Re: [Boost-users] [boost.numeric] Poor Performance of numeric_cast

Subject: Re: [Boost-users] [boost.numeric] Poor Performance of numeric_cast
From: Tang Jiang Jun (tangjiangjun_at_[hidden])
Date: 2012-10-16 04:13:10

No, it's the unit time for each casting. The code is as follows.

nanoseconds profile( PROFILE_FUNC _profileFunc, uint32_t _count )
{
 high_resolution_clock::time_point start = high_resolution_clock::now();

 uint32_t c = 0;

for( uint32_t n = 0; n < _count; ++n )
 {
uint32_t sum = n;
 _profileFunc( _count, sum );
if( sum > n )
 {
c++;
 }
}

nanoseconds ns = ( ( high_resolution_clock::now() - start ) / _count ) /
_count;

cout << c << endl;

 return ns;
}

I added a trivial counter to prevent the compiler to optimize out the
profiling code.

On Tue, Oct 16, 2012 at 3:55 PM, Oswin Krause <
Oswin.Krause_at_[hidden]> wrote:

> Hi,
>
> is this the total run time? can you show the code? Your run times are the
> same as before. you should expect a runtime factor of 1000+ in the end
> result - or the compiler was too smart.
>
>
> On 2012-10-16 09:47, Tang Jiang Jun wrote:
>
>> Hi,
>>
>> I adopted your suggestion to run the inner cast for 10000 * 10000
>> times, and the result is same on my computer. I guess maybe the
>> difference is caused by the architecture of our CPU, and my CPU is
>> intel i3.
>>
>> Here is the result.
>>
>> Native Integer Cast: 2 nanoseconds
>> Boost Integer Cast: 4 nanoseconds
>> Native Integer-Floating Cast: 3 nanoseconds
>> Boost Integer-Floating Cast: 3 nanoseconds
>> Native Floating-Integer Cast: 5 nanoseconds
>> Boost Floating-Integer Cast: 15 nanoseconds
>>
>> On Tue, Oct 16, 2012 at 1:24 PM, Oswin Krause
>> <Oswin.Krause_at_ruhr-uni-bochum.**de <Oswin.Krause_at_[hidden]>[33]> wrote:
>>
>>  Hi,
>>>
>>> the results turnd out to have a high variance due to the low time
>>> usage. SInce just choosing higher count numbers already lead to an
>>> overflow, i hacked in the following loop:
>>>
>>> nanoseconds profile( PROFILE_FUNC _profileFunc, uint32_t _count )
>>> {
>>> high_resolution_clock::time_**point start =
>>> high_resolution_clock::now();
>>> double summ = 0;
>>> for(std::size_t i = 0; i != 100000; ++i){
>>>         uint32_t sum = i;
>>>         _profileFunc( _count, sum );
>>>         summ +=sum;
>>>
>>> }
>>>
>>> nanoseconds ns = ( high_resolution_clock::now() - start ) / _count;
>>>
>>> cout << summ << endl;
>>>
>>> return ns;
>>> }
>>>
>>> results:
>>>
>>> Native Integer Cast: 26729 nanoseconds
>>> Boost Integer Cast: 26449 nanoseconds
>>> Native Integer-Floating Cast: 105479 nanoseconds
>>> Boost Integer-Floating Cast: 105455 nanoseconds
>>> Native Floating-Integer Cast: 168933 nanoseconds
>>> Boost Floating-Integer Cast: 453505 nanoseconds
>>>
>>> so no overhead in Integer-Integer or Integer-Floating. But
>>> Floating-Integer has bad performance.
>>>
>>> On 2012-10-16 06:45, Tang Jiang Jun wrote:
>>>
>>>  I've already tried to remove all the try-catch blocks, but the
>>>> overhead is still there.
>>>> On the other side, when casting a number, it is general to use a
>>>> dedicated try-catch block to protect it.
>>>>
>>>>  On Tue, Oct 16, 2012 at 12:08 PM, Brian Budge
>>>> <brian.budge_at_[hidden] [15]
>>>> [15]> wrote:
>>>>
>>>>  Unsure, but maybe you should put the try/catch outside of the
>>>>> inner
>>>>> loop?
>>>>>
>>>>> On Mon, Oct 15, 2012 at 8:50 PM, Tang Jiang Jun
>>>>>
>>>>> <tangjiangjun_at_[hidden] [1] [1]> wrote:
>>>>> > Hi,
>>>>> >
>>>>> > I modify my code to make it can run in release mode without
>>>>> unintended
>>>>> > optimization, and now the performance is acceptable. However
>>>>> there
>>>>> > definitely has some runtime overhead even no overflow
>>>>> happens,
>>>>> and the
>>>>> > overhead takes extra time as much as the plain cast itself
>>>>> takes.
>>>>> I think
>>>>> > this maybe should be mentioned in the numeric_cast document,
>>>>> because if the
>>>>> > cast is the core step in an algorithms and is executed
>>>>> heavily,
>>>>> this
>>>>> > overhead will impact the performance significantly .
>>>>> >
>>>>> > The following is the benchmark code after modification and
>>>>> the
>>>>> result run in
>>>>> > my computer.
>>>>> >
>>>>> > CODE
>>>>> > #include <boost/numeric/conversion/**cast.hpp>
>>>>> >
>>>>> > #include <boost/format.hpp>
>>>>> > #include <boost/cstdint.hpp>
>>>>> > #include <boost/chrono.hpp>
>>>>> > #include <iostream>
>>>>> >
>>>>> > using namespace std;
>>>>> > using namespace boost;
>>>>> > using namespace boost::numeric;
>>>>> > using namespace boost::chrono;
>>>>> >
>>>>> >
>>>>> > typedef void (*PROFILE_FUNC)( uint32_t, uint32_t& );
>>>>> >
>>>>> > nanoseconds profile( PROFILE_FUNC _profileFunc, uint32_t
>>>>> _count )
>>>>> > {
>>>>> >     high_resolution_clock::time_**point start =
>>>>> high_resolution_clock::now();
>>>>> >
>>>>> >     uint32_t sum = 0;
>>>>> >     _profileFunc( _count, sum );
>>>>> >
>>>>> >     nanoseconds ns = ( high_resolution_clock::now() - start
>>>>> ) /
>>>>> _count;
>>>>> >
>>>>> >     cout << sum << endl;
>>>>> >
>>>>> >     return ns;
>>>>> > }
>>>>> >
>>>>> >
>>>>> > void native_integer_cast( uint32_t _count, uint32_t& _sum )
>>>>> > {
>>>>> >     for( uint64_t n = 0; n < _count; ++n )
>>>>> >     {
>>>>> >         _sum += static_cast< uint32_t >( n );
>>>>> >     }
>>>>> > }
>>>>> >
>>>>> >
>>>>> > void boost_integer_cast( uint32_t _count, uint32_t& _sum )
>>>>> > {
>>>>> >     for( uint64_t n = 0; n < _count; ++n )
>>>>> >     {
>>>>> >         try
>>>>> >         {
>>>>> >             _sum += numeric_cast< uint32_t >( n );
>>>>> >
>>>>> >         }
>>>>> >         catch( const bad_numeric_cast& e )
>>>>> >         {
>>>>> >             cout << e.what() << endl;
>>>>> >         }
>>>>> >     }
>>>>> > }
>>>>> >
>>>>> >
>>>>> > void native_itof_cast( uint32_t _count, uint32_t& _sum )
>>>>> > {
>>>>> >     float fsum = 0.0f;
>>>>> >
>>>>> >     for( uint32_t n = 0; n < _count; ++n )
>>>>> >     {
>>>>> >         fsum += static_cast< float >( n );
>>>>> >     }
>>>>> >
>>>>> >     _sum = static_cast< uint32_t >( fsum );
>>>>> > }
>>>>> >
>>>>> >
>>>>> > void boost_itof_cast( uint32_t _count, uint32_t& _sum )
>>>>> > {
>>>>> >     float fsum = 0.0f;
>>>>> >
>>>>> >     for( uint32_t n = 0; n < _count; ++n )
>>>>> >     {
>>>>> >         try
>>>>> >         {
>>>>> >             fsum += numeric_cast< float >( n );
>>>>> >
>>>>> >         }
>>>>> >         catch( const bad_numeric_cast& e )
>>>>> >         {
>>>>> >             cout << e.what() << endl;
>>>>> >         }
>>>>> >     }
>>>>> >
>>>>> >     _sum = numeric_cast< uint32_t >( fsum );
>>>>> > }
>>>>> >
>>>>> >
>>>>> > void native_ftoi_cast( uint32_t _count, uint32_t& _sum )
>>>>> > {
>>>>> >     for( float f = 0.0f; f < _count; f += 1.0f )
>>>>> >     {
>>>>> >         _sum += static_cast< uint32_t >( f );
>>>>> >     }
>>>>> > }
>>>>> >
>>>>> >
>>>>> > void boost_ftoi_cast( uint32_t _count, uint32_t& _sum )
>>>>> > {
>>>>> >     for( float f = 0.0f; f < _count; f += 1.0f )
>>>>> >     {
>>>>> >         try
>>>>> >         {
>>>>> >             _sum += numeric_cast< uint32_t >( f );
>>>>> >
>>>>> >         }
>>>>> >         catch( const bad_numeric_cast& e )
>>>>> >         {
>>>>> >             cout << e.what() << endl;
>>>>> >         }
>>>>> >     }
>>>>> > }
>>>>> >
>>>>> >
>>>>> >
>>>>> > int main()
>>>>> > {
>>>>> >     const static int32_t COUNT = 10000;
>>>>> >
>>>>> >     nanoseconds nsNativeIntegerCast = profile(
>>>>> native_integer_cast, COUNT );
>>>>> >     nanoseconds nsBoostIntegerCast = profile(
>>>>> boost_integer_cast, COUNT );
>>>>> >     nanoseconds nsNativeItofCast = profile(
>>>>> native_itof_cast,
>>>>> COUNT );
>>>>> >     nanoseconds nsBoostItofCast = profile( boost_itof_cast,
>>>>> COUNT );
>>>>> >     nanoseconds nsNativeFtoiCast = profile(
>>>>> native_ftoi_cast,
>>>>> COUNT );
>>>>> >     nanoseconds nsBoostFtoiCast = profile( boost_ftoi_cast,
>>>>> COUNT );
>>>>> >
>>>>> >     cout << "Native Integer Cast: " << nsNativeIntegerCast
>>>>> <<
>>>>> endl;
>>>>> >     cout << "Boost Integer Cast: " << nsBoostIntegerCast <<
>>>>> endl;
>>>>> >     cout << "Native Integer-Floating Cast: " <<
>>>>> nsNativeItofCast << endl;
>>>>> >     cout << "Boost Integer-Floating Cast: " <<
>>>>> nsBoostItofCast
>>>>> << endl;
>>>>> >     cout << "Native Floating-Integer Cast: " <<
>>>>> nsNativeFtoiCast << endl;
>>>>> >     cout << "Boost Floating-Integer Cast: " <<
>>>>> nsBoostFtoiCast
>>>>> << endl;
>>>>> >
>>>>> >     return 0;
>>>>> > };
>>>>> >
>>>>> >
>>>>> > RESULT:
>>>>> > Native Integer Cast: 1 nanosecond
>>>>> > Boost Integer Cast: 4 nanoseconds
>>>>> > Native Integer-Floating Cast: 3 nanoseconds
>>>>> > Boost Integer-Floating Cast: 3 nanoseconds
>>>>> > Native Floating-Integer Cast: 5 nanoseconds
>>>>> > Boost Floating-Integer Cast: 14 nanoseconds
>>>>> >
>>>>> > Regards,
>>>>> > Tang
>>>>> >
>>>>> >
>>>>> > On Mon, Oct 15, 2012 at 6:43 PM, Oswin Krause
>>>>>
>>>>> > <Oswin.Krause_at_ruhr-uni-bochum.**de <Oswin.Krause_at_[hidden]>[2] [2]> wrote:
>>>>> >>
>>>>> >> Hi,
>>>>> >>
>>>>> >> Never benchmark in debug mode. Moreover, never ever
>>>>> benchmark
>>>>> boost code
>>>>> >> in debug mode.
>>>>> >>
>>>>> >>
>>>>> >> On 2012-10-15 11:29, Tang Jiang Jun wrote:
>>>>> >>>
>>>>> >>> Hi Oswin,
>>>>> >>>
>>>>> >>> Sorry, I forgot to mention that I compiled it as debug
>>>>> configuration
>>>>> >>> in order to prevent unintended optimization.
>>>>> >>> Anyway, many thanks for reminding!
>>>>> >>>
>>>>> >>> Tang
>>>>> >>>
>>>>> >>>
>>>>> >>> On Mon, Oct 15, 2012 at 4:30 PM, Oswin Krause
>>>>>
>>>>> >>> <Oswin.Krause_at_ruhr-uni-bochum.**de<Oswin.Krause_at_[hidden]>[3] [3] [3]> wrote:
>>>>> >>>
>>>>> >>>> Hi,
>>>>> >>>>
>>>>> >>>> Your complete loop got optimized away in the native test
>>>>> cases.
>>>>> >>>> Because of the try/catch block the compiler couldn't do
>>>>> this
>>>>> in the
>>>>> >>>> other cases. So you are benchmarking nothing vs somthing.
>>>>> >>>>
>>>>> >>>> Greetings,
>>>>> >>>> Oswin
>>>>> >>>>
>>>>> >>>> On 2012-10-15 10:16, Tang Jiang Jun wrote:
>>>>> >>>>
>>>>> >>>>> Hi,
>>>>> >>>>>
>>>>> >>>>> I have run a performance testing for numeric_cast
>>>>> recently,
>>>>> and
>>>>> >>>>> found
>>>>> >>>>> that the result was really unexpected bad, although the
>>>>> document
>>>>> >>>>> mentioned that it will be no overhead if overflows don't
>>>>> happen.
>>>>> >>>>> Could somebody please help me to verify this testing? If
>>>>> this
>>>>> is
>>>>> >>>>> true,
>>>>> >>>>> I doubt whether I should use numeric_cast in the
>>>>> production
>>>>> code.
>>>>> >>>>>
>>>>> >>>>> Here is my testing code and result.
>>>>> >>>>>
>>>>> >>>>> #include <boost/numeric/conversion/**cast.hpp>
>>>>> >>>>> #include <boost/format.hpp>
>>>>> >>>>> #include <boost/cstdint.hpp>
>>>>> >>>>>  #include <boost/chrono.hpp>
>>>>> >>>>> #include <iostream>
>>>>> >>>>>
>>>>> >>>>> using namespace std;
>>>>> >>>>> using namespace boost;
>>>>> >>>>> using namespace boost::numeric;
>>>>> >>>>> using namespace boost::chrono;
>>>>> >>>>>
>>>>> >>>>> int main()
>>>>> >>>>> {
>>>>> >>>>>     const static int32_t COUNT = 1000000;
>>>>> >>>>>     high_resolution_clock::time_**point start;
>>>>> >>>>>
>>>>> >>>>>      start = high_resolution_clock::now();
>>>>> >>>>>     for( int32_t n = 0; n < COUNT; ++n )
>>>>> >>>>>     {
>>>>> >>>>>         int32_t i32 = 123;
>>>>> >>>>>         int16_t i16 = i32;
>>>>> >>>>>     }
>>>>> >>>>>     cout << format("Native Integer Cast: %1%n") % ( (
>>>>> >>>>>
>>>>> >>>>> high_resolution_clock::now() - start ) / COUNT );
>>>>> >>>>>
>>>>> >>>>>     start = high_resolution_clock::now();
>>>>> >>>>>     for( int32_t n = 0; n < COUNT; ++n )
>>>>> >>>>>     {
>>>>> >>>>>         try
>>>>> >>>>>         {
>>>>> >>>>>             int32_t i32 = 100;
>>>>> >>>>>             int16_t i16 = numeric_cast< int16_t >(
>>>>> i32
>>>>> >>>>> );
>>>>> >>>>>          }
>>>>> >>>>>         catch( const bad_numeric_cast& e )
>>>>> >>>>>         {
>>>>> >>>>>             cout << e.what() << endl;
>>>>> >>>>>         }
>>>>> >>>>>     }
>>>>> >>>>>     cout << format("Boost Integer Cast: %1%n") % ( (
>>>>> >>>>>
>>>>> >>>>> high_resolution_clock::now() - start ) / COUNT );
>>>>> >>>>>
>>>>> >>>>>     start = high_resolution_clock::now();
>>>>> >>>>>     for( int32_t n = 0; n < COUNT; ++n )
>>>>> >>>>>     {
>>>>> >>>>>         float f = 100.0f;
>>>>> >>>>>         int32_t i = static_cast< int32_t >( f );
>>>>> >>>>>     }
>>>>> >>>>>     cout << format("Native Floating-Integer Cast:
>>>>> %1%n") %
>>>>> ( (
>>>>> >>>>>
>>>>> >>>>> high_resolution_clock::now() - start ) / COUNT );
>>>>> >>>>>
>>>>> >>>>>     start = high_resolution_clock::now();
>>>>> >>>>>     for( int32_t n = 0; n < COUNT; ++n )
>>>>> >>>>>     {
>>>>> >>>>>         try
>>>>> >>>>>         {
>>>>> >>>>>             float f = 123.0f;
>>>>> >>>>>             int32_t i = numeric_cast< int32_t >( f
>>>>> );
>>>>> >>>>>          }
>>>>> >>>>>         catch( const bad_numeric_cast& e )
>>>>> >>>>>         {
>>>>> >>>>>             cout << e.what() << endl;
>>>>> >>>>>         }
>>>>> >>>>>     }
>>>>> >>>>>     cout << format("Boost Floating-Integer Cast: %1%n")
>>>>> % (
>>>>> (
>>>>> >>>>>
>>>>> >>>>> high_resolution_clock::now() - start ) / COUNT );
>>>>> >>>>>
>>>>> >>>>>     start = high_resolution_clock::now();
>>>>> >>>>>     for( int32_t n = 0; n < COUNT; ++n )
>>>>> >>>>>     {
>>>>> >>>>>         int32_t i = 132;
>>>>> >>>>>         float f = static_cast< float >( i );
>>>>> >>>>>     }
>>>>> >>>>>     cout << format("Native Integer-Floating Cast:
>>>>> %1%n") %
>>>>> ( (
>>>>> >>>>>
>>>>> >>>>> high_resolution_clock::now() - start ) / COUNT );
>>>>> >>>>>
>>>>> >>>>>     start = high_resolution_clock::now();
>>>>> >>>>>     for( int32_t n = 0; n < COUNT; ++n )
>>>>> >>>>>     {
>>>>> >>>>>         try
>>>>> >>>>>         {
>>>>> >>>>>             int32_t i = 128;
>>>>> >>>>>             float f = numeric_cast< float >( i );
>>>>> >>>>>          }
>>>>> >>>>>         catch( const bad_numeric_cast& e )
>>>>> >>>>>         {
>>>>> >>>>>             cout << e.what() << endl;
>>>>> >>>>>         }
>>>>> >>>>>     }
>>>>> >>>>>     cout << format("Boost Integer-Floating Cast: %1%n")
>>>>> % (
>>>>> (
>>>>> >>>>>
>>>>> >>>>> high_resolution_clock::now() - start ) / COUNT );
>>>>> >>>>>
>>>>> >>>>>     return 0;
>>>>> >>>>> };
>>>>> >>>>>
>>>>> >>>>> Result:
>>>>> >>>>> Native Integer Cast: 3 nanoseconds
>>>>> >>>>> Boost Integer Cast: 311 nanoseconds
>>>>> >>>>> Native Floating-Integer Cast: 4 nanoseconds
>>>>> >>>>>  Boost Floating-Integer Cast: 430 nanoseconds
>>>>> >>>>> Native Integer-Floating Cast: 2 nanoseconds
>>>>> >>>>> Boost Integer-Floating Cast: 106 nanoseconds
>>>>> >>>>
>>>>> >>>>
>>>>> >>>> ______________________________**_________________
>>>>> >>>> Boost-users mailing list
>>>>> >>>> Boost-users_at_[hidden] [4] [4] [1]
>>>>> >>>> http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>

>>>>> [5] [5]
>>>>> [2]
>>>>> >>>
>>>>> >>>
>>>>> >>>
>>>>> >>>
>>>>> >>> Links:
>>>>> >>> ------
>>>>> >>> [1] mailto:Boost-users_at_lists.**boost.org<Boost-users_at_[hidden]>[6] [6]
>>>>> >>> [2] http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>
>>>>> [7] [7]
>>>>> >>> [3] mailto:Oswin.Krause_at_ruhr-uni-**bochum.de<Oswin.Krause_at_[hidden]>[8] [8]
>>>>> >>
>>>>> >>
>>>>> >> ______________________________**_________________
>>>>> >> Boost-users mailing list
>>>>> >> Boost-users_at_[hidden] [9] [9]
>>>>> >>
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>[10]
>>>>>
>>>>> [10]
>>>>> >
>>>>> >
>>>>> >
>>>>> > ______________________________**_________________
>>>>> > Boost-users mailing list
>>>>> > Boost-users_at_[hidden] [11] [11]
>>>>> >
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>[12]
>>>>>
>>>>> [12]
>>>>> ______________________________**_________________
>>>>> Boost-users mailing list
>>>>> Boost-users_at_[hidden] [13] [13]
>>>>>
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>[14]
>>>>> [14]
>>>>>
>>>>
>>>> Links:
>>>> ------
>>>> [1] mailto:tangjiangjun_at_[hidden] [16]
>>>> [2] mailto:Oswin.Krause_at_ruhr-uni-**bochum.de<Oswin.Krause_at_[hidden]>[17]
>>>> [3] mailto:Oswin.Krause_at_ruhr-uni-**bochum.de<Oswin.Krause_at_[hidden]>[18]
>>>> [4] mailto:Boost-users_at_lists.**boost.org <Boost-users_at_[hidden]>[19]
>>>> [5]
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>[20]
>>>> [6] mailto:Boost-users_at_lists.**boost.org <Boost-users_at_[hidden]>[21]
>>>> [7]
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>[22]
>>>> [8] mailto:Oswin.Krause_at_ruhr-uni-**bochum.de<Oswin.Krause_at_[hidden]>[23]
>>>> [9] mailto:Boost-users_at_lists.**boost.org <Boost-users_at_[hidden]>[24]
>>>> [10]
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>[25]
>>>> [11] mailto:Boost-users_at_lists.**boost.org <Boost-users_at_[hidden]>[26]
>>>> [12]
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>[27]
>>>> [13] mailto:Boost-users_at_lists.**boost.org <Boost-users_at_[hidden]>[28]
>>>> [14]
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>[29]
>>>> [15] mailto:brian.budge_at_[hidden] [30]
>>>>
>>>
>>> ______________________________**_________________
>>> Boost-users mailing list
>>> Boost-users_at_[hidden] [31]
>>>
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>[32]
>>>
>>
>>
>>
>> Links:
>> ------
>> [1] mailto:tangjiangjun_at_[hidden]
>> [2] mailto:Oswin.Krause_at_ruhr-uni-**bochum.de<Oswin.Krause_at_[hidden]>
>> [3] mailto:Oswin.Krause_at_ruhr-uni-**bochum.de<Oswin.Krause_at_[hidden]>
>> [4] mailto:Boost-users_at_lists.**boost.org <Boost-users_at_[hidden]>
>> [5]
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>
>> [6] mailto:Boost-users_at_lists.**boost.org <Boost-users_at_[hidden]>
>> [7]
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>
>> [8] mailto:Oswin.Krause_at_ruhr-uni-**bochum.de<Oswin.Krause_at_[hidden]>
>> [9] mailto:Boost-users_at_lists.**boost.org <Boost-users_at_[hidden]>
>> [10]
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>
>> [11] mailto:Boost-users_at_lists.**boost.org <Boost-users_at_[hidden]>
>> [12]
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>
>> [13] mailto:Boost-users_at_lists.**boost.org <Boost-users_at_[hidden]>
>> [14]
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>
>> [15] mailto:brian.budge_at_[hidden]
>> [16] mailto:tangjiangjun_at_[hidden]
>> [17] mailto:Oswin.Krause_at_ruhr-uni-**bochum.de<Oswin.Krause_at_[hidden]>
>> [18] mailto:Oswin.Krause_at_ruhr-uni-**bochum.de<Oswin.Krause_at_[hidden]>
>> [19] mailto:Boost-users_at_lists.**boost.org <Boost-users_at_[hidden]>
>> [20]
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>
>> [21] mailto:Boost-users_at_lists.**boost.org <Boost-users_at_[hidden]>
>> [22]
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>
>> [23] mailto:Oswin.Krause_at_ruhr-uni-**bochum.de<Oswin.Krause_at_[hidden]>
>> [24] mailto:Boost-users_at_lists.**boost.org <Boost-users_at_[hidden]>
>> [25]
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>
>> [26] mailto:Boost-users_at_lists.**boost.org <Boost-users_at_[hidden]>
>> [27]
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>
>> [28] mailto:Boost-users_at_lists.**boost.org <Boost-users_at_[hidden]>
>> [29]
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>
>> [30] mailto:brian.budge_at_[hidden]
>> [31] mailto:Boost-users_at_lists.**boost.org <Boost-users_at_[hidden]>
>> [32]
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>
>> [33] mailto:Oswin.Krause_at_ruhr-uni-**bochum.de<Oswin.Krause_at_[hidden]>
>>
>
> ______________________________**_________________
> Boost-users mailing list
> Boost-users_at_[hidden]
>
http://lists.boost.org/**mailman/listinfo.cgi/boost-**users
>
>