What could be wrong:
1. I removed an old copy of vtkdll.dll from the system32 directory
2. Ran vtk23.exe and saw a new copy of vtkdll.dll being placed in the system32 directory.
3. Compiled a simple cone example cleanly.
4. On the build process VC++60 Linker reports a corrupted vtkdll.dll again as in a previous release.
 
Has anyone validated the extracted vtkdll.dll that is extracted from vtk23.exe ? 
I don't think that any cleanly compiled code could in any way corrupt a dll.
Nor do I think the inclusion or exclusion of any static libs should cause the reporting of a corrupt dll.
BTW the same thing happened with vtk22.exe. By what do I know ! 
 
Was the vtk2.3 release of vtkdll.dll created with VC++ 60 ? Perhaps creating my own .dll might be a solution, but it still doesn't answer the question concerning  VC++ 60 reporting it as being corrupt.
 
 
Seeking your best advice,
Bob Palank