Guess my question would be what is size of the average raw and what is size of average xmp file? If the xmp file is only 1 - 10 % of the raw file size what would you gain? It seems like to be of any advantage the xmp would have to be close to the size of the raw file. Usually the xmp is quite small.
So to gain space you would be better served to unneeded images or put them in cold storage. HD space is cheap now, and if a comercial account storage would be cost effective if the image is needed in the future. If the chance of future use is slim then you have to ask yourself is it worth the storage price to me, or to the client?
I'm not working with raw files, I'm using tif images that are 100's mb each (these are our "raw" files). So just saving off teeny .xmp files would be a huge storage savings. And where I work trying to manage this data is like herding cats - we're talking storing and archiving terabytes per job. I still don't know how to get those xmps found/created/saved in mass quantities yet...
If you click View/show hidden files you can see the xmp files. If you have a 200 meg tiff how big is the xmp? My point is that if the xmp is 200k you could delete on tiff and it would be worth 1000 xmp files and make your life a lot simpler. If it is 20k then 10,000 xmp files = 1 image. Surely the xmp is not in the hundreds of megabyte range?
To save an xmp file you can choose manually, or else you will have to have a script written to do in bulk. But before you do this you have to answer the question "how will I pair up the xmp files to the original if they come from different folders when I want to reconstruct the image?".
"If you click View/show hidden files you can see the xmp files. " I have, and they don't show up. The example below is from a set of images already processed out to adjusted tifs, where I did not open them all up in Camera Raw and select save xmp manually.
Does Bridge/CR automatically generate xmps? The Bridge cache is on my PS scratch drive, maybe their there?
Depends on how you edit image. If you edit tiff in ACR an xmp file will be created. If you edit in PS then there will be no xmp, at least in my small test.
Original tiff 15 mb xmp 6kb. Edited in PS and added 2 layers size jumped to 48 mb.
Edit on in ACR just to check if xmp shows and how big it will be.
Curt (and anyone else interested in this thread) can you please follow this discussion in the CR forum? Follow the link above, as I'm starting to get conflicting information as to whether editing tif images in CR actually makes xmp files automatically OR do I have to create some type of automation to do this. Thanks again for your time!
I admit I made a mistake on the xmp. I had a tiff edited in ACR and a xmp was in place with same name. However that was the xmp for the nef of the same name. You can see if the image has been edited in ACR as it has a "badge" symbol in upper right corner.
But apparently there are no xmp files generated for jpg and tif files edited within ACR, the data is recorded inside the file in the same format as for xmp.
This imagery IS edited in ACR. After reading what you and others are now saying about ACR not automatically creating xmp files but embedding the adjustment data in a processed tif output - I'm stumped as to what to do next. I'm still thinking that it would be worth the effort to automate (now through scripting in Bridge) the creation of .xmp side car files that can be archived and "reunited" with original raw tif image data if needed vs. saving are these interim images. Any thoughts?
In you image from post 4 I see the badge of an ACR edit. But if this just contains data from ACR edit (exposure, tint, etc.) it may only less than 10 kb of information. The real volume of edits will be from the stuff you edit in PS and don't think it is included in xmp. I added 3 layers in a tiff image and can see no ACR edit badge, so assuming no xmp file was not created.
So therefore, you can save maybe 10 kb by stripping out the xmp file, but if the edits are 300 mb, and not in the xmp file, what is the point?
You need to try this on a test image and see what you come up with.
I have no idea how one would strip all the edit data from an image. More storage seems like the simple answer as seems like you are creating a nightmare.
Yes I will need to test this out. But I don't think you understand fully what I would like to do. Stick with me, I'll try again:
1. Images edited via ACR in Bridge are 300-500mb tif. Think of these as 'raw' camera files (unedited image data). These are flattened images - no layers, just big . This is geospatial image data. With a full set of four band channel information they can go over 1gb each.
2. I use ACR to do basic adjustment and use Photoshop to run a basic 'import from camera raw and save as' action - NO other PS adjustments are made. I do this because it is much more flexible vs. export in Bridge, and it frees up ACR so I can multi-task.
3. We do not want to save these images after we're done using them to make other things. If we can create individual .xmp data files that contain adjustment info that's ONLY 5k vs. 400+ mb per image, we can archive the xmp data with those original images from step #1. We should be able to reproduce imagery like #2 if needed in the future (this is an attempt @ saving archiving space) from the raw and xmp data if it placed in a common directory and opened up in Bridge? Right?
4. I'm open to ANY other ideas for achieving this. Saving lores jpgs and getting the adjustment data out of those somehow? Exporting DNG and try the same thing?
I hope this explaination is better than my previous attempts...
The issue I can not get beyond is this, as I understand your explaination.
1. You have a 400 mb image which is a tiff image.
2. You run it though the ACR edit in Photosop to adjust for color and whatever. This file is 5 kb.
3. You then save as TIFF.
You state "We do not want to save these images after we're done using them to make other things.".
My problem - if you can strip out the xmp file which contains the edit data and throw away the image how will that recreate anything. For instance say you just adjust the color temperature by 500 degrees negative. THis is recorded in the xmp file, but for it to be of any use it has to access the original image. So if that is thrown away what good is it to know the kelvin temp was 500 too warm?
My second problem - if you are starting with a raw image why convert to tiff and then do the ACR edits? Why not do the ACR edits on the raw, convert to tiff if needed for printing, then delete tiff? You are still left with original raw and xmp.
I didn't want to leave you hanging with your questions. If you check the thread on the CR forum you'll see that I don't have raw files to work with (not in the sense that you or ACR understands at least) - the tif image is the 'raw' image processed out from a proprietary image dataset.
I now have a way of stripping out the xmp data from these adjusted tifs - now I need to find a way to join them back other than to manually open each file in ACR and select them using 'load settings'. If you have any ideas along this line that would be great.
Thanks again for your time!