You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Use either arbitrary precision libraries or perturbation theory to allow deeper zoom levels.
The current limit is about 45 zooms, even with long doubles.
EDIT: I misunderstood the use of perturbation theory. It is effective at combating the slowdowns that come from using arbitrary precision libraries. It will still likely be slower than normal hardware floating point calculations.
Therefore, bignum libraries need to be implemented first.
The text was updated successfully, but these errors were encountered:
Use either arbitrary precision libraries or perturbation theory to allow deeper zoom levels.
The current limit is about 45 zooms, even with long doubles.
EDIT: I misunderstood the use of perturbation theory. It is effective at combating the slowdowns that come from using arbitrary precision libraries. It will still likely be slower than normal hardware floating point calculations.
Therefore, bignum libraries need to be implemented first.
The text was updated successfully, but these errors were encountered: