Kodak's new 50MP sensor spec sheet

[Graham Mitchell]

This is about the 50MP sensor which Kodak just announced. I just received the spec sheet today:

http://www.mediafire.com/?tzugyyqvzyy

FWIW, the dynamic range seems to be around 11.5 stops.

[BJL]

Kodak also has it up on its own site now: see
http://www.kodak.com/global/en/business/IS...938/12138/13219
and the links from there.

Yes, DR is 3280:1 (70.3dB, from 41 KE- max signal, 12.5e- read noise), a but under the 4096 of 12-bit A/D conversion, so a bit under 12 stops. No Kodak sensor has even gone beyond 4096:1 in DR.

Doe anyone care to explain again how 16-bit A/D conversion (65,636 levels) could be expected to improve IQ over 14-bit (16,384 levels)? I can see a use for some margin of error beyond what 12-bit conversion has, but I see no value to going beyond the safety factor of over four in 14-bit.

[Jack Flesher]

Doe anyone care to explain again how 16-bit A/D conversion (65,636 levels) could be expected to improve IQ over 14-bit (16,384 levels)? I can see a use for some margin of error beyond what 12-bit conversion has, but I see no value to going beyond the safety factor of over four in 14-bit.

I agree, and think even 14-bit is perhaps overkill.  I think the main issue is how much color is needed to generate a true continuous tone with no visible transition. The human eye is capable of distinguishing about 1 billion distinct shades.  Granted, many of these probably have to be right next to each other with a hard transition to detect, but if we accept that number as a constant, then:

1) 1BB colors = 1,000 per channel in RGB space.
2) 10-bit per channel color = 1024 distinct shades per channel or just over 1BB distinct shades.

So, in theory, we shouldn't need any more color than 10-bit, right?   Except if the adjacent hues colors are physically next to each other we can still detect the step in hue between them, so we need a "feather factor" to allow them to blend seamlessly if we want a truly continuous tone.  

So the real issue is what is an acceptable "feather factor" -  I submit that one tone separating the two adjacent colors we can visually separate should be enough to "feather" the transitions so we cannot see distinct transitions. (But since noise can act as a "feather-factor" too, perhaps having a little is not a bad thing... I feel the heat already! ) However in an ideal situation, if you want to build in a fudge factor then 2-bits/channel should be more than enough, and hence I will go out on a limb and suggest that 12-bit per channel color is about all we'll ever need to display a true continuous tone.  

Now enters DR.  Personally, I have yet to see any single image that contained a true 11 stops of DR that actually looked visually appealing to me, so I submit more than about 8 stops is not needed for output. (Again I feel the heat... )  However, to be able to capture it ALL, then manipulate it into a smaller output range of say 8 or so stops could definitely be an advantage, and in that respect I feel you will never have such a thing as "too much DR" in a capture medium. This of course assumes it will be manipulated well into the smaller output space, and with most of the currently used tools I don't think that happens very often...

Cheers,

[EricWHiss]

...

Now enters DR.  Personally, I have yet to see any single image that contained a true 11 stops of DR that actually looked visually appealing to me, so I submit more than about 8 stops is not needed for output. (Again I feel the heat... ) 

Cheers,
[a href=\"index.php?act=findpost&pid=206445\"][{POST_SNAPBACK}][/a]

Hmmm.... maybe you are mostly thinking of overdone HDR images that have been in vogue?   I think it will be necessary to have even more DR at capture to come close a true film look.  I'd also find more DR useful for editing purposes even if currently its hard to utilize that many stops in print.

[paul_jones]

I agree, and think even 14-bit is perhaps overkill.  I think the main issue is how much color is needed to generate a true continuous tone with no visible transition. The human eye is capable of distinguishing about 1 billion distinct shades.  Granted, many of these probably have to be right next to each other with a hard transition to detect, but if we accept that number as a constant, then:

1) 1BB colors = 1,000 per channel in RGB space.
2) 10-bit per channel color = 1024 distinct shades per channel or just over 1BB distinct shades.

So, in theory, we shouldn't need any more color than 10-bit, right?   Except if the adjacent hues colors are physically next to each other we can still detect the step in hue between them, so we need a "feather factor" to allow them to blend seamlessly if we want a truly continuous tone. 

So the real issue is what is an acceptable "feather factor" -   I submit that one tone separating the two adjacent colors we can visually separate should be enough to "feather" the transitions so we cannot see distinct transitions. (But since noise can act as a "feather-factor" too, perhaps having a little is not a bad thing... I feel the heat already! ) However in an ideal situation, if you want to build in a fudge factor then 2-bits/channel should be more than enough, and hence I will go out on a limb and suggest that 12-bit per channel color is about all we'll ever need to display a true continuous tone.   

Now enters DR.  Personally, I have yet to see any single image that contained a true 11 stops of DR that actually looked visually appealing to me, so I submit more than about 8 stops is not needed for output. (Again I feel the heat... )  However, to be able to capture it ALL, then manipulate it into a smaller output range of say 8 or so stops could definitely be an advantage, and in that respect I feel you will never have such a thing as "too much DR" in a capture medium. This of course assumes it will be manipulated well into the smaller output space, and with most of the currently used tools I don't think that happens very often...

Cheers,
[a href=\"index.php?act=findpost&pid=206445\"][{POST_SNAPBACK}][/a]

a lot of the most modern advertising photograhy is heavily colour and tone graded.

i think if you dont touch the shot, then 14-16 bit is overkill.

but if you change, twist grades and colours its very obvious that you need all the extra shades of tones. the 1dsmk2 files i use to shoot couldnt get any where near the grades as i can get out of my leaf or my phase. the mk3 is a world of improvement over it as well.
it was a brick wall we always hit when trying to get the most interesting colours, 12 bit aways pops into steps.
ask any decent retoucher if it matters- if the shot doesn't need much tweaking, then 12 bits fine, but any challenging grade, the more the better.

paul

[Anthony R]

a lot of the most modern advertising photograhy is heavily colour and tone graded.

i think if you dont touch the shot, then 14-16 bit is overkill.

but if you change, twist grades and colours its very obvious that you need all the extra shades of tones. the 1dsmk2 files i use to shoot couldnt get any where near the grades as i can get out of my leaf or my phase. the mk3 is a world of improvement over it as well.
it was a brick wall we always hit when trying to get the most interesting colours, 12 bit aways pops into steps.
ask any decent retoucher if it matters- if the shot doesn't need much tweaking, then 12 bits fine, but any challenging grade, the more the better.

paul
[a href=\"index.php?act=findpost&pid=206496\"][{POST_SNAPBACK}][/a]

Just going to pop in and state that Paul is very correct in his statement. The lack of this thinking or knowledge is the largest difference when it comes to those that try and state that a dslr can equal that of a mfdb or 'who needs it' etc, etc...etc.

[Jack Flesher]

the 1dsmk2 files i use to shoot couldnt get any where near the grades as i can get out of my leaf or my phase.
[a href=\"index.php?act=findpost&pid=206496\"][{POST_SNAPBACK}][/a]

and I agree wholeheartedly as I have noted the same thing, and this is precisely why I no longer even own a DSLR  

I am no sensor engineer, but I suspect it has to do with what actually comes out of the file at the demosaiced end, not the linear file off the sensor; IOW the ADC plays a strong role as does the demosaicing software.  So if the sensor of the top end DSLR's has the same theoretical bit-depth spec as that from a typical high-end MFDB, I think what you've stated and what I've seen is empirical evidence they do not deliver final output with comparable efficiency.

The original question was how much bit-depth is really needed, and the answer for me is I want 12-bits per channel at the output end and could care less what the sensor's theoretical maximum limit is.

Hope that clarifies,

[TMARK]

Man, this new 50mpx chip better be good at retoucjing, marketing to clients, archiving files, giving back rubs or at least sweeping up after a shoot, because otherwise, there is no way no how another 29 megapixels is going to pay its way in my business unless its a free upgrade. and even if it were free, storage is suffocating me.

[Guy Mancuso]

I'm pretty darn happy with my P25 plus at 22 mpx. Need more than that rent it

[paul_jones]

the only problem with the "22mp is enough" argument, is when the clients find out that 50mp is available, and insist on it. doesnt matter how much you argue, they will still want the best quality thing on the market.

paul

[mcfoto]

the only problem with the "22mp is enough" argument, is when the clients find out that 50mp is available, and insist on it. doesnt matter how much you argue, they will still want the best quality thing on the market.

paul
[a href=\"index.php?act=findpost&pid=206553\"][{POST_SNAPBACK}][/a]

Hi
With the exception of promotional images for camera/digital manufacturers i.e. CANON & LEAF we have never had a client request we shoot on a particular camera or the largest, greatest........
The physical  size of the chip still remains the same. Until there is an actual increase in the physical size of the chip I won't get excited.  It's like rearranging the furniture in a fixed sized room.

[klane]

Hi
With the exception of promotional images for camera/digital manufacturers i.e. CANON & LEAF we have never had a client request we shoot on a particular camera or the largest, greatest........
The physical  size of the chip still remains the same. Until there is an actual increase in the physical size of the chip I won't get excited.  It's like rearranging the furniture in a fixed sized room.
[a href=\"index.php?act=findpost&pid=206555\"][{POST_SNAPBACK}][/a]

Or cramming more furniture in it!

Id rather see one of the 33x44 sized chips developed with multishot, or just any new innovation in multishot or the like for that matter.

[marcmccalmont]

I wonder what their next micro-lens chip will be? I hope full frame.
Marc

[James Godman]

Man, this new 50mpx chip better be good at retoucjing, marketing to clients, archiving files, giving back rubs or at least sweeping up after a shoot, because otherwise, there is no way no how another 29 megapixels is going to pay its way in my business unless its a free upgrade. and even if it were free, storage is suffocating me.
[a href=\"index.php?act=findpost&pid=206534\"][{POST_SNAPBACK}][/a]

You're killing me!  Spot on, and don't forget it should wash your car too.  I don't even want 50 megapixels!  I agree with JR.  I'd rather pay for better higher ISO and more battery life and/or power management so as to be less tethered.

[NBP]

we have never had a client request we shoot on a particular camera or the largest, greatest........

Seconded.
I would probably diplomatically & politely tell them where to shove it if they did as well!