[Harbour] 2008-10-03 21:53 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)

Fri Oct 3 16:38:13 EDT 2008

Hi Przemek,

[resent to avoid line wrapping]

P4 2.66/800 HT, XP, MSVS2008 -gc3:

Startup loop to increase CPU clock...
10/03/08 22:23:07 Windows XP 5.1.2600 Service Pack 3
Harbour 1.1.0dev (Rev. 9539) (MT) Microsoft Visual C 15.0.26569 (32 bit)
N_LOOPS =    1000000

                                                      speedtst
                                            -gc0    -gc3    -gc0    -gc3
                                                              MT      MT
-----------------------------------------------------------------------
[ empty loop overhead ]....................0.13 ...0.03 ...0.09 ...0.06
=============================================== ======= ======= =======
[ x := L_C ]...............................0.20 ...0.13 ..29.09 ...5.27
[ x := L_N ]...............................0.05 ...0.05 ..11.69 ...1.53
[ x := L_D ]...............................0.02 ...0.03 ..16.94 ...3.92
[ x := S_C ]...............................0.17 ...0.14 ..27.41 ...5.06
[ x := S_N ]...............................0.06 ...0.03 ..26.14 ...3.95
[ x := S_D ]...............................0.05 ...0.00 ..23.11 ...4.23
[ x := M_C ]...............................0.20 ...0.11 ..24.00 ...6.58
[ x := M_N ]...............................0.00 ...0.13 ..19.19 ...6.88
[ x := M_D ]...............................0.06 ...0.05 ..23.89 ...4.41
[ x := P_C ]...............................0.22 ...0.13 ..35.14 ..12.75
[ x := P_N ]...............................0.03 ...0.08 ..36.55 ...4.50
[ x := P_D ]...............................0.06 ...0.02 ..23.91 ...5.83
[ x := F_C ]...............................0.77 ...0.39 ..55.25 ..41.00
[ x := F_N ]...............................0.55 ...0.30 ..47.52 ..23.03
[ x := F_D ]...............................0.23 ...0.08 ..24.73 ...7.06
[ x := o:GenCode ].........................0.53 ...0.16 ..38.81 ..19.94
[ x := 0 ].................................0.02 ...0.06 ..20.70 ...1.67
[ round( i / 1000, 2 ) ]...................0.64 ...0.47 ..43.63 ..23.92
[ str( i / 1000 ) ]........................1.61 ...1.27 ..81.23 ..53.89
[ val( s ) ]...............................0.80 ...0.78 ..39.22 ..28.03
[ val( a [ i % 16 + 1 ] ) ]................1.53 ...1.39 ..55.86 ..32.16
[ dtos( d - i % 10000 ) ]..................1.56 ...1.14 ..93.64 ..60.39
[ eval( { || i % 16 } ) ]..................2.53 ...2.06 .104.80 ..77.70
[ eval( bc := { || i % 16 } ) ]............0.83 ...0.83 ..53.44 ..14.50
[ eval( { |x| x % 16 }, i ) ]..............1.73 ...1.42 ..99.72 ..62.42
[ eval( bc := { |x| x % 16 }, i ) ]........0.88 ...0.78 ..46.69 ..12.03
[ eval( { |x| f1( x ) }, i ) ].............1.81 ...1.63 .102.41 ..58.31
[ eval( bc := { |x| f1( x ) }, i ) ].......0.45 ...0.80 ..56.55 ..11.64
[ x := &( "f1(" + str(i) + ")" ) ].........3.81 ...9.55 .149.52 .112.41
[ bc := &( "{|x|f1(x)}" ); eval( bc, i ) ..9.13 ..12.36 .150.80 .116.25
[ x := valtype( x ) +  valtype( i ) ]......1.00 ...0.94 ..88.27 ..56.63
[ x := strzero( i % 100, 2 ) $ a[ i % 16 ..1.59 ...1.58 ..95.48 ..69.97
[ x := a[ i % 16 + 1 ] == s ]..............0.70 ...0.73 ..46.44 ..24.19
[ x := a[ i % 16 + 1 ] = s ]...............0.73 ...0.70 ..56.73 ..27.17
[ x := a[ i % 16 + 1 ] >= s ]..............0.83 ...0.88 ..52.41 ..27.44
[ x := a[ i % 16 + 1 ] <= s ]..............0.94 ...0.84 ..53.38 ..26.97
[ x := a[ i % 16 + 1 ] < s ]...............0.73 ...0.92 ..54.41 ..27.22
[ x := a[ i % 16 + 1 ] > s ]...............0.78 ...0.70 ..54.78 ..27.38
[ ascan( a, i % 16 ) ].....................0.50 ...0.72 ..48.84 ..25.19
[ ascan( a, { |x| x == i % 16 } ) ]........6.92 ...6.58 .119.34 ..82.98
[ if i%1000==0;a:={};end; aadd(a,{i,1,.T...3.09 ...2.92 .126.52 ..89.03
[ x := a ].................................0.14 ...0.17 ..28.50 ..13.97
[ x := {} ]................................0.77 ...0.55 ..90.13 ..62.22
[ f0() ]...................................0.09 ...0.11 ..29.58 ..13.09
[ f1( i ) ]................................0.14 ...0.13 ..45.97 ..13.61
[ f2( c[1...8] ) ].........................0.31 ...0.28 ..44.81 ..18.41
[ f2( c[1...40000] ) ].....................0.28 ...0.23 ..28.41 ..17.34
[ f2( @c[1...40000] ) ]....................0.27 ...0.22 ..44.02 ..19.08
[ f2( @c[1...40000] ), c2 := c ]...........0.27 ...0.23 ..41.41 ..16.81
[ f3( a, a2, s, i, s2, bc, i, n, x ) ].....0.95 ...0.89 ..45.78 ..25.56
[ f2( a ) ]................................0.38 ...0.33 ..45.08 ..14.23
[ x := f4() ]..............................1.61 ...1.28 ..94.91 ..58.11
[ x := f5() ]..............................0.70 ...0.55 ..96.70 ..49.44
=============================================== ======= ======= =======
[ total application time: ]...............61.00 ..60.50 .154.64 .123.61
[ total real time: ]......................57.73 ..51.34 ..78.61 ..63.39

Brgds,
Viktor

On 2008.10.03., at 21:53, Przemyslaw Czerpak wrote:

> 2008-10-03 21:53 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
> - harbour/tests/speedtst.prg
> + harbour/tests/speedold.prg
>   * renamed old speed test - I left it because some other xbase  
> compiler
>     cannot compiple the new one due to not Clipper compatible PP
>
> + harbour/tests/speedtst.prg
>   * added new speed test which can also measure mutlithread  
> performance.
>     When linked with MT HVM version then it can be executed with  
> optional
>     parameter which causes that each test loop is executed by separate
>     thread. Please make such tests on different platforms.
>
> best regards
> Przemek
> _______________________________________________
> Harbour mailing list
> Harbour at harbour-project.org
> http://lists.harbour-project.org/mailman/listinfo/harbour