16 bit tiff from 12 or 14 bit raw

[ormdig]

I've been doing all my post processing in raw and never had occasion to use tiff files until now. I have a couple of editors that won't import raw files so have exported 16 bit tiff to them out of ASP. My question(s) is: How does a 12 or 14 bit raw file become a 16 bit tiff and is being 16 bit the reason the tiff files average 7x the size of the raw files? Where would be a good place to learn more about tiff files, what they are and how best to take advantage of their properties? Thank you.
Pete.

[afx]

12 or 14 bits fit easily in 16;-)

The reason 16bit TIFFs are so monstrous are twofold:
Nobody ever bothered with a decent compression for 16bit TIFF and AS does not support writing compressed TIFFs anyway.
(My slide scans are usually 120MB per slide)

Apart from that, make sure you use a wide working space (WideGamut, ProPhoto) for the fat TIFFs to not cripple your gamut for post processing. I think that is all one needs to know.

cheers
afx

[ormdig]

12 or 14 bits fit easily in 16;-)

Apart from that, make sure you use a wide working space (WideGamut, ProPhoto) for the fat TIFFs to not cripple your gamut for post processing. I think that is all one needs to know.

cheers
afx

So, is going from 12 to 16 bit just the opposite of compression, i.e. multiplying or expanding the color data by 1.33 say?
Thanks for the heads up on working space. It didn't even occur to me.

[afx]

So, is going from 12 to 16 bit just the opposite of compression, i.e. multiplying or expanding the color data by 1.33 say?

If it where a simple multiplication, going from 12 to 16 would be a multiplication by 4.
But that would then leave the bottom part where the shadows are empty as this is just a shift to the right.

The data from the raw file gets normalized to AfterShots internal 16bit representation.
That then gets converted to the TIFF representation and saved to the 16bit TIFF.
So you have two mapping operations (to put it simply), the first one could be an expansion, but I do not know enough about the internal representation of AfterShot to really say what happens there. AS could have a non-linear representation to actually squeeze in a higher DR than 16 bits would allow when used linearly.

cheers
afx

[grubernd]

AS could have a non-linear representation to actually squeeze in a higher DR than 16 bits would allow when used linearly.

it's called Color Management.

The really, really bad idea of 1 bit equals 1 stop of dynamic range was introduced by thoughtless engineers in the dark days of digital cameras where A/D conversion chips were used in a linear fashion. yuck.

that's about as clever as comparing the speed of cars by the amount you press on the gaspedal in centimeters.

everyone: get this link out of your head. please. now. thank you.

any digital color is just a number, either you have enough digits to encode your data a.k.a. image without visual artefacts or not. nothing more or less. how that data is distributed and related to colors is determined by color management.

[spoilerhead]

also, RAWs are smaller because they have one a single channel per pixel (as given by the bayer pattern), whereas a TIFF usually has 2 (R,G,B)

so its 14bit/pixel versus 3*16bit/pixel = 14 vs 48

[tintin]

If it where a simple multiplication, going from 12 to 16 would be a multiplication by 4.

hi guys,

just to make the statement mathematically correct - it would be a multiplication by 16 (2^4=16)
...but, of course, this doesn't change anything else you said...

[afx]

LOL, that's what I get when mixing shift with multiplication....

cheers
afx