Torque for Delphi/Kylix...
by Brett Fattori · in Torque Game Engine · 12/19/2003 (6:49 am) · 21 replies
Folks,
I know that a lot of people see Delphi (Visual Pascal) as a toy. That it can't do anything, any better, than C++ can. However, Delphi is a well done implementation of Pascal with objects. It is recognized by many people as THE way to rapidly develop applications. It seems to me that if it were allowable, it would be neat to port the entire engine over to Delphi. This may be a pipe dream, and it may not be allowed by the TGE license (I haven't checked into that) but there are many good reasons to port, and maintain the TGE, to Delphi/Kylix.
1) Delphi is based around a KNOWN learning language: PASCAL.
2) Delphi is object based, just like C++, so it can handle many of the implementation issues that would be impossible in a port to, say, Visual Basic.
3) Most of the engine is OpenGL, so we're not depending on how well it compiles into assembly.. Although, I have to say that Borland makes some amazing compilers.
4) Most of the engine depends on OpenGL, which (nowadays) is hardware bound. It doesn't matter how fast the CPU is (as much anymore) because the graphics are handled by graphics hardware.
5) Delphi/Kylix are available on the two main platforms: Windows and Linux. I am sorry that Borland hasn't made a "Kylix" for Mac, because I prolly never would have converted to C++ in the first place.
6) There's evidence that being able to maintain another codebase (albeit a little laggy behind the main codebase) would be effective. Looking at the Delphi JEDI projects, which maintain standard C++ libraries for Delphi, it would be entirely possible to maintain the code.
There are more reasons, that I can't think of right now, to maintain a Delphi implementation of the TGE. I won't kid anyone... This would be a MAJOR undertaking, it would require a team of people to do the conversion (and maintain it), and it would be fun to do.
I would be interested in working with a group of people to do the port, if people were interested. It would certainly open up some more doors (not everyone is a master C++ coder). And, again, I maintain it would be fun.
Any interest?
- Brett
I know that a lot of people see Delphi (Visual Pascal) as a toy. That it can't do anything, any better, than C++ can. However, Delphi is a well done implementation of Pascal with objects. It is recognized by many people as THE way to rapidly develop applications. It seems to me that if it were allowable, it would be neat to port the entire engine over to Delphi. This may be a pipe dream, and it may not be allowed by the TGE license (I haven't checked into that) but there are many good reasons to port, and maintain the TGE, to Delphi/Kylix.
1) Delphi is based around a KNOWN learning language: PASCAL.
2) Delphi is object based, just like C++, so it can handle many of the implementation issues that would be impossible in a port to, say, Visual Basic.
3) Most of the engine is OpenGL, so we're not depending on how well it compiles into assembly.. Although, I have to say that Borland makes some amazing compilers.
4) Most of the engine depends on OpenGL, which (nowadays) is hardware bound. It doesn't matter how fast the CPU is (as much anymore) because the graphics are handled by graphics hardware.
5) Delphi/Kylix are available on the two main platforms: Windows and Linux. I am sorry that Borland hasn't made a "Kylix" for Mac, because I prolly never would have converted to C++ in the first place.
6) There's evidence that being able to maintain another codebase (albeit a little laggy behind the main codebase) would be effective. Looking at the Delphi JEDI projects, which maintain standard C++ libraries for Delphi, it would be entirely possible to maintain the code.
There are more reasons, that I can't think of right now, to maintain a Delphi implementation of the TGE. I won't kid anyone... This would be a MAJOR undertaking, it would require a team of people to do the conversion (and maintain it), and it would be fun to do.
I would be interested in working with a group of people to do the port, if people were interested. It would certainly open up some more doors (not everyone is a master C++ coder). And, again, I maintain it would be fun.
Any interest?
- Brett
Torque Owner Sean Kerr
A few years ago I was looking into making a Borland C++ project, but then I realized the amount of work involved simply because all library files would need to be converted and on a daily basis. It's a large task.