ADFFS 2.64

I have the source to Wolf somewhere but I haven’t checked to see what language it’s in! I’ll try to dig it out when I get home.

Edit: As stated below, it’s BASIC assembly.

I don’t know much about the RISC OS port, but the MS-DOS original and other conversions I have seen were all written in C. So maybe it’s not entirely the programmer’s fault, but the compiler’s?

I have a very vague memory of seeing it described along the lines of “hand compiled” in an AU interview with the person who ported it to RISC OS. By that stage, I think we were having to pay more attention to performance than the developers on other platforms, so compiled C probably wouldn’t have worked too well on the hardware of the day.

The RISC OS version of Wolf 3D is written in BASIC assembler.

And yet ANSI C ftell() function returns signed integer 8-)

Which makes sense when you realise that the initial C libraries were created when the idea of a 2GiB file would have seemed ludicrous…

As Jeffrey has pointed out, its BASIC assembler.

753 signed shifts and 22 signed branches left to check, out of 1005 and 29 respectively! I’ve not looked at signed comparisons yet, so the problem is probably far worse.

I have a very vague memory of seeing it described along the lines of “hand compiled” in an AU interview with the person who ported it to RISC OS.

That would be Eddie Edwards. That interview was one of the 1995 issues of Acorn User. My only recollection of the article, is the £20,000 it cost him to purchase the licence!