M8 bit size and Photoshop CS2 & 3

[infocusf8@earthlink.]

I've been using PS 2 for some time and all of my M8 raw images appear as 16 bit files. I just upgraded to PS CS3 and my raw files are now appearing as 8 bit files. Is anyone else experiencing this and do you know the reason?
Thanks in advance.

 

[Matthew]

If you're using ACR to convert them it's possible that the setting to import them at 16-bit has been reset to 8-bit when you upgraded. I would check that first as it certainly possible to import M8 DNGs as 16-bit in CS3. Hope this helps.

 

[TJV]

What I never understood is why CS2 always compressed my DNGs, meaning I couldn't open them in other programmes like Capture One. For the life of me, I have not been able to find an option that stops this happening.

 

[RingoHung]

It does't matter. M8's natuve is 8 bit. 16 bit only makes the file larger and will not enhance the colors.

 

[jaapv]

It does matter. Processing in 8 bits will create posterizing.

 

[jackal2513]

in camera raw press the blue line of text at the centre bottom of the window and when the option window appears change the option to 16 bit


and Jaap is correct, it DOES make a difference A big one in fact.

you should always perform curves, color grading, levels etc.. in 16 bit.

try the same operation on an 8 bit image wnad watch all the values drop out !

 

[jaapv]

8 bits is indeed not equal to 8 bits. Quite apart from the bit depth discussion on the M8 - it is in fact a 16(=14) bit camera that stores in 8 bits using an obscure feature of DNG, but let's not go into that again, bit depth in processing is something quite different. Explained primitively: The data coming from raw in 8 bits are stored in 256 adjacent "buckets". Now when processing data are moved from one bucket to the next, making them indistinguishable from the data already there. Data lost forever. Move it on again and it happens again. The result is the effect of the density lines and jumps that we call posterizing.
If you process in 16 bits you add a large number of "empty buckets" between your 256 full ones. That means you can shift your data around without losing them, so when you convert to 8 bits after processing they are still there. And you have a smoother image.

 

[infocusf8@earthlink.]

M8 CS2 & 3 Redux

M8 CS2 & 3 Redux

I'm taking the images directly from the card into Bridge then opening them in CS3 to process them. It is the same process I've been using with CS2. Is there a setting in preferences that I missed when opening them from Bridge to CS3? I understand the differences in 8 bit and 16 bit that is why I'm trying to keep them 16 bit.:bang:

 

[jaapv]

in camera raw press the blue line of text at the centre bottom of the window and when the option window appears change the option to 16 bit

Try this. Alternatively, in MODE choose 16-bit. Then apply image.

 

[infocusf8@earthlink.]

jaapv -I've thought about that but isn't that adding 8 bits via an interpolation process that is random? I guess where I'm confused is I'm under the impression that the images come out of the camera in 16 bit and if that is the case then is PS3 converting them to 8 bit and why or am I incorrect about the images being 16 bit out of the camera?

 

[jaapv]

Don't confuse capture depth with processing bits. The camera has an 8-bit output, but points at a 16-bit image matrix through a logarithmic algorythm. The bits you add for processing are empty bits, which you need to give you elbow room for manipulation. Theoretically you should convert 16 bits camera output to 32 bits.

 

[Matthew]

in camera raw press the blue line of text at the centre bottom of the window and when the option window appears change the option to 16 bit

Yes, do this. It will open a pop-up window in which you can select 16-bit. If I recall, in previous versions of Photoshop it was a pull down menu and a slightly more obvious adjustment to make.

 

[jackal2513]

Try this. Alternatively, in MODE choose 16-bit. Then apply image.

no, better if its 16 bit staright from camera raw imo

 

[jaapv]

It is certainly more practical. The end effect should be identical.

 

[infocusf8@earthlink.]

Got it, thanks everyone I'll work on the suggestions. Out of curiosity does anyone know why CS2 was registering them as 16 bit and CS3 as 8 bit?

 

[DwF]

not to belabor but....

not to belabor but....

Don't confuse capture depth with processing bits. The camera has an 8-bit output, but points at a 16-bit image matrix through a logarithmic algorythm.

I understand more or less the empty buckets analogy but the sentence here regarding the camera is not clear, nor have I found a clear explanation elsewhere. So, when shooting M8 RAW, is the camera essentially capturing 16 bits and after this logarithmic algorithm, processing out (output) 8 bits which in CS2 I am sending on at 16 bits for processing there?

Thanks for your patience here. I am getting good results with my work (having just printed my first files), but I would like to understand this more. I also recognize that high contrast scenes need to metered very critically, but having shot slides for years, I am used to that.

David

 

[RingoHung]

Thanks all for educating me too...

 

[jaapv]

I understand more or less the empty buckets analogy but the sentence here regarding the camera is not clear, nor have I found a clear explanation elsewhere. So, when shooting M8 RAW, is the camera essentially capturing 16 bits and after this logarithmic algorithm, processing out (output) 8 bits which in CS2 I am sending on at 16 bits for processing there?



David

That is more or less correct There was an extensive article in LFI by Michael Hussmann, and he had graphs comparing it to a 12 - bit camera. (very few "16-bit" camera's produce more depth, and never more than 14 bits btw) The differences in output were marginal. The logarithm adapts the output to the non-linear response of our eyes. That is the story behind the amazing shadow detail of the M8. But forget about camera bits- look at the results 🙂

 

[jaapv]

Got it, thanks everyone I'll work on the suggestions. Out of curiosity does anyone know why CS2 was registering them as 16 bit and CS3 as 8 bit?

You probably processed a file as 16 bits at one point of time and canged the default.

 

[DwF]

That is more or less correct There was an extensive article in LFI by Michael Hussmann, and he had graphs comparing it to a 12 - bit camera. (very few "16-bit" camera's produce more depth, and never more than 14 bits btw) The differences in output were marginal. The logarithm adapts the output to the non-linear response of our eyes. That is the story behind the amazing shadow detail of the M8. But forget about camera bits- look at the results 🙂

Thank you Jaap!

and yes the results are there like I have never experienced with 35mm......of course there are qualities peculiar to film that we love. The fellow that printed my files was always partial to his 4x5 and Mamiya Press negs, and not much of a fan of the 35mm neg (or Leica).

When I returned to pick up my prints (B&W), he said "you got an M8" and praised the look of the files as well the resulting prints, saying that they were right in there tonally with medium format. I should add that he had a similar impression of one file that came from my R D1 (which I sold to fund the M8).

David