Update: Apparently this is by design and can't be avoided due to the changes between the processing versions. Wow, Lightroom 4 is really off to a less than stellar start if you ask me.
Doesn't anybody at Adobe work with custom tone curves? That's hard to believe. I have just found a second issue with them.
Adobe released Lightroom 4 and they pretty much messed up the migration of tone curves when converting the catalog from Lightroom 3 to Lightroom 4. You can read all about it here, the story is still ongoing as of writing this.
I use custom tone curves a lot, so I filed the original bug report right after finding out about this issue.
This is why I'm now part of a group of people alpha testing a fix that should recover the lost tone curves after an upgrade and that will hopefully make it into the full version of Lightroom and into an update for those who already upgraded.
As far as I can tell, Adobe hasn't issued a warning about this to their existing user base and we can only hope that power users with tens of thousands of pictures (and potentially with as many tone curve adjustments) won't get too many nasty surprises due to the bug.
Here comes issue #2:
During testing of the alpha script, I noticed something else, that I find quite disconcerting: I know changing to the new process will change the appearance of pictures, which is why Adobe suggests an A/B preview, but when I had the tone curve open when switching a picture from process version 2010 to 2012, I noticed this:
The curve does keep its overall shape, but the quite elegant few points of the curve get replaced by a ton of individual points.
WHAT .. ON .. EARTH .. IS .. THIS?!
Doing a quick change to the mid tones, or to how the shadows are rendered is a simple fix with the original curve. The replacement curve is 100% useless for that.
The only way to make the curve usable again is to start over and re-create it from scratch.
If this is by design, then it means that those of us who use custom tone curves extensively (I'm one of them) won't be able to benefit from the 2012 process for any of their existing images unless they are ready to start from scratch on them. In that case I'd really like a word with the person who made that decision.
Today I tried to pay for a train ticket online using my credit card. It was rejected. Of course I went online immediately to check my balance and the transactions, but nothing looked weird. So I called my bank to check what was wrong.
Twitter does have a sense of humor!
Long story short, I ended up on the phone with the credit card company's fraud department. Turns out their system found that me ordering several tons of concrete from a US company was implausible enough to not even let the transaction go through.
Instead they blocked the card and sent a new one my way.
On asking how my card number got into the wrong hands, they said it's pretty much futile to try to find out, likely cause would be a merchant being hacked and even then it's impossible to connect as these things can lie years in the past.
Oh well, let's hope those plausiblity detection systems do their job. And let's also hope that I'll never have to order concrete from a US company.
tl;dr: the LR3 to LR4 update ate my tone curve adjustments for breakfast.
Update 05/02/2012: It's been almost two months and Adobe still hasn't officially updated Lightroom 4.0. Here are some of my thoughts on it.
Update 03/06/2012: Here is the bug report that I filed with Adobe. Please consider to +1 it on their site to help raise awareness of the issue.
Update 03/07/2012: Tom Hogarty, Lightroom Product Manager has posted the following update as a response to the bug report: "Thanks Chris. We've been investigating this issue throughout the day and hope to have an update soon. Thanks for your patience and all of the detail you've provided."
Update 03/11/2012: It's 4 days later and the bug report now has 55 +1's. Some of the commenters are starting to get really impatient.
Update 03/12/2012: I received a mail with a script and test instructions from Tom Hogarty. The script recovers lost tone curves and it mostly worked. It's just an alpha test version, so it still has a few side effects though. Adobe is on the right track with this, but they're not there yet. This still has to be tested more and then needs to make it into an update. I posted my findings in the bug report. 62 +1s now by the way.
Update 03/14/2012: Since Adobe provided the test script to some of those who had the issue (this includes me), there have been mixed messages on the bug report thread. It works for some, some have found other issues. It generally worked for me, even though I ran into another potential problem during testing the fix. This bug is also on the top of the list of Adobe's Lightroom 4 Hot Issues, so even though it's been eight days, there's still hope they will eventually fix it. Until then I'm staying on LR3 and I'm just a little annoyed that I paid for an update that so far is unusable to me.
I just updated from Lightroom 3.6 to Lightroom 4. The new features and the 50% price drop made it look like a no-brainer. I loved the product from the first beta of version 1. I even recorded two video workshops for Lightroom (in German).
If you've listened to Tips from the Top Floor or Happy Shooting, you know that Lightroom is the hub for my photography and you'd have to dangle quite a juicy carrot in front of me to make me give it up. Very juicy.
Unfortunately I have to report that after the update, Lightroom 4 seems to have eaten up all my tone curve adjustments and replaced them with some defaults.
Here's a picture in Lightroom 3, note its histogram:
Now here's the same picture in Lightroom 4. The picture looks the same, the histogram doesn't:
At first I was confused: the LR4 histogram implies a much brighter black point and a much lower overall contrast. Then I realized that the picture shown in LR4 and the histogram must be disconnected in some way.
Next I went back to LR3 and went into the develop module. Opening up the tone curve adjustment tool, it shows the tone curve that I gave the picture.
I use tone curves a lot. I use them more than the other contrast management tools. I probably use tone curves on over 60% of my images.
Next I went to LR4 and opened the same picture in develop mode. This is not the tone curve that I originally gave my picture.
After a few seconds, the preview was updated to reflect the "new" tone curve. At least now the picture matches the histogram again…
NOTE: Adobe Photoshop Lightroom 4 took the liberty to swap out the tone curve that I spent time creating to some sort of default tone curve.
NOTE 2: I did not enable the new 2012 processing, the picture is still on the 2010 process version.
For now I'll revert to LR3 and wait until they ironed out the issues. I'm filing a bug with Adobe as we speak.
Wait. Should you use LR4? "I never used tone curves, I shouldn't have a problem then" I hear you say. Well, if Adobe borked something as essential as the tone curves, that has an effect on my personal trust in the overall product.
I love the new features in Lightroom 4, I think the new processing is awesome and the map feature is the first one in a product that looks like I might actually use it. Books is a great addition too.
But I just don't think Lightroom 4 is quite ready with an issue this big under the hood.
It's not the camera, it's the photographer. We all know that. Do we live it? Not always. Which is why I did a deliberate "lesser photographer" thing.
Where I would usually have the iPhone in my pocket as an emergency or backup camera, this time I made a deliberate decision to go out and shoot with nothing but the iPhone. No big medium format camera. No DSLR. Just the iPhone 4s.
Our creativity strives under constraints. Some of the greatest photography has been made with cameras that some of today's photographers wouldn't even touch with a ten-foot-pole. So I went an extra step and instead of using the iPhone's built-in camera app, I used one that most people would call crippled. Its name is NoFinder and it is pretty much what the title of the app says: a camera without a viewfinder.
Now adding that kind of a restriction might initially sound silly, but it has turned out to be surprisingly good for the creative side of things. Not being able to look through a viewfinder helped me concentrate on the actual scene a lot more than if I had looked through a viewfinder. Pointing the camera without a display also left a certain margin of error, but in the end for many shots that lead to interesting and unusual framing choices that I wouldn't have made with a viewfinder.
Most of those accidental choices of frame weren't that exciting, but then there were a few that I found really interesting. And again: I wouldn't have arrived at them any other way.
The last two constraints that I placed myself under turned out to be pretty much the most beneficial ones: my decision to set the app to only take square pictures and to work in black-and-white only.
The lack of a viewfinder initially made it harder for me to judge the angle of view, but after a few shots it became pretty clear how much would be in the picture. As an added benefit I now have a pretty good idea of the field of view that I can get from the iPhone. I didn't really have that angle visualized before.
And in the end that's how Frog Umbrella came into existence. Being able to see the entire scene with my two eyes, I could watch the umbrella kid walking away from the building and while it was doing so, I fired three shots while trying to anticipate the framing.
And the third shot was the charm. That's my kind of picture - everything fits nicely, the frog's eyes are doubled in the building, every element in the photo feels like it belongs exactly where I put it. I'll be happy when I bring home one single picture like this every time I go out shooting. I'm still working on that.
» Frog Umbrella on Flickr (leave a comment)
I did the sys admin thing where I'd call myself and ask what to do, and I found a partial solution to my DTS issue. I'm far from having this process completed, but I'm a few steps closer of getting my collection of 6-channel DTS 5.1 CDs ripped onto my Mac and into a somewhat useable format.
It's still a fairly manual process, but here's what I've done so far:
Step 1: rip the DTS 5.1 CD to iTunes using the WAV encoder as the input format (no MP3, AAC or similar). This process seems to get the track names right. Not sure if it takes them from Gracenote or if it reads the CD text, but the titles end up in iTunes.
At this point, the ripped files are recognized as 2-channel WAV files, and their content is DTS-encoded 5.1-channel audio. iTunes doesn't give me that information though, and it doesn't seem to be aware of this, so playing these over a normal system will result in noise. In order to play them properly, they have to be played over a DTS decoder, or they have to be decoded on the system.
Step 2: convert the ripped WAV files to 6-channel WAV files using VLC. This partially works, VLC can even be kind of coerced into converting them in a batch, using the Streaming/Transcoding Wizard, but for whatever reason VLC refuses to convert the last track.
My process: open the imported WAV files in VLC, File - Streaming/Transcoding Wizard, Transcode/Save to file, select items from playlist, only check "Transcode audio", set codec to Uncompressed, integer, 192 kb/s, select WAV, select output directory, go. Again, this process will not convert the last file in the list for whatever reason. My workaround is to add the last file twice.
The file names get messed up a bit in the process, spaces end up being replaced by 20, probably due to URL encoding.
At this point I have 6-channel WAV files that work. That's a pretty good start, because from here I should be able to convert the files to anything else I need in the future, using Compressor.
It's easy to visualize with Final Cut Pro X at this point, all 6 channels work and they seem to be even in the right order (something that surround audio fans know isn't always a given..)
Hattler Surround Cuts (also available as a DVD)
Step 3: fix the broken file names using an Automator action. Replacing 20 with a space in the filenames is easy this way, I do all other URL encoded characters manually for now. If this were a process that I'd need to do several hundred times, I would try to fully automate that. As it's now, I am okay with doing a bit of manual work here.
Next steps: find a way to make this process simpler and more automated. It's not that I have hundreds of DTS 5.1 CDs, but I'd still like to find an easier and less error-prone way to do this. I'm also going to need to find a good way to play or re-encode those files into something that I can easily play. But the 6-channel WAV gives me at least a great uncompressed starting point to continue using them down the chain.
And then: find a way to play all this back over my existing infrastructure. Fun times!
I've got a challenge: I have several DTS 5.1 audio CDs. It's a format that isn't too common, but I have them and they have surround audio on them. Note: they are CDs, not DVDs. That's an important detail.
I want to make digital copies of the CDs onto my hard drive. On my Mac and that turns out to be by far not as easy as it seems.
My first idea was to create disk images. Disk Utility doesn't create images of audio CDs. Next try.
VLC was the next thing I went for. Partial success here, it can transcode the CD to 6-channel WAV, but VLC won't let me do individual tracks, or at least I haven't found out how to do it. My tries ended in VLC doing all the tacks and pipe them into a single file, overwriting it with the next track, then with the next track, to end up with one WAV file that contains the last track of the CD.
Also VLC apparently doesn't read the track names (which I assume are on the CD). Instead I get Track 1, Track 2, …
All it seems is that I need to find a a way to rip those CDs to 6-channel WAVs using VLC and being able to batch this somehow. I haven't found that way just yet. I'm using VLC 1.1.12 on Lion.
*SIGH* … I had that a lot in my former life as a sys admin. After having tried 100 things, it sucks to be told you should contact your sys admin.
So I asked on Twitter.
Which is probably the wrong place to go to for things like this in the first place. Everyone is very sweet and wants to help, so they google this for me despite me telling them that I've already done so. Problem is, they google based on my 140 character question or come back with assumptions based on that limited knowledge, while I've already done the googling with the knowledge what I'm looking for. And I know my Google abilities. I believe they are quite good.
So while I'm thankful for people trying to help me, I get incredibly frustrated because Twitter is the wrong medium. The only reason I'm still using it for this is because it is very fast and I don't want to wait for a few days until someone in a newsgroup or on a forum answers my question.
I guess I'll have to keep on searching...
I just ran across another blog article that asked the question if mobile phones would take over in the long run and overthrow all other cameras because the sensor technology and the fact that you tend to have one with you all the time.
I'm not so sure for a two main reasons.
1. Control. Cameras tend to get better and better, but even the best automated decisions will not necessarily reflect your intentions.
An example: think about a backlit portrait. Without built-in intelligence, the camera's light meter willtell the camera that there's a lot of light and the image that comes out is likely to be a silhouette of a person. Most cameras nowadays will detect this and compensate for it, resulting in a well-exposed person (and most likely a slightly overexposed background). I guess in most cases that's what the person behind the camera wanted anyway, so it's okay.
But how about the times when a photographer intended to produce the silhouette picture but didn't have a way to tell the camera that that's what they wanted?
The way the current mobile phone cameras look, it's very hard for me to believe that they will get to this level of control any time soon.
2. Sensor size. Different sensor sizes result in different depths of field (DOF) and control over DOF is a very important tool for most photographers.
In-focus and out-of-focus areas in a picture are one out of a whole array of essential tools for photographers when it comes to telling a story in a picture. Focus will show or hide things, focus will help you guide the viewer's eyes through a picture.
Smaller sensors make it very hard to control DOF. Everything tends to be in focus. Bigger sensors make it easier to control DOF. A photographer can place focus where it's important. And as things look right now, mobile phone cameras are pretty unlikely to get larger camera sensors.
Even if mobile phone cameras got larger sensors, that would mean that the lenses needed to be bigger and further away from the sensors, adding bulk and size. Very unlikely.
Will newer technologies and computational photography replace the need for bigger sensors in the future?
Who knows, but at this point in time, even the Raytrix and Lytro cameras cannot do their job without a certain level of bulk, and the results are by far not where they'd need to be.
What do you think? Are we going to see DSLRs disappear any time soon?
Yes, 'tis the time where we say 'tis again. And it's the time where we bring out the box of Christmas tree ornaments and decorate the tree. Yes, the Brownie Tree is back! And finally.. FINALLY the Christmas spirit kicks in.. and it feels good again.
My goal for 2012 is to keep photography in the center of my life, and to look at my images at the end of the year and see that I've learned something new again. So far this has worked, so let's make it work again for next year.
Here's to a wonderfully photographic 2012!
I have used iTunes Match for a few days now, and it has quickly turned into one of the best things since sliced bread for me.
No more sync trouble, as everything is available on all devices, either to download or to stream*
The complexities of multiple-device sync and having to decide what to take with me on which of them turned into the main reason that I never rippedmy entire CD collection (and there are several hundred of them, I'm *that* old ;)) - in the end maybe only 20% of my music was on my computers, distributed between several machines and iTunes libraries. Bit of a mess.
As a result of the new convenience I have now gone back to rip all the music that was still sitting on my shelves, I'm about halfway through my CDs (the picture above shows a small fraction of it) and every time I open the music app on my iPhone and see new/old stuff popping up, it makes me smile as I rediscover some great music I almost had forgotten about.
Thanks for making things quite a bit easier, Apple!
* quick note about the "streaming" part: apparently it's not really streaming. There seems to be a fine distinction between streaming and the "playback while download" implementation that Apple chose, possibly for political reasons. Factually it doesn't make a difference to me as an end user. What I particularly like is that even though if you play an album from the cloud, the first title takes a second to begin playing, the subsequent tracks are 100% seamless. Apparently it's implemented so that it starts downloading the next track if that is less than one minute away. This should pretty much guarantee continuous playback even on slower connections. I like that a lot!
With Creative Suite 5.5 Adobe is introducing a new subscription pricing model. For many professionals this is a welcome way to spread out the cost for the software over a year instead of having to do the big upfront payment for the software.
Customers can still buy individual products or product suites, but you will now also be able opt for a monthly plan. I will mainly look at what this means for photographers and Photoshop. But just as an example, instead of buying the Design Premium Suite for a retail price of $1899, if you commit for a yearly plan, you'll apparently get it for a "rental fee" of $95 per month or $1140 per year. Mind you, this is nota payment plan, so you won't own the software at the end of the year. Adobe is offering upgrade pricing for those who paid for a year though.
As mentioned, you can still buy the products, but as I understand it, as opposed to being able to upgrade from older versions (I believe you could skip up to two versions), with the new pricing model you can't skip versions anymore to get upgrade pricing.
And this seems to be the biggest rub for a lot of people. Enough of a rub that Adobe went ahead and closed (and apparently even removed) the comments on the blog entry where they announced the change.
International pricing of Adobe products has always been one of my pet peeves. In Germany and other European countries, prices for Adobe products are dramatically higher than the US prices, in some cases we Europeans get to pay more than a 100% premium for the same software.
Back in 2007 when I interviewed Adobe product manager John Nack I brought it up, but mainly got an evasive answer.
This might also explain why a lot of people on this side of the pond appear to use pirated versions of Adobe's products.
Over the years a lot of photographers have become Photoshop users. Photoshop isn't the most intuitive product - I usually compare it to a huge toolbox full of tools but without a good instruction manual - but it is very powerful and many photographers have taken the effort to learn its intricacies, to adjust their workflow and to master it to a certain degree.
As I said, I'll mainly look at photographers in this article, but this might also be true for small agencies.
While Lightroom has pretty much taken over when it comes to 98% of my pictures, many photographers have spent years and year refining their Photoshop workflows, they have learned tricks and spent time learning from tutorials. The investment not only on the financial side is huge.
But for monetary reasons many individuals and agencies have also had to adopt a model where they would skip a version or two before they upgrade to a higher version. This possibility is now pretty much gone, so my guess is that the sentiment of many Photoshop users is that they are now expected to pay double or triple the amount they used to pay in the past.
Aside from that ecosystem, let's have a quick look at what makes Photoshop so great.
The thing that intimidates new users most is also one of Photoshop's greatest strengths. It is a collection of hundreds of powerful image manipulation and design tools and if you know how to use them, there is almost no limit to what you can do with it.
Layers, masks and layer modes let you do everything from complicated composites to things as simple as slapping a layer of text to an image. The mix of vectors and pixel graphics and the resulting flexibility is unsurpassed and I love being able to use smart objects to treat pixel graphics almost like vectors.
Profiles allow for a color-managed workflow in pretty much any color space you like and over the years many specialized tools have found their ways into Photoshop, from handling animations to stitching big panoramas to 3D and perspective work.
The plugin model is another part of that ecosystem, with a ton of add-ons available to do virtually anything you can imagine.
But its strengths can also be seen as weaknesses. Photoshop tries to be everything for everyone and its user base is so diverse that it is hard to find a common thread. Illustrators use it, it has its applications in the pre-press processes, it has even medical uses and of course there are the photographers.
Because Photoshop wants to be for everyone, it feels like a big piece of patchwork rather than an integrated application.
There are still a few areas where I tend to resort to Photoshop. These include simple illustrations that use layers and masks, adding text to images, more complex cloning operations, adding transparency and stitching images.
That's pretty much it. I do everything else in Lightroom.
I can only answer that question for myself, and it's pretty much a resounding no at this time. The few uses that Photoshop still has for me are easily covered in the CS4 version that I still own and there are a lot of great alternatives out there that cover a lot of Photoshop's bases.
One of the strongest alternatives on the Mac platform at this point is Pixelmator. In its new 2.0 version it supports layers, layer masks, over 100 file formats, plenty of filters, and even some of Photoshop's "killer features" such as content-aware fill. For €23.99 it's a bargain. Is it a full Photoshop replacement? No, but it covers 95% of what I need as a professional. The one item it doesn't have and that's high on my wish list is 16 bit support, but for most of the things I use it, I can live with that. If that's a must for you, I suggest you have a look at PhotoLine. It's not as pretty, runs on Mac and Windows, and it supports 16 bits and more, for a mere €59.
As a Mac user I can cover most of the remaining 5% with the tools that Mac OS X already has on board and I'd be surprised if Windows didn't have similar things on offer. I use the ColorSync Utility to do color space conversions, which includes converting pictures to CMYK, so they are ready for a printing house. Preview, one of the Mac's most underestimated apps, lets me use any ICC profile to soft proof images. And Image Capture (the second most underestimated OS X app) serves as a great front-end to any scanner.
When I got my MacBook Air with its 128 GB SSD, I went through a long software list to decide what I needed on the road and what I could go without. Lightroom made it onto that list, Final Cut Pro X did, Scrivener too, and even Apple's 4 GB heavyweight XCode development environment.
The one thing that I left off the system was Photoshop.
That was half a year ago. So far I haven't really missed it.
Haven't been up on my soapbox in a while…
I have taught photography to over a thousand of students, among them many really good photographers who often weren't aware why they were great, but I have also been surprised at times as some of the more professional appearing ones weren't even able to do basic things like setting up custom white balance for a specific light situation.
There is a part of me that loves to see all the nifty photo gadgets that brilliant people come up with, but I've also been watching the developmentof the camera landscape with a concerned eye.
There are a lot of automated sub-systems in our cameras. Focus, exposure and white balance are the important ones among quite a few.
But the smarter these systems seem to get, the more decisions they take away from the photographer, the more the photographers lose the ability to make the right decisions.
I've seen this over and over again this year during the workshops.
It's not the photographers' fault of course. The philosophy of the camera manufacturers is quite understandable: take as many of the complicated photography stuff as possible and make the decision (and set the setting) for the photographer. This way many of the less technically inclined people out there can pick up a camera and quickly get results, which will make them happy, and as a result they will buy more cameras.
The big issue with this approach is that even though the automatic systems get it right most of the time, the camera will never be able to know the photographer's intention. How can the camera know that I'm not at all interested in exposing for the face, but instead I want to show a silhouette? How should the camera know that I actually want this shot to be bluish cool and unfriendly instead of giving it a caribbean sunset white balance? And how should the camera be able to anticipate that I deliberately want to blow out the sky in this picture?
The philosophy of me as the photography trainer is substantially different from that of the manufacturer: if you want to tell a story (and let's face it, a good story is usually what makes a good photograph), you need to make the tools that help you tell that story do the right things. The tool in this case is your camera. And making it do the right thing means to know how to make it expose, focus and white balance in exactly the way you want.
And that's a skill set that more and more photographers have either lost, or they never had the incentive to learn.
Relying on the automatisms of the camera and getting it right 80% of the time might be good enough for many photographers.
I want those remaining 20% to be under my control too.