What are major incentives to upgrade to D2009 (Unicode excluded)?
I'm a hesitant upgrader when it comes to development tools. For roughly half of my product I still use D7, and for others D2006.
The truth is, although Unicode support is more than welcomed and very useful, it could cause me more troubles than gains with my current projects (they are more-or-less Unicode ready already). It's especially case with one of them who's performance would suffer a lot if each string takes twice as much memory as before.
So, Unicode aside, what are other major incentives to upgrade?
Answers
To put things in to perspective, look at the things that were added between Delphi 7 and Delphi 2007. This was a significant high water mark.
http://blogs.codegear.com/nickhodges/2007/03/28/33579
http://www.stevetrefethen.com/blog/VCLAndRTLEnhancementsSinceDelphi7D7.aspx
Delphi 2009 sets the bar even higher.
http://blogs.codegear.com/pawelglowacki/2008/11/03/38527
http://blogs.codegear.com/chrispattinson/2008/09/19/38897
Here are some of my favourites:
Generics (naturally) and generic collections in the RTL.
Improved build configurations where they inherit from a common base configuration.
DataSnap improvements, including removing COM dependencies.
Faster and more stable IDE over Delphi 2007, which was no slouch.
I'm not sure how I'm going to use them in production, but you have to admit that anonymous methods are really cool. I'm curious to see how people wind up using them with threading.
Just two things about Unicode support (another favourite of mine).
You will probably see a significant performance improvement when you convert your existing Unicode projects. I know I did.
You will need to be careful about converting any code that makes assumptions about character size. You probably won't see many problems if your existing code is Unicode aware.
http://dn.codegear.com/article/38437
http://dn.codegear.com/article/38498
http://dn.codegear.com/article/38693