Video Basics

Introduction

There is a moment that happens to almost everyone new to video work. A laptop is plugged in, the screen stays black, and four people look at you. Somebody suggests a different cable. Somebody else suggests restarting the laptop. Both suggestions are guesses, and one of them will eventually work, and nobody — including you — will know why.

That is what this book is for. Video has a reputation for being temperamental, and it is not. It is a chain of stages, each of which makes an agreement with the stage next to it about format, timing and protection, and almost every failure is one of those agreements not being met. Once you can name the stages and know what each one is agreeing to, a black screen stops being bad luck and becomes a question with about five possible answers.

What this book actually does

It gives you the chain and the vocabulary that goes with it. What a picture is made of and how it is described. How that picture becomes a signal, which connector carries which signal, and why a converter exists. What a switcher does, what a scaler does, and why the two are constantly confused. How the three kinds of display differ and what each one is good and bad at. And then how to put those stages together and trace them when a picture does not arrive.

Every chapter shows you the thing on a real job, walks you through the procedure, warns you about the failure you will actually meet, and gives you something you can take to work.

Who this is for

Somebody who will be handed video work in the next few months, or has already been handed it and is getting by on pattern-matching. You do not need to have run a switcher. You do need to be willing to look at a panel of connectors and work out what goes in and what comes out, which is a skill this book will build deliberately.

If you have read AV Fundamentals 101 you already have the shape of a signal chain. This book is the video half of it in detail. If you have not, you will still be fine; the ideas are introduced as they arrive.

What this book will not do

It will not teach you a particular switcher. Every manufacturer names things differently, and a book that taught you one console would be wrong in a year and useless on the next job. It teaches what all of them are doing, so that the manual for any of them makes sense.

It will not make you a video engineer, a projectionist or an LED technician. Each of those is a specialism with a course of its own. What it will do is take you from "video is the scary department" to being genuinely useful on a video crew, which is the step that actually gets you booked.

How to Use This Book

Each chapter follows the same shape, and the shape matters more than it might look. You read the ideas, then you see them play out in a real situation, then you get the procedure, then you find out what goes wrong, then you practice, then you get a checklist you can actually take to work.

1. Read the chapter body first. This is the understanding layer, and it is written to be read straight through.
2. Read On the job. The same material as a real situation on a real show, which is where it stops being abstract.
3. Work through How to do it. These are procedures you can follow on a job this week.
4. Study What goes wrong. Experienced technicians are not people who never make mistakes; they are people who recognise mistakes early.
5. Do the exercise before reading the answer key. Recognition feels like learning and is not.
6. Take the checklist with you. Photograph it on your phone. Use it until you no longer need it.

Keep a notebook

One habit is worth starting today, before the first chapter: keep a small notebook and write the technical details in it as you meet them. Video is full of numbers you will need exactly and will not remember — which resolutions are which, how long an HDMI run can be before it gets unreliable, what frame rate the country you are working in uses, which connector locks and which does not.

Write it by hand rather than photographing this page. Copying a number puts it on paper; rewriting it in your own words is what puts it in your head, and deciding how to phrase it is most of the learning. Add to it as you work: the adapter that worked when three others did not, the projector menu where the input setting hides, the laptop that always needs the display duplicated rather than extended.

Then use it twice. Read a few pages on the way to a job, which takes five minutes and turns things you half-know into things you know. And carry it on the job, because the moment you need to know whether a run is too long is never a moment when you can stop and look it up.

A note on the words

Video vocabulary is inconsistent and you should expect that rather than be caught by it. A switcher is a mixer to some people and a vision mixer to others. A scaler, a processor and a converter overlap and the boundaries move by manufacturer. Program and preview are PGM and PVW on one desk and something else on the next.

None of that is worth arguing about. What matters is that you can say what a box does, because a person who can describe the function will be understood anywhere, and a person who only knows one brand's word for it will be understood in one building.

A note on safety

Video work puts you near mains power, up ladders, and occasionally under things being flown. This book gives you awareness, not authorisation. If you are ever unsure whether you are qualified to do something, you are not — ask. Nobody in this industry has ever been fired for asking whether a thing was safe.

Where this book sits

This book assumes you have read AV Fundamentals 101, or that you already know what is in it.

Experience counts instead. If you have time on a crew where you have already met a signal chain, a source and a destination, and know what a switcher is for, start here.

If none of that is true yet, read AV Fundamentals 101 first.

Where it leads

These are not steps in an order. Take whichever one fits.

• Video Switching Fundamentals — the switcher in depth, which this book only opens
• LED Video Basics — if walls rather than projectors are where your work is going

How a Moving Picture Is Described

Everything in video rests on a handful of numbers, and almost every argument on a show floor is really an argument about one of them. How many pixels, in what shape, arriving how often, described in which range of colour. None of it is difficult, and all of it is specific, which is the opposite of how video is usually explained.

Pixels, resolution, and aspect ratio

A digital picture is a grid of coloured dots. That is the whole of it, and almost everything else in this book follows from that one fact.

Each dot is a pixel, and resolution is how many of them there are: 1920 across by 1080 down is the resolution most of the industry still runs on, usually called 1080p or HD. 3840 by 2160 is UHD, often called 4K, and it is exactly twice HD in each direction, which means four times the pixels and four times the data. 1280 by 720 is 720p, still common on older equipment and perfectly adequate on a small screen.

Aspect ratio is the shape of that grid — width against height, independent of how many pixels are in it. 1920 by 1080 and 1280 by 720 are both 16:9, the widescreen shape almost everything uses now. 1024 by 768 is 4:3, the older squarer shape, and you will still meet it on ageing projectors and on the occasional presenter's laptop.

The reason to care is that a mismatch between shapes is visible and a mismatch between resolutions usually is not. Send a 16:9 picture to a 4:3 screen and something has to give: black bars top and bottom, or a picture stretched until faces look wrong, or the sides cut off. Send 720p to a 1080p screen and a scaler quietly resizes it and most of the audience never knows. Shape is the one people notice.

One number that matters more than it looks: pixels are not a physical size. A pixel on a phone is a fraction of a millimetre; a pixel on an LED wall might be four millimetres across. Resolution tells you how much detail is in the picture, not how big the picture is. A wall built from panels with 4 mm between LEDs and a wall built with 2.6 mm can be exactly the same physical size and have wildly different resolutions, and the second one can be viewed from much closer before it stops looking like a picture and starts looking like a grid.

Frame rate

A moving picture is a sequence of still pictures shown quickly enough that the eye reads it as motion. Frame rate is how many of those arrive each second, in frames per second, and it is where a surprising number of live-event problems begin.

The common rates are 24, 25, 30, 50 and 60. Film has used 24 for a century and it still looks like film, which is largely why cinema looks different from television. Broadcast standards split the world in two: countries on 50 Hz mains generally use 25 and 50, and countries on 60 Hz mains use 30 and 60 — or, maddeningly, 29.97 and 59.94, a fraction left over from the introduction of colour television in the 1950s that has never gone away.

Higher rates look smoother and cost more data. Sixty frames a second is twice the information of thirty, and on a large system that is a real constraint rather than an abstract one.

What makes this a live-event problem rather than a trivia question is that a chain must agree. A camera running at 25, a switcher expecting 59.94 and a projector locked to 60 do not make a smooth picture between them; they make judder, dropped frames, or nothing at all. On a show with any broadcast element, one rate is chosen up front and everything is set to it, and "what frame rate are we?" is a question asked in the first ten minutes of a technical meeting rather than discovered at rehearsal.

For a presentation-only job it usually matters less, because a laptop and a projector will negotiate something workable between them. It starts mattering the moment there is a camera, a recording, a stream, or more than one processing box in the path. There is one more thing the numbers carry, and it is the letter after them.

A progressive frame — the p in 1080p — is a whole picture, every line of it, delivered at once. An interlaced frame — the i in 1080i — is delivered as two halves: first every odd line, then every even line, each half arriving in turn. Interlacing is a compression trick from the era of broadcast television, invented to halve the data a picture needed while keeping motion looking smooth, and it worked so well that it is still with us decades after the reason for it went away.

Almost everything you meet is progressive. Laptops, media players and projectors are progressive; the web is progressive; modern cameras are progressive unless somebody set them otherwise. Interlaced material turns up in broadcast, in older cameras, and in archive footage a client found somewhere, and that is where it becomes your problem rather than a piece of history.

What goes wrong is specific and recognisable, which is why it is worth knowing. An interlaced picture shown on a progressive display has to be de-interlaced — the two halves woven back into whole frames — and when that is done badly you see combing: fine horizontal teeth along the edges of anything moving, as though the picture has been shredded and misaligned. Once you have seen combing you never mistake it for anything else, and its cause is always the same. Somewhere a device is handling interlaced material and not handling it well.

The rule for a job is short. Ask what the sources are, prefer progressive throughout, and if something interlaced has to come into the show, de-interlace it once, as early as possible, in the best box you have. Two devices each doing half a job on the same picture is worse than either doing it alone.

Colour and brightness, simply

Colour in video is made from three primaries — red, green and blue — mixed at different intensities. Every colour on every screen you have ever looked at is those three, and white is all of them at full.

That is enough to work with, but three practical ideas sit on top of it and each of them has caused an afternoon of confusion for somebody.

Colour space is the agreed set of colours a signal is allowed to describe. Rec. 709 is the standard for HD and is what almost everything in the events world uses. Rec. 2020 is the wider set used for UHD and HDR material. A signal carried in one and interpreted as the other looks subtly wrong — usually oversaturated or washed out — and it is a common cause of "the picture looks fine on my laptop and strange on the screen".

Colour range is narrower and nastier. Video equipment has historically used a limited range, where black sits at value 16 and white at 235 rather than 0 and 255. Computers use the full range. Feed a full-range signal to a limited-range display and blacks crush and whites clip; feed a limited-range signal to a full-range display and the picture looks washed out and grey. This single mismatch accounts for a great many "why does it look flat" complaints, and the fix is a setting in a menu rather than anything on the picture.

Brightness is the one people argue about and the one physics decides. A projector's output is measured in lumens, and the number you need is set by the size of the image and how much other light is in the room. Doubling the width of a projected image spreads the same light over four times the area, so a projector that looked bright on a small screen is dim on a large one. A room with the house lights up needs several times the projector a dark room does. This is why a proposal that says "the projector was fine last time" is not evidence about the next room.

The numbers behind a picture

Rules of thumb, not specifications. Every one of them varies with equipment; all of them are the right shape.

On the job

A corporate breakfast in a hotel suite. One projector, one screen, one presenter with their own laptop, and forty minutes before doors.

The laptop connects and a picture appears, which is the good news. The bad news is that the slides have black bars down both sides and the presenter's photographs look slightly stretched, and the client has noticed both.

Nothing is broken. The laptop is an older machine running 4:3, and the projector and screen are 16:9. The projector is doing what it was told: fitting a squarer picture into a wider frame, which leaves bars. Somebody before you tried to fix it by setting the projector to stretch, which removed the bars and made everyone in the photographs eight percent wider.

The real fix is upstream and takes thirty seconds: change the laptop's display resolution to a 16:9 one, so it is sending the shape the screen actually is. The slides reflow, the bars go, and nobody is stretched. The lesson worth carrying is that a shape problem is almost always solved at the source, and almost never solved well at the display.

How to do it

What to check on any source, before you decide anything is wrong with it. Two minutes, and it settles most format arguments before they start.

1. Find out what the destination is: the native resolution of the projector, panel or wall, and its aspect ratio. This is on a plate, in a menu, or on the spec sheet, and it is the number everything else has to serve.
2. Set the source to that resolution if you can. A laptop sending exactly what the screen wants is the cleanest picture available and costs nothing.
3. Check the aspect ratio matches. If the source is 4:3 and the screen is 16:9, fix it at the laptop rather than by stretching at the projector.
4. Ask what frame rate the show is, and set anything that has a choice to match it. On a job with a camera, a stream or a recording, this is not optional.
5. Look at a full-black and a full-white part of the image. Crushed blacks or grey whites point at a range mismatch, which is a menu setting rather than a fault.
6. Note the working numbers in your notebook before you tear down. The same room next month will have the same answers.

What goes wrong

Practice it

Work these out before reading the answers. All five are questions you will be asked on a job, usually by somebody standing over you.

Sources: Where Pictures Come From

Everything downstream exists to carry a picture that something made. This chapter is about the things that make them, because a surprising share of video problems are not in the chain at all — they are at the source, and they are solved by knowing what the source is capable of before anybody plugs anything in.

The sources you will actually meet

On the vast majority of jobs there are exactly three kinds of source, and you will meet them in this order of frequency.

A laptop is the most common source in the industry by a wide margin, and it is the least predictable. It belongs to somebody who is about to speak, it has whatever is installed on it, and it will do something surprising at least once a week. It goes to sleep. It changes resolution when it is unplugged and back again when it is replugged. It shows a notification over the slides. It has one port and the wrong kind. Almost everything in this chapter about laptops is about reducing that unpredictability rather than eliminating it, because you cannot eliminate it.

A media player or playback machine is a computer that exists to play files reliably, which is the whole difference. It has no email on it, nothing pops up, it does not sleep, and it outputs one format all day. When a client wants a video played, this is what should play it — and getting the file onto it in advance is the single most valuable thing you can arrange.

A camera is the source that makes video feel like video. On a small job it is one camera on a tripod at the back giving an image-magnification feed of the speaker, so people behind row fifteen can see a face. On a larger job it is several cameras and somebody directing them.

There are others — a document camera, a phone, a games console, a video conference feed — and all of them are one of the three in disguise. Ask what it outputs, what format, and whether it is copy-protected, and any unfamiliar source becomes a familiar one.

Laptops, and why they cause so much trouble

A presenter's laptop is a source you do not own, cannot prepare, and meet fifteen minutes before it has to work. That is the entire problem, and the fixes are all about narrowing what can go wrong.

Start with the port. Modern laptops have USB-C, older ones have HDMI, business ones often have DisplayPort or Mini-DisplayPort, and older still have VGA. USB-C is the one that catches people, because a USB-C socket may or may not carry video and nothing about the socket tells you. This is a question to ask in advance rather than discover on the day, and the answer is worth having in writing.

Then the display arrangement. A laptop can mirror — the same picture on both screens — or extend, which treats the projector as a second desktop. Extend is what presentation software wants, because it puts the slides on the screen and the presenter's notes on the laptop. Mirror is what a nervous presenter wants, because what they see is what everybody sees. Neither is wrong, and the failure is having a different one than the presenter expects, which is why the answer is to ask them rather than to decide.

Then the settings that ruin a session rather than break it. Sleep and screensaver will both turn the main screen black in the middle of a slow question, and the fix is to turn them off, not to jiggle the mouse for ninety minutes. Notifications will put somebody's private message on a four-metre screen; every operating system has a do-not-disturb switch and it should be on. Automatic updates have ended keynotes.

And the resolution, which is the one thing you can fix rather than merely prevent. Set the laptop to the resolution the screen actually is, and the picture is as good as it can be. Leave it on whatever it was and something in the chain will scale it, and the slides will be softer than they need to be for no reason at all.

Media players, and why they are the easy one

A media player is a computer whose only job is to play files out as video, and it is the most reliable source you will ever be handed. Everything good about it comes from what it does not have.

It has no email on it, so nothing pops up. It does not sleep, because sleeping is not part of playing a file. Its resolution and frame rate are set once and stay set, so nothing renegotiates halfway through the afternoon. It has no updates waiting to install themselves during a keynote. A presenter's laptop is a general-purpose machine you are borrowing; a media player is a single-purpose machine you own, and that difference is the whole of it.

On a small job it is a compact box with an HDMI or SDI output and a folder of files, driven from a remote or a laptop. On a larger one it is a media server, which is the same idea with far more of everything — multiple outputs, content spread across several screens, and a cue list that a show operator triggers. Both belong to video rather than to control, whatever anybody tells you, because what they do is make pictures.

The one thing that goes wrong with a media player is not the machine, it is the file. Content that arrives on the day, in a codec the machine does not like, at a resolution nobody agreed, is the failure you will actually meet — and it is a scheduling problem rather than a technical one. Ask for content early, play every file all the way through before doors rather than skipping through it, and check that the last frame is what you expect, because a clip that ends on a black frame and one that ends on a copyright card look very different on a screen behind a speaker.

And this is the machine that rescues the laptop problem. When somebody's laptop turns out to have no video output at all, the deck running from a media player you control is the answer — which only exists if you asked for the deck in advance.

What a camera actually gives you

A camera turns light into a video signal, and for a technician there are about five things that matter about it, none of which are the ones camera enthusiasts argue about.

What it outputs is first. A professional camera outputs SDI on BNC, which locks and runs a hundred metres, and this is the reason professional cameras are easy to work with. On a broadcast or outside-broadcast job you will also meet triax, a thicker cable that carries the picture out, the power in, the talkback both ways and a return feed to the camera operator, all down one line over very long distances. It is why a camera at the far end of a stadium needs one cable rather than five. A consumer or prosumer camera outputs HDMI, which does not lock and does not travel, so it needs a converter at the camera. Some cameras output a clean feed and some output their own menus and focus indicators over the picture, and finding out which before the show is worth doing, because a focus rectangle on the main screen during a keynote is memorable for the wrong reasons.

What format it is set to is second, and it matters more than anything else in this chapter. A camera has a resolution and a frame rate, and both have to agree with the rest of the show. This is the single most common camera problem on an event: a camera at 1080p59.94 on a show running 1080p50, which switches badly or does not switch at all.

Whether it is genlocked is third, and it is worth knowing the word even if you never touch it. Genlock is a shared timing reference sent to every camera so they all start their frames at the same instant. Without it, cameras are running at the same rate but not in step, and cutting between them can glitch. Professional switchers have a frame synchroniser on every input which hides this, which is why you can usually get away with ignoring it — until you are on a job where you cannot.

Power and mounting are fourth and fifth and they are unglamorous and they are what actually goes wrong. A camera on battery will run out during the session it matters most in; use mains where you can and check the battery where you cannot. And a camera on a tripod that somebody can walk into is a camera that will be walked into.

The three sources, and what each one costs you

Ask the same three questions of anything unfamiliar and it becomes one of these: what does it output, what format, is it protected.

The laptop conversation, before the day

Five questions. Asked a week out they cost nothing; asked at fifteen minutes they cost the rehearsal.

Ask the presenter

• What ports does your machine have, and does that USB-C port carry video?
• Do you want your notes on your screen, or the same picture as the audience?
• Is anything in the deck a video, and is it embedded or a separate file?
• Is any of it streamed from a service rather than saved on the machine?
• Can you send the deck in advance, so it can run from our machine if yours will not?

On the job

A half-day conference with two speakers and a camera at the back for image magnification, so the room can see faces on the main screen.

The camera goes in first and behaves, because it is a professional body on SDI and it was set to the show format — 1080p at 50 — before it left the shop. That decision, made in the shop by somebody who thought about it, is the reason the camera is a non-event all day.

The first speaker's laptop is a business machine with Mini-DisplayPort, and a passive adapter handles it. It is set to extend, which is what they want because they are using presenter notes, and the resolution is set to match the screen. Sleep and notifications are turned off with the speaker's permission, which takes a minute and is asked rather than done quietly.

The second speaker arrives at fifteen minutes with a laptop that has two USB-C ports and nothing else, and neither port carries video. There is no adapter in the world that fixes that, because the machine cannot output a picture at all.

What saves the session is that the deck was asked for in advance and is sitting on the media player. It runs from there, the speaker clicks through it with a presenter remote, and the audience never knows. The thing that worked was not equipment or skill on the day — it was a question asked a week earlier.

How to do it

Taking in a source you did not prepare. Ten minutes, and it prevents most of what goes wrong at the source end.

1. Look at the actual ports on the actual machine. Not what it should have — what is on it. If it is USB-C, confirm that port carries video before you rely on it.
2. Connect it and see a picture on your own monitor before you send it anywhere. Prove the source in isolation first, exactly as you would anywhere else in the chain.
3. Set the resolution to match the destination, so nothing downstream has to scale it for no reason.
4. Ask the presenter whether they want mirror or extend, rather than deciding for them. Getting this wrong is the difference between their notes being private and being on the main screen.
5. Turn off sleep, screensaver, notifications and automatic updates, with their permission. These do not break the show, they ruin moments in it.
6. Ask whether anything in the deck is a video, and whether it is embedded or streamed. Streamed is a copy-protection conversation, and it is one to have now.
7. Strain-relieve the cable at the machine. It does not lock, the presenter will move the laptop, and it will come out at the worst moment.
8. For a camera: confirm the format matches the show, confirm the output is a clean feed with no menus, and put it on mains power if there is any.

What goes wrong

Practice it

Seven questions about the source end. Every one of them is something that has cost somebody a session.

Video Signals, Formats & Connectors

This is the chapter that decides whether you are useful on a load-in. Not because the theory is deeper, but because the work is physical: you will be standing in front of a panel of sockets with people waiting, and knowing what each one is and what it can and cannot do is the difference between solving the problem and going to find somebody who can.

Analog and digital video

Every video signal you meet is one of two kinds, and knowing which you are holding tells you most of what you need to expect from it.

An analog signal describes the picture as a continuously varying voltage. The voltage rises and falls in step with the brightness of the picture, and the cable carries that shape from one end to the other. Nothing in the middle knows it is video; it is just a waveform on a wire. VGA and composite are the analog signals you will still meet, and both are on their way out.

A digital signal describes the picture as numbers. The same brightness that was a voltage is now a value, and the cable carries those values as a stream of ones and zeros. HDMI, DisplayPort and SDI are all digital.

The practical difference is how they fail, and it is stark enough to be worth memorising. Analog degrades. A long analog run, a poor connector or interference makes the picture soft, or ghosted, or gives it a colour cast — it gets worse gradually and it keeps working while it does. Digital does not degrade. It arrives perfect or it does not arrive: a clean picture, or sparkles and flashes, or nothing at all. There is very little middle ground, and that cliff edge catches people out. A digital run that worked yesterday and fails today did not slowly get worse; it crossed a line.

This is why "the picture looks a bit soft, it must be a long cable" is a sensible thought about VGA and a nonsense one about HDMI. Softness on a digital signal is a resolution or a scaling problem, never a cable-length one.

The common signal formats and connectors

There are eight you need to recognise on sight, and recognising them is genuinely most of the job on a load-in.

HDMI is what almost every consumer device outputs — laptops, media players, cameras, games consoles. It carries picture, sound and control information down one cable. It does not lock, which is its central weakness on a live job: an HDMI connector will fall out of a socket if somebody catches the cable with their foot, and it will do it during the keynote. It is reliable to about 15 metres for HD on a good cable, less for 4K, and after that it needs help.

DisplayPort looks superficially similar and is a different standard, common on business laptops and desktop graphics cards. Many DisplayPort outputs will speak HDMI through a cheap passive adapter, which is why those adapters are in every technician's bag. Mini-DisplayPort, the same signal on a smaller connector, is what Apple laptops used for years and Thunderbolt ports still carry.

DVI is the older digital computer connector and it is far from dead. It is a wide plug, usually white, with a block of pins and often a flat blade with four pins around it for the analog signal it can also carry. It gives picture only and no sound, which catches people out when they swap an HDMI source for a DVI one and lose the audio. Electrically its digital side is the same as HDMI, so a passive adapter between them works and costs almost nothing. You will meet it on installed projectors, on older graphics cards, and across a great deal of university and corporate kit that was bought once and never replaced.

SDI is the professional one, and it is the format the events and broadcast industries actually run on. It uses a BNC connector — a small round bayonet that twists and locks — over ordinary coaxial cable, and it goes a long way: over 100 metres for HD without any assistance. It carries no copy protection and no negotiation, which is exactly what makes it dependable. If you learn to love one connector in this trade, love BNC.

Fiber is the one that goes further than any of them, and it is how a signal crosses a building. Light down a glass strand does not care about distance or interference the way copper does, so a fiber run is measured in hundreds of metres and often in kilometres. You will meet it as a pair of small connectors clipped together, usually LC, and as a transmitter at one end and a receiver at the other converting to and from SDI or HDMI. Two things to know on a job: fiber is far more fragile than coax, so it does not like being trodden on or bent tightly, and the connector ends are optical, so a fingerprint on the end face is a fault. Keep the dust caps on until the moment you connect.

VGA is the old analog computer connector, a blue 15-pin plug with thumbscrews. It carries picture only, no sound, and it is dying but not dead; you will still find it on hotel lecterns and university podiums for years yet.

Composite is the older analog video connector, usually a yellow RCA phono, and it is standard definition however short the cable is. You will meet it on legacy equipment, on the back of old players, and on very little that was made this decade. It is worth recognising mainly so you know that finding one means the source is old, and the picture will be soft whatever you do to it.

USB-C is the newest arrival and the most confusing, because the connector tells you almost nothing. Some USB-C ports carry video and some do not, and there is no way to tell by looking. On a job this is worth knowing in advance rather than discovering ten minutes before doors.

Two more things sit on top of the connectors and cause more trouble than the connectors do.

EDID is the conversation between a source and a display. When you plug a laptop into a screen, the screen sends back a small block of data saying what resolutions and rates it accepts, and the laptop chooses from that list. It happens in under a second and you never see it, and when it goes wrong the symptoms are unforgettable: no picture, or the wrong resolution, or a laptop that suddenly offers only 640 by 480. Anything in the middle of the chain — a switcher, a splitter, a long-distance sender — is passing, blocking or inventing that conversation, which is why professional equipment lets you choose what EDID it presents.

HDCP is copy protection, and it rides on HDMI and DisplayPort. If the source insists on it, everything downstream must support it or the picture goes black — and it goes properly black, with no error message worth reading. This is why a Blu-ray player or a streaming stick will work into a hotel television and refuse to work through a professional switcher. SDI has no HDCP at all, which is one more reason the professional world prefers it.

Choosing and converting

Most video jobs come down to a chain of sources that do not natively speak what the destination wants, which is what converters are for.

A converter changes one format into another: HDMI to SDI, SDI to HDMI, VGA to HDMI. The good ones are small metal boxes with a power supply and they are entirely unglamorous, and they are also the reason a show happens. Every working technician's bag has at least an HDMI-to-SDI and an SDI-to-HDMI in it, because between them they let you put any consumer source onto a professional system and take any professional feed back to a monitor.

Adapters are different and worth distinguishing. An adapter changes the shape of a connector without changing the signal — a Mini-DisplayPort-to-HDMI dongle works because the laptop can already speak HDMI and just has the wrong socket. It is passive and it costs almost nothing. A converter changes the signal itself, needs power, and costs real money. Asking for "an adapter" when you need a converter is one of the more common ways to lose an hour on site.

The choosing rule is short. Go digital wherever you can. Use SDI for anything that has to travel — across a room, across a stage, anywhere near a floor people walk on. Use HDMI at the ends of the chain, where the sources and the monitors are, and keep those runs short. Convert once, as early as you can, rather than converting back and forth along the way, because every conversion is a device that can fail and a place where the format contract can be broken.

And when a long run is unavoidable, use fiber or an AV-over-IP encoder rather than a longer copper cable. Distance is a problem with known solutions, and "a longer HDMI cable" has never been one of them.

The eight connectors, at a glance

The panel in front of you on a load-in. Distances are for HD on decent cable and are the point at which sensible people stop trusting it.

Adapter or converter?

Getting this wrong costs an hour on site and sometimes a show.

On the job

A product launch in a conference centre. A presenter's laptop at the lectern, a media player for the launch film, a camera on a tripod at the back, and a switcher feeding two projectors 40 metres away.

The camera and the switcher are SDI, and those runs go in and work first time, because BNC over coax at that distance is not asking anything difficult. The laptop and the media player are HDMI at the lectern, both converted to SDI at the lectern itself rather than run long, which is the right call and takes two small boxes.

The launch film is on a streaming stick, and it goes black through the switcher while working perfectly on a monitor beside it. That is HDCP: the stick insists on copy protection and the professional switcher does not carry it. There is no setting to fix this and no cable that solves it. The fix is to get the film as a file and play it from the media player, which is what should have been asked for in the first place.

Then, twenty minutes before doors, the presenter's laptop shows the wrong resolution — a stretched 1024 by 768 where there was 1080p an hour ago. Somebody reseated the HDMI cable and the EDID conversation happened again with a different answer. Setting the switcher input to present a fixed 1080p EDID makes it permanent, and it is the last time that input argues all week.

How to do it

Getting an unfamiliar source onto a professional system, in the order that saves the most time.

1. Look at what the source actually has. Not what it should have — what is on the back of it. Photograph the panel if you are quoting for a job you are not on yet.
2. Decide where the conversion happens, and make it as close to the source as possible. Convert at the lectern and run SDI, rather than running long HDMI and converting at the far end.
3. Check whether an adapter will do. If the source can already speak the format you want and only the socket is wrong, a passive adapter is the cheaper and more reliable answer.
4. Ask whether the source uses copy protection. Anything consumer and streaming probably does, and that is a conversation to have with the client days before, not on the day.
5. Set a fixed EDID on the input if the equipment allows it, so the source is told what to send rather than guessing every time it is plugged in.
6. Strain-relieve anything that does not lock. Tape, a cable tie or a loop through the lectern shelf. HDMI falls out; assume it will.
7. Test with the real source and the real content, not with your test pattern. Half of these problems only exist when the actual laptop is on the actual cable.

What goes wrong

Practice it

Six situations from real load-ins. Answer them before you look.

Switching, Scaling & Processing

The middle of the chain is the part people are most frightened of, and it is mostly two ideas wearing a lot of different badges. A switcher chooses. A scaler translates. Almost every box between a source and a screen is doing one of those two jobs, and the ones that confuse people are the ones doing both quietly at the same time.

The video switcher

A switcher takes several video sources and decides which one goes to the screen. That sentence is the whole function, and everything else a switcher does is an elaboration of it.

The core idea is program and preview. Program is what the audience is seeing right now. Preview is what you are lining up to show next, visible only to you on your own monitor. You set up the next shot in preview, check it is right, and then take it to program. The audience sees a finished picture and never sees you choosing it, and that gap between preview and program is where a video operator actually does their job.

How the picture changes from one source to the next is a transition. A cut is instant and is the default for almost everything; it is invisible, which is why it is right most of the time. A dissolve or fade is a soft blend over perhaps half a second to a second and reads as gentler and slower, which suits a change of mood or a move between very different pictures. A wipe pushes one picture across another and looks dated on a corporate stage in a way it does not on a broadcast sports programme. There are dozens of other effects and you can safely ignore all of them for years.

Beyond that, the useful features on a live stage are a picture-in-picture, so a presenter's camera can sit in the corner of their slides, and a downstream key, which lays a logo or a lower third over whatever program happens to be. Both are worth learning early because clients ask for them constantly.

Consoles vary enormously — a compact four-input unit for a breakout room, a large multi-screen presentation system for a conference stage — and the vocabulary shifts by manufacturer. What does not shift is the model: sources in, one chosen, a transition between, one or more outputs. Learn the model and any desk becomes a question of finding where the buttons are.

Scaling and the format contract

A scaler changes the resolution and frame rate of a picture so that a source can meet a destination that wants something different. Almost every source and screen you will ever connect disagrees about something, and scalers are why that is a non-event instead of a crisis.

The idea underneath is worth naming, because it explains most of the department. Every connection in a video chain is a contract: this many pixels across, this many down, at this rate, in this colour range. Both ends have to agree. A scaler is the device that renegotiates the contract when the two ends want different things — it takes what is arriving and rebuilds it as what the next stage asked for.

Scaling up means inventing information. A 720p picture on a 1080p screen has more than twice as many pixels to fill as it arrived with, and the scaler is guessing at the ones in between. Good scalers guess well and the result looks clean and slightly soft. Cheap scalers guess badly and the result looks blurry or blocky. Scaling down means throwing information away, which is easier and generally looks fine.

Two practical rules follow, and both are worth carrying. Scale once. Every scaling stage is a generation of guesswork, and a picture that has been scaled three times along the chain looks noticeably worse than one that was scaled once at the end. And scale as late as you can — keep the picture at its native resolution as far along as possible and let the final device match it to the screen.

What matters most, though, is knowing where scaling is happening. A switcher may be scaling every input silently. A projector will happily scale anything you send it. A long-distance sender may scale to fit its bandwidth. When a picture looks softer than it should, the question is not "is something scaling it" — something almost certainly is — but "how many times".

Other processing you will meet

Between the switcher and the screen sits a handful of other boxes, and knowing what each one is for stops them being mysterious.

A distribution amplifier, universally called a DA, takes one input and produces several identical outputs. It exists because you cannot split a video signal by making a Y-cable; a video output drives one input, and splitting it passively degrades or kills it. Whenever one source needs to reach three destinations, a DA is the answer.

A frame synchroniser lines up a source that is not running in step with the rest of the system. Two devices at nominally the same frame rate are not necessarily starting their frames at the same moment, and switching between unsynchronised sources gives a visible glitch at the cut. A frame sync buffers a frame and releases it in time with everything else. On a professional switcher this is built into every input and you never think about it, which is exactly why it is worth knowing it is there.

A matrix router is a switcher without the production features — many inputs, many outputs, and any input can be sent to any output. It is the plumbing of a large installation: it does not do transitions or effects, it just decides what goes where.

A media server plays content. It is a computer built to output video reliably, holding the show's clips, stings and backgrounds, and on a larger job it drives content across multiple screens at once. It sits upstream of the switcher as one more source, and it belongs in video rather than in control, whatever anybody tells you.

An LED processor is the specific box that takes a normal video signal and turns it into what LED panels need. Every LED wall has one, it is the thing that knows the wall's exact pixel layout, and it is a specialism in itself.

What each box in the middle is for

One line each. If you can say the line, you can work out where the box belongs in a chain.

Program, preview, and the words for them

The same two things under four sets of names. Nobody is right and everybody is confident.

On the job

A single-day conference in a ballroom. One screen, a four-input switcher, and three sources: the presenter's laptop, a media player with the videos on it, and a camera at the back for the speaker's face.

Setup goes cleanly until the first rehearsal, when cutting from the camera to the laptop produces a visible glitch — half a frame of tearing at the moment of the cut. The camera is 1080i at 59.94 and the laptop is 1080p at 60, and the switcher is having to rebuild one of them on the fly. Setting the laptop to 1080p at 59.94 to match everything else removes the glitch entirely, and it takes one menu.

Then the client asks for the speaker's camera in the corner of their slides. That is a picture-in-picture and the switcher does it, but it exposes something: the laptop is 1080p and the screen is 1080p, and the picture-in-picture is scaling the camera down, which is fine, while the switcher is also scaling the laptop for no reason because an input setting was left on automatic. Turning that off makes the slides visibly crisper. Nothing was broken and nothing was fixed; a redundant scaling stage was removed, which is most of what improving a picture actually consists of.

How to do it

Setting up a small switched system so that it behaves. Half an hour, and it prevents the problems that otherwise appear at rehearsal.

1. Decide the show format first — resolution and frame rate — and write it on the case. Everything else is set to serve that one decision.
2. Set every source to the show format if it can be set. A source that arrives already correct is a source the switcher does not have to rebuild.
3. Set each switcher input's scaling off where the source already matches, and on only where it genuinely does not.
4. Confirm program and preview are on separate monitors and you know which is which. Getting these the wrong way round is a mistake you make exactly once.
5. Set the default transition to a cut. Choose a dissolve deliberately when a moment calls for it, rather than living on one permanently.
6. Build any picture-in-picture or key you have been asked for, and rehearse recalling it, rather than building it live while a client watches.
7. Cut between every pair of sources once and look for glitches at the cut. A glitch here is a format or timing mismatch, and now is when it is cheap to fix.
8. Count the scaling stages between each source and the screen. If any source is scaled more than once, find out why.

What goes wrong

Practice it

Six questions about the middle of the chain. Answer before you look.

Displays: Where the Picture Appears

Three technologies do the same job, and each of them is the wrong answer in circumstances where the other two are right. What decides between them is almost never a preference — it is the size of the image, the light in the room, and what the building will let you hang. Being able to say why is what turns a quote into advice.

Projection

A projector throws light at a surface, and everything about how projection behaves follows from that one fact — including all of its weaknesses.

Brightness is measured in lumens, and it is the number that decides whether the job works. The lumens you need depend on two things: how big the image is, and how much other light is falling on the screen. Doubling the width of an image quadruples its area, so the same projector spread over twice the width is a quarter as bright. A room with the house lights up or daylight through the windows can need several times the projector that a blacked-out room does. Anybody who has watched a presentation wash out at 11 a.m. in a hotel with one uncovered window has seen this in action.

Throw ratio is the second number, and it is the one that decides where the projector goes. It is the distance from the lens to the screen divided by the width of the image. A 2.0 throw ratio means the projector sits at twice the image width — a 4-metre image needs 8 metres of room. A short-throw lens does the same image from much closer, and a long-throw lens does it from the back of a ballroom. Getting this wrong means arriving with a projector that physically cannot make the image the client asked for in the room they booked, which is a phone call nobody enjoys.

Then there is the practical business. Keystone correction squares up a picture projected at an angle, and it does it by distorting the image digitally, which costs sharpness — so it is a rescue, not a plan. Getting the projector square to the screen mechanically is always better. Anything in the light path shows up: a person walking through the beam, a chandelier, a truss. And projection needs a surface — a proper screen, ideally, because a hotel wall is rarely as white or as flat as it looks.

What projection is good at is size. Nothing else makes a 10-metre-wide image for the money, and that is why it survives.

It is also the display type with by far the widest range, and a table row cannot capture it. At one end is a breakout-room projector: three or four kilograms, one person, out of the case and onto a table in ten minutes. At the other is a large-venue machine of fifty kilograms or more that takes two people to lift safely, arrives with its lens as a separate decision, and is usually flown from truss or stacked on a tower — which means rigging, which means qualified people, a structural sign-off and a crew. Between those sit the ones on a tall stand or a scaffold, where the projector is light enough to carry and still ends up above head height with everything that implies.

So when somebody asks how long a projector takes to set up, the honest answer is another question: which projector, how big is the image, and where is it going. A 3-metre screen in a meeting room and a 12-metre screen in a ballroom are the same technology and completely different jobs, and quoting the second at the price of the first is one of the more expensive mistakes available in this department.

Flat-panel displays

A flat panel is a television or a large professional monitor, and it makes its own light, which changes everything about how it behaves in a room.

Because the picture is generated rather than thrown, ambient light barely matters. A panel in a bright foyer looks like a panel; a projector in the same foyer looks like a rumour. Panels are sharp, consistent, need no throw distance, and nobody can walk through the image. For anything up to about 2 metres wide they are usually the better answer and increasingly the cheaper one.

Their limit is size and weight. Panels get expensive fast above 85 inches, and a large one is heavy, awkward and fragile in a way a projector is not. They also come on stands or brackets that need thinking about — a floor stand takes floor space and can be knocked, and a wall mount means somebody drilling somebody else's wall.

For a technician the practical points are short. Check the panel's native resolution and feed it exactly that. Turn off every consumer picture mode you can find — motion smoothing, dynamic contrast, anything with the word "enhance" — because they are designed for films at home and they make a presentation look strange. And check the overscan setting: some panels crop a few percent off every edge by default, which is invisible on a film and cuts the edge off a slide.

LED walls

An LED wall is built from panels of individual light-emitting diodes, tiled together to make a single picture. It is the format that has taken over the top end of the events industry in the last decade, and it is worth understanding even before you work on one.

Pixel pitch is the number that defines a wall: the distance in millimetres between one LED and the next. A 2.6 mm wall has LEDs 2.6 mm apart; a 4 mm wall has them further apart. Smaller pitch means more pixels in the same area, more detail, and more money. The practical consequence is viewing distance — as a rough guide, the closest a wall looks like a picture rather than a grid of dots is somewhere around one metre per millimetre of pitch. A 4 mm wall is fine from four metres and looks like Lego from one. This is why a wall behind a stage can be coarse and a wall people walk past in a foyer cannot.

LED walls are extremely bright — brighter than any projector, bright enough for daylight — which is their headline advantage. They come in modular panels, so a wall can be almost any shape and size, and they have no throw distance, so they fit rooms projection cannot.

What they cost you is weight, power and time. LED is heavy and usually flown, which means rigging, which means qualified people and structural calculations. It draws real power and needs distribution planned for it. And it takes a crew and hours to build, where a small projector takes one person and twenty minutes — though a large-venue projector, flown with its lens chosen, is a rigging job of its own. Every wall also needs its processor — the box that knows the wall's exact pixel layout and turns an ordinary video signal into what the panels want — and the wall's real resolution is whatever the panel count works out to, which is almost never a standard number.

Choosing a display

The honest comparison. Most arguments about which to use are settled by the room and the budget rather than by preference. Read the projection column as a range rather than a figure: it covers more ground than the other two put together.

Pixel pitch and viewing distance

A rule of thumb of roughly one metre per millimetre of pitch. Treat it as the point where the picture stops looking like dots, not as a specification.

On the job

A client wants a 6-metre-wide screen for an awards evening in a hotel ballroom, and asks for a quote on all three options. The room has a ceiling at 4.5 metres, chandeliers, and a bar at the back that stays lit all evening.

Panels are out on size before anything else — 6 metres would be four panels tiled with visible bezels through the middle of the picture, which is not what anybody means by a screen.

Projection is possible and is the cheap answer, but the room fights it. At 6 metres wide in a room with a lit bar you need serious lumens, and the throw distance means the projector goes at the back where the chandeliers are. A short-throw lens from a rigged position solves the chandeliers and costs more than the projector saved.

LED gives the brightest and cleanest picture, ignores the bar entirely, and needs no throw. It also needs rigging at 4.5 metres, a power distribution the hotel may not have near the stage, and a crew for most of a day.

There is no correct answer here — there is a conversation. What makes a technician useful in it is being able to say why each option costs what it costs, which is what turns a quote into advice.

How to do it

Working out whether a display choice will actually work in a room, before anybody quotes it.

1. Get the image size the client actually wants, in metres, and the shape. Everything else follows from this number.
2. Find out how much light is in the room and whether it can be reduced. A lit bar or an uncovered window rules out projection sizes that would otherwise be fine.
3. Measure or ask for the throw distance available, and the ceiling height. A projector that cannot reach and a wall that cannot be flown are both discovered here rather than on the day.
4. Check the destination's native resolution and aspect ratio, and for an LED wall get the actual panel count rather than assuming a standard number.
5. Ask how close the nearest audience will be. For LED this decides the pitch; for projection it decides whether the surface quality matters.
6. Confirm power and rigging early. LED almost always turns into a structural conversation, and so does any projector heavy enough to be flown or stacked — better a week out than on load-in.
7. On the day, feed the display its native format, turn off consumer picture processing, and check for overscan before anybody sees a slide with its edge cut off.

What goes wrong

Practice it

Six real decisions. Work them out before reading the answers.

Putting a Video System Together

Everything so far has been parts. This chapter is the assembly, and the skill it teaches — tracing a chain until you find where the picture stops — is the one that gets a technician called back. It is not a talent and it is not experience. It is a method, and it is short enough to learn today.

The full chain

Every video system, from a laptop and a projector to an arena, is the same five stages in the same order. Learning to see them is what lets you walk into an unfamiliar room and be useful in ten minutes.

Source is where the picture is made: a laptop, a camera, a media server, a playback machine. Transport is how it travels: a cable, a converter, a fiber link, a network. Processing is what happens in the middle: switching, scaling, keying, synchronising. Distribution is how one picture becomes several: a DA, a matrix, an LED processor feeding a wall. And destination is where it lands: a projector and screen, a panel, an LED wall, a recorder or a stream.

A small job has all five stages and some of them are invisible. A laptop into a projector is source, transport, destination — and the projector is quietly doing the processing by scaling whatever it receives. A conference stage has all five explicitly, with named boxes for each. Nothing changes about the model; only how many boxes are in each stage.

The reason to hold this model in your head is that it makes an unfamiliar system readable. You do not need to have used a particular switcher to work out that it is the processing stage, that its inputs come from the transport stage, and that its output goes towards distribution. Systems you have never seen become systems you can reason about, which is a genuinely different feeling from memorising setups.

And it makes the failures readable too. When no picture arrives, the fault is at one of five stages, and each stage has a small number of things that can be wrong with it. That is a much better position than "video is broken".

Tracing and troubleshooting video

Troubleshooting is not a talent. It is a method, and the method is the same every time: find the last place the picture was known good, then move one stage downstream.

The trap that catches everybody at the start is changing more than one thing at a time. You swap the cable, change the input, restart the laptop, and a picture appears — and you have learned nothing, because you do not know which of the three fixed it, and it will happen again next month. Change one thing, look, change it back if it did not help. It feels slower and it is much faster.

The second trap is starting in the middle. The urge is to go to the switcher, because that is where the buttons are. But the switcher will tell you nothing you cannot learn faster by starting at the source and asking, at each stage, whether the picture got that far. A confidence monitor plugged in at each point along the chain answers the question in seconds.

A small number of tools make this trivial and they cost very little. A portable monitor with SDI and HDMI inputs is the single most useful thing in a video technician's bag, because it turns "is the picture here?" from a guess into a look. A pattern generator gives you a known-good source, so you can prove a path without depending on somebody's laptop. Known-good short cables let you substitute one variable at a time. And a torch, because half of this happens under a stage.

What you are really building is the habit of asking what changed. A system that worked an hour ago and does not now has had something done to it — a cable reseated, a device power-cycled, a laptop woken from sleep, a resolution changed by somebody being helpful. The question "what changed?" solves more video faults than any piece of test equipment.

A worked example

A breakout room. One laptop at a lectern, a 5-metre screen, a projector rigged from the ceiling, and a confidence monitor on the floor facing the presenter. Twenty minutes before the session and there is no picture on the screen.

Start at the source. The laptop's own screen is on and showing slides, so the machine is awake and running. Its display settings show a second display connected — which is worth noticing, because it means the EDID conversation happened and something downstream answered.

Move one stage. The HDMI from the laptop goes into a converter at the lectern, which outputs SDI. The converter has a power light and a signal light, and the signal light is on. So the picture reached the converter and it thinks it has something to send.

Move again. Plug the portable monitor onto the SDI at the lectern end — picture, correct, 1080p. The source and the first transport stage are proven good, which has taken about ninety seconds and eliminated half the system.

Now the run. The SDI goes 25 metres to the rack. Plugging the monitor onto the same cable at the rack end gives nothing. The fault is in that run, and it is now a two-minute job rather than a mystery: a known-good cable in its place gives a picture, and the original cable goes in the bin with tape around it rather than back in the box to fail on somebody else's show.

Nothing about this required understanding the projector, the switcher, or the room. It required starting at one end and moving one stage at a time, and it is the same method whether the system has three boxes or three hundred.

The five stages, and what goes wrong in each

Every video system is these five, in this order. When there is no picture, the fault is in one of them.

What to carry

A short list. Between them these turn most video faults from a discussion into a two-minute answer.

• A portable monitor with SDI and HDMI inputs, which is the most useful item on this list by a distance
• An HDMI-to-SDI and an SDI-to-HDMI converter
• A Mini-DisplayPort-to-HDMI and a USB-C-to-HDMI adapter
• Two known-good short HDMI cables and two short BNC cables
• A pattern generator, or an app on your phone that does the same job
• Gaffer tape, a torch, and a marker for labelling

• The show's resolution and frame rate
• Which sources are copy-protected, and what the plan is for them
• The native resolution of every screen on the job
• Which connectors in the system do not lock

On the job

A three-camera panel session, twenty-five minutes before doors, and the centre camera has vanished from the switcher. The other two are fine.

The temptation is to start at the switcher, because that is where the problem appeared. Resist it: the switcher is the last place the picture would arrive and the worst place to start looking for it.

Go to the camera. It is powered, it is on, and its own viewfinder shows a picture — so the source is good. The portable monitor on the SDI at the camera end shows the picture too, so the camera's output is good. That is ten seconds and the whole source stage is eliminated.

The run goes to a floor box and from there to the rack. The monitor at the floor box shows a picture; at the rack it shows nothing. The fault is in the second half, and the second half is a patch through a floor box that somebody moved a riser over during the set change. The cable is crushed.

A replacement run down a different route takes four minutes, and the note in the notebook — "floor box 3, riser sits on it, route round" — is the part of this that saves an hour on the next job in that room.

How to do it

No picture on the screen. This is the order, and it works on any system of any size. Change one thing at a time and never skip to the middle.

1. Ask what changed. A system that worked an hour ago has had something done to it, and the answer is often in that sentence alone.
2. Start at the source and confirm it is making a picture — the laptop is awake, the camera's viewfinder is live, the media player is playing.
3. Confirm the source is set to the right output. A laptop that is extending rather than duplicating, or a camera on the wrong output, looks exactly like a dead cable from further down the chain.
4. Put a monitor on the signal at the source end. This is the moment the guessing stops: either the picture is there or it is not.
5. Move one stage downstream and look again. Converter output, then the far end of the run, then the switcher input, then the switcher output, then the display input.
6. When the picture disappears between two points, the fault is between those two points. Substitute one thing — the cable, then the converter — and change only that.
7. When it works, do not stop. Find out why it failed, because a fault you fixed without understanding is a fault that comes back during the session.
8. Label or bin whatever failed. A bad cable that goes back into the box unmarked will fail again on somebody else's show.
9. Write the room's quirks in the notebook before you leave. The next job in this room is the one that benefits.

What goes wrong

Practice it

The last six. These are the questions that decide whether you get called back rather than just booked once.

Where to Go Next

Video is the widest department in the industry and the one with the most doors into it. What you have here is the trunk; every direction below is a branch, and all of them start from the same understanding of the chain.

• If the switching interested you most, go to Video Switching Fundamentals and then to directing.
• If the pictures interested you, go to camera work — operation first, then direction.
• If the screens interested you, go to LED Video Basics and then to LED processing, which is its own trade.
• If the signal path interested you, go to video engineering and then to AV-over-IP, which is where the whole department is moving.
• If the problem-solving interested you most, that instinct is worth more than any of the above. It is what makes a technician the person who gets called.

Choosing without guessing

Those are honest directions and they are not a decision. What decides whether a route suits you is more ordinary than the subject: how much of your week is travel, whether you would rather be judged on taste or on reliability, whether you like the show or the problem, and what you already have that you are discounting because it did not happen in AV.

Start with the Opportunity Map, at https://reboot-opportunity-map-v1.vercel.app, which draws the industry as a transit map and shows what is genuinely next to the station you are standing on. Somebody who can trace a video chain is closer to a video engineer's seat than they usually believe. Looking around costs nothing and takes a few minutes, which is why it comes first: it is far easier to judge a route once you can see where it goes.

Then, if something on the map has your attention and you want to know whether it actually suits you, take the AV Pathfinder, at https://reboot-pathfinder.vercel.app. It is a short personal quiz, and what comes back is three AV paths that suit you with a Reboot course journey mapped to each — the specific courses, in order, between you and that path.

Both live inside the Reboot community, which is also where the rest of the course library sits, so you will be asked to sign in.

About Reboot AV

Reboot AV exists because the audiovisual industry has a map problem. The work is there, the pay is real, and the ceiling is high, but almost nobody entering it can see the route from where they are to where they want to be. People do not fail in this trade for lack of talent. They fail because nobody ever showed them what the path looked like.

So Reboot AV builds the thing that was missing: a course library covering the full breadth of live-event and installed audiovisual work, an interactive map of the roles in the industry and the routes between them, and an assessment that helps people work out where they actually fit rather than guessing.

Glossary

	Adapter	Changes the shape of a connector without changing the signal. Passive, cheap, and not the same as a converter
	Analog video	A picture carried as a continuously varying voltage. Degrades gradually with distance
	Aspect ratio	The shape of a picture — width against height. 16:9 is standard, 4:3 is the older squarer shape
	BNC	The small round locking connector used for SDI over coaxial cable
	Clean feed	A camera output carrying only the picture, with no menus, focus marks or overlays on it
	Composite	Old analog video on a yellow RCA connector. Standard definition only
	Converter	Changes one signal format into another, such as HDMI to SDI. Needs power
	Combing	Fine horizontal teeth on moving edges. The signature of interlaced material de-interlaced badly
	Cut	An instant change from one source to another. The default transition
	DA / distribution amplifier	Takes one input and produces several identical outputs
	Digital video	A picture carried as numbers. Arrives perfect or fails outright — no gradual degradation
	De-interlace	Weaving the two halves of an interlaced frame back into whole frames. Do it once, early, in the best box
	DisplayPort	A digital video connector common on business laptops and graphics cards
	Dissolve	A soft blend from one source to another over roughly half a second
	Downstream key	A logo or lower third laid over whatever is on program
	DVI	The older digital computer connector. Picture only, no sound, and adapts to HDMI passively
	EDID	The conversation in which a display tells a source what formats it accepts
	Extend	Treating the screen as a second desktop, so slides go out and presenter notes stay on the laptop
	Frame rate	How many still pictures arrive each second. 24, 25, 30, 50 and 60 are the common ones
	Frame synchroniser	Lines up a source that is not running in step with the rest of the system
	Genlock	A shared timing reference that makes every camera start its frames at the same instant
	HDCP	Copy protection on HDMI and DisplayPort. If the source insists and the path cannot carry it, the picture goes black
	Fiber	Video carried as light down a glass strand. Hundreds of metres, immune to interference, and fragile
	HDMI	The common consumer digital connector. Carries picture, sound and control, and does not lock
	Image magnification / IMAG	A camera feed of the speaker put on the main screen so the back of the room can see a face
	Interlaced (the i)	A frame sent as two halves — odd lines, then even. A broadcast-era trick to halve the data
	Keystone	Digital correction of a picture projected at an angle. Costs sharpness — a rescue, not a plan
	LED processor	The box that turns a video signal into what a specific LED wall's panels need
	LED wall	A screen tiled from panels of light-emitting diodes. Very bright, modular, heavy
	Lumens	The measure of a projector's light output
	Matrix router	Any input to any output, with no production features. The plumbing of a large system
	Media player	A machine that exists to play files reliably. No notifications, no sleep, one format all day
	Media server	A computer built to play the show's content out reliably as a video source
	Mirror	Showing the same picture on the laptop and the screen. What a nervous presenter usually wants
	Overscan	A display cropping a few percent off every edge by default. Invisible on video, fatal on a slide
	Picture-in-picture	A second source shown in a window inside the main picture
	Progressive (the p)	A whole frame, every line, delivered at once. What almost everything modern uses
	Pixel	One dot of the grid a digital picture is made of
	Pixel pitch	The distance in millimetres between one LED and the next. Smaller means more detail and more money
	Preview / PVW	What you are lining up next, seen only by you
	Program / PGM	What the audience is seeing right now
	Rec. 709	The standard colour space for HD, and what almost all events work in
	Resolution	How many pixels a picture has — 1920 x 1080 for HD, 3840 x 2160 for UHD
	Scaler	Changes a picture's resolution or frame rate so a source and its destination can agree
	SDI	The professional digital video format, on BNC over coax. Runs 100 m and more, no copy protection
	Switcher	Chooses which source goes to the screen, with a transition between
	Throw ratio	Lens distance divided by image width. It decides where a projector can physically go
	Transition	How the picture changes from one source to the next
	Triax	One thick cable carrying a broadcast camera's picture, power, talkback and return feed over long distances
	USB-C	A connector that may or may not carry video. Nothing about the socket tells you which
	VGA	The old blue 15-pin analog computer connector. Picture only, no sound
	Wipe	A transition that pushes one picture across another

What the gear actually looks like

Every device named in this book, in one place. Recognition is the whole point of Chapter 4 — use this page to attach a picture to each name before you meet it on a dock at six in the morning.

Audio

Dynamic microphone

Condenser microphone

Wireless handheld

Radio rack

Mixing console

Stage box / snake

Power amplifier

Main loudspeaker

Subwoofer

Stage monitor wedge

In-ear monitors

DI box

Video

Camera

Video switcher

Scaler / processor

Projector

LED wall panel

Confidence monitor

Lighting & control

Conventional fixture

Moving-head fixture

Lighting console

Dimmer

Show controller

Media server

Networking & infrastructure

Network switch

Patch bay

Equipment rack

Power distribution

Cable tester

A properly coiled cable

Connectors, close up

XLR

TRS / TS ¼-inch

SDI on BNC

HDMI

Cat5e / Cat6 (RJ45)

Fiber optic

DMX (5-pin XLR)

AC power — not signal