and a camera like the K1 is very limited in that department.
That's not true because the chip has more access to more data about the camera than the external software does.

The "chip" may have access to full-frame calibration files (such as the hot-pixel map and dust map) that the external software does not.

The "chip" may have access to detailed IBIS system sensor measurements that the external software does not.

The "chip" may have access to detailed sensor and ADC operating data (internal voltages, etc.) that the external software does not.

The "chip" may have access to factory calibration data that the external software does not.

Some of these things may be in EXIF but I doubt they all are are (for example, the full-frame calibration files could potentially double the size of the RAW image).
03-28-2018, 10:57 AM   #41
Veteran Member




Join Date: Jan 2009
Location: London
Posts: 559
I meant software running on the camera's CPU.
03-28-2018, 11:08 AM   #42
On Hiatus. Back Later.
Loyal Site Supporter
monochrome's Avatar

Join Date: Mar 2008
Location: Flyover Country
Photos: Gallery | Albums
Posts: 24,447
QuoteOriginally posted by peterh337 Quote
I meant software running on the camera's CPU.
Video functions can run in software on a PC CPU, yet we have dedicated NVIDIA dedicated video cards to handle those functions
03-28-2018, 11:23 AM   #43
Pentaxian




Join Date: Apr 2014
Location: Baltimore
Posts: 1,110
QuoteOriginally posted by monochrome Quote
This is a perfect example of how Ricoh’s lack of Pentax brand marketing is abysmal. It is such a clever use of their tech. They should be brand-managing all over this. Real Resolution, AstroTracer, PixelShift, IBIS (SR) - nobody knows what they are. And they all ‘shift pixels’ or photosites (ambiguously).
Yes. And I suspect a lot of it has to do with Pentax being insularly Japanese, Ricoh being rather Japanese, but also divided in their interests (and the left hand not talking to the right hand, which of course never ever happens in western corporations, ahem...), plus typical corporate group think BS. But I want to really highlight the Japanese-ness of this. Lack of western (really, American...) branding "skills", insularity, real differences in the way the languages work, leading to bad translations misunderstandings &etc on the rhetoric front.

The language difference is a very salient point. Japanese is the most simple advanced language out there, I think. A lot is conveyed by context and nuance. And the way English is taught in Japan is apparently abysmal, even though English instruction is ubiquitous. No surprise to me therefore that none of this is being described well. And based on what I've seen with my own eyes with Pentax U.S., no surprise that it's not being well presented by North American Pentax....
03-28-2018, 11:49 AM   #44
On Hiatus. Back Later.
Loyal Site Supporter
monochrome's Avatar

Join Date: Mar 2008
Location: Flyover Country
Photos: Gallery | Albums
Posts: 24,447
QuoteOriginally posted by texandrews Quote
Yes. And I suspect a lot of it has to do with Pentax being insularly Japanese, Ricoh being rather Japanese, but also divided in their interests (and the left hand not talking to the right hand, which of course never ever happens in western corporations, ahem...), plus typical corporate group think BS. But I want to really highlight the Japanese-ness of this. Lack of western (really, American...) branding "skills", insularity, real differences in the way the languages work, leading to bad translations misunderstandings &etc on the rhetoric front.

The language difference is a very salient point. Japanese is the most simple advanced language out there, I think. A lot is conveyed by context and nuance. And the way English is taught in Japan is apparently abysmal, even though English instruction is ubiquitous. No surprise to me therefore that none of this is being described well. And based on what I've seen with my own eyes with Pentax U.S., no surprise that it's not being well presented by North American Pentax....
Canon, Nikon and Sony seem to do it. Fuji uses viral and YouTube. Only / Panny do it.

Why can’t Ricoh?

Last edited by monochrome; 03-28-2018 at 11:56 AM.
03-28-2018, 12:24 PM - 2 Likes   #45
Pentaxian




Join Date: Apr 2014
Location: Baltimore
Posts: 1,110
QuoteOriginally posted by monochrome Quote
Canon, Nikon and Sony seem to do it. Fuji uses viral and YouTube. Only / Panny do it.

Why can’t Ricoh?
It's a reasonable question. And a damned good one. I think there are several possible answers, and probably combinations of answers (for Pentax):
1. Sony is a huge company, and has way more practice. But remember Betamax? How about those memory sticks?
2. Nikon is bigger than Pentax, and has more practice. Although, it's not like Nikon isn't flubbing stuff lately...and their corporate speak is pretty bad
3. Canon is bigger, has more practice. But see above----and their corporate speak is atrocious!
4. Fuji was a bigger company. They are doing well with what they are doing. But x-trans?
5. Oly/Panny don't look all that hot to me, although better than Pentax.
6. Hoya truly was a disaster for Pentax. None of these other companies suffered that.
7. Ricoh I don't think gets it with the consumer market. They were selling copiers to businesses, and had a good support reputation. It totally mystifies me why they have failed in that aspect with Pentax, that should have been easy. But connecting with businesses is different than with consumers.
8. I wonder if Ricoh actually knew what to do with Pentax after acquisition.

My guesses.

Last edited by texandrews; 03-28-2018 at 01:52 PM.
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!
camera, dslr, full frame, full-frame, ii, k-1, k-1 mark ii, k1, mark, pentax k-1, pentax k-1 mark, photographers, pictures, pixels, pp, psr, results, reviews, sensor, shift, spectra, term, utility, weeks
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Pentax K-1 Mark II DSLR camera detailed specifications leaked Craig66 Pentax News and Rumors 389 02-19-2018 10:34 AM
Pentax K3 ii vs Canon 5D mark ii mogge Welcomes and Introductions 49 03-27-2016 02:39 AM
Lens Reviews - Camera used not listed on 3rd party lens reviews? ak_kiwi Pentax SLR Lens Discussion 3 02-15-2014 01:34 PM
Reviews really missed the mark on Kr AWB in tungsten light LRB21 Pentax K-r 3 09-06-2011 03:12 PM



All times are GMT -7. The time now is 10:24 PM. | 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