Marc Sabatella August 24th, 2007
The Workflow
The workflow I am about to describe differs slightly from the one in Peter Krogh’s The DAM Book. Some of the differences are due to the fact that Krogh is dealing with at least two different applications – Adobe Bridge and iView MediaPro. He therefore tries to do things in an order that minimizes the need to switch back and forth between them, and has to take some extra steps in order to get them to share information. Because we will be working within a single application – ACDSee Pro – we don’t have to make those sorts of compromises. On the other hand, some of the differences between my workflow and his are due to other factors. For instance, I do not assume you will want to convert your RAW files to DNG, and indeed, I do not assume you are shooting RAW at all. I also am trying to be pragmatic about how much backup people are willing to do, and how much hardware they are willing to dedicate to DAM. A few changes are also to facilitate some things I find useful, such as having easy access to my images on my notebook computer even when my primary storage disks are not attached.
Here, then, is the list of steps to perform for each new cardful of images:
- get images
- rename files
- RAW: convert to DNG (optional)
- rotate images as necessary
- enter high level metadata
- rate images
- enter low level metadata (optional)
- RAW: custom process high rated images as necessary (optional)
- RAW: batch process low rated images as necessary (optional)
- export metadata
- generate editable copies (optional)
- edit copies as necessary (optional)
- generate proofs (optional)
- move files to archive buckets
- export database (optional)
I will now explain each of these steps in more detail, beginning with the things you must to do to set up before you can start processing images using this workflow. I am including a lot of explanation, and it may appear at first like a lot of work, but to put things in perspective, I can process a card with a 100-200 images on it in just a few minutes if I put my mind to it. I do not always pressure myself to work so fast, though, so I may have several folders representing several cardfuls of images that are at various stages in this process. It is important to keep in mind that while I refer to these steps as a “workflow,” the time you spend with your images can and should be fun as well.
Get Images
Once everything is set up and you are ready to start working with a cardful of new images, your first step is to copy them to your computer. Connect your camera or insert your card and go to File->Get Photos->From Camera or Card Reader. Select your device and press next, then next again (no reason not to just import all the images). I recommend the options to keep the original filenames (for now), to automatically rotate images. Under “Location”, choose your Working folder (this will be selected by default in the future). I also recommend letting ACDSee create a subdirectory with today’s date. Depending on how your camera names you files and how you want them renamed on import, you may be able to use the renaming facility of the Get Photos wizard. But it is also possible you will have to rename the images as a separate step using Tools->Batch Rename, because this provides more flexible naming options than the Get Photos wizard.
I should mention that I do not normally actually use ACDSee Pro to import my images. Instead, I use a free utility called ImageIngester. I do this in part because it has very flexibile renaming options, but also because ImageIngester will run the Adobe DNG converter on my RAW files for me, meaning I can skip both of the next two steps.
Rename Files
Now that you have the files on your computer, you should rename them to something more useful (assuming you were not able to do this on import). As I mentioned previously, this is not going to be anything having to do with the subject of the photo, but it will still be more useful. What you want is a name like MJS_070224_3642.PEF. “MJS” are my initials, and you will of course replace that with your own initials or name or something to identify these images as yours. “070224” is the date in YYMMDD format. We use this format so the files will line up in the correct order even when sorted by filename. “3642” is the number that was part of the original name generated by the camera. My files start out with names like IMGP3642.PEF. I can turn this into what I want by selecting all images and going to Tools->Batch Rename. I created a template by typing “MJS_” and then inserting the metadata Image:Date/time original, then typing “_” and insert the metadata File Properties:Filename (w/o extension). On the Search and Replace pane, I replace IMPG with nothing. This template is available to me every time I enter Batch Rename. Actually, as I mentioned previously, ImageIngester can do this renaming for me, so I normally do not need to rename my files from within ACDSee Pro.
RAW: Convert to DNG (Optional)
Peter Krogh, in The DAM Book, makes a strong case for converting your proprietary RAW images to DNG and then throwing away the originals – treating the DNG versions from then on as if they were the originals. The main reason he favors this is the DNG allows a preview of the image that includes your adjustments to be stored in the image itself. Metadata such as IPTC can also be stored in the DNG file itself, and Krogh believes that an image in DNG format is more likely to be understood by more programs and longer into the future than an image in a proprietary RAW format. These issues are of concern primarily if you are trying to share files between applications, but since ACDSee will be handling virtually everything for us, the arguments in favor of using DNG are not so strong. Furthermore, since ACDSee Pro does not currently support including a preview in the DNG file, and metadata such as IPTC can be stored in sidecars for proprietary RAW files, and ACDSee Pro does not actually support DNG files from cameras for which it does not also support the proprietary RAW files, there is less to be gained by using DNG with ACDSee Pro. However, if you are concerned with future portability of your proprietary RAW files, or if your camera does not produce compressed RAW files, or if incorporating metadata into the image files themselves rather than sidecars is important to you, then you may well find that DNG gives you an advantage. Since my camera does not compress RAW files, converting to DNG cuts the file size roughly in half, and that is an important factor in why I personally convert to DNG and delete my PEF files.
While ACDSee Pro does not offer DNG conversion itself, the Adobe DNG Converter is a free download. You can install it as one of your external editors via Tools->Open in Editor->Configure Editors; this need be done only once. You can then select images and then run the DNG converter from within ACDSee Pro. I recommend not including a full size preview in the DNG file, since any preview generated by the Adobe DNG Converter will not reflect any processing you do in ACDSee Pro.
After the DNG file has been generated, I feel I can safely delete my original PEF files. You may prefer to archive them first and then delete them, or move them to another folder. Again, I normally use ImageIngester to download my images, and it can run the DNG converter for me and move my PEFs to another folder where I can delete them or archive them at my leisure. This means I can skip directly to the next step.
Rotate Images As Necessary
If your camera is capable of detecting when you are shooting vertically and recording this fact in the image file, and you had ACDSee Pro (or ImageIngester) automatically rotate the images accordingly on import, then this step will never be necessary. I am not so lucky. All images I shot vertically need to be rotated at this time. Images shot as JPEG files can be rotated directly from ACDSee Pro – simply select them and go to Tools->Batch Rotate / Flip Images (available as a button on the toolbar as well). Images shot RAW files cannot be rotated in this manner. But there is another fairly simple solution: create RAW presets for rotation, and apply these to images in need of rotation. To create a preset to rotate clockwise, enter the RAW processor on a RAW image that has had no processing applied, go to the Crop/Rotate tab, hit the “Right 90” button, and save this as a preset. Do the same for counterclockwise rotation using “Left 90”. These presets can then be applied directly to a group of images from the browser. Note that when you apply one of these presets, it will override any previous processing you have done on that file, so be sure to do your rotation at the beginning of your workflow as I am suggesting. Also note that because this rotation is performed using RAW processing, it does take a few seconds per image for the thumbnail to be updated to reflect the change. However, this work is done in the background, so you can continue working while the images are processed.
Enter High Level Metadata
The metadata you should enter at this stage is that which applies to large numbers of images – especially adjacent images. For instance, if the card I have downloaded contains pictures from a birthday party as well as pictures from a friend’s garden, I will start by selecting all of the birthday party pictures and entering the metadata that applies to all of them, then selecting all the pictures from the garden and entering the metadata that applies to all of them. I use the Notes field of the ACDSee Pro database to enter a short description of the event or location, like “Wendy’s birthday party” or “Jill’s garden.”
I also have keywords and categories I assign to my pictures, and I will try to get started on that now as well. In general, keywords are what I expect other people may someday use to search my images, and I try to choose terms with this in mind. The keywords I enter into the database will be copied to the IPTC Keywords field in the “Export Metadata” step (this is what makes the keywords accessible to others as well). Categories are things that I use to keep track of images for my own purposes, and while I have been copying these to IPTC fields as well, there is no real standard for using this information, and I do not depend on others being able to use it. In practice, the distinctions between what makes sense to enter as a keyword versus a category are admittedly often fuzzy, so some examples may help.
I always try to assign all my images a keyword of either landscape, cityscape, seascape, portrait, candid, interior, macro, animals, plants, or perhaps some other term that basically describes the subject matter of the picture at the highest possible level. In addition to this, I assign keywords for each person pictured in a photo if it is someone who is well known or whom I expect I or someone else may want to search for later. If the photo depicts a significant identifiable place, object, activity, and/or emotional state, I may enter a keyword for these, too. Most pictures end up with 3-6 keywords. I try to assign the first two or three – which, except for names of individual people, are often the same for many if not all pictures shot in a given session - at this stage in my workflow. After rating the images, I may get pickier and enter more keywords for the best of the images, to make them easier to find in the future.
The categories I define are harder to pin down. In general, I try to follow the advice of Peter Krogh in The DAM Book when he says to create categories because you need them, not just because you can. For instance, I took a lot of pictures of my fellow students while working on my Masters degree in Jazz Composition at the University of Denver, and I knew I would be creating a slide show of these at the end of the year. So I created a category called “DU Jazz”, and each time I shot pictures of DU musicians, I entered them into this category. When it was time to create the slide show, I started by creating a subcategory called “2007 Candidates,” and I put the highly rated photos from “DU Jazz” into this new category. I then created another category for the images I selected for the actual show. But the only category to which I would assign images as I shot them was the overall “DU Jazz” category, and I did that only because I knew in advance I would be wanting to create a slide show of these images.
For important events, I have categories that essentially duplicate the type of information I enter in the Notes field. For instance, I have a category “Weddings” with subcategories for each wedding I have attended and shot pictures. Any picture from Jessica and Wouter’s wedding will probably say something in “Notes” that reflects this, and I can certainly search on those terms, but it becomes that much easier to find these images if I have assigned them to their own category. I only bother doing this for images I expect to want to find later as a group. If I were a professional photographer, I would have categories for each job and would assign my photos to them immediately.
I have categories for groups of people for whom I do not feel like entering individual keywords. For example, if I take a picture of a friend of a friend I meet at a party, and her five children are in the picture, I may just create a category with the name of the family rather than enter keywords for each family member.
I have also used categories to record location information, but given that I want this to eventually live in IPTC, and there is no simple way to automatically transfer location information from categories to IPTC, there is really nothing to be gained by having this information duplicated in categories. Since I shoot most of my pictures in or around my home city, I have this set as the default during the “Export Metadata” step. I can easily override this location for a specific batch of pictures taken elsewhere. Another possibility if you travel more than I do is to use the Author field of the database to record location information, and copy this field to IPTC location fields when exporting metadata (I recommend the Author field simply because it is the only field I never use otherwise).
I should note that copyright and contact metadata will also be exported later, but on the assumption that this does not tend to change from one batch of images to the next, you do not need to do anything about that at this stage – this information can also be incorporated into the defaults for the “Export Metadata” step.
So to summarize, I am primarily concerned with entering notes, keywords, and to a lesser extent, categories (most images will have at most one or two categories; many will have none). These first few steps are very easy and quick to do – especially if you enter just the bare minimum at this point, and save keywording tasks such as identifying individual people in each photo for later. I would recommend trying to get at least this far immediately upon downloading your images, and consider backing them up (by running File->Sync to copy them to your external drive, or copying them to rewritable DVD) at this point, especially if you know you are not going to complete the full workflow at this sitting.
Rate Images
At some point, after entering the most basic of the metadata I want to enter for my images, I start rating them. There are two main reasons to rate images. One is so you can then concentrate most of your subsequent work on the higher rated images. The other is so that when you are searching for images later, you can constrain your search to just the images you previously identified as the better images.
The rating scheme I use is based on the ideas of Peter Krogh as expressed in The DAM Book. My scheme differs somewhat because of differences between ACDSee Pro and Adobe Bridge but also simply because I had some some ideas of my own in this department. I assign ratings as follows:
1: This image has no redeeming qualities. If I deleted it, I wouldn’t be likely to ever miss it.
2: This is a perfectly acceptable picture, but is not one of the best ones I took of that subject in that setting. I would show it to someone who wanted to see everything I had of that subject or that setting.
3: This is one of the best pictures I took of that subject in that setting. These (and higher rated pictures) are normally the only ones I show people who are interested in a subject or setting, unless they specifically ask to see more.
4: This is one of the best pictures I have taken of any subject in any setting. I would show it to someone who wanted to see some good examples of the kind of pictures I take in general, whether or not they had specific interest in that subject or setting. These are the ones I would be most likely to consider uploading to a photo sharing site, use as a screensaver, or print at some size larger than 4”x6” (you have to realize these are all relatively rare events for me).
5: This is not only one the best pictures I have taken; it is one I think compares well with pictures I have seen from other photographers I admire. I would consider entering this into a photo competition.
When I first started trying to organize my pictures, I think I took too long rating my pictures, in part because I did not have a clear idea of what my ratings should really mean. I was asking myself, “is this a 2 or is it a 3” without giving myself anything concrete on which to base this decision. On the other hand, by asking myself, “is this the best picture I took of that subject in that setting or is it not”, it is at least clear what I am trying to determine. It may still take a bit of comparing to decide which are in fact the best of the images of a given subject in a given setting, but at least it is then clear what ratings to assign the the images. I also have keyboard shortcuts defined to make assigning ratings even easier: 1, 2, 3, 4, and 5 (plus 0 to clear the rating). Note if you do define keyboard shortcuts for rating, be sure to do so both from within the browser and within the viewer, since you may find you do rating from the both windows, and keyboard shortcuts defined in one window do not apply to the other.
The vast majority of the pictures I take are either 2 or 3, with many more 2s than 3s. I try not to give out too many 4s – this is reserved for the pictures that are truly memorable. I have yet to give out any 5s. Of the pictures I assign a rating of 1, I actually do delete quite a few – ones that contain major mistakes (eg, out of focus, subject moved, completely wrong exposure) that were fixed in other shots. But if there is anything even remotely usable about the shot, it doesn’t hurt to keep it – disk space is cheap. By giving it a rating of 1, I will know not to bother doing any post processing work on the image unless I decide to revisit it later, and I’ll only do so if I should have reason to want to see everything I have ever shot on that topic.
I may also have images that are so bad they should simply be deleted. Disk space is cheap enough that if there is even the slightest doubt, I keep an image, but some clearly have no value whatsover. Even so, rather than deleting images when you first look at them, I recommend getting in the habit of just marking them for deletion later, perhaps with a custom category you create and assign a keyboard shortcut. Then you can review your choices when you have finished rating all your images to make sure you are not deleting images you should have kept.
After I have rated my images, I usually sort them on rating. This gives me a chance to make sure I have really made the right choices and not missed anything, and it also allows me to easily identify the images rated 3 and higher so I can concentrate on them for the next steps. I have keyboard shortcuts assigned to the sorting criteria I use most often – rating and filename (recall the filename includes the date, so that sorting on filename is equivalent to sorting on date).
Enter Low Level Metadata (Optional)
This is just an extension of the metadata entry I do before rating. But since I have now identified the images I think are the best of each subject, I can limit my effort to those. Realistically, I often treat 2s and 3s pretty much the same int his respect – which is to say, I may not enter any additional metadata at this point. But if I shot any 4s on that card – and often, I have not – I can consider entering some more esoteric keywords, like ones depicting emotions of abstract concepts. Although I have not tended to enter captions on individual photos unless there is a specific reason to, this is the point where I would consider doing so, at least for the 4s. In short, there is no specific reason to split your metadata entry into two phases, but I often find it useful to perform some before rating (when the images are sorted in there order taken) and some after rating (when the images are sorted by rating). There is, of course, nothing stopping you from switching between sorting by rating and by filename / date after doing the rating.
At this point, we can turn our attention to any processing we want to do on our images. This is where the workflow will differ the most depending on whether you shoot RAW or JPEG, and as you will see, can actually be easier if you shoot RAW.
RAW: Custom Process High Rated Images As Necessary (Optional)
Because RAW processing is non-destructive – meaning that changes you make are stored in the database and do not require modifying the actual file – we can do this work without the need for generating copies of the files to work on. And because the ACDSee Pro RAW processor is so fast and can perform much of its work in the background, we can easily move from image to image, and back and forth between the RAW Processing and the browser, without having to stop to save a file each time we move from one image to another. In fact, because it is sometimes difficult to compare images to see which is truly the best of a given bunch without doing some processing (such as improving exposure or adding sharpening), we can quickly do a little of this as desired while rating our images, and find that any processing we started to apply during that step is still there for us now waiting for us to fine tune – again, all without the need to ever explicitly save a new version of the file.
Another great thing about RAW Processing in ACDSee Pro is how easy it is to copy settings from one file to another. Once you have the exposure, color, sharpening, and noise reduction for one image looking the way you want, you can easily apply these settings to other similar images.
While RAW Processing does not give you access to all of the editing tools that ACDSee Pro provides – no red-eye reduction or photo repair, for example – I think you will find that many of your photos can be improved significantly in just a few seconds of your time in the RAW Processor, and that you will often not feel the need to perform any further editing. For the photos that do require more work, you can always convert them to JPEG or TIFF and edit them later.
Depending on the type of shooting you do and how you feel about post processing your photographs, you may not feel the need to do any custom processing at all to even your highly rated images, and that is fine. For me, the benefit of shooting RAW is still how easy it is to do when necessary. I do a lot of concert photography in which the lighting and exposure are very difficult to control (flash is usually inappropriate in these settings). The amount and color of light often varies drastically from one area of a stage to another, and the levels of light are often such that, even if using a tripod or some form of image stabilization and the highest ISO setting the camera supports well, the shutter speeds necessary to get a good exposure often result in blur from subject motion. So I find it more convenient to simply set my camera WB to the generic “tungsten” or “incandescent” setting, and set the exposure to something that has at least some chance of freezing my subject and still produce some sort of image, and shoot away, confident that the images can be improved through RAW Processing. Some people find this attitude offensive, but it works for me. On the other hand, most of my pictures taken outdoors need no further processing to make me happy, even with the camera on automatic WB and exposure settings. Anyhow, assuming you will want to process at least some of you images, I will describe a general strategy for doing so.
Because we chose at least one picture of each subject to rate as a 3, we can concentrate our attention on just the 3s. As mentioned before, I have my images sorted by rating at this point. I select all the images rated 3 or higher and enter the RAW Processor (and because this is an operation I use a lot, I have it assigned to a simple keyboard shortcut: R). I usually begin with the first image in line that needs work. If I have previously saved a preset that I think will apply well to this image, I select it. Otherwise – or if the preset does not work out as well as I had hoped – I spend a few seconds each on the exposure, color, and detail tabs to custom process the file.
Most sources recommend adjusting white balance first. I may try one of the presets, or I may use the eyedropper to select a neutral area. I may then adjust the sliders slightly to improve the color further. If you shoot a frame with a gray card or other calibrating device, you can set the white balance on that image and then copy it to the others shot in the same light. On the exposure tab, I usually start by adjusting exposure compensation as necessary to make the exposure look right without introducing clipping (be sure to have the clipping warning turned on). Then I may adjust the other controls to improve the exposure and contrast further. I usually look at the details tab only for high ISO images that need a lot of noise reduction and/or sharpening.
I seldom use the crop / rotate tab, because in general, if I am cropping, it is for a specific print size, and I would rather generate new conversions of the file for each specific print size as necessary. And rotation to orient the image correctly is done earlier. However, I do occasionally make use of cropping or slight rotation to fine tune horizontals, such as when shooting my own paintings, you are welcome to use this tab as you see fit.
When I have an image looking pretty good, I move on to the next. Again, one of the benefits of RAW Processing in ACDSee Pro is that I can move on instantly without stopping to save my work, and I can always come back to the image later if I change my mind, and my settings are intact so I can pick up where I left off if I like.
In many cases, I do not need to actually custom process images in order to improve them. Many will need no processing at all, and I will simply skip these. For the ones that do need work, I use the “Last Used” preset a lot, as well as the ability to select specific RAW settings for an image (by right clicking its thumbnail in the RAW Processor) and paste them on to other similar images. Often this is good enough. At most, I may individually adjust the curves for individual images after copying settings from another image. If, after processing an image, I think I may be shooting more images under similar conditions that will need similar processing in the future, I may save a preset for use later.
When I am done, or if I decide to return to the browser to work on some more files, I hit Done. Once again, the fact that I do not need to explicitly save my work, and the fact that ACDSee Pro is able to perform much of its work in the background, means I can immediately resume browsing and return to RAW Processing whenever I desire, without ever having to wait for anything.
RAW: Batch Process Low Rated Images As Necessary (Optional)
Here again, ACDSee Pro provides facilities that make it simple to improve the images we have decided are not worth spending as much time on – the 1s and 2s. For starters, you can apply RAW processing presets directly from the browser. If I do not have a preset I think is applicable, I will often return to RAW Processing on one of the higher rated images I had processed in the previous step, then immediately hit “Done”. This loads the “Last Used” preset with settings that may work for other images. One drawback of applying presets is that if you have already used a RAW Processing preset to rotate images, this will be lost when you apply the new preset. So instead, you may prefer to simply copy and paste white balance settings from one of the custom processed images to the lower rated images – this can also be done from the browser (Edit->White Balance). Or, if you wish to copy settings other than white balance, you can return to RAW Processing to copy all settings except rotation. Because these are lower rated images, I do not worry about making them perfect, but if they are severely underexposed or the white balance is so far off that they are not pleasant to look at, I will spend a few moments at least bringing the 2s into the right ballpark. Again, since this non-destructive, I know I can always revisit the processing for any given images later.
Export Metadata
As mentioned previously, keywords and other metadata should live in the IPTC fields, so they can be accessed by others browsing your images. We enter them into the database first because it is far easier and faster, but at some point after you finish your metadata entry, you should export it from the database into the image files themselves (or sidecar XMP files for proprietary RAW images). Because the RAW Processing parameters are also a type of metadata that can also be exported from the database, and because performing RAW Processing may cause me to change my mind about ratings and other metadata, I usually wait until I finish my RAW processing before exporting metadata.
The database fields can be written to IPTC fields using Tools->Batch Set Information, and this same command can fill in other IPTC fields. I have created a template that loads into Batch Set Information by default with all the fields set up the way I like them. Unfortunately, there is no one absolutely correct way to do this, so different people end up doing it differently. Note that if you are shooting RAW, it is important to use IPTC fields that are part of the XMP standard for IPTC, since this is how IPTC information is stored for RAW files. I have the Copyright Notice, Photographer, Credit, Source, Writer, and Special Instructions fields filled in with my name and contact information. The database Keywords field is copied to the IPTC Keywords field. The database Notes and Caption fields get copied to IPTC Headline and Caption respectively. I also record the filename in the IPTC Object Name field, so that I can still identify the image should someone else to whom I send it change its filename. If you record location information in the Author field of the database, you can copy this to the IPTC City field; otherwise you might have the location fields default to your home city.
This is information we record into IPTC in order to make it available to others. The other database fields are also worth writing to IPTC, simply to make sure they are preserved should your ACDSee Pro database be inaccessible, or should you eventually want to migrate your image catalog to another application. I suggest copying the database Ratings field to the IPTC Urgency field and the database Categories field to IPTC Supplementary Categories field. Both of these IPTC fields are considered “Deprecated” by the XMP standard for IPTC and hence are unlikely to be used for other purposes, but still exist and can be written to by ACDSee Pro, and both can be read by other applications if necessary.
With proprietary RAW files, exporting metadata to IPTC actually writes it to a sidecar file with the extension “.xmp,” not to the RAW file itself. Since we will seldom if ever be giving our RAW images to anyone else, this should not be a problem. Just be sure to always keep the sidecar files with the originals as you move them around. As you will see, we will do very little moving around of our RAW files, so this should not be a problem, either. If you use the DNG format for your RAW images, the metadata is written directly to the files themselves; no sidecar files are involved.
Note that the Beta 2 version of ACDSee Pro 2 does not actually support writing metadata to DNG files, nor will it create XMP sidecar files for these files. However, I have determined that you can force it to create sidecar files by simply running Tools->Batch Rename to change the file extension from “.dng” to the extension used for your camera’s proprietary RAW file, and you can change it back after running Tools->Batch Set Information to create the XMP files. This step will not be necessary in the final version of ACDSee Pro 2; metadata will be written directly to the DNG files. You may wait until the new version is released before exporting metadata if you are using DNG, or you may use the renaming workaround to generate sidecar files if you wish.
As mentioned above, RAW Processing parameters can be exported at this time as well. To do so, select your files and run Database->Export->RPP Files. This creates a text file with the extension “.rpp” in the same directory as your files if you are using proprietary RAW files. This step is not necessary if you are using DNG, as the RAW Processing parameters are already stored directly in the DNG file. Again, the Beta 2 version of ACDSee Pro 2 does not support writing to DNG, so I do recommend exporting RPP files for now.
This would be a good time to backup your images again, especially if you do not plan to finish the workflow in this session.
Popularity: 57% [?]