It's very interesting to compare different programming languages benchmark on the HP 50g using the N-Queens Problem:

userRPL:

22.6sHRAST Basic:

7.2ssysRPL: 2.32s

HP Pascal Studio 2.1 / Cross Compiler:

0.895sHP Pascal Studio 2.1 / Cross Compiler / Fast Mode x1.5 (75->203 MHz):

0.622snewRPL:

0.106sHPGCC 3.0 / RegVars / Cross Compiler / Slow Mode x12.7 (75->6.0 MHz):

0.00613sAssembly / ARM9 @ 6.0 MHz / Slow Mode x13:

0.00524sHPGCC 3.0 / RegVars / Cross Compiler:

0.000484sAssembly / ARM9 @ 75 MHz:

0.000404sAssembly / ARM9 @ 192 MHz / Fast Mode x2.7:

0.000150sHPGCC 3.0 / RegVars / Cross Compiler / Fast Mode x2.5 (75->192 MHz):

0.000192suserRPL

`<<`

0.

DO

8. SWAP 1. +

WHILE

DUP2

DO 1. -

UNTIL

DUP2 5. + PICK - ABS

DUP2 - * NOT

END

REPEAT

DROP

WHILE SWAP DUP 1. SAME

REPEAT -

END

1. - SWAP

END

DROP

UNTIL DUP 8. SAME

END

->LIST

>>

HRAST Basic`CALL QUEEN,8`

PROC QUEEN(INTEGER N) INTEGER A[N+1],D,S=1,X,Y A[1]=N

FOR X=2 TO N S> A[X]=N FOR Y=X-1 D=A[X]-A[Y]

IF NOT D OR X-Y=ABS D@ A[X]< WHILE NOT A[X]@ X< A[X]< ENDWHILE S> Y=X

NEXT NEXT ? S

Reference:

**Calculator Speed Benchmark using the N-Queens Problem**http://www.hpmuseum.org/cgi-sys/cgiwrap/hpmuseum/articles.cgi?read=700