How can CM's standard pink mask work on a mac but not on a pc?

[the cube]

I'm having a problem with a menus mod (designed to be less intrusive than the original and shown in the left and middle panels of the jpeg as the modded menus should appear, except of course for the compression-induced blurriness), made on a mac G4 running macos 9.1 and using macromedia's fireworks 4. The problem, as illustrated in the right panel of the jpeg, is that the masking of the color pink doesn't seem to carry over properly to pc's.

Further, as illustrated for the 'pause' command in the right panel of the jpeg, when a bmp made using GraphicsConvertor is substitued for the version made using fireworks, pc's seem to ignore it and instead access the originals directly from the CM disk. Interestingly, my mac behaved in much the same way after I installed a modded version of 32000.bmp (from DeanCo's gunmetal interface mod) meant to change text color and font.

My point is that if there are practical work-arounds they may involve more than just coordinating masking colors between mac and pc. The mask was defined by me to be, and therefore displays on my mac as, the standard shade (255,255,0) of pink. The files were then converted from the png format in which I built them to the standard bmp format required for public consumption with the masking color remaining unchanged at (255,255,0). After running these bmps through my macmod convertor app as usual they appear without problems on my mac (as illustrated in the left and middle panels of the jpeg). The problem is that despite the way that the masking shade (255,255,0) of pink shows up apparently unchanged on pc's, the masking effect is nevertheless destroyed.

In anticipation of the oft made claim that pc's can "see all the colors that a mac can and more" let me just say that it is not clear to me how this would bear on any of this. So why doesn't the mask carry over to pcs properly? I suppose it is possible that the problem may be the result of some relatively benign and thus easily correctible incompatibility between color tables as they are arranged in macs vs pc's and their applications. But based on the foregoing, I'm skeptical of this. Also, it is not clear to me how such a solution would explain in any obvious way the anomalous "ignoring" behaviour described above, though this may be unrelated to the problem at hand. Maybe this has something to do with the fact (DeanCo, who has also been spending time on this told me this) that the windows version of CM is a port of the macintosh code in which CM was originally written. In any event, any insights offered would be appreciated.

jeff.c1123 ]

[ 05-28-2001: Message edited by: jeff.c1123 ]