Lets see if I get it right...
The Sint32 which we are using as Object::ID (only the positive values)
has the range:
-2,147,483,648 to 2,147,483,647
That gives up 2147 Million IDs we can use for Objects.
Lets say one object has only 100 bytes data.
(it has a lot more, maybe even like 1000 on average, but anyway):
100 x 2,147,483,647 = 214,748,364,700 bytes
Now we all know that 1 Gigabyte (GB) = 1,073,741,824 Byte.
214,748,364,700 / 1,073,741,824 = 199.9999999068677425384521484375 GB =
200 GB RAM needed.
I hope thats right. :P
Anyway, we should be safe using only the positive range value of the
unsigned INT32 Object IDs
til at least 2020, and then I will flip a switch and make it an unsigned
INT32 and we are good til
at least 2040.