Are resource files compiled as UNICODE or ANSI code-page?

First - my apologies if this has been answered a hundred times over! D'oh!

But my search-fu apparently sucks, as I'm having no luck answering this basic question:

How are resources stored in the EXE/DLL? As UNICODE (UCS-2, Windows native internal character format), or as multibyte characters using the code-page of the resources block?

• How does one embed UNICODE strings into one's resources (.rc)?
• Can UNICODE (UCS-2) text be inserted into the language strings from within VS 2012?
• Is Windows still using UCS-2, or is it using UTF16 internally?

I'm just looking for general answers, or links to details, rather than a detailed how-to for putting a UNICODE string into an .rc string table. Thanks!

All resource strings in WIN32 are compiled as Unicode. See here for more info. The .rc script itself can be ANSI (using the local codepage) or UCS-2 with the appropriate BOM (reference).