[mark_k]

Thanks, now it almost builds! I'm getting an error at the linking stage: LINK : fatal error LNK1181: cannot open input file '..\hq2x16.obj'. That was already mentioned above so I'll install Nasm and see if I can fix that.

I've attached the current build log in case it might be of any use (maybe VS 2013 gives more warnings than VS 2012?). Here are a few lines extracted from it:

  NoisePacker1.c
..\..\prowizard\rippers\NoisePacker1.c(33): warning C4018: '>' : signed/unsigned mismatch

  cpuemu_22.cpp
..\..\cpuemu_22.cpp(38822): warning C4146: unary minus operator applied to unsigned type, result still unsigned

  gfxboard.cpp
..\..\gfxboard.cpp(722): warning C4244: '=' : conversion from 'uint64_t' to 'uae_u32', possible loss of data
..\..\gfxboard.cpp(727): warning C4244: '=' : conversion from 'uint64_t' to 'uae_u32', possible loss of data
..\..\gfxboard.cpp(732): warning C4244: '=' : conversion from 'uint64_t' to 'uae_u32', possible loss of data

  cirrus_vga.cpp
..\..\qemuvga\cirrus_vga.cpp(1153): warning C4018: '<=' : signed/unsigned mismatch
..\..\qemuvga\cirrus_vga.cpp(2082): warning C4244: 'argument' : conversion from 'uint64_t' to 'uint32_t', possible loss of data
..\..\qemuvga\cirrus_vga.cpp(2107): warning C4244: '=' : conversion from 'uint64_t' to 'uint8_t', possible loss of data
..\..\qemuvga\cirrus_vga.cpp(2114): warning C4244: 'argument' : conversion from 'uint64_t' to 'uint32_t', possible loss of data

I'm assuming all those warnings are actually harmless, but wonder whether in some cases (e.g. gfxboard.cpp) they could indicate a possibility for optimisation? If a function returns a 64-bit integer, could modifying it to instead return a 32-bit int improve speed of the x86 build?