I finally bought a circular polarizer for my camera. Instantly, it was one of those purchases that I wished I had made much earlier. When shooting around scenes where sky and/or water predominate, it is a must. It saturates skies and knocks down glare on water. You can actually see the magic happen right inside your viewfinder.

With a circular polarizer you rotate the outer ring. While doing this you see glares and reflections slowly vanish and the saturation of other objects begin to increase. If you continue to turn the filter glares creep back in. When these reflections begin to return you probably want to reverse the rotation slightly, back to the sweet spot, where the reflections are reduced as much as possible.

In this sample image I had forgotten to “dial it in” so I was seeing a lot of reflections. Then after the snap, I remembered to make the adjustment and dial down the reflected light. Since the camera was locked off it makes a nice before-and-after image that really demonstrates the power of the polarizer.

Hover your mouse over the image to see the polarized image. Then mouse off again to see the un-polarized, original image. The appearance of the water changes dramatically. Also you can see some increased saturation in the polarized imaged as well. What a difference.

For example’s sake: let’s assume you are involved with productions that are transmitted from a small TV studio. Further, let’s suppose that in addition to a live broadcast you also want to record your shows digitally, for archival purposes or to post to a website later. And finally, let’s assume that you want to do this at regular/predictable times each and every day.

If you are a Mac user in this position, you likely went in search for third-party applications. And they are certainly out there. However, if you have QuickTime Pro on your Mac, then it can be coupled with OS X’s other built-in tools to accomplish this task. Here is what you need: Automator, iCal, QuickTime Pro and maybe Applescript. Beyond that, as long as the source signal, coming into the Mac, is something that is recognizable by QuickTime Pro, then these programs already hold everything that you need to perform scheduled audio or video recordings. Let’s jump right in.

1.) Set up your QuickTime recording preferences. Launch QuickTime Pro, then go to Preferences. Of the tabs at the top, click on “Recording”. Set this up according to your own particular audio/video interfaces.

2.) Develop the Automator actions. You will need two—one to start the recording and another to stop the recording. When you launch Automator, it will looking something like the following image. (I am on 10.5.8, at the time of this post, so this might be slightly different for other versions of OS X.)

We are only concerned with Actions that are specific to QuickTime. To narrow the list of actions down simply type “quicktime” into the search field atop the Library area:

Now the middle pane, the list of actions, is only displaying actions that pertain directly to QuickTime Pro:

The first Automator action needs two instructions from the list: “New Video Capture” and “Start Capture”. (Of course, select “New Audio Capture”, if you intend to only record audio, and not video.) You drag these, one at a time, from the Actions list to the right-hand panel. This is where you build up your Automator workflow. The result should look like this:

Initiating a recording is as simple as this. Now to save our workflow. When you bring up the “Save As” dialog there are two options under the “File Format” area: Workflow or Application.

Here is the difference: With a Workflow, you double click it. The Workflow opens up in Automator. Then to actually perform the steps it contains, you click the “Run” button in the top-right corner. All the steps are executed. If, alternatively, you save your action set as an Application, then to run it, you simply double-click the Application. Applications run upon being launched, while Workflows require you to manually push the “Run” button.

In this case we want an Application, so save the Automator routine as such. For this example I am going to put mine on the Desktop. You will likely want a more permanent location.

That is it for starting a recording. We now need a “stop recording” function. Even easier—one step in the workflow. In Automator, go to the File menu, at the top, click and come down to “New”. (Keyboard Shortcut: ⌘+N) Limit the actions to QuickTime actions, like before, then locate the “Stop Capture” task and drag it to the right pane. Looks like this:

Again save this as an Application, in the same location as the “Start Recording” application. That’s it for Automator. Now for the scheduling…

3.) Schedule iCal to start/stop recording. When you set up a new iCal event, iCal allows you to set “alarms” for that event, that can take various forms. For example, an alarm could be: None, Message, Email, etc. But the alarm option you would want, in this particular instance, would be “Open File”.

Click the word “None”, to the right of alarm, to see the options for your alarm:

Immediately after setting the alarm to be “Open File”, a new line appears under Open File. It will say iCal.

Click the little up/down arrows to the right of the word iCal and then select the option “Other”. Then you simply navigate to wherever you stored your “Start Recording” application and select it.

Then finally, click on the word “15 minutes before” and change this to “on date”, from the bottom of the list. More details will appear allowing you to set precise times for the alarm to occur. (The “from” and “to” dates at the top of this same panel, will not matter. You designate the alarm time in the alarm area, you can ignore the times at the top.)

I usually do this with some headroom. ie, start early. Now you want to finish this off by setting up one more iCal event—this time, to stop the recording. Now the alarm would be to open the file that is your “Stop Recording” application. So navigate to it like we did before.

So that’s it. iCal will now launch your “Start Recording” app., triggering a QuickTime recording, whenever you told it to do so and will, later, launch the “Stop Recording” app. to halt the recording. And the recording parameters will be defined by the QuickTime recording preferences that you assigned in the first step of this guide.

We’re done. Unless, that is, you want to change the recorded file’s name to something more meaningful than the generic “Movie.mov” or “Audio.mov” that are the defaults. If that sounds helpful, read on:

*Bonus: I mentioned in the beginning that you would need, amongst other tools, Applescript to make this work. Yet, I have not mentioned it and we seem to be done. Well, with my particular needs I wanted a time-stamp for my recordings. Otherwise, QuickTime will save the file to wherever you specify, however, it is going to assign it a generic name: Movie.mov or Audio.mov (Depending on whether you have recorded a video signal or audio only.) Note that if you perform multiple recordings you will not have worry about QuickTime overwriting “yesterday’s” recording with today’s. The new movie would be called Movie 2.mov, Movie 3.mov, and so on. Obviously though, this is not the most helpful naming scheme always. So I set up an Applescript to append a date-stamp to the front of my recordings. Plus I have the Applescript add some additional descriptive text to the movie title as well. Here is how the date-via-applescript can work.

There are numerous ways that Applescript can format dates. I chose to go with the format of YYMMDD, and to put this at the beginning of the file name. That way, if you sort the folder of recordings on their names, you’re also sorting them chronologically at the same time. So the Applescript for this kind of formatting looks like this:

# Date
on zeroPad(theNumber)
set theString to “”
if theNumber is less than 10 then set theString to “0”
set theString to theString & theNumber
return theString
end zeroPad

set theDate to current date
set dayString to zeroPad((day of theDate) as integer)
set monthString to zeroPad((month of theDate) as integer)
set yearString to zeroPad((year of theDate) mod 100)

set theDateString to yearString & monthString & dayString

You can copy/paste this code into the Script Editor (which should be in a folder called “Applescript” in your Applications folder) push “Run” and the results box, at the bottom should display today’s date.

Now we need to append this bit to the front of our “Movie.mov” file, plus replace the “Movie” part with something more meaningful. I will show you a renaming script without the date portion, then we will sew everything together. Here’s the script.

#path to file
set myDesktop to (“Macintosh HD:Users:YOURUSERNAME:Desktop:”)

tell application “Finder”
if exists (files of folder myDesktop whose name contains “Audio.mov”) then
set name of (files of folder myDesktop whose name contains “Audio.mov”) to “_myRecording” & “.mov”
end if
end tell

You can copy/paste this script into Script Editor and run it, as well. However, there are critical assumptions that are being made that you must be aware of here. This scripts assumes:

A.) Your boot drive is named “Macintosh HD”. If not, rename that area of the script to the appropriate name.

B.) Where I have “YOURUSERNAME”, you will need to insert your user name. I’m making a call to a very specific path. To my Desktop. You need to alter the script for the specific path to the recorded movie, for your machine.

C.) Finally this assumes that you have a QuickTime movie, already on your Desktop, that is currently named “Audio.mov”.

If you set all those conditions properly, then run the script, you will find that the “Audio.mov” file on your Desktop is now called “_myRecording.mov”

Hopefully that all makes sense so far. If not, you might want to looking into Applescript resources because this guide is not meant to be a comprehensive discussion on Applescripting.

Now, finally to piece the two scripts together. Date formatting + File renaming =

# Date
on zeroPad(theNumber)
set theString to “”
if theNumber is less than 10 then set theString to “0”
set theString to theString & theNumber
return theString
end zeroPad

set theDate to current date
set dayString to zeroPad((day of theDate) as integer)
set monthString to zeroPad((month of theDate) as integer)
set yearString to zeroPad((year of theDate) mod 100)

set theDateString to yearString & monthString & dayString

# Rename the file that contains the keyword
set myDesktop to (“Macintosh HD:Users:YOURUSERNAME:Desktop:”)

tell application “Finder”
if exists (files of folder myDesktop whose name contains “Audio.mov”) then
set name of (files of folder myDesktop whose name contains “Audio.mov”) to theDateString & “_myRecording” & “.mov”
end if
end tell

Again, all the same assumptions are being made in this script as alluded to before. You can, and should, test this before trying to implement it into your Scheduled tasks. If there is a file on your Desktop name “Audio.mov” and you compile and run the above script, then you should now find that the file in question is named “100919_myRecording.mov” (of course with the appropriate date for whenever you actually ran the script.) If there are errors, my bet would be that there is something amiss in the file path that you specified. Check that first.

Now that we have a means of date-stamping via Applescript, how do we implement this into our previous processes? Again it’s Automator. Simply, Automator can run Applescripts.

To see it in action, return to Automator, then perform a search for the word “Run”. Of the few limited actions that reveal themselves, one will be “Run Applescript.”

Drag this action to the right pane. Notice the pre-built text in the script dialog box area. There is a section that says (* Your script goes here *). Maybe there is something in how I write my Applescripts, but I have not had luck with simply trying to paste my script into the prescribed area, and having it work properly. Instead, I have always deleted ALL the text in that box and then pasted my script into that area instead. Pre-filled script area:

Emptied of text:

My Applescript pasted in:

Now you can save the process, but like before be sure to save the file as an Application, not a Workflow. Now to add this final phase to our entire process you have options. From where we stand right now the easiest thing to do would be to schedule one final iCal alarm even that would call up this Application. You would schedule this a few seconds or minutes after the Stop Recording procedure has run. Now you would have three “Applications” called on by iCal: start recording, stop recording, rename movie.

But there is a slightly simpler, more consolidated way to do this. We can append the naming action, the “Run Applescript” task, to the end of the stop recording task that we already built, in Automator. Thus, we handle both tasks with a single iCal event. We would end up then, with two alarms: one to start, one to stop/name. Let’s do that now.

In Automator, go to “Open” and navigate to the Stop Recording application that you saved out before. Add the “Run Applescript” task after the “Stop Capture” task that is already in place. Paste your Applescript into the Run Applescript text area. Save this an application again, maybe as a new one this time, with a revised name.

One final thing. I’ve had issues when attempting to trigger an Applescript immediately after some other event that might be in the process of trying to finish up. To avoid this, add a “delay” to the very beginning of the Applescript. Simply saying “delay 15” tells an Applescript to wait for
15 seconds before beginning. This seems to cut down on launch errors. In this case I would certainly add a short delay, I will stick with a 15 second delay. Like this:

That’s it. It might sound complex, (especially the AppleScript bit) but you should be able to set this all up in only a few minutes. To recap:

1.) Set up QuickTime recording preferences.
2.) Build Start Capture and Stop Capture “Applications” in Automator.
3.) Schedule alarms in iCal. Sets alarm to open files: Applications built above.
*Extra) Employ Applescript to rename saved recording files with time-stamp and custom name. (This could be integrated into step #2, as part of the Stop Capture process. Don’t forget your delay if you go this route.)

The real final step: we have set up a single recording. The whole point of this entire exercise is to generate recordings that are regular and repeatable. There’s a final move in iCal that handles this. Particular events there will have an additional option. Repeat:

Click this area to expose the repeat parameters and set the alarm to repeat as you need:

I hope this helps you out with your scheduled recording needs!

 

Edited 8/5/15:

Many of you have mentioned errors with the Stop Capture action, try this workaround instead:

Solution :
Delete the “Stop Capture” action.
Add the “Run AppleScript” action, clear all text in the action.
Put this script in the action

on run {input, parameters}
tell application “QuickTime Player”
try
stop document 1 — to stop capture
end try
end tell
end run

I can’t take credit for this particular fix. I found this on the Apple Forums and hope it helps! Thanks to Jacques Rioux.

 

Mattes, masks, keys, alpha channels… They all roughly relate to the same concept: the transparency of an image. When you have a picture, be it a still image or moving (video/film), and you want to reveal part of the image but hide another part—you need a matte. (Or a mask. The terms matte and mask are often used interchangeably.)

In it’s most basic form a matte is an image that tells a second image where to be opaque and where to be transparent. Sometimes this opacity information comes in the form of an “alpha channel”. In this special instance, the transparency information is embedded inside the matte image in question. So, a RGB image, with an embedded alpha channel, actually has 4-channels, not just three: RGBA (Red, Green, Blue and Alpha). The alpha channel embedded in the one file can be used to dictate the transparency in another file. So let’s talk about mattes and alpha channels. Specifically, luma mattes and alpha mattes—what’s the difference? And why would you need to invert a matte?

Luma Mattes
When you assign a luma matte to your subject image, you are telling the subject to use the lightness values of the matte image as it’s source of transparency. Typically, the areas of the matte that are perfectly white will cause the subject to be completely opaque: 100% visible. Conversely, areas of the matte that are completely black force the opposite: those areas of the subject will disappear completely. Then, if areas of the matte are gray, there will be partial transparency in these areas of the subject.

So let’s look at some examples of the luma matte. Let’s suppose we have the (perhaps oddly shaped) luma matte image like this:

And we have a “subject image” that looks like this:

If we use the first image as a luma matte for the subject then the result would look like this:

Note: For all examples, the checkerboard pattern represents transparency: areas of the image that are completely invisible.

We see that where the matte is pure black, the subject is invisible. Where the matte is pure white, the subject is 100% visible. This simple demonstration conveys how luma mattes control opacity, given that the matte contains areas that are either purely black or white. Though, what if areas of the matte are not purely one or the other, but are instead partially black or white? Those gray areas of the matte will partially reveal the subject image. A luma matte of this kind may look like this:

A matte that looks like this, produces a subject image like this:

Areas that are partially black/white in the matte cause the same amount of partial transparency for the subject image. Moving on. Now, here is a common use for the luma matte—white text on a black background.

And the resulting image:

Building the composite up with additional layers is now relatively simple:

So hopefully the concept of how to deploy a luma matte is making good sense now. But there is one more thing we need to understand before moving on to the alpha matte. And that is “inverting” mattes. This topic is going to apply to alpha mattes as well, so I might as well discuss it now.

You might recall, from earlier, I said, “Typically, the areas of the matte that are perfectly white will cause the subject to be completely opaque.” Most software I have encountered assumes this to be the case. And that is the assumption that I have made up until now. However, some applications, (I have heard—I have never encountered one…) actually expect luma mattes that are the opposite. Instead of white representing fully opaque areas, black represents fully opaque areas. So if that were the case, and you fed the very first luma matte pictured at the top, to your subject, and your application was anticipating a black=opaque matte, then the resulting image would look like this:

What is now transparent or opaque is the opposite of what we really want to achieve. All compositing apps that allow you to designate a matte, at all, also allow you to invert that same matte. So now, making the assumption that our software was expecting black=opaque and we were given the white-on-black matte—we need to invert the matte. That way what was black becomes white and vice versa. The “proper” luma matte in this instance would be the exact inverse of what we started with and it would look like this:

So, sometimes black=opaque, sometime white=opaque. Either way, with luma mattes, it always the lightness, the range of blacks and whites in an image, that define the transparency of your subject image. It all depends on what your specific software expects, coupled with what the matte looks like; is it white on black? Or is it black on white? Regardless, all software has the means of inverting a matte. Swapping white for black and vice versa. In After Effects, for example, you can assign a Luma Matte or an Inverted Luma Matte. The After Effects panel:

Hopefully now, the topic of inverting the matte makes more sense. I hope so, because this same concept will apply to our upcoming conversation on alpha mattes.

Alpha Mattes
An alpha matte is just like a luma matte—it controls another image’s opacity. However, the difference is that instead of the subject using the mattes lightness values to drive it’s opacity, the subject uses the built-in alpha channel in the matte image. In this instance, it will not matter if the matte image is black and white or if it’s rainbow-colored. What will matter is that the matte image is carrying it’s own transparency information in an embedded alpha channel. It essentially passes this built-in opacity information, down the line, to the subject image.

Now let’s assume we had a different (wacky looking) matte image. This time we will use a star shape for the matte. (I hope you realize that after you get a grip on this article, with a little extrapolation on your own, you’ll be rolling your own star-wipes for your video. Always a crowd-pleaser!) Here is our alpha matte image:

Yep, this is a silly image to use for a matte. And this particular image might be misleading because of the black areas. But most software, out there, has some sort of default background color. In this circumstance, it just so happens to be black. But it’s artificial. If I force my software to show me transparency, our checkerboard pattern demonstrating what is really visible, then the image looks more like this:

When we now assign this star-on-transparent image to be an alpha matte, there is a different kind of transparency assignment made to our subject image. Instead of black or white=transparent, it’s transparency in the matte=transparency in the subject. Further, any sort of opaque areas of the matte yield equally opaque areas in our subject. Where the alpha matte is partially opaque, the subject will be partially opaque. The colors and patterns of the alpha matte do not matter. The built-in transparency is reigning over the opacity of the subject image:

When the textured star is assigned as an alpha matte, the subject then looks like so: (Checkerboard in the matte yields checkerboard in the subject.) The red color and texture are disregarded. The embedded alpha channel, in the matte, is what matters.

Again, it’s irrelevant that our alpha matte image has color and patterns. Further, if the matte is semi-transparent:

The resulting subject image will use the alpha matte’s partial transparency for it’s own transparency:

And finally, inverting the alpha matte. Sometimes, because of what your software expects or because of what you have been given, you’ll need the inverted version of the alpha matte. So, in some instances, an alpha matte like this:

(Again, usually, checkerboards—invisible areas in the matte—yield checkerboards in the subject. While, wherever pixels exist in the matte, image is revealed in the subject.) So this may then result in your subject image looking like the next one below. But if you’re seeing the exact part of the image that you, instead, wish to be invisible, then you need to invert the matte. Inverting simply flip-flops the transparent and opaque areas of the subject.

In this circumstance, if you wanted the to reveal the center balloon—not cut a hole in it— you would want the inverted version of the alpha matte.

So there you have it. Luma Mattes, Inverted Luma Mattes. Alpha Mattes and Inverted Alpha Mattes. This topic seems to come up quite a bit on the video editing forums. Folks want to use one image to “cut a hole” into another image. This is a 101 version of how to get the results you are seeking, using the fundamentals of how mattes work. All the applications I have used, for even the most basic compositing, work on these general principles. In, particular, with Photoshop, if you are painting onto an image mask, that is a luma matte at work. Sometimes you will use a clipping group, to define transparency, and this is a basic alpha matte scenario. Other software have their own means for allowing you to set a matte. Consult your particular software’s help manual if necessary. Here is a quick look at how a few different apps deal with transparency assignment, via their own respective matte controls:

This post is not meant to address all the in-and-outs of every software packages’ particular matte handling characteristics. This is meant to simply give a fundamental understanding of the simple luma matte, it’s alpha matte counterpart and both of their respective “inverted” modes. I hope this helps define what these things are and how you might get started using them.

Happy compositing!

There are two kinds of quote marks commonly used out there: Curly Quotes and Straight Quotes. If you push Shift + “that little key just left of the Return, or Enter key”, on your keyboard, you probably just made straight quotes. Regardless, here’s what straight quote marks look like:

Unfortunately, we typically use these incorrectly. We will usually put the above marks around text to illustrate that it was a statement. However, those instances, should actually be framed by curly quotes. To be proper.

But this is not a guide to being completely proper. This is a guide on forcing your quote marks to be straight, if that is indeed what you want. Best I can tell, WordPress, through the publishing process, does it’s part to put those proper curly quotes around your text. Although, you typed straight marks. That’s fine, that WordPress “dots your i’s and crosses your t’s” for you. I appreciate the accuracy. Though sometimes you specifically need straight quotes. For instance, if you are trying to share code via your site—you likely need straight quotes.

In a recent post, I wanted to share some Applescript for copying-and-pasting. The initial tests with my article, produced issues with my Applescript. Here were my steps: I copied the text from a working script into my article and previewed it, then I pasted the resulting WordPress-produced text, back into a new AppleScript. When I compiled it, I had errors. Upon inspection I realized that the quotes in my new AppleScript were, all curly. Because I had pasted them from my WordPress page. So, in this case I needed to force straight quotes into WordPress. That way, others can copy and paste my code, with no issues. To do this you need to substitute:

"

for all the straight quote marks. The way I did that was to use a simple text editor to perform a “Find and Replace” function. Any text editor worth it’s salt can quickly and easily perform this swapping task. You copy your working code, then paste it to the text editor. Tell it to find all occurrences of " and replace them with " This new version of your code is what you would paste back into your WordPress document. Then all the code-related quotes will be straight quotes, not curly quotes. Your visitors should be able copy and paste your code into their target apps with no problems now.

One of the great things about OS X is the ease with which you customize the interface to your liking. You can dictate the appearance of any given Finder window: icon view, list, view, column view or the iTunes-inspired “Cover Flow” view. You can then force a given appearance setting to become the default for a folder. Further, in icon view and list view you can set the size of your icons. Also, in icon view you can color the background of the window, or even add an image as the background. Too, you can easily swap icons for most any element in the OS. The list goes on…

However, sometimes you might want just a little bit more than what seems to be offered by the OS. Luckily, the results you are after can often be reached by firing up the Terminal and inputting a line of code. Terminal is the OS X command line interface that allows great flexibility and power with regard to altering your system. But with great power comes great responsibility. There is no hand-holding in Terminal. For example, if you are on the brink of executing a code that will wipe your hard drive clean, you will not be prompted to check if you are absolutely certain about this move. Push return, and the command is executed. When going into the Terminal, it is best to be cautious. But the Terminal codes I will discuss here are quite benign. They will not lead you into accidentally erasing your daughter’s wedding photos.

Scroll Bars: Double Arrows on Both Ends
When I set up a new Mac—as I just helped someone do—there are three terminal codes I typically use. The first one involves altering the scroll bar characteristics for all windows. The conventional way to do this is to go into the System Preferences and click on “Appearance” in the upper left hand corner:

The Appearance pane has an area that addresses scroll bar behavior. The standard options listed here are either to Place Scroll Arrows: Together or At top and bottom. The pane in question looks like this:

Given these options, your scroll bar arrows—in every application and window—will look like one of these examples:

That’s great. But what if you wanted the best of both worlds? What if you wanted to have double scroll bar arrows, not just at the bottom, but at the top as well? A little Terminal code can make this happen. (The Terminal is in the Utilities folder, which is inside the Applications folder.) Fire up the Terminal and type this in:

defaults write “Apple Global Domain” AppleScrollBarVariant DoubleBoth

Push return to execute the command. Nothing will appear to have happened. In order to make the change take effect you either need to reboot or simply log out and log back in. Once you have done that your scroll bar arrows should now look like this:

If for some reason you want to revert back to one of the standard configurations, simply open the System Preferences and make the change. It will take effect immediately. No need to log out/in. I have used this code on all my Macs for years now. In fact, it is odd for me to use someone else’s Mac and not have this option enabled.

Keep Widgets visible on the Desktop
The second Terminal code that I run on my Macs allows the ability to keep a Widget persistently on the Desktop. Frankly, I do not run a ton of widgets and typically when I do the process is: call up the widget, use it, hide all widgets again. However, every once in a while I find the need to have a Widget remain in front of me, on the Desktop. Sometimes it is a count down timer, or maybe a timecode calculator. To accomplish this you need to put your Widgets into “Development Mode”. This mode keeps Widgets visible for the sake of developers who are building/tweaking them. But you might find it handy as well. Here is the Terminal Code:

defaults write com.apple.dashboard devmode YES && killall Dock

This is actually two commands in one, linked by the “&&”. The first part invokes the Development mode. However, as with many of these types of changes, you need to take additional steps to force your tweak to take effect. In this case, you need to restart the Dock. So, the final part of the code, after the &&, handles this. It restarts the Dock.

Then, this is the simple procedure to keep your desired widget up front. Call up your widgets. (On most Macs you bring widgets up, and also hide them, with the same keyboard shortcut, F4 or F12.) Click and drag on the widget that you want to remain on your Desktop. While dragging, hide your widgets again. The widget you were dragging should remain visible on the Desktop while all others retreat to their hiding spots.

To return the widget to a hiding mode, reverse the process: while dragging it, call up your widgets. Once they have appeared, release the widget. Now hide your widgets. The previously persistent widget should now hide itself, alongside all the others.

Should you want to revert to the default behavior and get out of development mode you would then change the YES to NO in the code. Like this:

defaults write com.apple.dashboard devmode NO && killall Dock

Create Recent Applications Stack in the Dock
I just found this handy Terminal code that allows you to create a “Stack” of Recent Applications in the Dock. Stacks are Dock items that offer quick access to files and folders. (More information on Stacks.) You can call up a list of Recent Applications, and other recent items, by going to the little Apple symbol in the uppermost left corner of your interface. Click the Apple and one of the options listed there is, Recent Items. By the way, preferences for how many items are listed in each group here, can be adjusted in the same Appearance panel, in System Preferences mentioned at the top of this article.

The Recent Applications list looks (something) like this. I have mine set up to show 20 recent apps:

The problem for me is that I only remember to use this list about once a month. Realizing this, I developed an alternative technique. I would make a folder full of aliases to applications that I use all the time. I put this folder of aliases in the Dock. That way I could quickly use the Dock to access the apps I was using most. However, I found two things that kept this from being ideal:

1.) I would put an application’s alias into this folder because I was using a respective app heavily. But heavy use of some programs, for me at least, often wanes and I would end up with several aliases to applications that I had not used in 3 months. So what’s the point of having it so handy?

2.) This might sound silly. But I had developed the habit of calling up apps using Spotlight. In Leopard, Spotlight got so snappy, that by the time you type p-h-o, it was pointing to Photoshop. So I began launching most apps from Spotlight. Maybe this is strange. But the speed involved made it practical.

Then I found a Terminal code that puts a Stack of Recent Applications in the Dock. This is my standard set-up now. The code looks like this:

defaults write com.apple.dock persistent-others -array-add ‘{ “tile-data” = { “list-type” = 1; }; “tile-type” = “recents-tile”; }’

Then you need to restart the Dock:

killall Dock

So the programs I am using most often are now readily accessible via the Dock. And if for some reason the Application I need, at any given moment, happens to not appear in my list of 20, then I just revert to the Spotlight search technique. But my Dock Stack looks like this:

There are additional codes you can employ to remove these special Stacks from the Dock. The downside is that they tend to remove all Stacks from the Dock. So I simply, click and drag the Stack out of the Dock—should I decide to remove it.

It’s worth noting that once this Stack is in place in the Dock it can be changed to show you the other “Recent Items” types: Documents, Servers, Volumes, Favorites. Right-click the Stack to reveal the other options:

Search the web for more Terminal codes to customize your Mac. There are many out there that will help you you perform “hidden” tricks. Feel free to share any Terminal codes, you use regularly for OS customization, in the comment area below.

In effort to help out with some troubleshooting over at Apple’s Final Cut Pro Discussions I wrote a few posts about the “order of operations” of effects. The goal was to demonstrate that by adding effects in a certain order, it would be possible to color correct keyed footage, without disrupting the key. I wanted to show how stacking your effects up into the proper hierarchy can deliver the end result you’re after. After writing the first post I realized it needed modification. The follow-up post, likewise, had some anomalies that prevented it from being 100% accurate too. So here’s the clean slate, from the top: How to Color Correct keyed footage in Final Cut Pro.

1.) I’m going to start with a brand new Final Cut Project. There will be a single sequence already in the browser. Immediately I’ll rename this sequence to “Main Key Sequence”.

2.) Gather assets. We need two items: footage to key and a background plate to put it on. For the clip to key, I used footage from a Chris Meyer tutorial on Keylight.

Then we need something to put our footage on. I made a quick gradient in Photoshop:

3.) Back in FCP, make a new Bin in the Browser for the incoming media and give it a meaningful name. Right-click the bin and then hover over “Import”, then click on file. Point the navigator to your file(s). Both elements should end up in this folder.

4.) Edit the background image onto V1 at the very beginning of the timeline. Then edit your footage to key into V2, also putting it’s In-point at the beginning of the sequence. Trim off any excess footage at the tail of the background clip. Your timeline should now look like this:

5.) Pull the key. Drop the “Chroma Keyer” effect onto the green screen footage on V2. Adjust it to your liking. Sine this guide is not meant to be an exhaustive guide to keying, I only played with the key settings for a few seconds. This key is, by no means perfect, but is adequate to move on. My result so far:

6.) Nest the green screen shot. To maintain our current look we need to nest the footage. Click (only) on the green screen shot on V2. Make sure the clip on V1, beneath, is not selected. Then either push option+c or go to the Sequence menu and come down to “Nest Item(s)”.

This action is going to create a new sequence and put the selected clip inside that new sequence. A dialog box is going to present itself. This is your chance to name the new sequence. I chose to name mine, “Man_ongreen KEYED”.

Now the original sequence, “Main Key Sequence” contains the newly created sequence, not just the clip. You’ve got a Sequence-inside-a-Sequence. The project now has one clip, one graphic plate and two sequences. But we’ll continue to work in the Main Key Sequence.

7.) One frame of the footage has some green spill on the subject’s body. This is the spot we want to try to color correct. So drop the 3-way Color Corrector onto the clip on V2 in the Main sequence. Here’s the spill I want to eliminate:

8.) Limit the color. We really only want to affect those greenish areas on the man’s back. In the Viewer, click on the tab for the 3-way Color Corrector to reveal it’s controls. In the very bottom of this panel you’ll see the words “Limit Effect”. Then off to the far left of this there will be a small disclosure triangle. Click this to turn it down and reveal the controls for limiting the color corrector.

On the right of this extended panel are three small icons: an eyedropper, a key, and rhombus shape.

Ensure the Canvas is zoomed in close enough to plainly see all the areas of green that need clean-up. Click that little eyedropper in the Limit controls then click on the offensive areas on your target. I tried to click right on the greenest spot I could find on the subject’s back.

Boom! What happened? Did your background stop showing through? Mine did. Instead of now seeing my red/orange gradient, I only see my subject on black instead. Well this is the “gotcha”! We’ve just broken our order of operations. But let’s not fix it yet. Let’s continue to repair this green spill.

We need to make our selected area larger. When we clicked with the eyedropper we sampled a single spot. We can return to the image and select more green spill areas and add these to our current selection.

Click the eyedropper again. Hover over the green areas and before you click hold down the Shift key. By holding Shift, you’re adding to the prior selection. So hold Shift, click and drag around other green areas to include them in this area that we hope to fix.

There are a couple of ways to check your work thus far. After that initial click, the rainbow-colored bar should be changed to show you the color you selected. This might be a tiny sliver of color. As you add to your selection, this colored region should grow, reflecting the broader range of colors that you are adding. My initial range looked like the image on the left. Subsequent clicks opened it up to what you see on the right.

The other way to review your work is to view the mask you’re building. I mentioned those three icons on the right of the limit area controls. The one in the middle is the “View final/matte/source” toggle button. If you click it you’re toggling between states: Final footage, the matte only, or the original footage state. By toggling to the matte you see white areas that represent your selection. So far, my selection’s mask look like this:

Use these two methods to review your selection and clean it up as much as you’d like. Once you know you’re only affecting the offending areas you can apply whatever color corrections are needed. Again, this is not meant to be a comprehensive course in color correction. This guide is to help you get to the point where it’s even possible. But I’ll say that for my particular clip, I desaturated it some and moved the midtones color picker away from green to offset the green spill. This is as far as I took mine. Spill suppressed on the right, original, still green shot on the left:

10.) Where’s my background? So what happened to that background plate? Now our subject seems to be married to a black background. Well again, it’s that order-of-operations. When you use the limiter area of the 3-way Color Corrector it seems to break the render chain. How do we fix it?

In our main comp, select the bottom clip, the background plate on V1. Right-click it and select “Cut”. While we are here we can drop the color-corrected clip down onto V1. (If you hold Shift while dragging it downwards, it constrains it in time: it won’t move left or right.)

Now double click that clip. Since it’s really a nested sequence, that “other” sequence – where we keyed the footage to begin with – should be visible with your keyed shot on V1. Drag this clip upward to V2. Put the playhead at the beginning and paste your clip from the clipboard onto V1. This should be the background clip that you just cut from the Main sequence.

That should be it. If you return to your Main sequence you should have your keyed clip, now with limited color corrections, sitting atop your background plate.

I recently wrote a post discussing Final Cut Pro’s order of operations in regards to effects. To quickly review, the prior scenario was this: A user on the Final Cut Pro discussions forum, at apple.com, wanted to color correct a clip that also needed to be keyed. So the question was, key first? Or color correct first? After some quick testing I arrived at the conclusion that by stacking your effects up, in the proper order, one could in fact reach the desired result. If you pulled the key first, then applied the 3-way Color Corrector, the latter effect would consider the result of the first effect (because of their “stacking order”) and you could, theoretically, freely color correct the footage after pulling the key. My initial tests supported this. However, the original poster on the discussion forum found a problem with this workflow. This individual, “Toolius” started a follow up thread that said this:

I can corroborate this behavior. When I use the same JPEG image as before I can completely replicate the issue that the poster was seeing. “Limiting” the effect of the 3-way Color Corrector does seem to completely override the Chroma Key, even though the 3-way CC was being applied after the Keyer. Odd. But that’s how it goes. So I wanted to try to find a fix or a workaround.

I encountered some anomalies with the prior JPEG image so I switched media and used a “movie” for this second round of testing. The clip is one that I downloaded from a Chris Meyer’s Keylight tutorial.

Here’s what I did:

1.) In Final Cut Pro, start a new project. Rename the single, default sequence to something meaningful like”Main Key Sequence”.

2.) Import the footage item, “Man_ongreen.mov” into the project and insert it into my only sequence, the aforementioned, “Main Key Sequence.”

3.) Apply the Chroma Keyer to the clip. Adjust settings. The cleanliness of the key isn’t exactly crucial to the proof of concept here so I didn’t fiddle with it for too long. The resulting, keyed image, looks like this:

4.) Click the keyed clip once, to select it (if not already selected) and then “Nest” it. So either push option+c or go to the Sequence menu and come down to “Nest Item(s)”.

This action is going to create a new sequence and put the selected clip inside the new sequence. A dialog box is going to present itself. This is your chance to name the new sequence. I chose to name mine, “Man_ongreen KEYED”.

Now the original sequence, “Main Key Sequence” contains the newly created sequence, not just the clip. The project now has one clip and two sequences. We’ll continue to work in the Main sequence.

5.) Apply the 3-way Color Corrector to the media in the Main sequence.

6.) My test clip had some subtle green reflections on the subject’s back. This sounded similar to the forum poster’s issue so I attempted to “limit” the 3-way Color Corrector to this green area.

7.) Before, when I got to this stage, the single click in the image (with the Color Picker) to select the colors to limit, immediately revealed the green background. However, the act of nesting the clip, first, seems to “fix” things. Limiting the 3-way Color Correctors range did not bring back my previously keyed material. And the matte for the area that I’m trying to affect looks like this:

Looks promising!

8.) Further adjustments made to remove the green spill do not reveal the green background. On the left you will see the green spill, evident on the subject’s back. On the right we have suppressed the green spill somewhat. But the good news is: my green screen never reappeared.

Hopefully this workaround, with nesting, will fix the forum poster’s problems. It seems to work for me. I’ll post back the findings from the Apple forums once I know the original poster’s results. There definitely was a breakdown in the normal “order of operations” that nesting may fix.

OOPS!…I’ve noticed an error with this technique. I was never using a background plate and all my work was on FCP V1 track. So I was not able to see, and did not investigate deep enough to uncover, the fact that limiting a Color Correction will always force the respective clip to disregard it’s alpha channel.
Issue: Limit a color correction with a Track on V2 and V1 will no longer show through.
Fix: In the main comp, CUT the background plate, open the sub-sequence where the keyed footage is. Raise the keyed clip to V2, paste the background plate to V1.

When you return to the Main Sequence everything should look as expected. I think it’s time to write the 3rd chapter to this series and put to bed. It should’ve like this up front: “How to Color Correct keyed footage in Final Cut Pro”.

One of the more remarkable features of Mac OS X is the search function, referred to as Spotlight. It’s the little magnifying glass icon in the uppermost right-hand corner of the screen. You click there to begin searching for things: anything. This search is powerful: it will even index and recall specific text inside documents. You know there’s a PDF invoice, somewhere, on the system, let’s say, but all you know is the dollar amount on the bottom line. Put those digits into Spotlight and chances are, it’s going to turn up your hiding document.

Further, all items in the Finder can have “Spotlight Comments” added to them: files, folders, apps, aliases. And of course, when it comes time to search, one thing that Spotlight goes to work, looking for, are these Spotlight Comments. A simple way to add Spotlight comments to an item would be to right-click (control-click) it and select “Get Info”. A panel will fly out. There are many parameters that can be reviewed there and there are small gray disclosure triangles that allow you to turn up/down particular groups of parameters, if you’d like. The uppermost region is important. This is where you can enter your own Spotlight Comments. This makes finding your file/folder much easier later on.

However, what if you have 1,156 new pictures from a recent trip and all of them could reasonably take on the same set of Spotlight comments. You probably would not want to add the comments one-at-a-time. I sure wouldn’t! And if that were the only way to do it, it would be a deal-breaker for most folks. Luckily though, you have them option to add Spotlight Comments to a group of items, all at once, using the Automator application.

When you launch Automator it starts with an empty “workflow” that looks something like this:

There are a few ways to get things going: If you know of particular phrases that are part of the Action that you want to employ, you could start typing them in the area where it says, “Name.” Alternatively, you could immediate thin things out a bit if you know that the Action in question will be used for “Fonts” or “Mail” or “Files and Folders”, for example. The Automator Workflow that we’re going to build applies specifically to “Files and Folders” so if we click that the list of Actions will be thinned down to reveal acts that are only applicable to Files and Folders:

At this point, it’s probably worth mentioning, that you could use both search techniques mentioned above: you could first click on the proper category of Actions; then begin typing more information into the “Name” area.

The two actions we’re looking for should now be visible in your Actions list. They are “Get Selected Finder Items” and “Set Spotlight Comments for Finder Items”. You can click the first one of these, hold ⌘ (the Command or Apple key), and then click on the other Action.

With both of them highlighted, drag them both to the right. Your Automator panel hopefully now looks like this:

The “Get Selected Finder Items” action should be on top. If, for some reason, it’s not – simply drag it above the other Action and you’ll be ready to save. Go to File > Save as and your options will look like this:

The File Format needs to be “Workflow” so change this if need be. I called my workflow, “Add Spotlight Comments.” The save location is very particular. Save this workflow into the Folder:
Users : your_username : Library : Workflows : Applications : Finder :
(If that complete path doesn’t exist, use the Finder to manually create it.) Save your workflow here.

You’re done. All you have to do is call the Workflow up. Go the Finder and right-click on a file or folder. At the bottom of the resulting fly-out panel you should see an option that says “More”. Hover over that and you will see “Automator”. Hover over that and you should “Add Spotlight Comments” – if you named your workflow as I named mine – otherwise, the workflow will be called whatever you named it. Click it.

This final dialog box, that flies out, will be where you type your comments. If you wan to add onto any existing comments, leave the box checked for: “Append to existing comments”. If you want to overwrite any existing comments then un-check this box. That’s it.

If you now go to the Finder and invoke the “Info. Panel” you will see that the Spotlight Comments area is now filled with whatever comments you just added. Further, if you do a Spotlight search for any of the respective comments you should see the appropriate files show up pretty high on the “found” list.

The greatest thing about this is that process is that it applies to a multi-file selection. So now if you go to that folder full of recent pictures, select them ALL, right-click – Add Spotlight Comments – and wait. In only a few seconds, all 1,156 of your pictures will inherit your custom Spotlight comments. And they will found so much quicker, in 4 years, when you can’t remember what you called them or where you put them.

I hope you find this Mac tip handy!

Currently, Final Cut Pro is my NLE of choice. (For what it’s worth, I’m also proficient with Media 100.) I try to stay active on the Final Cut Pro forums on Apple’s website. It’s a great place to go for help, hand out help and check in on the basic “state of the union” regarding Final Cut and post-production.

Today an interesting scenario came up. A forum user, ‘Toolius’ was asking:

My knee-jerk response was: Yes. Chroma Key first, Color Correction second. However, I figured that I should test my theory before posting my suggestions. And by the way, the idea that this would work, as I was hoping to demonstrate, was quickly contested by another user on the forum.

So whose idea will prevail?

By doing a Google Image search for “chroma key”, I quickly found this image on dvinfo.net:

I imported this JPEG into Final Cut and applied the “Chroma Keyer” effect. For proof-of-concept I did not require a very clean key, so I only played with the key settings for a few seconds. I did no garbage matting, at all. The rough key looked like this:

Then I applied the 3-way Color Corrector, knowing one of two things would happen here:

• This color correction effect would consider everything, and if I pushed the mids towards green, the previously keyed material would start to reveal itself – ruining my work. Or…

• The color corrector would act upon the image AFTER considering the Chroma Keyer (since the Keyer was applied first in the “stack.”) and therefore color correction should not have an effect on my key.

So what happens? What is the “order of operations”? As I was expecting, the 3-way Color Corrector “looks at” the source file AFTER the prior effects are considered. So, in essence, color correcting after keying does not seem to harm or contaminate the key. The keyed material, with it’s midpoint pushed towards extreme green, look like this: (Not revealing the keyed material.)

This all makes sense if you view the Alpha channel for the clip in question:

The white regions represent fully opaque areas, while the black regions show areas that will be completely invisible. This demonstrates that the white/fully visible areas are the only spots that will be effected by the next-effect-in-the-stack. In this case that happens to be the 3-way Color Corrector.

So next time this comes up: Key first? Color Correct first? You can proceed with keying your material first, then dropping on the 3-way Color Corrector. Tweak the colors to your heart’s desire – knowing you’re not going to override all that hard work you put into pulling your subject off his/her background.

The order of operations makes a difference!

Typefaces are a major component of any designers’ toolkit. Be it web design, letterheads, logos, motion graphics…the list goes on. Fonts play a key role in any design where the written word is employed.

There are countless third-party applications out there for font management on the Mac. However, I typically try to use OS X’s built in tools, to their full extent, to handle as much as possible. The only application I use to manage typefaces is Font Book.

There’s no built-in magic with this application: you only get as much out of it, as you put in. However, if you put in a little work up front, then the next time you have to go hunting for that perfect font, you won’t have to start at the top of your list and go through them all one-at-a-time. Consider putting your fonts into “Collections”. This should speed up your font selection process. And there’s a bonus: these Font Collections are portable! Moving them from machine to machine is a snap.

The tedious part comes right up front – classifying all of your fonts. The more fonts you have, the longer this will take. Though once you’ve done this once then managing newly added fonts is quick and easy. The largest hurdle is sitting through that first round of sorting. By default, there are a few font collections built into Font Book. These are: Fixed Width, Fun, Modern, PDF, Traditional and Web. Seen here on the left panel:

To add your own your own Font Collection you simply click the plus symbol in the lower left corner of the Font Book and give your new collection a meaningful name. You can have as many of these collections as you like and can define them as you see fit. You may want to go with some of the “classic” type classifications like Old Style, Transitional, Modern, Slab Serif and San Serif. It’s likely though that you would want more classifications than this. So add in Blackletter, Oldstyle, Italic, Script, Grunge, etc. if you’d like. This is all up to you. Define the classifications however you would like. Then, in Font Book, click on “All Fonts” in the upper left hand corner. The next column to the right will then be displaying, you guessed it, all the fonts on the computer in question. You simply drag a font from this center column, to the left, onto the appropriate Font Collection. Then repeat. (Personally, I like to precede the Collections’ titles with ‘My Fonts – ” then Scripty or Grunge or whatever title is appropriate for the group. That way when I see “My Fonts – ” I know it’s one of my own personal classifications.) It might look something like this:

Again, this process of sorting all of your fonts into classifications might take a while, depending on how many fonts you have on your system. But the good news is that you can put a single font into as many collections as you like. Which can be very handy. Let’s assume you’ve got a font like Geo Sans Light:

This font would obviously be classified as a “Sans Serif”. However, couldn’t we also call this particular specimen “thin” or “light” too? So you could make a Font Collection called “Thin” and one called “Sans Serif” and then put this font into both Collections. There is no penalty for putting any one font into as many collections as you might like. Especially when many fonts are clearly candidates for multiple classifications like a Geo Sans Light.

This technique should help you save time when you need to fish out the ideal font for your project(s). For example, if you know you want a “Scripty” font, then why not look (only) at your 20 Scripty fonts instead of your entire catalog of 4,000 fonts? I know Font Collections have certainly helped me to find the right font, many times, without having to filter through every font to get there.

Now for the bonus: moving Font Collections from one system to another. I’m a freelancer that typically works from my home studio. Though sometimes, I work at a colleague’s studio where we have multiple Macs. When moving complex projects (After Effects, Photoshop, Final Cut Pro, etc.) from one machine to the next, there are many considerations. Among other things, both Macs need the same plug-ins, codecs and fonts. We’ve always been quite diligent about maintaining the exact same fonts on all the production machines. This eliminates the “gotchas” when a project is transplanted from one system to the next. Font Collections are one more item that can be moved from Mac to Mac.

Font Book stores your Font Collections in a specific place on your hard drive. The file path is:

User > Library > FontCollections

In that folder you will find all the Apple default collections as well as any custom-built collections you’ve created. (Again, this is why I precede my personal collections’ name with “My Fonts – “. That way when I dive into the font collections folder on my drive, I can immediately determine which collections were built by me.)

To move Font Collections around, simply copy the collections from the FontCollections folder on the first Mac, then move them to the same folder (User > Library > FontCollections) on the second Mac. When you launch Font Book the custom collections should now be visible. And they will hold the same sets of fonts as the original machine – again, as long as both machines had the exact same fonts to begin with.

Note: If for some reason Font Book doesn’t immediately display your new collections you may need to delete the preference file for Font Book. The file is named com.apple.FontBook.plist and located in the folder: User > Library > Preferences. For safety, you may want to simply move this file somewhere else, like the Desktop, instead of deleting it outright. Then once things are OK, delete it.

Hopefully this will help you to better organize typefaces on your Mac. And the next time you need to select the ultimate font for your design this tactic should speed the process. If you can plow through this process once, the work’s 99% done. You’re not likely adding 100’s of fonts per day. If so, wow! But for most folks, adding newly acquired fonts to the appropriate Font Collections should only take a matter of seconds.

 

[UPDATE: 2016-12-06] As of Mac OS X 10.7 the Users’ Library folder is hidden. It’s still there, just not visible. To reveal it, in more recent versions of the OS, click on the “Go” menu in the Finder. Then push the option key. The Library folder should reveal itself. Then you can proceed as outlined. Hope that helps.