Last week I had the pleasure of attending a performance by the Rochester Philharmonic Orchestra with my parents, and my 8th grade son Lewis. Lewis plays the cello and has recently spent dozens of volunteer hours playing in the “pit orchestra” for his school’s production of Oliver (not to be confused with my other son of the same name). He and I are the lone torchbearers of classical music in our household of six.
I had not been to a proper orchestral performance in many years, and as things got underway, something struck me in a way it never had before.
I played the violin from first through eighth grade, and over those eight years I tuned my violin so many times that, to this day, I can still hum a pretty perfect violin “A” on command.
The violin’s A is the note (A above middle C) by which the entire orchestra make sure it is “in tune.” However the A they use, does not come from the violin, it comes from the oboe.
The oboe is a woodwind instrument that is considered to have the least variability of all the instruments in a full orchestra, so before an orchestra begins playing a piece, the principal oboe player toots out an steady A, from with the concert master (head violin player) tunes his or her violin and then the entire orchestra joins in to tune their instruments.
I described the oboe as having the least amount of variability, not because it is perfect, quite the opposite. It is simply the best choice for the task given the options.
So, the orchestra tunes to the oboe’s A, the conductor ascends the riser and off they go into the piece. Does it matter if the oboe’s A is absolutely perfect? Nope, not at all. If the oboe’s A is a little off a theoretically perfect A, every other instrument will have the same degree of shift and the orchestra will sound great together.
Click the play button if you don’t know what I’m talking about:
As the performance unfolds, strings stretch, pegs move, horns fill with spit and reeds transform under the rigors of being played.
By the time a typical 20-30 minute orchestral piece is finished, every instrument has drifted out of tune, at least a little.
The conductor takes a bow, the audience applauds and the musicians stand up to be recognized. However, before they move on to the next piece, they go through the whole tuning routine again. Even if the oboe blows a slightly different A this time, it doesn’t matter because all of the instruments are using it. The A that is played becomes the one thing all the musicians know to be true.
Do you remember that game called “telephone” we used to play in elementary school?
All the kids in the class sat on the floor in a big circle. The teacher would start the game by whispering a sentence or phrase into the ear of one student. That student would then whisper it into the ear of next student, and this would repeat until the message made it all the way around the circle.
The fun of the game comes when the last kid in the circle blurts out what they heard from the second to last kid, and then the teacher reveals what it started out as. I’ve never seen the game played with anything but hilarious results.
The reason the message gets so distorted, is each student has only the one whisper on which to base his or her understanding before passing it along. The process is designed to introduce errors at every step, and although it is fun, it gives the kids something to think about with regard to the accuracy of the things they hear, and how much faith to put in things as told by others.
In 2015, we all enjoy a variety of digital video products. From digital cable and satellite television to Netflix, YouTube and Vine, very few people get through a modern day without some exposure to digital video.
Most people don’t think about it (because they don’t have to) but digital video is incredibly complicated to create and deliver. The amount of math and programming that goes into the 30 seconds of a dancing baby goat you saw on Facebook, would melt your brain if you had to try and understand it.
One of the things that makes digital video possible is something called compression. Video in its purest form uses an incredible amount of data. I won’t get into specifics, but in general terms without compression your Internet connection would need to be about 500 times faster than it is to watch an HD program on Netfilx.
Luckily we do have compression and the most common kind of compression is called MPEG or temporal compression.
In simple terms MPEG compression works like this:
- Software goes through every frame of a video (typically 30 per second) and figures out the differences from one to the next.
- Then, it only stores the information dictated by those changes, so for every pixel that doesn’t change from one frame to the next, it simply repeats what those pixels looked like in the previous frame. Basically, “make these changes and leave everything else alone.”
- By saving the change information only, (the delta for you scientists and engineers) this process can result in a file size as much as 700 times smaller.
- The reference frames from which the software starts keeping track of the changes are called “keyframes,” and they contain a complete set of all the pixel information. Because if this, they are much larger than the frames for which the file has only the change data.
- The trick is figuring out just how many keyframes the video really needs to produce an acceptable level of quality.
- Compression is a balancing act between quality and file size, using all that aforementioned brain melting math and programming.
I am sure you have seen videos where the image starts to get all blurry and pixilated (those are called compression artifacts) and then all of a sudden it snaps back into focus and gets clear again. In that moment you experienced a keyframe. A reset of what the compression software actually knows to be true.
The following is a six-minute video about video compression. I’ve set it up to start playing a little more than half way through where it shows an example of what I’m talking about. If you really want to geek out and watch the whole thing, simply slide the controller back to the beginning.
So what do these three things have to do with each other?
In all of our daily dealings at home, at work, on the Internet and in social situations, everything we say and do is based solely on the information we have.
The question is… how good, or fresh, or accurate is that information?
Like the orchestra that checks-in with the oboe before every piece, or the video compression algorithm that knows exactly how often to use a keyframe to keep things sharp; we too need to be cognizant of how frequently we should check the quality of the sources of information on which we are basing the things we say and do.
Take the time to make sure your actions are rooted in things you actually know to be true, not just what somebody said, somebody said, somebody said. After all, the last kid in the circle of the telephone game gets the laugh… but he doesn’t feel good about it.
Copyright © 2015 – Stephen S. Nazarian – All rights reserved.(Check-In With The Oboe)