• Please use real names.

    Greetings to all who have registered to OPF and those guests taking a look around. Please use real names. Registrations with fictitious names will not be processed. REAL NAMES ONLY will be processed

    Firstname Lastname

    Register

    We are a courteous and supportive community. No need to hide behind an alia. If you have a genuine need for privacy/secrecy then let me know!
  • Welcome to the new site. Here's a thread about the update where you can post your feedback, ask questions or spot those nasty bugs!

How is Ricoh-Pentax "Multizone" WB delivered to RAW images?

Asher Kelman

OPF Owner/Editor-in-Chief
It's really impressive that Ricoh has applied a very sophisticated machine learning game to the challenge of areas of a scene being lit by light of markedly different hues and intensities. The technology is called multi-pattern WB, but really it should be simply considered as multi-zone WB, as that's what is achieved.

Michael Nagel pointed to this unique WB processing option here, post # 20.

But can we shoot with a Ricoh or Pentax camera set to this more intelligent WB and still have the benefits applied in processing RAW images, as opposed to using the very well processed JPG's straight from the camera.

Any idea on this?

Asher

BTW, in case Cem, (or anyone else), thinks my actual photography is being sidetracked these days by my enthusiasm for everything Ricoh-Pentax, let me assure you that my Canon cameras are also getting ample workouts with my muses almost every day to such an extent that I have recently taken to walking around the neighborhood with no camera in my hand, just to unwind!
 

Jerome Marot

Well-known member
But can we shoot with a Ricoh or Pentax camera set to this more intelligent WB and still have the benefits applied in processing RAW images, as opposed to using the very well processed JPG's straight from the camera.

I don't think that any software other than Ricoh's own supports that feature. Therefore, I would think that the only solution would be to use manual masks to the same effect.


BTW, in case Cem, (or anyone else), thinks my actual photography is being sidetracked these days by my enthusiasm for everything Ricoh-Pentax, let me assure you that my Canon cameras are also getting ample workouts with my muses almost every day to such an extent that I have recently taken to walking around the neighborhood with no camera in my hand, just to unwind!

That is good to hear. Did the workouts produce anything worth showing?

 

Asher Kelman

OPF Owner/Editor-in-Chief
I don't think that any software other than Ricoh's own supports that feature. Therefore, I would think that the only solution would be to use manual masks to the same effect.




That is good to hear. Did the workouts produce anything worth showing?


Jerome,

I have read their white paper on this process and it reveals a major sophisticated game of learning and correcting mistakes. The areas are then "grown" into one another to leave no seams. I don't see it stated anyway that their own provided software can use the information. It might be just an in camera feature for jpgs. This is an opportunity for someone to devise an action which creates a mask of zones from the jpg and from the same size uncorrected file from the DNG. I think it could be a valuable tool.

I'd hope someone has experimented with the RAW files using Ricoh's provided software version of Silky Pix. I've never even looked at it!

Asher
 

Michael Nagel

Well-known member
Asher,

Short answers:
It affects the jpeg of course, but there is also information in the EXIF. LR5 seems to have issues with it.
It is not 100% effective, but not hat bad either. You can see the effect quite well if you take a picture of someone/something in a room illuminated by light at - let's say 3300k and a portion of it by a screen/TV.
The blueish effect by the screen/TV is (almost) eliminated and you only see the brightness impact..

Best regards,
Michael
 

Andrew Rodney

New member
Asher,
Short answers:
It affects the jpeg of course, but there is also information in the EXIF. LR5 seems to have issues with it.

It is proprietary metadata so it's not just LR but every raw converter. Of course, this proprietary metadata can be used within the camera itself to produce a JPEG. This is nothing new either. All kinds of cameras produce proprietary metadata that can be accessed within the manufacturer's converters but not in other converters. It's questionable if this is even useful. Raw is raw. There's absolutely nothing to WB in a raw file (it isn't color as yet).
 

Michael Nagel

Well-known member
Andrew,

Thanks for chiming in. Strictly from an engineering and informatics point of view this is information, proprietary or not.

It is in the hands of the manufacturers of the cameras and the companies/people writing raw converter software to agree on a way to share that information ina sensible way to make use of it.
I wish for more communication in that area....

Best regards,
Michael
 

Andrew Rodney

New member
Andrew,
It is in the hands of the manufacturers of the cameras and the companies/people writing raw converter software to agree on a way to share that information ina sensible way to make use of it.
I wish for more communication in that area....

They can't even agree on an open, non proprietary (with private tags) raw format. Hell will freeze over when the big two (Nikon, Canon) by option on the camera, allow us to save a DNG. The universe will have to freeze over before they open up their proprietary processing, which they think is so great. Meanwhile, it appears few users are impressed enough to use these camera manufacturer's products.
 

Michael Nagel

Well-known member
Andrew,

Pentax/Ricoh is not Nikon or Canon. There is the possibility to use DNG for a while now - I started using it with the K200D...

It would be useful that the companies using DNG would agree on some features of that format that includes the feature which created this thread. If the camera manufacturers serving the MF market together with Ricoh/Pentax could agree on some things, maybe it could be enough that others may join in. I do not put a lot of hope on that. If Olympus, Panasonic and Sony join in it could become more interesting, but not much hope for that either...

Best regards,
Michael
 

Andrew Rodney

New member
Andrew,
Pentax/Ricoh is not Nikon or Canon.

Thankfully not. My point is, it's doubtful we'll see any kind of cooperation in sharing this data.
The DNG spec allows for proprietary metadata yet the big guys still want nothing to do with it. So the idea that they will not support an open format because they fear their secret sauce will be understood is nonsense. It's simply political. I'd love to see what you propose but I'm not holding my breath.

Whatever Ricoh is doing here is proprietary and perhaps they will share it with Adobe and other's so it can be accessed in LR or ACR. Can't hurt to ask. But I doubt it will happen. So if you use Ricoh's software, the feature may be really nice. If you don't use Ricoh's software, all bets are off.
 

Michael Nagel

Well-known member
Andrew,

I do not think that this will happen either, but here is an example from the world of electronics and semiconductors (I feel home here work-wise):

For simulations of analog circuits there is SPICE used in many cases (except the very high-speed stuff). Now the models of the circuits to be simulated are described in plain text, there is not much hidden. There is the possibility to use encypted models to keep things secret. Now this would not help as the data would have to be decrypted before use and would contain all information.
The other approach for higher speeds which was taken in that domain was IBIS, which made it possible to use information without revealing IP. If there is a venue to use a similar approach for raw files - this could be the way to go.

I know that there is a lot of politics involved...

Best regards,
Michael
 

Asher Kelman

OPF Owner/Editor-in-Chief
Andrew,

I'd like just the result of the zones created by their intelligent analysis. That would be already pretty damn useful. Disclosing the local WB correction for each zone would be nice too, but not really necessary as we can do that ourselves.

For Ricoh they use Silkypix. Now is this special version and in any case, can it access the metadata that's used for processing the jpgs in camera?

Asher
 
Top