Here is a calibration run I made plus one someone else made.
The first image shows that lowering temperature eliminates stringing. I used 4.5mm retraction. I printed slowly only because the part was so small and hollow that I needed time to adjust the temp and mark the lines with a sharpie and take notes all at the same time. I wasn't able to go down to 170C because when I tried, Cura turned off the extruder (that part not shown) and ruined the test.
The next image was created by someone else
How does loose belts (backlash) cause gaps in diagonal lines you might ask? Here are some diagrams to help explain. Please don't tell me the diagrams are wrong without spending at least 30 minutes drawing it out very slowly and carefully like I did. Start with the Y axis backlash only diagram as that is the simplest one, the move onto X and then XY. In parts of these, the head isn't moving at all, or retraces it's path so the picture doesn't tell you everything that is going on but it does s