One of our goals in doing .NET was to provide our customers with a number of different
languages so they can choose the ones that best meet their needs, and the individual language
teams are tasked with providing products that best serve their features. From my perspective,
customers should move when the benefits of the managed environment make it worth their while,
and not because we think they should move. Though I'm in the C# team now, I still have a soft
spot for the C++ product, and I think it will be around for quite some time.
Should companies start migrating C/C++ based applications to C#?
I think it's much better, however, to do a few small new applications in C#. Porting an
existing app is always a complex undertaking, and not something you want to undertake until
you're really familiar with a platform, and what your architectural options are.
One area of C# development that has not been explored deeply is game development. How do you see
C# fit into the game industry?
Currently, the goals of a managed environment are productivity and robustness, while many games need as much raw speed as possible, and often with some "near real time" types of execution guarantees. So, for games where frame rate is the driving factor, I think C++ is still the language of choice. Other games I think are already practical in C#.
In the future, it will be easier to use DirectX from C#, and as hardware gets faster and the
run-time gets better, I'd expect this gap to close. I think this question is a lot like the
assembly language vs high-level language tradeoff that used to exist in games, and over time
the advantages of compiled languages became more important than the raw speed of assembly. I'd
expect most of the guidance in this area to come from the DirectX team.
| Here's the follow-up discussion on this article: | View all related messages |
For a complete listing of article comments, questions, and other discussions related to Jose Aniceto's C# Programming topic, please visit the Discussions page.