Steve'sDigicams

photosbyvito · Jun 27, 2004, 6:32 PM

nice shot eric

that's some focal length there!

btw....anyone care to explain wat practical differences there are between
8 bit and 16 bit images?

Vito

eric s · Jun 27, 2004, 9:14 PM

ohenry,

So now I have a reason to go back to using levels. I've been trying to use curves more. I'd seen that middle slider, but never fully understood what it was for. I should read about it some more.
photosbyvito
The debate between 8 & 16 bit photos has been going on for years. Basically you have more colors (well, more shades of colors but the same dynamic range) and more data to represent those colors available. The most obvious way to demonstrate the different (in writing) is to image doing math and rounding the value. Imagine the numbers0 to 100 represend shades of blue. Now imagine that 8-bit values is thesame as rounding up to the nearest 10. 0 and 100 will still represent the same shades of blue, but the middle shades are gone (since you can't represent them.)

So85 rounds up to 90, but it really was 85. Well, if you could make 85 map to a slightly different shade of blue then the rounded up 90 shade of blue... well, you can see the difference. Or maybe you can't, if the shades are close enough together.

And that is the rub. Some say "you can't see the difference" and others say they can.

Oh, another way they are different is when you do lots of calculations on them. Say you alter sharpening,contrast, and brightness... all those change the data. If you have only 8-bit you are effectively rounding during each calculation... and in each calculation the rounding "error" grows larger and compounds. If you start with 16-bits you have more shades to do your adjustments with, and less error.

This is why you shift to 8-bit at the end, when you're done all your editing.

I hope that helped.

Eric

photosbyvito · Jun 27, 2004, 10:06 PM

oh ok....thanks eric

that makes sense....so the dynamic range would be, for example, between 40-80...or 20-50....these aren't calculated numbers or anything...just wondering if that's where dynamic range fits in....

am i close? lol

Vito

ohenry · Jun 27, 2004, 10:09 PM

I use levels and curves both. I use the levels layer to remove color casts and adjust the tonal range. I'll use curves to tweak selected tones in the image and add some contrast. There are lots of ways to achieve things with PSCS, some easier, some better. I continue to learn new tools all the time. Some I like, some I just say "ok...that's cool, but I like to do it another way".

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Steve'sDigicams

Recent Camera Reviews

Most Active Forums

Most Popular Deals

Latest Feature Articles

Jun 27, 2004, 6:32 PM	#11
photosbyvito Senior Member Join Date: Dec 2003 Posts: 5,710	nice shot eric that's some focal length there! btw....anyone care to explain wat practical differences there are between 8 bit and 16 bit images? Vito

Jun 27, 2004, 9:14 PM	#12
eric s Senior Member Join Date: Dec 2002 Posts: 5,803	ohenry, So now I have a reason to go back to using levels. I've been trying to use curves more. I'd seen that middle slider, but never fully understood what it was for. I should read about it some more. photosbyvito The debate between 8 & 16 bit photos has been going on for years. Basically you have more colors (well, more shades of colors but the same dynamic range) and more data to represent those colors available. The most obvious way to demonstrate the different (in writing) is to image doing math and rounding the value. Imagine the numbers0 to 100 represend shades of blue. Now imagine that 8-bit values is thesame as rounding up to the nearest 10. 0 and 100 will still represent the same shades of blue, but the middle shades are gone (since you can't represent them.) So85 rounds up to 90, but it really was 85. Well, if you could make 85 map to a slightly different shade of blue then the rounded up 90 shade of blue... well, you can see the difference. Or maybe you can't, if the shades are close enough together. And that is the rub. Some say "you can't see the difference" and others say they can. Oh, another way they are different is when you do lots of calculations on them. Say you alter sharpening,contrast, and brightness... all those change the data. If you have only 8-bit you are effectively rounding during each calculation... and in each calculation the rounding "error" grows larger and compounds. If you start with 16-bits you have more shades to do your adjustments with, and less error. This is why you shift to 8-bit at the end, when you're done all your editing. I hope that helped. Eric

Jun 27, 2004, 10:06 PM	#13
photosbyvito Senior Member Join Date: Dec 2003 Posts: 5,710	oh ok....thanks eric that makes sense....so the dynamic range would be, for example, between 40-80...or 20-50....these aren't calculated numbers or anything...just wondering if that's where dynamic range fits in.... am i close? lol Vito

Jun 27, 2004, 10:09 PM	#14
ohenry Senior Member Join Date: Dec 2003 Posts: 1,676	I use levels and curves both. I use the levels layer to remove color casts and adjust the tonal range. I'll use curves to tweak selected tones in the image and add some contrast. There are lots of ways to achieve things with PSCS, some easier, some better. I continue to learn new tools all the time. Some I like, some I just say "ok...that's cool, but I like to do it another way".