My friend
I will not argue on the advantage to work with sprite object, suffice to say that it is just like working with multiple transparent layers in PSD, and when you are pleased with everything then you merge (flatten) all the layers.
Vectorial objects could be antialiased or NOT, using the ZS_DRAFT style to work like GDI32.
See the attached screen shot
top DRAFT mode
bottom antialias mode
and this could be changed on the fly, something impossible when you work directly with a GDI32 DC.
Thus better to work with vectorial sprite objects, and flatten them once to create the final texture file, but only when the whole editing process is over, this is exactly what gMap64 does when you use the "Create a project image" button.
Morever gMap64, shows always the real mouse cursor coordinates (in pixel) whatever the zoom being used to help the whole editing process, and for each of the sprite their respective XY top left coordinates. In GDImage, vectorial objects could be moved arround and manipulated just the same than raster bitmap. You can even move them pixel by pixel, using the arrow keys, for ultimate precision.
BTW, the main window of gMap64 could be moved inside of OR, and the splitters could be used to enlarge the view port just like gMap64 does. Then you could put it aside of the current unused splitter, but we will need also the windowtool zoom.
If you haven't tried gMap64, perhaps you can give it a run to better understand how it works.