So here's the idea:

Pt-Off clears an 8x8 area, and draws to the screen.
Pt-On only draws over what's there.

Neither way really lets you draw something without a box around it, but that has white inside of it. For instance, a Character on an RPG with a Black outline, but white inside. If you could have a value for transparent pixels (probably by using nybbles, or just 4 bits per 2 Pixels) Then the (other, not changing pt-off and pt-on) command would draw a sprite to the screen, erasing where a "White" pixel is, but leaving transparent marked pixels alone. Also, when using both buffers with B/W, Transparent should let the backbuffer through, but White shouldn't (I.E. A cursor)

What do you think?

yeah that's what pt-mask does, I use it quite a bit, very helpful

and pt-on doesn't draw a box around it, it actually OR's it to the screen (draws only the black parts) to the buffer, and pt-off XOR's it (basically, kinda like inverting it onto the screen kinda sorta)

Umm, I didn't say anything about a boxfrom Pt-on. You just can't have any white inside of it.

But anyways, that's sort of what I had in mind. Does it act like Pt-On (just drawing over it) or Pt-Off? (Where it erases a square in the buffer)

My end goal was basically to be able to leave the backbuffer alone, while the front render can show objects with white pixels, even if the backbuffer has a black pixel there.

SirCmpwn doesn't your sprite 2 also need to be inverted too? Else, I think you also have to Pt-Change sprite 1 after displaying Sprite 2, making it a bit slower.