Allow deeper plots
Bug #783092 reported by
Ross Younger
This bug affects 1 person
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
brot2 |
Confirmed
|
Undecided
|
Ross Younger | ||
0.5-beta |
Won't Fix
|
Undecided
|
Unassigned | ||
1.0-gamma |
Won't Fix
|
Undecided
|
Ross Younger | ||
2.0-delta |
Confirmed
|
Undecided
|
Ross Younger |
Bug Description
This probably means abstracting it out.
rjk has some fixed-point assembler code. We might also use GMP/CLN/MPFR.
Could we swap multiple maths modules in/out dynamically and still keep performance? Possibly, if we use traits.
After some thought, I couldn't figure out a reliable way to predict a good MINIMUM_PIXEL_SIZE for different FP types; try the epsilon value at 3.0 but be prepared to rejig.
Changed in brot2: | |
status: | New → Confirmed |
Changed in brot2: | |
importance: | Undecided → Medium |
Changed in brot2: | |
importance: | Medium → Undecided |
Changed in brot2: | |
status: | Confirmed → Triaged |
status: | Triaged → New |
status: | New → Confirmed |
description: | updated |
description: | updated |
description: | updated |
To post a comment you must log in.
This might be best done as part of the fractals-by-formula blueprint which will completely overhaul the maths engine.