Performance Update

• Should be able to handle much larger canvases without slowdown
  • Target a maximum canvas size increase to 4000x2500px
• Operations that currently cause massive slowdown (fill, wand, move selection, auto-) should be parallelized or otherwise employ multi-threading while still ensuring thread and memory safety. Operations include:
  • fill, wand
  • move selection, rotate selection, stretch selection, reflect selection
• Should demand a lot less from the device and use less power, memory, and CPU