Michael Webber

Category: Television

  • Trim video without re-encoding using ffmpeg

    I’ve been taking the time to learn ffmpeg and some of the really useful stuff you can do with this command line tool (CLI).

    Here’s one example of a situation I find myself in frequently and have not had a great solution until now.

    The ability to trim a video file without re-encoding!

    “Re-encoding” means loss of video quality in my mind. Not quite as bad as copying a VHS tape, known in the analog world as a “generation loss” Having the ability to trim a video file WITHOUT suffering any quality loss, that was a WOW ffmpeg moment for me.

    This is an issue that comes up frequently in community media and I’m sure all video professionals have delt with this issue.

    You have some raw footage you wan to keep for the future, but the camera operator forgot to stop recording and the file has a bunch of “junk” at the end.

    Or you digitized some analog tapes and during the process you walk away only to come back to “snow” or “junk” at the end of the tape that you have now just captured.

    Or you have a video file with the standard leader of color bars and countdown at the beginning. You want to trim the footage, but don’t want to lose video quality in the process…

    So you either keep all the extra “junk” taking up space on your drive or you open up your editing software and edit out the junk and take the hit and re-encode.

    Right now, I’m cleaning up a 70TB Synology NAS that is nearly full. I found several large files that fit the above examples.

    Using ffmpeg and a rather simple command line, I was able to clean up some of these files and free up some space.

    Here’s an example of the command line I used:

    ffmpeg -i BestVideoEver.mov -ss 00:37 -t 51:14 -c:a copy -c:v copy BestVideoEver-trimmed.mov

    To break this string down just a bit, here’s what’s happening

    -i = Input Video
    -ss 00:00 = In-Point (everything before will be removed)
    -t 00:00 = Out-Point (everything after will be removed)
    -c:a copy = Copy Audio (Copy NOT re-encode)
    -c:v copy = Copy Video (Copy NOT re-encode)

    Finally “BestVideoEver-trimmed.mov” is the ffmpeg output file, which you can set to a specific destination if you want, for example:

    ffmpeg -i BestVideoEver.mov -ss 00:37 -t 51:14 -c:a copy -c:v copy /users/michaelwebber/Desktop/BestVideoEver-trimmed.mov

    Anyway, that’s it! Something I learned how to do recently and wanted to share just in case others find it useful.

    If you want to learn MUCH more about ffmpeg I recommend reading this blog post. Covers a whole bunch of topics including how to install ffmpeg and many of the cool features. Never knew you could edit video and even add text graphics with ffmpeg. I really recommend this blog post at img.ly, an in-depth, well-organized post.

  • High School Intern Returns

    Such a cool moment to have a former High School volunteer and intern stop in and do a TV hit from our studio because his flight was canceled.

  • Zoom TV Interviews

    Feeding our Blackmagic ATEM Switcher with Zoom Participants

    Four computers running Zoom feeding Blackmagic ATEM

    A few weeks ago I had a blog post on my Zoom and Stream Deck configuration to support TV interviews. The setup continues to evolve and I wanted to take a moment to share an update.

    The biggest problem we have with Zoom is the Audio-Follow-Video auto-switching within Zoom. We prefer to see people full screen and don’t want to depend on the gallery view, in fact I try to avoid it.

    Using the spotlight or pin tools in Zoom just doesn’t work fast enough for our purposes. We want to cut between participants just like we would in a traditional in-studio interview show. This includes reaction shots and cutaways.

    Multiple Zoom Calls Into Switcher

    The new setup we’re experimenting with uses multiple PCs, all with separate Zoom connections, each set to full screen with a single participant “pinned”. The output of each PC feeds the Blackmagic ATEM via HDMI as a source.

    I was a little concerned about lip-sync or slight delays between PC connections to Zoom. Thankfully that’s been a non-issue, everything syncs!

    Primary Audio

    To be clear, we’re only pulling audio from ONE of the PCs (input #1) with the gallery view display. All other switcher inputs are video only.

    Once setup and configures, which does take a few minutes to establish all the Zoom calls, it works great! We can now switch between headshots, in the same way we would if guests were in-studio. I can bring in lower third graphics without fear the Zoom auto-switching and don’t need to chase pinning and spotlighting people.

    The only downside is the number of PCs needed… To have a true single shot of everyone you need one more PC than you have participants. (one is always dedicated to the gallery or multiview.)

    So in this case I have FOUR workstations running Zoom for a three-person show.

    Couple of Notes & Tips

    • “Hide self-view” on the PC displaying the gallery view.
    • All other Zoom workstations should have their mic and cameras OFF as they are for view only purposes of feeding the switcher.
    • All Zoom workstations must be set up in a dual-monitor setup, with the pinned zoom person on the second monitor. Then you want to open up the chatbox and place it on the first (primary monitor) THIS IS CRITICAL! Placing the chat box open on the primary monitor avoids pop-up alerts on your full screen pinned video. (Hopefully, that makes sense)
    • You also want to “Hide all non-video participants” on the main gallery view feed so you don’t see all the extra zoom caller PCs feeding the switcher.

    Happy to answer any questions and would love to hear anyone with suggestions on how I might improve things.

    What a mess! Our experimental COVID-19 rig is constantly changing. The lessons learned from this will no doubt alter our planned studio and control room upgrades scheduled for this year.
  • Production Switcher to Zoom

    Production Switcher to Zoom

    Blackmagic Web Presenter, the perfect tool for the job.

    After several weeks on backorder, our Blackmagic Web Presenter arrived from BH Photo!

    The “black box” is a simple device that works out of the box. No drives, no software. Simply take the output of your video production switcher, professional SDI camera or HDMI source into the BM Web Presenter. Then connect a USB cable to your PC/Mac and the audio/video signal appears as a 720p HD webcam to the computer.

    We had ours installed and working with Zoom within minutes.

    Program Return for Zoom

    On a daily basis, we’re producing live interview programs and government meetings using Zoom. We need the guests and participants on Zoom to be able to see our in-studio hosts, graphics, roll-ins, and graphics.

    Feeding program video from our production switcher through the Blackmagic Web Presenter into Zoom gets this done.

    Avoiding Video Feedback on Zoom

    If you are not careful you can easily create a video feedback situation in Zoom. Once you have the production switcher feed working with Zoom I suggest clicking the three dot menu and select “Hide Self View” to prevent those issues.

  • Stream Deck for Zoom Meetings

    Stream Deck for Zoom Meetings

    UPDATE: My setup for Zoom has evolved and somewhat less dependent on StreamDeck now. You may want to check out my new Blog Post with our updated Zoom setup for TV interviews.

    This week I purchased a Stream Deck XL ($235) to help streamline our video productions with Zoom remote guests.

    Since the COVID-19 crisis hit mid-March our community television station scrambled to adapt and help disseminate important information. Zoom quickly became the go-to platform for setting up meetings and remote show guests.

    In the past few weeks, we’ve produced more than a dozen programs using Zoom. Initially, we had an in-studio host and multiple remote guests displayed on an in-studio monitor. Now, most of our interview shows a produced entirely on Zoom.

    Despite the limitations, we are making every effort to maintain high production values, focusing on:

    • Good Clean Audio
    • Well Composed Shots (eyes on the upper third!)
    • Shot sequencing, Close Ups / Multi-Box Switching
    • Graphics, Lower Thirds, Etc.
    • Open/Close Music
    • Still Store for Inserts

    Our existing television studio was not set up well for the shift in production style. This prompted us to build a temporary video production console on several folding tables right inside our studio.

    Adding Steam Deck XL into the Workflow

    Using the Zoom keyboard shortcuts certainly helps the production value. I find myself using Shift-Command-W (Mac) constantly to switch between Speaker view and Gallery View in Zoom. To me that’s the key to using zoom for video production. You can force a better cadence of switching that matches the conversation.

    I do wish there was a way within Zoom that I could force a certain camera view to appear full screen, similar to the normal workflow of a video production switcher. Zoom does offer a “spotlight video” option which does this, but it’s not mapped to a keyboard shortcut and requires too many mouse clicks to make it useful.

    Another shortcut I use often hides the control panel buttons on the lower part of the screen.

    The Stream Deck XL simply automates the keyboard shortcut process and reduces keystrokes to a single button. Allowing new users in the video production environment to get up to speed faster and with better results.

    TIP – ENABLE GLOBAL SHORTCUT

    Enabling global shortcuts really helps the Stream Deck configuration. This setting allows the shortcuts to work even when Zoom is not in focus.

    We’re also using VLC shortcuts to play intro/outro music.

    Bottom row of buttons on our Stream Deck are all shortcuts for Zoom.

    Then I added several shortcuts to websites we use for our live broadcasts.

    A work in progress, but thought I’d share what we have setup so far. If you’re using Stream Deck for video production I’d be curious to hear your use case.

  • Backing Up Large Video File Collections to AWS Glacier Deep Archive

    Part 1 – Why AWS Glacier Deep Archive?

    Backing up large collections of raw video footage and edit masters remains a real challenge for anyone working in the video production world. As the Executive Director of a local community media station, I’m responsible for maintaining a Synology NAS which currently holds 55TB of final edit master videos. The idea of incorporating “Cloud Storage” into our backup procedures has always interested me, but the expense has held me back.

    Until recently, our backup was rudimentary. We utilized Archive.org, which is a wonderful organization operating an online digital library of print, audio, and video works. They allow users to upload high-quality MPG2 files to be added to the collection and unlike YouTube, they offer the ability to download your original upload at any time.

    For us, this was a win-win. Archive.org provided a FREE way for us to share our content, preserve it for the future, and have an offsite backup if we ever needed it.

    We love the Archive and will continue to support them. The main goal of keeping our content open and available to the public for decades to come is just amazing.

    That said, the Archive is NOT a backup solution, but given our budget constraints, it was a quasi backup. We frequently said, “If the Synology NAS went up in flames, at least the videos would not be lost forever”…but the recovery process might take that long.

    When I learned about Amazon’s Glacier Deep Archive service earlier this year I was instantly intrigued and thought this might finally be a perfect solution for our needs. At “$1 per TERABYTE per month” they certainly had my attention.

    Glacier Deep Archive is a new product offering by Amazon Web Services (AWS) that falls under their S3 Storage product line. Deep Archive is the lowest cost storage class. When Amazon released the new storage class on March 27, 2019 their press release highlighted several use cases including media and entertainment companies:

    “there are organizations, such as media and entertainment companies, that want to keep a backup copy of core intellectual property. These datasets are often very large, consisting of multiple petabytes, and yet typically only a small percentage of this data is ever accessed—once or twice a year at most. To retain data long-term, many organizations turn to on-premises magnetic tape libraries or offsite tape archival services. However, maintaining this tape infrastructure is difficult and time-consuming; tapes degrade if not properly stored and require multiple copies, frequent validation, and periodic refreshes to maintain data durability.”

    Amazon Web Services Press Release, March 2019

    There is some “fine print” to be aware of, although none of it’s a real concern for me. There are additional charges for recovering your data and the data is not instantly available, retrieval time is up to 12hrs, but that’s the trade-off for the low cost. Again, not a big deal for my use case. You can check out the Amazon S3 Website for more specifics. The whole idea of Glacier Deep Archive is LONG TERM storage, files that you need to keep and don’t want to lose but may never actually need to access if your local files remain intact.

    For media professionals, I see Glacier Deep Archive as a great tool for:

    • Wedding and Event Videographers who want to backup raw footage, master files, and other assets.
    • Community Media Stations (PEG Access) looking to backup programs and raw footage.
    • Local Production Companies, again for all the same reasons.

    Before I share my workflow and experiences with AWS Glacier Deep Archive let’s jump back for a second and talk about backup best practices for just a minute.

    3-2-1 Backup

    Peter Krogh’s 3-2-1 Backup Strategy is a well-known best practice adopted by IT professionals and the government. The 3-2-1 concept is rather simple:

    3. Keep at least three copies of your data
    The original copy and two backups.

    2. Keep the backed-up data on two different storage types
    Multiple copies prevent you from losing the only copy of your data. Multiple locations ensure that there is no single point of failure and your data is safe from disasters such as fires and floods.

    1. Keep at least one copy of the data offsite
    Even if you have two copies on two separate storage types but both are stored onsite, a local disaster could wipe out both of them. Keep a third copy in an offsite location, like the cloud.

    With the 3-2-1 Backup goals in mind, I’d like to share my experiences with AWS Deep Archive in Part 2 of this blog post. I’ll share the workflow I’ve established after running into some issues initially.

    Keep in mind, I’m new to the AWS platform and I’m a media professional, not an IT genius. I am a tech geek and enjoy the challenge of learning new things. If you have any feedback, tips, or suggestions please feel free to post in the comments.

    Part 2 – Backup to AWS Glacier Deep Archive using CLI
    Coming Soon

  • Just Another Game

    Insights from Super Bowl XLIII TV Director Mike Arnolds

    From a Television Production standpoint treating the BIG GAME like it’s “just another game”, as much as possible, might sound strange at first but the approach makes sense. In a recent article on thebroadcastbridge.com, Super Bowl LIII Television Director Mike Arnold shared some insights on how he approaches the game.

    Failure is NOT an Option

    With over 100M television viewers and advertisers taking chances with $5M ads, the stakes could not be higher for all the creative professionals involved with the production of the Super Bowl.

    CBS will be deploying 115 cameras for the coverage this year, which is truly incredible and part of the mystique for viewers. I think we all have a vision in our mind’s eye of a single director staring at a wall of 115 monitors faced with the impossible task of choosing the exact right camera at just the right moment.

    Obviously, it’s not quite like that but we know the task requires great experience and skill developed over many years in the industry. Although there’s 115 cameras, many of them of specialty cameras such as the pylon cams and extra angles for special situations. Director Mike Arnold points out that from his standpoint, there’s no need to keep eyes on all 115 cameras.

    “I try to my focus on my key 20 cameras and let other people worry about the rest,”

    Super Bowl LIII Director Mike Arnold (via thebroadcastbridge.com)

    Just as the football teams should avoid getting overly hyped up or alter established routines, nor should the TV Production Staff. Treating the BIG GAME like it’s “just another game”, as much as possible, is logical and the recipe for success.

    Focus on Story

    Storytelling is always the key a concept that is focused on in the classroom and what drives every script, every outline…what is the story we’re telling. Although Football is a live event and unscripted, there’s always story.

    The Broadcast Bridge article highlights storytelling, explaining how shot sequencing and the use of close-ups help to expose the emotion, the anguish, the elation of the game.

    Paying Attention to the Announcers

    Lastly, I want to point out the importance of having the visual match what the viewer is hearing…from the broadcast team in the booth.

    Arnold said this year his shot selection will often shadow CBS Sports on-air announcer Tony Romo’s commentary, because the former quarterback has gained a reputation for seeing the entire field from a player’s perspective and how plays develop.

    The Broadcast Bridge

    You can check out the full article on The Broadcast Bridge.com

  • Video Production Trade News – Week of November 12, 2017

    Each week I share select trade news articles with students enrolled in my New England Tech video production classes. It’s critical in any field to remain current, follow emerging trends, examine best practices, and learn about various challenges that face the industry.

    Below are some articles from the past week that were discussed during class:

    1. NBC to Make SkyCam Main Camera for Titans-Steelers

    Trade Publication Source:
    TVTechnology – http://www.tvtechnology.com/business/0011/report-hdr-tv-shipments-to-reach-nearly-48m-in-2021/282201 

    2. Snapchat Usage by Media Companies

    There are two announcements this week by major media companies that plan to use the Snapchat platform. 

    Conan O’Brien Launches Comedy Show on Snapchat
    Trade Publication: Broadcast & Cable
    http://www.broadcastingcable.com/news/currency/conan-o-brien-launches-comedy-show-snapchat/170052

    ESPN Launches ‘SportsCenter’ Segments on Snapchat
    Trade Publication: Multichannel News
    http://www.multichannel.com/news/content/espn-launches-sportscenter-snapchat/416529

    2. On this Date in Radio history – Lee de Forest

    On November 13, 1906 Lee de Forest patented the Audion tube.

    Trade Publication: Media Confidential
    http://mediaconfidential.blogspot.com/2017/11/november-13-radio-history.html

  • Video Production Trade News – Week of July 31, 2017

    Each week I share select trade news articles with students enrolled in my New England Tech video production classes. It’s critical in any field to remain current, follow emerging trends, examine best practices, and learn about various challenges that face the industry.

    Below are some articles from the past week that were discussed during class:

    1. PBS Federal Funding

    “There is no PLAN B” if $450 Million federal budget for PBS is cut.

    Read Full Article: http://www.latimes.com/entertainment/la-et-entertainment-news-updates-july-there-is-no-plan-b-for-public-1501448678-htmlstory.html  

    2. Follow UP on NFL Ratings Discussion

    Since our discussion last week, a J.D. Power 2017 Fan Experience Survey reveals other possible reasons for the dip in ratings.

    Read More: https://www.sbnation.com/2017/7/27/16050800/nfl-viewership-national-anthem-protests-jd-power-survey-2017?yptr=yahoo

    3. Drone UAV Pilot Training

    Website: https://www.zephyr-sim.com/ 

    4. July 30th – Vladimir Zworykin Birthday

    In 1889…Vladimir Zworykin, called the “Father of Television” was born in Russia. He invented the iconoscope in 1931 while in the employ of RCA, the parent company of NBC. He died July 29, 1982 on the eve of his 93rd birthday.

    Read More: http://mediaconfidential.blogspot.com/2017/07/july-30-radio-history.html

  • Video Production Trade News – Week of July 24, 2017

    Each week I share select trade news articles with students enrolled in my New England Tech video production classes. It’s critical in any field to remain current, follow emerging trends, examine best practices, and learn about various challenges that face the industry.

    Below are some articles from the past week that were discussed during class:

    1. Sony’s Electronic Variable ND Filter

     

    Electronic Variable Neutral Density (ND) Filter is the name Sony is giving a new feature now available on certain Sony Cameras.

    Controlling our environment and camera exposure settings to produce professional level compositions is a topic we’ve discussed in class many times. Developing a solid understanding of exposure is critical to controlling Depth of Field, a technique used is frequently to direct viewers attention and allow our subjects to stand out from the background.

    Sony’s auto ND filter makes controlling DOF even easier, assuming you have enough light.

    2. NFL Making Changes to Commercial Breaks – Testing Split Screen Ads

    The NFL conducted extensive research last Fall, measuring viewer metrics including eye tracking, heart rates, and skin response. All in an effort to address the 8% decrease in TV ratings last fall.

    Among the changes coming, the NFL will look to make the following changes:

    •  30 percent fewer promotional messages (ie. 60minutes coming up next)
    • Split screen Ads, with an ad on one side and what’s happening in the stadium on the other

    The league is also making changes with an eye on cutting down the length of the game. Officials will use tablets while consulting with officials in New York. Overtime will be cut to 10 minutes from 15.

    Source: https://www.bloomberg.com/news/articles/2017-07-17/nfl-puts-fans-in-lab-to-see-what-works-as-football-seeks-answers 

    3. Media Composer | First

    The first 30 days (or so) of Media Composer First