I think the move to "managed code" is inevitable for most programmers - unless you're working on a very low level. Even then it might end up that you get microchips that run the appropriate "flavor" of bytecode in a native form. Defeats the point of virtual machines but there were LISP machines back in the day too...

Whether it's C# or C++ or Python, I dunno. All three are ALGOL derivatives, so there seems be a natural progression in going from one to the next; Python is the "slowest" at the moment of course, but I'm a big fan of it for hitting the target of "simple yet powerful," and it lets you put in C modules, so with you can get equivalent performance in the input/output area as any project in C using the same libraries. For an indie working in 2d, PyGame is a perfectly acceptable library, though software rendering performance at high resolution/color is limited if you make a scrolling or other render-heavy game.

The problem i see..is like someone else already mentioned in this post.Its the time it takes to learn c# and porting my existing codebase from native c++.

To the other poster...i didn't find it easy to port my code(a lot of it) to the .NET enviroment.In fact,i gave up because it was too much hassle.Iv'e spent over 2 years writing it,and i'm good with c/c++...
so why would i want to switch to another language now? after spending almost 2 straight days of head f***s trying to port it to .NET?

3D concepts,i'm good with,so picking up the basics of c# took literally hours.But as i said that isn't the problem,i'm definately not going to re-write my whole codebase in c#.

If it was a simple matter,i would have done it long ago,but it isn't.I don't know what you refer to as simple(a pong game maybe)fair enough but if your source code weighs in at around 2.5MB (or over)(in just .h and .cpp files)what do you do then..say it doesn't really matter,and start from scratch? (my code also using several other external libs).

Reading what most here wrote...i agree with most part of it,if your starting a new project(or just learning),but if your far with an existing project in c/c++ ....?????

Quote:Original post by dirtbag
The problem i see..is like someone else already mentioned in this post.Its the time it takes to learn c# and porting my existing codebase from native c++.

To the other poster...i didn't find it easy to port my code(a lot of it) to the .NET enviroment.In fact,i gave up because it was too much hassle.Iv'e spent over 2 years writing it,and i'm good with c/c++...
so why would i want to switch to another language now? after spending almost 2 straight days of head f***s trying to port it to .NET?

3D concepts,i'm good with,so picking up the basics of c# took literally hours.But as i said that isn't the problem,i'm definately not going to re-write my whole codebase in c#.

If it was a simple matter,i would have done it long ago,but it isn't.I don't know what you refer to as simple(a pong game maybe)fair enough but if your source code weighs in at around 2.5MB (or over)(in just .h and .cpp files)what do you do then..say it doesn't really matter,and start from scratch? (my code also using several other external libs).

Reading what most here wrote...i agree with most part of it,if your starting a new project(or just learning),but if your far with an existing project in c/c++ ....?????

To me, the few advantages of C# don't seem to warrant a total move away from C++ especially if you have a reasonable amount of codebase.Also, i depend on many other libraries out there which are written in C/C++ with portability in mind.

The main advantages of C# come from the huge .NET Framework which can be used in Managed C++ as well.

I have written a tile based level editor in C# and i fail to see how it has made me any more productive besides not having to cleanup memory myself. The only reason it was faster to do than in C++/Win32 was because of Winforms and GDI+, both of which are also available in C++.

Quote:Original post by ggs
The old axiom of Microsoft takes 3 versions to get it right comes to mind. Version 2.0 isnt due out to next year.

I don't think you can call that an axiom. It's a Law.