Mult3 - emulator
Vectrex calculates its vector positioning using mainly two factors:
the scale (length of integration)
the strength (voltage) of integration values
So a position X, e.g. 6000 (vectrex pixel position) (y the same) can be calculated with:
strength 60 x scale 100 = 6000
or
strength 100 x scale 60 = 6000
So I thought - and stated that about 18 years ago in my vectrex instructions. While in "general" this is true and can be used, specifically this is NOT true. In many cases even IF there is a slight difference - it does not matter. Since usually object are allways drawn with the same scale factor and the same strength value.
If you use different values, you usually draw different things on screen no one ever notices a difference. Like drawing "backgorund" with large scale factors and sprites and such with small factors.
BUT there are a few programs which DO draw "combined" "objects" using different values. E.G. Cleansweep (see section why emulators hate cleansweep).
For all people who have access to a vectrex I have written an easy (quite a few actually but one ni particular) example program to show that the difference is very real!
See: TestMultiply3.asm
Let me in short explain the program:
two dots are drawn near the center
using joystick X you can position the dot from 0-$ff (although only up to $7f make sense)
one dot is drawn with a scale of 1 and strength of position
the other is drawn with scale of position and strength of 1
Mult3 - emulator
Mult3 - vectrex
While on these pictures the difference may not seem all that substantial, I might remind you, that we only look at coordinates "+127" (from about a maximum of 20000).
The difference DOES add up.
I don't know if that is the only "CleanSweep" fault, but this is a major player!
I am pretty sure there is at least one other "undiscovered" feature. Mostly visible with "optimized" raster displays. I will sometime in the future further research that!