Creating multiple Selection sets

[pBe]

Three more variations of a possible XOR function, for fun:

 

I was looking for the thread where you explain logand/Boole/Logior. it took me a while to find it:

 

http://www.cadtutor.net/forum/showthread.php?61237-I-want-just-Blocks-and-not-Xrefs/page2&highlight=LOGXOR

[Lee Mac]

Ah yes... bitwise logic

 

Here's another:

 

http://www.cadtutor.net/forum/showthread.php?45257-Simple-Setvar-request...&p=306339#post306339

[irneb]

Nice explanation Irne

Thanks Lee, nice alternatives. Though "logically" ... the one I've shown "should" be more efficient, since it uses only 4 function calls internally. Those you've shown uses 5, 5 & 7 respectively. I think we have discussed this previously, and come to the conclusion that practically there's not much in it - the differences are minute.

Benchmarking ...................Elapsed milliseconds / relative speed for 65536 iteration(s):

   (XOR nil nil)......1435 / 1.04 <fastest>
   (XOR1 nil nil).....1435 / 1.04
   (XOR3 nil nil).....1466 / 1.02
   (XOR3 T nil).......1466 / 1.02
   (XOR nil T)........1466 / 1.02
   (XOR2 nil T).......1466 / 1.02
   (XOR2 nil nil).....1467 / 1.02
   (XOR2 T T).........1467 / 1.02
   (XOR T nil)........1482 / 1.01
   (XOR1 T nil).......1482 / 1.01
   (XOR2 T nil).......1482 / 1.01
   (XOR1 nil T).......1482 / 1.01
   (XOR T T)..........1482 / 1.01
   (XOR3 nil T).......1498 / 1
   (XOR1 T T).........1498 / 1
   (XOR3 nil T).......1498 / 1 <slowest>

I've renamed yours to xor1, -2, & -3.

 

As you see, the one I gave tends to be faster. But only fractionally (seems to be under 1%), the actual values seem to have a greater impact.

[Lee Mac]

Oh of course, mine aren't too readable either - I was just having fun with it