Why This Class Exists

Content.

The whole module is won or lost here. The goal is not to teach a protocol. The goal is that a year one lighting student leaves the room believing that a computing class is their class, not something bolted on by the curriculum committee.

Learning outcomes#

By the end of this session a student can:

1. State the spine sentence in their own words and give one example from their own specialism.
2. Sort any piece of show traffic into one of the Four Flows and say what would break it.
3. Explain the difference between latency and jitter, using a physical example.
4. Estimate whether a given signal will fit down a given wire, to an order of magnitude.
5. Trace one signal in their own specialism end to end, naming what it is, where it goes and what its deadline is at every stage.
6. State the six working habits of productive diagnosis and explain why each one exists.
7. Say what they will be assessed on and what to bring to the production visit.

Block plan#

Block 1: Cold open#

This class opens with a rig already running, before anyone has said a word about a syllabus.

What is set up: one laptop, one switch, three destinations.

• Laptop sends sACN or Art-Net to a node, which drives a moving light. The light is moving.
• Laptop sends audio via Dante Virtual Soundcard (or an analogue send if Dante is not available) to a speaker. Music is playing.
• Laptop sends NDI to a second machine driving the projector. Video is on screen.

It runs for thirty seconds. Then the question: how many cables are carrying the show right now?

Most people count three or more. It is one: a single Cat lead from the laptop to the switch. Light, sound and picture, all on one wire.

Then the lead comes out. Everything stops at once: the light freezes or goes to its last look, the audio drops, the video freezes on a frame.

It goes back in, and the show recovers. The sentence that follows is the spine of the whole module:

Every signal in a modern show is, at some point, a number in a computer's memory, travelling over a shared network, with a deadline.

That is the module.

Watch how the three departments failed differently, because it is the first real thing this class teaches. The light held its last value, the audio stopped dead, the video froze on a frame. Three identical cable pulls, three different failures. Why that happens is Class 4, and it is worth carrying the question until then.

Block 2: The argument#

The three claims#

Deliver these as claims to be defended, not facts to be copied.

Claim 1: your instrument is now a computer.

Walk the arc for each specialism. Keep it fast, one minute each.

• Lighting. Resistance dimmer, to analogue 0 to 10 V, to DMX512 in 1986, to networked control in the 2000s, to a modern fixture that boots, runs firmware, has an IP address and can be updated over the network. Your moving light is a computer with a lamp in it.
• Audio. Analogue desk, to digital desk with a DSP engine, to a mixing surface that is a control surface for a computer somewhere else in the building. Every plugin you love is a program. The desk is a computer with faders on it.
• Video. Slide projector, to VHS, to a media server that is explicitly and openly a computer with a professional graphics card. Nobody pretends otherwise here.

Claim 2: your cable is now a network.

The industry replaced dedicated single purpose cables with a shared general purpose one.

The trade is always the same: flexibility bought with configuration risk. That is the sentence worth writing down.

Claim 3: your craft judgement is now a systems judgement.

Concrete examples, one per specialism, phrased as artistic decisions:

• Choosing a Dante latency setting of 0.25 ms or 5 ms is deciding whether a musician's in ear monitor feels tight or feels wrong.
• Choosing 8 bit or 16 bit dimmer control is deciding whether a slow fade to black is smooth or visibly stepped.
• Choosing a codec is deciding whether a fast camera move on a 12 m LED wall is crisp or mushy.

None of those are IT decisions. All of them are made by the artist, and all require the numbers.

Block 3: The Four Flows#

The model is worth having in your head before the examples, because you will use it immediately.

Card sort exercise (15 minutes)#

A deck of about 24 cards, one item per card, sorted in threes into the four columns and then defend two of their choices.

Suggested cards, with the intended answer and the teaching point:

Debrief question, the one that matters: "Which flows can share a wire safely, and which cannot?" That one is not answered yet. Class 3 answers it. A real question left open across three weeks is good pedagogy.

Break#

Block 4: The two numbers#

Data rate maths#

Work it through slowly, out loud, with the numbers in front of you.

Audio, one channel, CD style quality but professional rate:

48,000 samples per second  ×  24 bits per sample  =  1,152,000 bits per second
                                                  =  1.152 Mbit/s per channel

So 64 channels is about 74 Mbit/s of actual audio, call it 100 Mbit/s once the network overhead is added. A 1 Gbit/s Cat lead swallows that without noticing. This is why a single Cat5e can replace a 48 pair analogue multicore that took two people to coil.

Video, one HD picture:

1920 × 1080 = 2,073,600 pixels per frame
× 60 frames per second = 124,416,000 pixels per second
× 20 bits per pixel (10 bit, 4:2:2 colour) = 2,488,320,000 bits per second
                                           ≈ 2.5 Gbit/s

That does not fit down a 1 Gbit link. One uncompressed HD picture is bigger than a thousand channels of audio. Say that out loud, because it rearranges their intuitions permanently.

Lighting, for scale:

DMX512: 512 slots at roughly 44 refreshes per second ≈ 0.25 Mbit/s per universe

A hundred universes of lighting control is still smaller than one channel pair of audio. Control is tiny. Media is enormous. Clock is nothing. Management is unbounded.

The ranking, worth keeping in front of you: clock < control < audio < video.

Latency budget, and the physical anchor#

Sound travels at about 343 metres per second, so roughly 3 milliseconds per metre.

Ask a student in the back row how far they are from the speaker. Say 15 m. That is 44 ms of delay they are currently experiencing and have never once complained about.

Here is a latency chain for a typical digital audio path, added up:

The lesson lands by itself: the entire digital signal chain costs less than three metres of air. Latency, when it is known and constant, is a budget you spend deliberately.

Latency versus jitter#

• Latency is delay. Constant delay can be measured and compensated. A delay tower is literally a latency compensation device, and nobody calls it a fault.
• Jitter is delay that changes. You cannot compensate for a number that will not sit still. Your only defence is a buffer, and a buffer costs latency.

Perceptual anchors worth carrying from here on:

Note the asymmetry and explain it: we are evolved to accept sound arriving after picture, because thunder always follows lightning. Sound arriving before picture is unnatural, and we catch it three times faster. This is a good example of a technical spec derived from perception.

Block 4b: Trace one signal, all the way#

This is the exercise the rest of the module refers back to, and it is worth doing properly rather than describing.

The three questions. At any point in any chain, in any department, there are only three questions worth asking, and you will still be asking them in twenty years:

1. What is it here? Air pressure, a voltage, a number in memory, a packet on a wire, a photon.
2. Where is it going next? Which box, over what, sharing with what else.
3. What is its deadline? Now, this frame, this millisecond, or whenever.

Answer those at every stage of a chain and you can diagnose it. Fail to answer them and you are guessing, and guessing on a show is expensive.

The exercise, in pairs, on paper#

Take the chain from your own specialism. Draw it as boxes and arrows, left to right, from the human at one end to the human at the other. For every arrow, write:

• what the signal is at that point
• which of the Four Flows it belongs to
• what happens if that arrow is cut

Fifteen minutes, then swap with a pair from a different department and find one thing in their chain that they have not accounted for. You will find one. Everybody does, and it is almost always the clock, because clock is the flow with no obvious cable.

What the exercise is really teaching. Nobody can hold a whole show system in their head. What you can hold is a method for walking one, and the method is the three questions. That is why this appears in Class 1 rather than at the end.

Block 4c: How to be wrong productively#

Everything after this class is diagnosis, so the working method is worth stating on day one. These are not study tips. They are how the job is done, and a first year who works this way is more useful in a technical rehearsal than a third year who does not.

Evidence beats intuition, every time. "It should work" has never fixed anything. Look at the link light, read the address, check the number. Your intuition is a way of choosing what to look at, not a substitute for looking.

Change one thing. Change two and you have learned nothing, whichever way it goes. This is slower for the first ten minutes and much faster for the next hour.

Write down what you changed. A show has fifty people touching it. Half of all "mystery faults" are somebody else's undocumented fix from yesterday, and the other half are your own.

Halve the problem. Do not walk a chain end to end. Test in the middle: is it working here? Now you have eliminated half the system with one measurement. Six of those and you are inside any rig in the world.

Know what normal looks like. Record the healthy state while it is healthy. "Is 12 percent bad?" is unanswerable unless somebody wrote down what it was on a good day.

Say what you actually know. On headset, "the node has no link light" is useful and "the network is broken" is not, because the second one is a conclusion and it might be wrong. Report the observation, then your best guess, and mark which is which.

These six lines are examined. They are also the shortest description of the difference between somebody who is trusted with a rig and somebody who is not.

Block 5: Course map and expectations#

The shape of the module, quickly:

• The shape: 2 intro, 2 visit, 16 content, 4 practical exam.
• The assessment map and the four criteria rubric. Spend the most time on criterion 3, "can someone else run it", because it is the one most people underestimate.
• The glossary is a working tool. Bring it every session. Add to it.
• The Numbers card is a memorisation task. There is a five minute verbal quiz at the start of every session. It carries no marks and it is not optional.
• Software install list, due before Class 2.

Block 6: Visit briefing#

• Venue, meeting point, travel time, arrival time. Arrive early, a production visit does not wait.
• Dress: closed shoes, dark clothing, no loose jewellery. Bring a torch if they own one.
• Etiquette: touch nothing, ask before photographing anything, never step over a cable, do not stand in a doorway, keep off headset channels, and if someone says move, move first and ask after.
• Teams of three, allocated now, not on the day.
• The observation sheet comes with the production visit brief, and it is worth reading through for two minutes so they know what they are hunting for.

Common misconceptions in this session#

Reading a wrong belief with its correction attached does very little. Spot the myth, on the Practice tab, puts each of these claims in front of you bare and asks you to commit before the correction appears.

• "So it is all just IT." No. IT optimises for throughput, security and uptime averaged over a day. Show networks optimise for a guaranteed deadline at 19:32 tonight. The priorities are genuinely different, and that difference is why show networks are configured by MDT people.
• "Wireless would be easier." Wireless is a shared, contended medium with no delivery guarantee and a variable delay. It is acceptable for an operator's tablet. It is not acceptable for anything that must happen.
• "Latency is bad." Latency is a budget. Say it every time you hear this.
• "The audience cannot tell." The audience cannot name it. They can absolutely feel it.

Sources and further study#

If this class interested you, the map of where each direction leads, what is free and what costs, is on Where to go next.

The index this module checks itself against#

showstack, the open index of live entertainment technology. Protocols with their ports and multicast ranges, standards with their designations, hardware and software with what each one actually speaks, and bilingual EN / 繁中 vocabulary, with a citation on every fact and a free JSON API.

Three ways to use it in this module:

• Check a number before you repeat it. Every port and date in these notes was verified against it.
• The interop checker: pick two products, get the protocols they share and in which direction. That is the question this whole module is teaching you to ask.
• The field tools, 62 of them, for the calculations this module does not examine but the job needs: DIP switch addressing, bridle angle, voltage drop, noise dose, RF intermod, beam angle.

Everything here is free and checkable. Two habits worth starting now: read the actual standard rather than a summary of it, and read the manufacturer's own documentation rather than a forum post.

Where the rules actually live. Somebody decided each of these and wrote it down, and knowing that the documents exist is itself professional knowledge.

The ESTA one matters most to you. Every published ESTA standard is free, and the two you will use for the rest of your career are linked in Class 4.

A note on the links in this module. Standards bodies and manufacturers keep their documents at stable addresses. Third-party videos do not always survive. If a link here is dead, search the title and the channel rather than assuming the resource has gone.

Homework before the production visit#

1. Read the study guide section for Class 1 and self test.
2. Learn the first block of the Numbers card (the data rate numbers).
3. Read the observation sheet. Write down three things you expect to see at the venue and one question you want answered. Bring both to the visit.