Jump to content

Leica M Monochrom DNG raw question


Recommended Posts

Advertisement (gone after registration)

i get all that, but an explanation is still missing.

thinking of raw as a negative is alright, but its far from true.

 

the way i see it is that a mono-dng holds a number for each pixel.

now, if you convert it to a tif file, or any other uncompressed format, it will still hold the same numbers for each pixel.

 

this is not true for a regular RGB-RAW file, where the final image pixels are not "ready" because they still needs to be demosaic and only after that process the pixels becomes numbers = image is ready to view.

 

so as i said i get everything you all say, but an explanation is still missing.

another thing i don't get -

whats the difference between the "shadows" (low numbers) and "highlights" (high numbers) ? i mean, in real world of negative once you "burn" a photo you can't go back to dark but if its too dark you can make it brighter. but with digital this is not the same, you treat low numbers just as you'd treat high ones, its all mathematics and not chemestry any more.. and yet another question, what can possibly be more black (blacker) then black itself ?

Link to post
Share on other sites

i get all that, but an explanation is still missing.

thinking of raw as a negative is alright, but its far from true.

 

the way i see it is that a mono-dng holds a number for each pixel.

now, if you convert it to a tif file, or any other uncompressed format, it will still hold the same numbers for each pixel.

 

this is not true for a regular RGB-RAW file, where the final image pixels are not "ready" because they still needs to be demosaic and only after that process the pixels becomes numbers = image is ready to view.

 

so as i said i get everything you all say, but an explanation is still missing.

another thing i don't get -

whats the difference between the "shadows" (low numbers) and "highlights" (high numbers) ? i mean, in real world of negative once you "burn" a photo you can't go back to dark but if its too dark you can make it brighter. but with digital this is not the same, you treat low numbers just as you'd treat high ones, its all mathematics and not chemestry any more.. and yet another question, what can possibly be more black (blacker) then black itself ?

 

 

Please see this:

 

Leica M Monochrom – performing recovery on my exposure mistakes | Bo Photography

Link to post
Share on other sites

whats the difference between the "shadows" (low numbers) and "highlights" (high numbers) ?

 

:confused:

 

If all the answers didn't help, maybe the question was not clear. Again, perhaps you want to get a DNG file yourself, convert to TIFF and try to answer the question. Then you can enlighten us with the answer.

 

Sorry for not being helpful.

 

cheers

Link to post
Share on other sites

Advertisement (gone after registration)

The question is very simple: does the DNG hold the pixel values as a TIFF does or not. If not, it must be some kind of vector-like information. Maybe the difference between DNG and TIFF is comparable to a pixel grafic and a vector grafic...?

 

In my imagination the sensor gives some kind of matrix with pixels and values for the pixels. The difference between DNG and TIFF becomes evident when processing: if you make more than one step of changing the brightness of a pixel, in a TIFF you would end up with a result as follows:

 

original pixel -> result of first edit -> second edit based on result of first step -> end result.

 

In DNG I hope it's different - but I don't know if my theory is true:

 

original pixel -> result of (first edit + second edit) -> end result.

 

So the difference in my theory would be that with DNG all calculations are carried out on the original pixels, while with TIFF the latest edit will always be based on the result of the former edit. In theory this should lead to better results with DNG. Also this would explain why something can be darker than black. Example: 0 = black and 100 = white. You have a TIFF and make it darker, say -5. Black pixels will remain black (ie 0) because there is nothing below 0. In the next step you make the pixel brighter, say +10. The result will be that the pixel now has a value of 10. In case my theory is true with DNG it would be different: the pixel has 0 at the beginning, then both calculations are carried out based on the original pixel value, so you end up with 0 - 5 + 10 = 5. Result: the DNG offers better fidelity. My experience does confirm this...

 

Who can confirm if this theory is true?

Edited by jpk
Link to post
Share on other sites

 

Who can confirm if this theory is true?

 

The .dng file has to become something after the first adjustments are made in RAW processing software, usually it is a TIFF or JPEG. There are no other steps possible with .dng, no second or third edit, you can't keep going back and forth in .dng format during post processing, once it is in Lightroom or Photoshop it is not a .dng file anymore. Equally you can't print a .dng file, it has to become a TIFF or JPEG. So it isn't even an academic question you are raising over the advantages of .dng over TIFF, they both exist in different universe's, the .dng can be adjusted before conversion, and the other adjusted after conversion. And if you want the black and white points to remain exactly the same in the TIFF file as the .dng file you have to pay attention to what you are doing in post processing.

 

Steve

Link to post
Share on other sites

jpk my man :)

finally!!!!

your theory sounds correct, i think!

remember this is only mono-dng-raw we're talking about..

 

so, if its indeed true, what user "250swb" said was right:

"If you have processed the .dng file with the maximum amount of detail in the highlights and the shadows (no clipping) the resulting TIFF will be the same"

 

but this is of course only theory since its only if you "got it right".

so, i think i can say this now and relax finally:

 

A mono-dng-raw which was perfectly edited with maximum details, no clipping of highlights and shadows, is in fact, the same for editing as its tif exported twin.

 

both files offers everything the sensor had to offer.

 

what do you guys think ? is this about right ?

Link to post
Share on other sites

The .dng file has to become something after the first adjustments are made in RAW processing software

 

Can you explain how actually the sensor output is stored in a Mono-DNG? Is ist similar to a TIFF?

 

iirc DNG is just an archiv file format which can contain an image file (RAW, TIFF, JPG) plus side car files with meta data such as processing steps, exif data and other...

Edited by jpk
Link to post
Share on other sites

A mono-dng-raw which was perfectly edited with maximum details, no clipping of highlights and shadows, is in fact, the same for editing as its tif exported twin.

 

both files offers everything the sensor had to offer.

 

what do you guys think ? is this about right ?

 

But what if the DNG contains a vector-like description of what the sensor saw...?

Link to post
Share on other sites

Raw is not the sensor output, but the output of the amplification/adc stage. Thus not vector-like. The DNG is very similar to TIFF, to the extent that Sandisk Rescue Pro stores the data as .TIF, which can simply be renamed .DNG.

Edited by jaapv
Link to post
Share on other sites

Raw is not the sensor output, but the output of the amplification/adc stage.

 

So the RAW of the Monochrome is indeed practically like a TIFF but saved in a different file format...

Link to post
Share on other sites

A mono-dng-raw which was perfectly edited with maximum details, no clipping of highlights and shadows, is in fact, the same for editing as its tif exported twin. both files offers everything the sensor had to offer.

 

Not necessarily: as 250swb said the DNG needs to be interpreted by the RAW converter to make it's content usable. So any use of the RAW means to extract the actual image from the DNG by using the converter. Every software does this in it's own way, so the remaining question is:

 

Can TIFF store more information than DNG? If not, the converter has to cut off some of the RAW information offered by the DNG in order to get the TIFF! I always would prefer DNGs. It even makes sense to generate DNGs out of JPGs or TIFFs if you don't have RAWs: so you keep the original JPG/TIFF and store additional data with it. If you later open the file you can undo your processing without losing quality...

Link to post
Share on other sites

:confused: Both formats have no practical upper limit in the amount of data. Image TIFFs can easily be hundreds of MB. Btw, it is raw as in uncooked, as opposed to the capitalization of an acronym.
Link to post
Share on other sites

just another reminder -

 

it is very clear to all of us what is the difference between a regular RGB raw file and a tiff.

 

what is less clear is the difference between the mono channel raw and his tiff export.

 

one thing is for sure - there is no vectors in all of this.. :)

and also, the fact that you can not "change" a raw file, doesn't mean it's better. you can always reopen (for that matter) a first generation well edited tiff and get the same results, i think..

Link to post
Share on other sites

A raw file is NOT RGB. Colour information will only appear after conversion of the file in a raw converter.

 

Basically the output after conversion of the raw is anything you like. Photoshop CS ACR mistakenly defaults to greyscale with Monochrom files, which is quite unpractical if one wants to use a plug-in like Silver Efex Pro, which needs RGB input.

 

Best is to change the ACR default and to convert to 16-bits RGB -the colour space is irrelevant for the Monochrom (but essential for Bayer filtered cameras)- which will produce three identical colour channels.

 

Lightroom will be working in an RGB environment (Prophoto colourspace- again irrelevant for the Monochrom) by design, so there is no choice there. Three channels again, enabling toning effects.

 

Just to be akward - even in Photoshop and after conversion one does not need to be in a file on RGB basis. For instance one can convert to LAB and back again, which is to all intents and purposes lossless, or convert to CMYK, or to Greyscale.

Edited by jaapv
Link to post
Share on other sites

you can call it whatever you like as long as you and the rest understands.

you don't have to call it rgb-raw,

call it a data file that came from a sensor with a bayer filter on him.

some call it a triple channel raw - since the filter is made out of 3 colors - you guessed it red green blue which in short - rgb-raw for this discussion.

 

besides that,

im having another thought (please do tell me what you think) -

 

they could've made it a tif (and maybe jpg as well, for the kids) only camera without(!) "raw", and by that, ending this entire raw shenanigans. raw belongs to the color-digital world.

 

with the leica monochrome, everything is more simple!

more like film! no weird "raw" computer terms, -

 

B&W film = original = viewable = adjustable = original tif from mono-sensor

 

there is no digital "negative" or "positive", just a tif :)

 

i sure hope they will bring tif to this camera through firmware or maybe in the next version.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...