If you’re reading this, then I am already de… no, wait, I’m still getting a hang of these scheduled posts. If you’re reading this, then Adobe has released Lightroom CC, the latest in what has become the massively dominant industry standard of professional RAW processing and photo organizing. I was honored to be selected as a member of Adobe’s beta-testing team, and I can say that I have been loving every moment of testing a pretty solid product, and that I haven’t gone back to Lightroom 5 for months.
Lightroom CC has several major new features and enhancements, the most obvious being in-application panorama rendering, HDR, face recognition, and speed increases. These enhancements and others are diverse enough that most people will be really excited for some new features, and care little for others (although we all love more speed). Given that I am best known in some circles for inventing and popularizing a panorama technique, it’s not hard to guess that I was most keenly interested in the panorama features. But the devil is in the details, and after a few weeks I found myself using Lightroom in ways I hadn’t imagined.
The “YESSSSSSS”: Better speed
Speed has been the biggest complaint I hear among Lightroom users, and while whether Lightroom CC is “fast” depends on your system and subjective opinion, it is noticeably faster. In Lightroom 5 I usually would change my iMac 5K resolution to 25 percent of its full capability, just to keep things snappy. I’ve never felt the need to do that in Lightroom CC. I haven’t run numbers, but in practice it took away my biggest frustration with an otherwise great program.
Panoramas: Not quite there yet
Disclaimer: I have been testing beta software. The final version should be better at everything.
Seven years ago, I popularized a panorama technique that has come to be known as “The Brenizer method.” But before it caught on to the extent it has today, I almost stopped doing it completely.
Why? Because working with the software at the time was extremely annoying. You see, Brenizer method images are often produced with 50 or more images, and the stitching software at the time struggled to keep up. Photoshop CS could produce beautiful results, but it would seemingly freeze up two-thirds into a panorama for 15 or 20 minutes at a time, and you had to leave, go watch some TV, and hope. This is painful enough with one panorama, but if you shot five to 10 panoramas per job processing them could take all day. Things got a bit better until by CS3 they were working pretty well … and then CS4 came out. They had solved the progress bar issue, but for this particular type of panorama, the success rate dropped precipitously, and there were no easy fixes. I turned to third-party software partially for better results but, even more importantly, because you can set up a batch of panoramas and leave your computer to work them all out in peace.
Sadly, even though Lightroom is the general class leader in “setting up a bunch of batch processes and walking away while your CPU spins at 400 percent for a while,” there still is no way to batch process panoramas. And unless you are a careful, tripod-using sort of pano shooter, the results are … less than stellar.
Lightroom is and has been my choice for processing panorama pieces for a long time. The “match total exposure” feature is particularly valuable for any times where you couldn’t perfectly pre-set all of your parameters. The photos above were originally four shots taken on an iPhone 6 Plus, each at a slightly different exposure — which doesn’t make for a great pano. But Lightroom not only is able to do lens corrections on the iPhone camera, it can automatically adjust the exposure variation if you know how to find the surprisingly secretive menu item.
Four images, no major wind — it wasn’t the hardest panorama to stitch, even though there was no tripod. But Lightroom didn’t do too well at it, and there was no easy way to fix it. When I tried the same pano with Autopano Giga, it didn’t break a sweat.
HDR fans will be glad to hear that the HDR function works better in my testing, and HDR haters will be glad to hear that it works well at producing HDR photos that don’t look extremely “HDR-y”. The result pops back into the Lightroom catalog as a fully adjustable DNG file, with no inherent way to turn the tone-mapping dial up to 11. It just allows you to create a RAW file that has more bits and dynamic range than you could have made in a single shot, and then process to your choice from there. I haven’t shot much HDR over the years, but the speed, ease, and natural results of this means that I may try a bit more here and there.
Surprisingly great: Face recognition
There are some industries where face recognition can be extremely helpful, but wedding photography isn’t generally one of them. The best potential application is answering the question “Do you have any more photos of …” but the trade-off is that creating a library of fans even for a single wedding shoot can take hours. But Lightroom’s face recognition is so great that I have made it the primary way of collecting my personal photos, even though I am a devout “new catalog for every shoot” guy who has always used other programs for this purpose.
But face recognition becomes extremely handy when you are dealing with a giant collection of photos of people you really care about. Just the main folder of my friends and family photos has nearly 30,000 finished photos, more than enough to become unwieldy. But when a friend asked me two days ago whether I had a particular photo of her, Lightroom was able to find it in a few seconds.
Now, by “surprisingly great” I don’t mean that the actual face recognition algorithms are any better than Apple’s or Google’s — they all work in a way that seems fundamentally like magic, but they can all also be thrown off in amusing ways, such as Lightroom thinking that the faces of 20 or so of my friends live in this Christmas wrapping paper.
No, what’s great about Lightroom’s face-recognition is the implementation. Labeling the faces in 30,000 photos individually sounds like actual torture. It’s really important for programs to have very well-worked out systems for batching as many photos correctly as possible, and Lightroom does that much better than Apple’s new Photos app.
In the Faces section you will see the confirmed faces of any individuals you have named — and these names seem to only exist on a per-catalog basis. You can drag or drop either individual images or “stacks” that Adobe has automatically created, throwing more photos on the “Ryan” pile, for example. And you can also double-click on any confirmed individual and it will start looking through whichever folders you have selected for photos that look like it might be the same person.
This allows for a very efficient batch-labelling process. For example:
Looking for photos of my father shows that I have already found every photo that looks like him in the selected folder. But I also see a photo of my great-uncle Victor, who I haven’t created a folder for yet. Typing his name in will add him to the list of confirmed people.
With just one or two images Lightroom already has a good enough idea of what he looks like to find more photos. I can shift-click on the images below to select them all, and with one drag not only will I have more photos labelled correctly, but Lightroom will automatically and quickly use them to get a better sense of what he looks like, and find even more photos. Lightroom will start with its best guesses, and then guess more and more wildly. Given all this, you can very quickly fill out someone’s labelling folder by starting with even just one photo of them, and selecting the photos that appear before Lightroom starts guessing wrong, even if those aren’t all of the photos you see. With this process, the guesses will quickly just get better and better.
I’ve used iPhoto, Photos, and Picasa, and Lightroom’s implementation is the quickest and most intuitive. This alone has taken me from ignoring the cataloging features to being my primary way of collecting photos of my friends and family.
There are so many things to discover in Lightroom CC. I encourage you to try it out and see which ways you’ll be surprised.