Feel free to shove the code below in a system.vc and give it a go. Basically:
-Seeding the randomiser with a constant doesn't give a constant output
-The interface between b1 and key[SCAN_ENTER] is unreliable, and there's no way of 'pressing' the buttons apart from using key[]
-Oh, and is the switch() return bug gonna be fixed? :D

#define I_LOVE_LIFE 50


int numbers[I_LOVE_LIFE]; // Some numbers





void autoexec()


{


	int i; // A looper


	// If the next line is in, enter will automatically be pressed once,


	// but ignore the press after that. If it's commented out enter


	//  works as would be expected, there is no Press(1)


	key[SCAN_ENTER] = 1;


	while (!b3) // If escape is not pressed


	{


		if (b1) // If enter is pressed


		{


			// Set the seen to a constant


			// (so the next number should always be the same)


			SetRandSeed(1);


			// Generate a string of random numbers


			for (i = 0; i < I_LOVE_LIFE; i++)


			{ numbers[i] = Random(0, 100); }


			// Randomize the seed again (keep up that good practice)


			SetRandSeed(0);


			unpress(b1); // Unpress enter


		}


		// Reset the screen buffer to plain black


		RectFill(0,0,ImageWidth(screen),ImageHeight(screen),0,screen);


		// Print our random numbers


		for (i = 0; i < I_LOVE_LIFE; i++)


{ PrintCenter(ImageWidth(screen)>>1, i * 10, screen, 0, str(numbers[i])); }


		ShowPage(); // Displays the buffer on the screen


	}


	exit("Bye!");


}

Zip

[Edit: Fixed the display on my 1024x768]

>> is bitshift right - it's used to divide by multiples of 2. He shifted to the right once, so it has the same effect as dividing by 2, only a whole lot faster.

thats awsome

=> is not a valid operator as far as I know. The verge compiler dosen't catch it as an error, but try and put that in a c compiler and it will accuse you of rape

Umm, seeding the random number generator is a way to help it produce more random output that by not seeding it. Random number generators shouldn't produce constant output ;)

PRNG?

PRNG = Pseudo-Random Number Generator


Edit: Everything that follows below is likely much better explained by the Wikipedia article that Zip posted, that (were I smart enough) I would have just pointed you towards rather than writing my own pamphlet on it. -_-;

---

It is impossible to generate true random numbers on a computer, since they are deterministic machines. (Hell, it's pretty much impossible for humans to come up with random numbers too.) The closest they've gotten (far as I know) is machines that produce numbers by performing maths on certain atmospheric conditions that change constantly (like various kinds of particles striking the outer atmosphere, the levels of magnetism in the Earth's magnetosphere, whatever). Obviously something like that is somewhat out of the realm of possibility for a VERGE game.

Anyway, so we're stuck with pseudo-random number generators. They take a seed number, and perform wacky arithmetic to them (I've seen random number generator algorithms before, and let me tell you: they make no sense) taking advantage of modulus, big numbers, the wrapping around of values, etc. etc. If you provide the same seed, then you will always get exactly the same string of numbers out of the generator. This is useful if you wanna test something and always get consistent results (which is why we were asking for a seed-specifier).

Specifying a number is not so useful if you actually do want it to be random. The classic way, and the easiest way (and what I assume to be the VERGE way) is to assign the current second count since 1980 (or 1960 or whatever the epoch date for your operating system is) as the seed. This pretty much guarantees that any non-trivial program will not execute the same twice. On the other hand, if you run a program twice in the same second, it will be the same.

I'm sure that was more response than you intended to elicit from saying "PRNG?" but I tend to ramble sometimes. Have a good day. :)

And anyone who sees that I said something wrong, please correct me, as I like to know what I'm talking about. ^_^

Hehe, I just about beat Kildorf's polite, clear and well explained post with my rude and silly one. :D

For the curious, a PRNG I use in c:

//picks a random number from 0 to max-1


unsigned long randomn(unsigned long max)


{


	r_num();


	long double rand_y;


	long double rand_i;


	rand_i = 2147483648L;


	rand_y = max*((long double)seed/rand_i);


	return((unsigned long)rand_y);


}





//sets seed to a random number from 0 to 2 billion


unsigned long r_num(void)


{


	seed=(seed*907725L+99979777L)%2147483648L;


	return seed;


}

Zip

Yes, and I presume vec's SetRandSeed(0) does something along those lines - you certainly don't want to use the time as a factor DURING any number generation though, otherwise getting constant output is impossible (and you wan it sometimes). In the example above, for instance, the seed is a global variable so I can set it based on time if I want, but will always generate the same string of numbers if given the same seed.

Zip

No no, it's up-up-down-down-left-right-left... wha... nuclear deployment plans for Turbekistan, timetable urgent? This isn't infinite lives!

(closes program)