Hum.. Anyone? ;-)
On Fri, Feb 19, 2010 at 10:46 AM, Patrick Mast, xHarbour.
<patrick.m...@xharbour.com> wrote:
> Hey Andi,
>
> Thanks!
> Wow, scaring.. These numbers say that Harbour is TWICE as fast??
>
> Patrick
>
> On Fri, Feb 19, 2010 at 10:36 AM, Andi Jahja <xharb...@telkom.net.id> wrote:
>> Patrick and All,
>>
>> OK, here you are:
>>
>> ----- cut -----
>> OS: Windows XP Professional 5.01.2600 Service Pack 3
>> Compiler: Microsoft Visual C++ 16.00.30128.01
>> xHarbour build 1.2.1 Intl. (SimpLex) (Rev. 6686)
>> Harbour 2.0.0 (Rev. 13372)
>>
>> THREADS: 0
>> N_LOOPS: 1000000
>> XHARBOUR HARBOUR
>> ======== =======
>> [ T000: empty loop overhead ]...................................0.11 ...0.05
>> ==================================================================== =======
>> [ T001: x := L_C ]..............................................0.08 ...0.06
>> [ T002: x := L_N ]..............................................0.05 ...0.05
>> [ T003: x := L_D ]..............................................0.05 ...0.05
>> [ T004: x := S_C ]..............................................0.06 ...0.09
>> [ T005: x := S_N ]..............................................0.05 ...0.06
>> [ T006: x := S_D ]..............................................0.05 ...0.06
>> [ T007: x := M->M_C ]...........................................0.08 ...0.09
>> [ T008: x := M->M_N ]...........................................0.05 ...0.06
>> [ T009: x := M->M_D ]...........................................0.06 ...0.06
>> [ T010: x := M->P_C ]...........................................0.06 ...0.13
>> [ T011: x := M->P_N ]...........................................0.05 ...0.06
>> [ T012: x := M->P_D ]...........................................0.06 ...0.06
>> [ T013: x := F_C ]..............................................0.61 ...0.27
>> [ T014: x := F_N ]..............................................0.31 ...0.30
>> [ T015: x := F_D ]..............................................0.20 ...0.17
>> [ T016: x := o:Args ]...........................................0.53 ...0.23
>> [ T017: x := o[2] ].............................................0.11 ...0.17
>> [ T018: round( i / 1000, 2 ) ]..................................0.56 ...0.31
>> [ T019: str( i / 1000 ) ].......................................1.75 ...0.77
>> [ T020: val( s ) ]..............................................0.64 ...0.36
>> [ T021: val( a [ i % 16 + 1 ] ) ]...............................0.88 ...0.63
>> [ T022: dtos( d - i % 10000 ) ].................................0.97 ...0.61
>> [ T023: eval( { || i % 16 } ) ].................................1.31 ...0.67
>> [ T024: eval( bc := { || i % 16 } ) ]...........................0.80 ...0.47
>> [ T025: eval( { |x| x % 16 }, i ) ].............................1.03 ...0.56
>> [ T026: eval( bc := { |x| x % 16 }, i ) ].......................0.77 ...0.47
>> [ T027: eval( { |x| f1( x ) }, i ) ]............................1.42 ...0.64
>> [ T028: eval( bc := { |x| f1( x ) }, i ) ]......................1.13 ...0.56
>> [ T029: eval( bc := &("{ |x| f1( x ) }"), i ) ].................1.09 ...0.56
>> [ T030: x := &( "f1(" + str(i) + ")" ) ].......................10.02 ...4.86
>> [ T031: bc := &( "{|x|f1(x)}" ), eval( bc, i ) ]...............10.20 ...4.61
>> [ T032: x := valtype( x ) + valtype( i ) ].....................1.48 ...0.50
>> [ T033: x := strzero( i % 100, 2 ) $ a[ i % 16 + 1 ] ]..........1.86 ...1.05
>> [ T034: x := a[ i % 16 + 1 ] == s ].............................0.55 ...0.47
>> [ T035: x := a[ i % 16 + 1 ] = s ]..............................0.55 ...0.52
>> [ T036: x := a[ i % 16 + 1 ] >= s ].............................0.52 ...0.52
>> [ T037: x := a[ i % 16 + 1 ] <= s ].............................0.58 ...0.48
>> [ T038: x := a[ i % 16 + 1 ] < s ]..............................0.55 ...0.50
>> [ T039: x := a[ i % 16 + 1 ] > s ]..............................0.55 ...0.56
>> [ T040: ascan( a, i % 16 ) ]....................................0.81 ...0.50
>> [ T041: ascan( a, { |x| x == i % 16 } ) ].......................8.52 ...4.98
>> [ T042: iif( i%1000==0, a:={}, ) , aadd(a,{i,1,.T.,s,s2,a2 ]....2.53 ...1.19
>> [ T043: x := a ]................................................0.06 ...0.06
>> [ T044: x := {} ]...............................................0.38 ...0.17
>> [ T045: f0() ]..................................................0.38 ...0.09
>> [ T046: f1( i ) ]...............................................0.45 ...0.19
>> [ T047: f2( c[1...8] ) ]........................................0.44 ...0.16
>> [ T048: f2( c[1...40000] ) ]....................................0.44 ...0.17
>> [ T049: f2( @c[1...40000] ) ]...................................0.42 ...0.13
>> [ T050: f2( @c[1...40000] ), c2 := c ]..........................0.52 ...0.20
>> [ T051: f3( a, a2, s, i, s2, bc, i, n, x ) ]....................0.91 ...0.55
>> [ T052: f2( a ) ]...............................................0.45 ...0.16
>> [ T053: x := f4() ].............................................1.95 ...0.80
>> [ T054: x := f5() ].............................................1.47 ...0.39
>> [ T055: x := space(16) ]........................................1.00 ...0.28
>> [ T056: f_prv( c ) ]............................................2.09 ...0.58
>> ==================================================================== =======
>> [ total application time: ]....................................70.59 ..35.91
>> [ total real time: ]...........................................70.63 ..35.91
>>
>> ----- cut -----
>> --
>> Andi
>>
>> On Fri, 19 Feb 2010 09:43:51 +0100
>> "Patrick Mast, xHarbour." <patrick.m...@xharbour.com> wrote:
>>
>>> Hey Andi,
>>>
>>> >> Ron, any clue why this is? This is not the first time I hear people
>>> >> talk about a faster Harbour. ;-)
>>> >> Do we have numbers on this? Andi?
>>> >
>>> > What numbers are you referring to? We do not have an agreed benchmarking
>>> > tool except speedtst. If you are referring to this tool, I can post it
>>> > here.
>>>
>>> Exactly what I was asking Miguel. ;-)
>>>
>>> Do we have a speedtest where we can see the faster speed of Harbour.
>>> It's not a matter of not believing Miguel, not at all! But a speedtest
>>> would show us in what field Harbour is better and thus we can finetune
>>> xHarbour to match Harbour's speed.
>>>
>>> I mean, if Harbour can do those things, we can! ;-)
>>>
>>> Patrick
>>
>>
>>
>> ------------------------------------------------------------------------------
>> Download Intel® Parallel Studio Eval
>> Try the new software tools for yourself. Speed compiling, find bugs
>> proactively, and fine-tune applications for parallel performance.
>> See why Intel Parallel Studio got high marks during beta.
>> http://p.sf.net/sfu/intel-sw-dev
>> _______________________________________________
>> xHarbour-developers mailing list
>> xHarbour-developers@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/xharbour-developers
>>
>
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
xHarbour-developers mailing list
xHarbour-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/xharbour-developers
