Ad
Ad
Ad
Pages: [1]   Bottom of Page
Print
Author Topic: Spot removal tool question... on its impact  (Read 2135 times)
Phil Indeblanc
Sr. Member
****
Offline Offline

Posts: 1107


« on: March 08, 2013, 06:17:07 PM »
ReplyReply

I've read that the spot removal tool when used often will make file catalog larger.  Is there a feature you can "finalize" imbed that correction?  I know its non destructive, but I wish there was a way to override it...Anyone?
Otherwise it maybe a better choice to do those edits in PS than LR.
Would this be something important enough to add in a future "enhancement"? Your thoughts....?
Logged

If you buy a camera, you're a photographer...
Tony Jay
Sr. Member
****
Offline Offline

Posts: 2121


« Reply #1 on: March 08, 2013, 06:54:11 PM »
ReplyReply

Phil, are you not majoring on minors here.
Are a few kilobytes of data really stressing the system?

If there is more to this then feel free to elaborate.

Tony Jay
Logged
Glenn NK
Sr. Member
****
Offline Offline

Posts: 290


« Reply #2 on: March 09, 2013, 01:12:56 AM »
ReplyReply

 I have two almost identical images of one flower in the range of 22,000 KB, one of which has had extensive Adjustment Brush editing (two pins on the whole background).

The XMP file of the non-edited image is about 7 KB; the XMP file of the heavily edited one is 189 KB.  Although there is a twenty-seven fold increase in the XMP file size, 27 times almost nothing is still almost nothing.  There is almost as much variation in the respective RAW files:  175 KB.

If the size of an XMP file with heavy Brush adjustment is insignificant, I'd think that spot removal would have virtually no effect on file size.

Glenn
Logged

Economics:  the study of achieving infinite growth with finite resources
Simon Garrett
Sr. Member
****
Offline Offline

Posts: 371


« Reply #3 on: March 09, 2013, 03:28:47 AM »
ReplyReply

To answer the first part of the question, no I don't think there's any way of "finalising" any adjustment in LR.  Remember that LR merges all the edits anyway.  For example, if you increase exposure 0.3, then another 0.3, then 0.5, then -0.3 the result is that LR stores an exposure value of +0.8, the net value.  

PS - the history panel stores all the individual steps of course, but also stores the net result of the edits.  Only the net result of the edits is stored in the .xmp sidecar file. 
« Last Edit: March 09, 2013, 03:31:00 AM by Simon Garrett » Logged
BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3655


« Reply #4 on: March 09, 2013, 03:39:39 AM »
ReplyReply

I have two almost identical images of one flower in the range of 22,000 KB, one of which has had extensive Adjustment Brush editing (two pins on the whole background).

The XMP file of the non-edited image is about 7 KB; the XMP file of the heavily edited one is 189 KB.  Although there is a twenty-seven fold increase in the XMP file size, 27 times almost nothing is still almost nothing.

Hi Glenn,

That's +1.82GB per 10000 images. I wouldn't call that insignificant, not the end of the world, but still. Add to that perhaps some 10-20% of additional space is wasted on the fact that data is stored in full clusters of sectors, and therefore small files waste relatively more so-called slack-space. Add to that the waste cumulating in storage (mirroring). The additional real data (not the slack-space) also adds to required bandwidth for data transmission (and cloud storage fees).

So I wouldn't dismiss Phill's question that fast, although I do not think it is possible to integrate the corrections in the Raw file, because it wouldn't be Raw anymore (unless the spot correction data is contained within a partially processed DNG envelope).

Cheers,
Bart
Logged
Tony Jay
Sr. Member
****
Offline Offline

Posts: 2121


« Reply #5 on: March 09, 2013, 03:58:44 AM »
ReplyReply

The math is interesting Bart but I would clarify the detail.
That increase would pertain to 10 000 heavily edited images surely?

Tony Jay
Logged
BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3655


« Reply #6 on: March 09, 2013, 04:53:57 AM »
ReplyReply

The math is interesting Bart but I would clarify the detail.
That increase would pertain to 10 000 heavily edited images surely?

Hi Tony,

Obviously, yes, as in Glenn's example.

Cheers,
Bart
Logged
Tony Jay
Sr. Member
****
Offline Offline

Posts: 2121


« Reply #7 on: March 09, 2013, 04:56:28 AM »
ReplyReply

Thanks for the confirmation Bart.

Tony Jay
Logged
elied
Sr. Member
****
Offline Offline

Posts: 263


« Reply #8 on: March 09, 2013, 05:53:55 AM »
ReplyReply

I wonder if the net catalog size increase would be less if after doing the spotting or brushing one were to make a VC, make the VC the master version, and then remove the original master. Of course you would lose the history.
Logged

Roll over Ed Weston,
Tell Ansel Adams the news
PeterAit
Sr. Member
****
Offline Offline

Posts: 1911



WWW
« Reply #9 on: March 09, 2013, 07:38:10 AM »
ReplyReply

To answer the first part of the question, no I don't think there's any way of "finalising" any adjustment in LR.  Remember that LR merges all the edits anyway.  For example, if you increase exposure 0.3, then another 0.3, then 0.5, then -0.3 the result is that LR stores an exposure value of +0.8, the net value.  

Are you quite sure about this? When I make multiple adjustments to exposure, or any other slider setting, the history lists each and every one. While the image shows the net value, it seems that each individual change is stored in the catalog.
Logged

Peter
"Photographic technique is a means to an end, never the end itself."
View my photos at http://www.peteraitken.com
Simon Garrett
Sr. Member
****
Offline Offline

Posts: 371


« Reply #10 on: March 09, 2013, 04:44:44 PM »
ReplyReply

Are you quite sure about this? When I make multiple adjustments to exposure, or any other slider setting, the history lists each and every one. While the image shows the net value, it seems that each individual change is stored in the catalog.
AFAIK, Lightroom stores the edits in two ways.  The complete list is stored in the catalogue for the history.  The summary effect of the edits is also stored in the way that I described.  This is stored in the catalogue and also in the .xmp file alongside the raw image (or embedded in metadata in the image file if it's DNG or it's not raw).  That means you have every edit, so you can undo one-by-one, but LR doesn't have to render each edit individually, but only the sum total effect for each control. 

You can demonstrate this.  Remove a raw file from the catalogue, but don't delete the raw or .xmp file.  Then re-import it.  The sum total effect of the edits will be there (as that was stored in the .xmp file, and imported into the catalogue when the image was re-imported), but not the history (as that was stored only in the catalogue, and was lost when the file was removed from LR). 

If I'm wrong, I'm sure someone will come along and explain it better!
Logged
PeterAit
Sr. Member
****
Offline Offline

Posts: 1911



WWW
« Reply #11 on: March 09, 2013, 08:59:10 PM »
ReplyReply

AFAIK, Lightroom stores the edits in two ways.  The complete list is stored in the catalogue for the history.  The summary effect of the edits is also stored in the way that I described.  This is stored in the catalogue and also in the .xmp file alongside the raw image (or embedded in metadata in the image file if it's DNG or it's not raw).  That means you have every edit, so you can undo one-by-one, but LR doesn't have to render each edit individually, but only the sum total effect for each control. 

You can demonstrate this.  Remove a raw file from the catalogue, but don't delete the raw or .xmp file.  Then re-import it.  The sum total effect of the edits will be there (as that was stored in the .xmp file, and imported into the catalogue when the image was re-imported), but not the history (as that was stored only in the catalogue, and was lost when the file was removed from LR). 

If I'm wrong, I'm sure someone will come along and explain it better!

You may well be correct, but still all individual adjustments are saved and take up space, although this is, as others have pointed out, a non-issue.
Logged

Peter
"Photographic technique is a means to an end, never the end itself."
View my photos at http://www.peteraitken.com
Glenn NK
Sr. Member
****
Offline Offline

Posts: 290


« Reply #12 on: March 10, 2013, 12:08:30 PM »
ReplyReply

Hi Glenn,

That's +1.82GB per 10000 images. I wouldn't call that insignificant, not the end of the world, but still. Add to that perhaps some 10-20% of additional space is wasted on the fact that data is stored in full clusters of sectors, and therefore small files waste relatively more so-called slack-space. Add to that the waste cumulating in storage (mirroring). The additional real data (not the slack-space) also adds to required bandwidth for data transmission (and cloud storage fees).

So I wouldn't dismiss Phill's question that fast, although I do not think it is possible to integrate the corrections in the Raw file, because it wouldn't be Raw anymore (unless the spot correction data is contained within a partially processed DNG envelope).

Cheers,
Bart

Bart:

This was an extreme example (an outlier really) using the brush on about 80 percent of the image, not removing a few dust bunnies.  The percentage of extra space required was (189-7)/22,000 which is about 0.83 percent.  On this particular image, I had brushed out the entire background of a flower - twice to increase the effect.  The analogous technique with dust bunnies would be to erase each one twice.

I can't even say I've noticed the size of the XMP files increasing when I apply other edits - perhaps they do, but the XMPs are so small to start with that a few culls of garbage would seem to easily compensate for the size of the XMP files, let alone the difference in size due to edits (once an image is opened, there is a minimum size XMP created for it).  Note that I don't ever convert my RAW files to any other format.

Glenn
Logged

Economics:  the study of achieving infinite growth with finite resources
BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3655


« Reply #13 on: March 10, 2013, 02:02:06 PM »
ReplyReply

Bart:

This was an extreme example (an outlier really) using the brush on about 80 percent of the image, not removing a few dust bunnies.  The percentage of extra space required was (189-7)/22,000 which is about 0.83 percent.  On this particular image, I had brushed out the entire background of a flower - twice to increase the effect.  The analogous technique with dust bunnies would be to erase each one twice.

Hi Glenn,

Thanks for that additional info.

Cheers,
Bart
Logged
JRSmit
Sr. Member
****
Offline Offline

Posts: 375


WWW
« Reply #14 on: March 10, 2013, 04:38:25 PM »
ReplyReply

Hi Glenn,

That's +1.82GB per 10000 images. I wouldn't call that insignificant, not the end of the world, but still. Add to that perhaps some 10-20% of additional space is wasted on the fact that data is stored in full clusters of sectors, and therefore small files waste relatively more so-called slack-space. Add to that the waste cumulating in storage (mirroring). The additional real data (not the slack-space) also adds to required bandwidth for data transmission (and cloud storage fees).

So I wouldn't dismiss Phill's question that fast, although I do not think it is possible to integrate the corrections in the Raw file, because it wouldn't be Raw anymore (unless the spot correction data is contained within a partially processed DNG envelope).

Cheers,
Bart
In relation to the image itself, the raw file plus the previews, the variation in xmp file size, thus also its record in the lrcat, is pretty insignificant. The variation raw file size is not that much different.
Logged

Fine art photography: www.janrsmit.com
Courses and workshops: www.centrumbeeldbeleving.nl

Jan R. Smit
Pages: [1]   Top of Page
Print
Jump to:  

Ad
Ad
Ad