Many months ago, I wrote a post similar to what this one is about my set up an workflow for image and screen capture turnaround. It's pretty much what I'll say in this post, but with many revisions and out of date info, but you can read it here should the feeling take you.
So, what do I actually do. I record live broadcasts, and create on-the-fly gifs of moments during those broadcasts. That could be a podcast, Formula 1 race, TV show, anything. If I'm watching it, I'm giffing it.
So, how do I do it.
To start, 2 monitors. Right screen is whatever I'm watching, left screen is for everything else. And I mean everything else. The way I make sure I miss nothing is by never having my right screen interrupted. So I make sure that all my programs are either open on my left screen, or will open on my left screen before I start recording.
Second, software. For capture, I use Mirillis Action!. It's a robust screen and game capture, record and streaming program, which can record in either MP4, for small file sizes, or in AVI for uncompressed capture. Both formats support up to 1080/60p, or can be set to match the resolution of your screen or game. I use AVI because I have the storage capacity to handle the large files; 15 minutes of 720/30p comes to around 4GB, which for broadcasts that can run up to 3 hours can become quite the issue if you don't have the storage.
What Action! also offers is single button hit screen shots. Getting the perfect frame to post can be tricky, especially if you have to try and preempt a moment for print screen. But, by mashing an assigned key(F10 in my case), Action! takes a screenshot with each press. That makes it quite easy to quickly react to whatever's happening, or just guess what's going to happen, and have a swathe of screenshots to choose from. And if I miss it, I can always export an image sequence from the recording to get the perfect frame.
Third piece of the puzzle, editing. For all my videographer work, I use Adobe Premier. Fantastic software, but the file load in process is too clunky and time consuming for the short turnaround time I need for livestreams. So for that I use Vegas Pro 14, by Magix(older versions were built by Sony before they sold it to Magix). I can drag and drop the capture files onto a timeline and instantly get to editing the moment I want. No sequence to set, no bins to fill, just video on a timeline.
And that typically looks like this:
In Vegas Pro, I'll find the moment I want, cut around it, do whatever other editing I need to do to it; pan and zoom, adding text or time remapping, and export it out.
So how does that video become a gif. There are multiple ways of doing it, but I'll explain the 2 I use.
1. Adobe Media Encoder: This is a file conversion program that comes bundled with Adobe Premier, and will convert almost any video file type to almost any other video or motion image file type. It has some very low quality Animated gif presets, but I made my own which balances image quality with file size.
2. A freeware video to gif converter: this is a very lightweight program that converts any video file to an animated gif. Rather than using presets, you can select the resolution of the gif, and the frame rate up to a max of 20fps. It converts that into an image sequence, which you can then click through and delete frames as necessary. There are 2 output options, based on matching the colour of the gif to the input video. Normal will give a lot of compression artifacts in the gif output and a small file size, and Excellent will give an accurate colour match but a larger file size. Luckily, Twitter supports gifs of up to 15mb, so there's a lot of scope for high quality gif output.
Because of its ease of use, and fast output, I use Option 2 the most. Once you're used to how the program UI works, going though the settings becomes almost like muscle memory. You can download that piece of software from cnet here
That's basically all of the how covered. So here's how to make the how work fast and efficiently.
- Know where everything is coming from and where everything is going.
I know my capture is set to desktop one, I know what folder on what hard drive I'm recording to, and have a file explorer window to it open. I know what screen my editor open on, I know what output preset I'm using, what semi descriptive file name I've given the clip, and I know what folder it's exporting to. I know the gif converter will open within that output folder, and I know what resolution I want the gif to output at(550x309 is the best balance of image quality and file size, primarily down to there being no large gif viewing support on Twitter, and the sites which specialise in large gifs take the files in as the video file, which is 1280x720). I use the same file name for the gif as I do the video, and it goes to the same output folder.
- Know your presets
This is specifically for reducing the time it takes to make the clip or gif. I've made an output preset in Vegas which is the same resolution and frame rate as my capture, so that there are no compression or resampling issues that will need me to reexport a clip. I know what gif size output I want, as explained above. This adds to the streamlining of the production pipeline between recording and tweeting a gif.
- Always be listening to the broadcast, even if you're not watching it
If you're focussed on editing a clip, always be paying attention to the audio of the broadcast, because you might miss seeing a moment, but you'll hear it happen, and hear the aftermath of it, so you know you need to break capture and find that moment.
I've omitted some of the more boring tech parts, primarily because they're different for pretty much everyone, and they're covered in the old post too. But that, in a nutshell, is how I do what I do. It's all about paying attention, spotting the moment, and having an optimised and streamlined way of getting that moment from a broadcast to a postable piece of media.
Ask any and all question either below in the comments, or directly to me on twitter @peterhayesf1.