GDL> TIME_COMPARE, ['idl7','g09', 'g101018','g110330','g111011']
      idl7: |TIME_TEST3 performance for IDL 7.1:
      idl7: |       OS_FAMILY=unix, OS=linux, ARCH=x86_64
      idl7: |	Tue Oct 11 23:45:49 2011
       g09: |TIME_TEST3 performance for IDL 6.0:
       g09: |       OS_FAMILY=unix, OS=Linux, ARCH=x86_64
       g09: |	Tue Oct 11 23:56:50 2011
   g101018: |TIME_TEST3 performance for IDL 6.0:
   g101018: |       OS_FAMILY=unix, OS=Linux, ARCH=x86_64
   g101018: |	Tue Oct 11 23:57:18 2011
   g110330: |TIME_TEST3 performance for IDL 6.0:
   g110330: |       OS_FAMILY=unix, OS=Linux, ARCH=x86_64
   g110330: |	Tue Oct 11 23:45:33 2011
   g111011: |TIME_TEST3 performance for IDL 6.0:
   g111011: |       OS_FAMILY=unix, OS=Linux, ARCH=x86_64
   g111011: |	Tue Oct 11 23:43:49 2011

    Time    idl7     g09 g101018 g110330 g111011
    0.04     102     102     842*    102     100^  Empty For loop, 2000000 tim
    0.01     100^    181*   1238*    197*    181*  Call empty procedure (1 par
    0.01     100^   2048*   3102*   2297*    166*  Add 200000 integer scalars 
    0.01     100^   1817*   2749*   1946*    249*  50000 scalar loops each of 
    0.00     320*    980*   2696*   1240*    100^  Mult 512 by 512 byte by con
    0.01     100^   1384*   2974*   1901*   1373*  Shift 512 by 512 byte and s
    0.01     250*    803*   2450*    947*    100^  Add constant to 512x512 byt
    0.01     234*    575*   2013*    728*    100^  Add two 512 by 512 byte arr
    0.00     362*   1495*   3331*   1555*    100^  Mult 512 by 512 floating by
    0.01     100^    528*   1037*    915*    507*  Shift 512 x 512 array, 60 t
    0.00     477*   1493*   3761*   1671*    100^  Add two 512 by 512 floating
    0.01     119*    100^    175*    112     101   Generate 1000000 random num
    0.01     100^    178*    185*    182*    182*  Invert a 192^2 random matri
    0.00    9355* 141575*   1925*   1500*    100^  LU Decomposition of a 192^2
    0.02     100^    634*   2619*    654*    196*  Transpose 384^2 byte, FOR l
    0.02     100^    280*    833*    279*    278*  Transpose 384^2 byte, row a
    0.03     139*    109     715*    137*    100^  Transpose 384^2 byte, TRANS
    0.02     100^    736*   2853*    798*    537*  Log of 100000 numbers, FOR 
    0.00     344*    102     183*    134*    100^  Log of 100000 numbers, vect
    0.01     160*    121*    155*    125*    100^  131072 point forward plus i
    0.03     100^    726*   1827*    813*    520*  Smooth 512 by 512 byte arra
    0.01     100^    598*   2251*    610*    604*  Smooth 512 by 512 floating 
    0.02     258*    247*    388*    245*    100^  Write and read 512 by 512 b
    0.40     100^    430*   1119*    475*    217*  Total Time
    0.01     100^    317*    663*    294*    101   Geometric mean
 
^ = fastest.
* = Slower by 15% or more.
Tue Oct 11 23:58:46 2011