Re: 16 bits = 15 bits in Photoshop?
Re: 16 bits = 15 bits in Photoshop?
- Subject: Re: 16 bits = 15 bits in Photoshop?
- From: Jim Rich <email@hidden>
- Date: Wed, 20 Apr 2005 08:41:10 -0400
I was contacted off list on and given what looked like Photoshop generated
files. Not a scan. One was 8 bit and ther16 bit. And clearly the 16 bit
file was a keeper.
Hey I asked for it. And it was no problem. However, I did ask the question
of how often that happened. The answer I got back was one in ten, or in
laymans terms every one image out of ten was broken if 8 bits was used.
I personally think that 10 % of this workflow is broken but I am not there
doing the counting.
The result I saw was like a very few (very few) other scanned files I have
come across and have seen once in my own tests. When I did my original
test it turns out the photo I used was either over exposed or processed
incorrectly. Because of that the actual image had banding in the original.
The scan of this image was adversely affected by editing it in 8 bits
and less in 16 bits. So yes I have seen this but when shown to experts
they could not clearly say it was an 8 bit or 16 bit image.
Having seen that in a test. I have to say that having a renegade image is
more the exception than the rule. I estimate under 3% of all images in
that category.
The point I made to the person who contacted me off line was that the
decision to use 8 bit or 16 bits images always comes down to money.
And that when I consult I ask the client questions like:
How much is it worth for you to work with 8 or 16 bit images?
Is it worth having your complete workflow 16 bit when less than 2 % or
3% of your images might require 16 bit processing?
What is it going to cost you in time and frustrating phone calls to send
16 bit images to clients who are used to 8 bit images?
What is it going to cost you in production time to send 16 bit files
down stream to a rip that will break?
The answers I get back vary and might be yes for some and maybe no for
others.
In my second post on this topic I stated:
" The tools are available to edit 16 bit images. However if a business is
considering doing the 16 bit workflow for what ever reason, a few tests
need to be done to determine things like are there are hits on processing
speed, will image quality be worse or the same and will the workflow have
the right capabilities. Based on those test, it is necessary to determine
what makes sense."
Lets face it, after seeing these latest threads, this is not a black and
white issue, we all have to use some common sense. And my common sense might
be different that someone elses.
If you have a lot of images that you believe require 16 bit processing or
you are just plain paranoid that something will break downstream then go
for the 16 bit approach. But if you only have a few images that require
16 bit processing then it does not make sense.
As for the folks who have commented about banding in skys and gradient areas
of photos. I would agree that if you have that particular problem, a 16 bit
approach is one type of solution.
However, if banding is happening on a regular basis, there is a high
probability that you have a capture problem with your scanner or camera.
And the reality of that situation is that the 16 bit workflow is a
workaround and not a solution for a broken input device.
One solution would be to get a new scanner or camera that works and provides
both the best 8 bits and of course 16 bits during the capture process. But
then again that costs money.
I know from a practical point of view that capturing in 16 bits or more is
a solid strategy. Where I depart from the beliefs of this list is that
editing in 8 bits will work for the majority of work in most workflows! It
has proven to be just plain cost effective while providing high quality
results in the majority of cases!
So let me end this with some middle of the road advice.
It makes sense to capture everything as high bit images.Everyone seems to
agree on this.
Save the high bit parent file in your archive.
Create derivative 8 bit files from the parent file for production.
Use Adjustment layers.
If the 8 bit file breaks ( as they can in rare cases), then go back to
the parent 16 bit file and use the previous Adjustment Layers.
In this scenario, the best of both worlds are available. A streamlined
workflow that uses lower cost 8 bit tools, that are based a good solid
archive of high bit images.
I am sure mileage will vary.
Jim Rich
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden