Speed up sin and cos by a factor 6.5 #117
Conversation
dpsanders
changed the title
dpsanders
force-pushed
the
faster_trig
branch
2 times, most recently
from
72ea4dd
to
b7b1d03
Compare
|
I have not found a sensible way to do tan correctly. Hence leaving tan aside for now. |
dpsanders
changed the title
|
Is this ready for review? Ping me when it is ready. (Regarding |
|
Yes, ready for review. |
Codecov Report
@@ Coverage Diff @@
## master #117 +/- ##
==========================================
- Coverage 92.81% 92.65% -0.16%
==========================================
Files 21 21
Lines 960 967 +7
==========================================
+ Hits 891 896 +5
- Misses 69 71 +2
Continue to review full report at Codecov.
|
| Multiply an interval by a positive constant. | ||
| For efficiency, does not check that the constant is positive. | ||
| """ | ||
| multiply_by_positive_constant(α, x::Interval) = @round(α*x.lo, α*x.hi) |
There was a problem hiding this comment.
Perhaps instead of defining multiply_by_positive_constant we should rather define the methods *(a::T, x::Interval) and *(x::Interval, a::T) (in arithmetic.jl), where T<:Real (or maybe <:Number), or even unspecified, as it is here. Currently we use promotion to an interval and rounding, which may slow down things.
Question: Below you use this function for to define half_pi(::Type{Float64}), but not for other types; can you explain why?
There was a problem hiding this comment.
I use it for two_pi as well.
Actually the half_pi function is no longer used, I believe.
We may well want to define explicit methods for *(constant, Interval). I've opened an issue: #120
There was a problem hiding this comment.
That was actually the suggestion.
There was a problem hiding this comment.
Right, but I'd prefer to leave it for a separate PR.
Also, if you actually know that the constant is positive then this specialised version is even slightly quicker.
| temp = atomic(Interval{T}, x) / half_pi(x) | ||
| (floor(temp.lo), floor(temp.hi)) | ||
|
|
||
| return SVector(floor(temp.lo), floor(temp.hi)) |
There was a problem hiding this comment.
What about returning a tuple here, instead of a SVector? (Same comment applies in line 44.)
There was a problem hiding this comment.
It used to be a tuple; changing it to an SVector had a ridiculously big impact on the speed due to the minimum and maximum below that are not specialised for tuples!
In any case all of this code can be simplified a lot when #119 is possible, I believe.
There was a problem hiding this comment.
OK, I found an alternative solution: return a tuple and do
max(tuple...)
Too late...
Current master:
This PR: