I love that the image preview I get for the first image is just "icant decimal"
This eventually got fixed, but some of the discussion is 
. FP is approximate, so introducing extra error is ok!
https://bugs.chromium.org/p/v8/issues/detail?id=3006 …pic.twitter.com/DwHzClQfLt
-
-
- 1 more reply
New conversation -
-
-
.
@danluu@ffee_machine ahh yes, cubic splines, my method of choice for trigonometric f'n approximation impls. still max WTF on every read.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
not sure if you disagree with extra error, range you get huge gaps (ulp) and range reduction throws them away
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
discussing it as something you 'can't hear' is a huge unforced error though :]
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
interesting discussion, reflecting several misconceptions. BTW, I added mod2pi to Julia wrapping Sun’s rem_2pi range reduction :-)
-
“proper” range reduction (inside sin/cos) reduces not to [0,2pi), but +/- pi/4 and info on quadrant, giving 3 extra bits precision
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.