Ad
Ad
Ad
Pages: [1]   Bottom of Page
Print
Author Topic: ACR vs PS on TIFF files.  (Read 6674 times)
Paul Osgood
Newbie
*
Offline Offline

Posts: 2


« on: January 24, 2009, 08:33:52 AM »
ReplyReply

First the facts:
I've been shooting MF and having drum scans made. These have been 8bit/200MB, but now that they're available for the same price, I'm getting 16bit/400MB. All scans are Tiff files. My volume is small, so the ability to handle a large volume of images is not needed. I've just recently become familiar with how intuitive ACR is and am contemplating doing most of my adjustments there rather than PS.  I currently have PS CS3/ACR 4.6.  
Reason for the query:
In "Real World Camera Raw with Adobe Photoshop CS3", on page 112: "The jury is  out on whether the ability for Camera Raw to open and adjust JPEG and TIFF files will end up being considered a good thing." (I am only concerned with Tiff files.) Yet, when I look at the newer edition for CS4, this is no longer mentioned. The reason for this possible limitation is not given. I just read a post on this forum indicating ACR is only non-destructive on RAW scans and IS destructive on Tiff files. This would seems to be a sufficient reason not to use ACR on Tiffs.  
Now the questions:
1) Is it true that ACR is destructive on Tiffs? If so, wouldn't PS (with layers) be the preferred tool for all adjustments?
2) Is the above mentioned statement (from page 112) only true for ACR4.6 but no longer true with 5.2? If so, what has changed in ACR?
3) If ACR is appropriate for 16bit Tiffs, is it also appropriate for 8bit Tiffs?

Thanks,
Paul

Logged
Chris_Brown
Sr. Member
****
Offline Offline

Posts: 792



WWW
« Reply #1 on: January 24, 2009, 10:24:06 AM »
ReplyReply

Quote
1) Is it true that ACR is destructive on Tiffs? If so, wouldn't PS (with layers) be the preferred tool for all adjustments?
The definition of "destructive" is when adjustments (curves, levels, etc.) cause a re-distribution of image data and thus a loss of tones in various parts of the image (as seen in a histogram with combing). Within that definition, then yes, ACR is destructive because there is no "undo" after the ACR adjustments have been applied. If you save out the ACR result as a smart object, then the image is not permanently changed but there will still be a redistribution of image data. Adjustments made in PS are also considered destructive.

Since you're working with 16-bit files, there will still be noticeable histogram combing (indicating a loss of tones at certain levels) until you convert to an 8-bit file.

Quote
2) Is the above mentioned statement (from page 112) only true for ACR4.6 but no longer true with 5.2? If so, what has changed in ACR?
I never used ACR 4.6 so I can't comment directly to this.

Quote
3) If ACR is appropriate for 16bit Tiffs, is it also appropriate for 8bit Tiffs?
I wouldn't consider ACR appropriate because adjustments can be made with far more accuracy in Photoshop. Within PS I can make an adjustment layer then tweak its blending mode, its opacity and fill. This provides much more capability than ACR and any adjustments can be changed later when keeping all adjustments as discrete layers.
Logged

~ CB
Eyeball
Full Member
***
Online Online

Posts: 106


« Reply #2 on: January 25, 2009, 10:02:26 AM »
ReplyReply

Paul,

I think I found the post you are referring to regarding the "destructive" TIFF processing of ACR  ( http://www.luminous-landscape.com/forum/lo...php/t25437.html ) .  I believe part of the problem is with the definition of "destructive".

My definition of "destructive" would revolve around the question "Does the newly created file allow me to get the original pixels back if I want to?".  If the answer is "No", then the process was destructive.  If the answer is "yes", then the process was not destructive.

First example using a scanned TIFF and Photoshop:

I bring the TIFF into PS with 16 bits as a background layer.  I then do a variety of edits on layers and save the newly-created image as a PSD, including my background layer.  This could be considered "non-destructive" since I have the option of opening the PSD, removing all the edited layers, and getting back to my original TIFF pixels.

Second example using ACR:

I bring the TIFF into ACR.  I make edits, which are stored as commands within the TIFF file.  The file will change but the pixels will not.  If I want to return to my original pixels, I can re-open the TIFF, remove the edits (Selecting Camera Raw Defaults from the fly-out menu should do the trick), and I am back with the original pixels.  Again, non-destructive.

Now there are some things to be aware of when using ACR to edit TIFFs and Jpegs:

- As I mentioned above, ACR will modify the file even though the pixels themselves are unchanged within the file.  This will cause the filesize and filedate to change.  This could also introduce the possibility of damaging the file if Adobe happened to have a bug in its software or if something happened to your PC while writing the file (like a power outage).  Only you can assess if that is a big risk or a small one.

- The commands that ACR inserts into the TIFF or Jpeg will be understood only by certain Adobe products.  Non-Adobe products will be able to read the file but without the edits applied by ACR (this is actually another indication that the ACR process is non-destructive to the original pixels and might even be handy under certain situations).

Regarding why Jeff left out the "doubt" in the latest version of his book, only he can answer definitively.  Maybe he will if he is back now from playing with penguins.   If I were to speculate, however, I suspect Jeff's logic was something like the following:
- RAW HAS to be interpreted so RAW+recipe of some type is a given as well as the non-destructive nature of such an arrangement.
- Jpegs and TIFFs can be interpreted "as-is" and do not NEED a recipe.
- Tools like Photoshop already provide a means to do non-destructive editing of TIFFs and Jpegs using layers as I mentioned in my first example above.
- Is it really worth it to duplicate that non-destructive functionality when you don't have to?  What about the file change caveats I mentioned above? (RWCRwAP CS3)
- Ah screw it.  Choice is good.  ACR now has some nifty and easy-to-use tricks like gradients and localized brushes.  Let users decide if it's better to use ACR or PS for TIFFs and Jpegs. (RWCRwAP CS4)

Here is a link to John Nack's blog that provides some additional info:

http://blogs.adobe.com/jnack/2007/02/nondestructive.html

If you take the time to read through the comments on John's blog, including his replies, you will note that Adobe's intent was to make ACR into a tool to read and edit camera-generated files.  That is why Raw, TIFF, and Jpegs are included but PSDs, for example, are not.  Since in your case, the TIFFs are scanned images, not pics from a camera, that might be worth a conceptual ponder.  In the end though, choice is good and make up your own mind.
« Last Edit: January 25, 2009, 10:23:11 AM by Eyeball » Logged
Paul Osgood
Newbie
*
Offline Offline

Posts: 2


« Reply #3 on: January 27, 2009, 08:09:07 AM »
ReplyReply

Eyeball - Thanks for the reply.  
I am using the same definition and understanding of 'nondestructive' - whether you can return the image to its original state.   You have dispelled my fears (generated by the statement in "Real World ACR with PS" - see original post) that processing TIFF files with ACR might not be considered a good thing.

Since I find ACR more intuitive than PS, I believe I'll start using ACR for first stage processing of my scanned images and only use PS (with layers) when that does not accomplish the ask.  From ACR, I'll open in PS (as smart obj), do any layer adjustments and save the adjusted image as PSD or DNG.  When actually using the image I'll flatten and save as appropriate.  In all cases I'll also keep the original scan as a TIFF.   In all this gives me 2 stored files at full resolution (the original TIFF and the PSD/DNG) and additional files for each output size (appropriately sharpened).  Since this is for a small number of images (<200), this is not a problem.

Not having heard anyone say that ACR is inappropriate for first stage adjustments on 200MB, 8bit Tiffs (and PS would be better), I'll use the above workflow on both my older 8bit and newer 16bit images.
   
Perhaps Jeff will respond regarding the meaning of the statement in the book, when he returns from Antarctica.

Paul
Logged
Eyeball
Full Member
***
Online Online

Posts: 106


« Reply #4 on: January 27, 2009, 09:45:43 AM »
ReplyReply

Paul,

I came across a post on Digital Outback that you might want to take a look at.  It contains some additional tips and gotchas related to using TIFFs with ACR and Lightroom.  It was written for the previous versions of the ACR/LR engine so it is possible some things have been fixed/changed in the v5.x version of the engine but probably worth a read-through.

http://www.outbackphoto.com/artofraw/raw_3...s.html#20070417

By the way, you can clear the ACR settings in a TIFF file while in Bridge by using a right-click>Develop Settings>Clear Settings.  That will allow the file to be opened directly by PS again without going through ACR.  This is similar to what is explained in the DO article but he gives the instructions for Lightroom.

Dennis
Logged
Schewe
Sr. Member
****
Offline Offline

Posts: 5472


WWW
« Reply #5 on: January 27, 2009, 11:40:41 AM »
ReplyReply

Quote from: Paul Osgood
In "Real World Camera Raw with Adobe Photoshop CS3", on page 112: "The jury is  out on whether the ability for Camera Raw to open and adjust JPEG and TIFF files will end up being considered a good thing."


The jury is no longer out...the processing pipeline in Camera Raw 5 is very good for TIFFs & JPEGs if it suits your workflow.

The implementation of TIFF/JPEG editing in Camera Raw 4 and CS3 sucked...it required the correct preferences set in three different applications in order to work correctly. In CS4 all control over ACR's ability to edit TIFF & JPEG is now under the direct control of ACR.

As for whether or not the Camera Raw pipeline is useful to YOU for processing TIFFs and/or JPEGs, that's up to you to decide. However, the Camera Raw processing pipeline has proven to be equal or better than Photoshop for those tasks that Camera Raw can do. And workflow wise, it can be a very efficient process.
Logged
Eyeball
Full Member
***
Online Online

Posts: 106


« Reply #6 on: January 27, 2009, 03:09:53 PM »
ReplyReply

Has the problem with layered TIFs been remedied in 5.x, Jeff?
Logged
Schewe
Sr. Member
****
Offline Offline

Posts: 5472


WWW
« Reply #7 on: January 27, 2009, 03:13:15 PM »
ReplyReply

Quote from: Eyeball
Has the problem with layered TIFs been remedied in 5.x, Jeff?


No...it's not a problem...Camera Raw will ONLY work on flattened TIFFs. I don't see that as a problem.
Logged
MichaelEzra
Sr. Member
****
Offline Offline

Posts: 655



WWW
« Reply #8 on: January 29, 2009, 09:27:43 AM »
ReplyReply

My 2c: I would suggest to have any future scans made in RAW DNG format that can be further processed using ACR.
Vuescan is capable of DNG output from most scanners.
Logged

Pages: [1]   Top of Page
Print
Jump to:  

Ad
Ad
Ad