I've read the thread with interest several times now (and a few in the past). I notice very few references to what the maps will be used for (certainly decides what I will use) or the actual workflow used (and many of us go from one prog to another to another and back again etc in doing a single map).

If I were writing a book that needed a lot of maps (I'm not), I'd want my maps to be 'beautiful' and with a clear and consistent style. They would be worth spending a lot of time over.

If I were planning a RPG campaign and wanted a world or regional map, I'd mostly want accuracy of scale and detail (and zoomability would be good). If I were going to let players see it or have a copy, then attractiveness and style would become more important.

Dungeon plans for/by the GM also mostly need accuracy of scale and detail.

If I were designing a battlemap, then atmosphere and usability would be more important. And ease and speed of creating the map would be very important.

And if the battlemap were for use in a VTT, then I would want all or most of the components of the map to be separable because I may well need to apply or change them from within the VTT.

And if designing components for any of the above, the amount of time I'd be prepared to spend depends on how much I would use it (and what is already out there that would do).

Most of my mapmaking is for battlemaps for VTTs. As far as I can see, the cost benefit of the various types of software (and I use virtually all of them for one thing or another) in terms of time spent Vs atmosphere/attractiveness is strongly in favour of the software designed for this use (Dundjinni, Dungeonforge, MapX, DungeonCrafter in the past, CC3?). If someone just wants to do battlemaps, and is not already experienced in use of the other software (or has another use for it) then this is probably the best way for them to go especially since the learning curve for most of them is small (not sure about CC3 for this). Despite these progs not being the most used or recommended on this site, and not producing the most individually styled or 'beautiful' maps (though this is partly dependent on the images used - and the final and/or intermediate stages can be tweaked in other progs). And this probably also stands for battlemaps that will be printed for use (a main reason for the initial popularity of Dundjinni). Because so many battlemaps may be needed, the time taken to do the map is a major issue.

And vector approaches (CC3, Inkscape, Xara etc) have real advantages for the world/regional design maps for the GM. Rasters only compete here by having lots of maps at different scales.