Just press the green flag. It says a random number from 1 to 20 every second.
You might be wondering how to get random numbers without the random number block, being that a scratch script will do the same thing each time. Therefore, I would need to find an external variable that would be different each time it was referenced. Scratch only pulls 2 things from outside the project: username, cloud variables, and time. The username wouldn't work because it would always pull the same random numbers each time the user ran the project. You could always have a cloud variable that changed based on an in-project variable, such as mouse position, but the player could find a way around that and control the random number generator. That leaves one option: time. Scratch can return the year, month, day, hour, minute, and second. This means that when the player ran the program, the random number generators would be different each second. However, if you wanted to use random numbers more than once per second, this would not work. There is still one more option that changes every frame, and the user has no control over: Days since 2000. Yes, days since 2000. The seemingly most useless block in all of scratch is the foundation of this project. While days is a large unit of time to measure, days since 2000 is calculated to 12 decimal points. That's to one-trillionth of a day. Since each day is less than 100000 seconds, that means that anything past one millionth of a day (6 decimals) will be changing with every frame. Using this I can create a list of 100 random digits fetched from the days since 2000 value, and use them to generate the random number. Now that I have my random seed (the list of digits), I can use it to find random numbers. When I need to fetch a random number, I find the difference between the minimum and maximum values. I take the logarithm of it (finding how many times it can be divided by 10 until it is less than or equal to 1), add 1 to the logarithm, and fetch that many digits. When I join the digits, I come up with a number that is well above the difference. I divide said number by the difference between the minimum and the maximum, at which point I have a remainder between 0 and said difference. Since the minimum may be higher than said number, I keep adding said difference to it until it is within the bounds of the minimum and maximum.
