Swap  Reaction race

Simon Says needs four lights, four buttons, four resistors and a buzzer, joined by about twenty wires. That is thirteen parts, and if it does not work you have thirteen suspects.

This build needs one light, one button and a buzzer. It teaches every single idea in Band 5:

• Functions you write yourself, with values handed in and answers handed back
• tone and pitch
• random, and why it needs seeding
• millis, and the difference between waiting and watching
• A job broken into pieces small enough to name

If four buttons is more wiring than you want to fight with today, come here. You are not taking an easier band. You are taking the same band with less to go wrong at the bench, which is a reasonable thing to want.

The light's pin wire goes in b1, not a1: the resistor leg is already there and a hole takes one leg only.

a. Read loop first

Go to the bottom of the sketch. The whole round is there and it reads like a description of the game: get ready, wait a random time, check for cheating, light on, time the press, compare with the best. Everything complicated is behind a name.

b. The function that hands back an answer

unsigned long timeOnePress() {
  unsigned long startedAt = millis();
  while (!isPressed()) {
    if (millis() - startedAt > 10000) {
      return 0;
    }
  }
  return millis() - startedAt;
}

This one starts with unsigned long rather than void, because it hands something back. That is what return does, and it is the half of functions that Band 5 Part 5 introduces.

Inside it is a while loop that does nothing but watch. That is blocking, on purpose, and it is fine here because there is genuinely nothing else the game needs to do while it waits for a press. Band 5 taught you when blocking is a problem; this is what it looks like when it is not.

The if in the middle is the one escape. After ten seconds with no press it gives up and hands back 0, and loop reads that 0 as “no press seen” and starts the round again. Without it, a badly wired button would leave the board sitting there for ever looking dead, with no message and nothing to work from. Any loop that waits for the outside world needs a way out that does not depend on the outside world. Write that in your log; it is one of the most useful sentences in the course.

c. The function that catches cheating

bool waitWithoutCheating(unsigned long howLong) {
  unsigned long startedAt = millis();
  while (millis() - startedAt < howLong) {
    if (isPressed()) {
      return false;
    }
  }
  return true;
}

d. Why A0 must be empty

randomSeed(analogRead(A0));

random is not really random. It runs a calculation, and from the same starting point it produces the same sequence every time. Without seeding, this game would wait exactly the same series of times after every power-up, and you could learn them.

analogRead on a pin with nothing connected gives a drifting, meaningless number, because the pin is floating. You met floating pins in Band 3 as a problem. Here it is the solution. Meaningless is exactly what a seed needs to be.

Bronze

Make it a game of five rounds that then prints your average as well as your best. You will need somewhere to keep a running total and a count, and you will meet whole-number division from Band 4 on the way.

Silver

Make it a two-player game: a second button, and whoever presses first wins the round. This is harder than it sounds. You cannot use timeOnePress as it stands, because it watches only one button, and the honest fix is to write a function that watches both and hands back which one won.

Gold

Add a false start that is properly fair. First move the LED's pin wire from pin 7 to pin 6 and change ledPin to match, because pin 6 has a ~ and pin 7 does not, and you cannot dim a light on a pin without one. Then light the LED dimly for a moment at a random point before the real signal, and count a press during that as a fault. Then, using millis throughout, make sure the game never stops responding, even during the penalty.

This Swap replaces Simon Says as the build you are assessed on. To claim Band 5 you still need:

• A video of somebody playing it, which may be you if you are working alone, and your own best time
• The list of your own functions with one sentence each saying what they do
• Your notes from the counting activity in Part 1
• Your before-and-after waits from the randomSeed experiment
• Your build log
• Your completed timing table from Part 10a, done on paper with b5_04. This Swap does not replace it
• Five or six correct answers in Part 13, the Check yourself questions, including question 4

Part 13's question 4 is the one that matters most in the band: a delay hidden inside a function that looks like it is using millis properly from outside. You have now built both kinds of waiting on purpose, so you are in a better position to answer it than somebody who only built the game.