jeremiah johnson <naikrovek@gmail.com> wrote:
> That compilation takes place with a JIT (Just In Time) compiler, and
> this adds time to the execution. It is noticeably slower.

> How much slower though, depends on what you're doing. Some argue half
> as fast, some argue 1/4 as fast.

"Michael A. Covington" <look@ai.uga.edu.for.address> wrote in message
news:ekIjWd8xFHA.2312@TK2MSFTNGP14.phx.gbl...
>
> "jeremiah johnson" <naikrovek@gmail.com> wrote in message
> news:uXvVv37xFHA.356@TK2MSFTNGP10.phx.gbl...
>> Michael A. Covington wrote:
>>> "Roy Gourgi" <royng@videotron.ca> wrote in message
>>> news:4X%%e.7653$rA4.250843@wagner.videotron.net...
>>>
>>>>Hi,
>>>>
>>>>How much is C# slower than C++?
>>>
>>>
>>> Depends on what you're doing. There no inherent reason for it to be
>>> slower at all. MSIL is compiled into native machine code before it
>>> executes.
>>
>> That compilation takes place with a JIT (Just In Time) compiler, and this
>> adds time to the execution. It is noticeably slower.
>
> Only the first time through a method. Loops - which are the
> time-consuming part - run at full speed after the first iteration.
>
>

"Scott" <sdcoonce@gmail.HEY_YOU.com> wrote in message
news:ePyr8uAyFHA.1256@TK2MSFTNGP09.phx.gbl...
> As a "débutant" in c++, I did the following in c# and in c++, just to see
> for myself:
>
> namespace PerfTest
> {
> struct Obj {
> public int a,b,c,d,e,f,g,h,i,j, k, l, m,n,o,p,q,r,s,t,u,v,w,x,y,z;
> int get_i() { return i;}
> }
>
> class Class1
> {
> static void pass_by_value(Obj o) {
> o.i++;
> }
> static void pass_by_ref(ref Obj o) {
> o.i++;
> }
>
> [STAThread]
> static void Main(string[] args)
> {
> Obj obj = new Obj();
> DateTime start, finish;
> TimeSpan duration;
>
> Console.WriteLine("CSHARP test.");
> Console.Write("How many? ");
> int count = Convert.ToInt32(Console.ReadLine());
>
>
> Console.Write("Test pass by value: ");
> start = DateTime.Now;
> for(int i=0; i<count; i++)
> pass_by_value(obj);
> finish = DateTime.Now;
>
> duration = finish - start;
> Console.WriteLine("{0} ms.", duration.Milliseconds);
>
> Console.Write("Test pass by ref : ");
> start = DateTime.Now;
> for(int i=0; i<count; i++)
> pass_by_ref(ref obj);
> finish = DateTime.Now;
> duration = finish - start;
> Console.WriteLine("{0} ms.", duration.Milliseconds);
>
>
>
> Console.Write("[Enter] to quit."); Console.ReadLine();
> }
> }
> }
>
>
> c# turned out to be 10x faster over 10e6 loops, but I suspect it was
> because JIT (or whatever) optimized away a lot of the work I was trying to
> make the computer do. I'm not saying c# is faster most of the time, but
> in my little test, it was. Of course, startup time for c# apps is longer
> than for c++ for the reasons mentioned elsewhere in this thread.
>
> scott
>
>
> "Roy Gourgi" <royng@videotron.ca> wrote in message
> news:4X%%e.7653$rA4.250843@wagner.videotron.net...
>> Hi,
>>
>> How much is C# slower than C++?
>>
>> TIA
>> Roy
>>
>
>

"Scott Coonce" <sdcoonce@gmail.HEY_YOU.com> wrote in message
news:eYk7Bx43FHA.3888@TK2MSFTNGP10.phx.gbl...
> just in case anybody comes along and reads this thread again, i made the
> classic mistake of comparing the "debug" builds. so as i've read in other
> threads, all timing comparison bets are off.
>
> scott
>
>
> "Scott" <sdcoonce@gmail.HEY_YOU.com> wrote in message
> news:ePyr8uAyFHA.1256@TK2MSFTNGP09.phx.gbl...
> > As a "débutant" in c++, I did the following in c# and in c++, just to
see
> > for myself:
> >
> > namespace PerfTest
> > {
> > struct Obj {
> > public int a,b,c,d,e,f,g,h,i,j, k, l, m,n,o,p,q,r,s,t,u,v,w,x,y,z;
> > int get_i() { return i;}
> > }
> >
> > class Class1
> > {
> > static void pass_by_value(Obj o) {
> > o.i++;
> > }
> > static void pass_by_ref(ref Obj o) {
> > o.i++;
> > }
> >
> > [STAThread]
> > static void Main(string[] args)
> > {
> > Obj obj = new Obj();
> > DateTime start, finish;
> > TimeSpan duration;
> >
> > Console.WriteLine("CSHARP test.");
> > Console.Write("How many? ");
> > int count = Convert.ToInt32(Console.ReadLine());
> >
> >
> > Console.Write("Test pass by value: ");
> > start = DateTime.Now;
> > for(int i=0; i<count; i++)
> > pass_by_value(obj);
> > finish = DateTime.Now;
> >
> > duration = finish - start;
> > Console.WriteLine("{0} ms.", duration.Milliseconds);
> >
> > Console.Write("Test pass by ref : ");
> > start = DateTime.Now;
> > for(int i=0; i<count; i++)
> > pass_by_ref(ref obj);
> > finish = DateTime.Now;
> > duration = finish - start;
> > Console.WriteLine("{0} ms.", duration.Milliseconds);
> >
> >
> >
> > Console.Write("[Enter] to quit."); Console.ReadLine();
> > }
> > }
> > }
> >
> >
> > c# turned out to be 10x faster over 10e6 loops, but I suspect it was
> > because JIT (or whatever) optimized away a lot of the work I was trying
to
> > make the computer do. I'm not saying c# is faster most of the time, but
> > in my little test, it was. Of course, startup time for c# apps is longer
> > than for c++ for the reasons mentioned elsewhere in this thread.
> >
> > scott
> >
> >
> > "Roy Gourgi" <royng@videotron.ca> wrote in message
> > news:4X%%e.7653$rA4.250843@wagner.videotron.net...
> >> Hi,
> >>
> >> How much is C# slower than C++?
> >>
> >> TIA
> >> Roy
> >>
> >
> >
>
>