mzan has quit [Read error: Connection reset by peer]
pera has quit [Ping timeout: 252 seconds]
lavaflow has quit [Read error: No route to host]
lavaflow has joined #racket
pera_ has joined #racket
crsc has quit [Quit: leaving]
mzan has joined #racket
ZombieChicken has quit [Ping timeout: 256 seconds]
lavaflow has quit [Read error: No route to host]
lavaflow has joined #racket
davidl has quit [Ping timeout: 252 seconds]
iyzsong has joined #racket
crsc has joined #racket
davidl has joined #racket
ubLIX has joined #racket
YuGiOhJCJ has joined #racket
joebobjoe has joined #racket
joebobjoe has quit [Ping timeout: 246 seconds]
pera_ has quit [Ping timeout: 244 seconds]
pera has joined #racket
jsomedon has joined #racket
selimcan has joined #racket
selimcan has quit [Ping timeout: 264 seconds]
iyzsong has quit [Ping timeout: 252 seconds]
maximjaffe has joined #racket
jsomedon has quit [Quit: jsomedon]
_whitelogger has joined #racket
pera has quit [Ping timeout: 240 seconds]
joebobjoe has joined #racket
YuGiOhJCJ has quit [Quit: YuGiOhJCJ]
joebobjoe has quit [Ping timeout: 268 seconds]
dan_f has joined #racket
dan_f has quit [Client Quit]
dan_f has joined #racket
dddddd has joined #racket
joebobjoe has joined #racket
joebobjoe has quit [Ping timeout: 252 seconds]
jao has joined #racket
jsomedon has joined #racket
jsomedon has quit [Client Quit]
ziyourenxiang has quit [Ping timeout: 268 seconds]
joebobjoe has joined #racket
joebobjoe has quit [Ping timeout: 272 seconds]
ng0 has quit [Quit: WeeChat 1.6]
hjek has joined #racket
ng0 has joined #racket
hjek has quit [Quit: Leaving.]
hjek has joined #racket
joebobjoe has joined #racket
ubLIX has joined #racket
joebobjoe has quit [Ping timeout: 268 seconds]
davidl has quit [Ping timeout: 240 seconds]
davidl has joined #racket
hjek has quit [Quit: Leaving.]
wingsorc has joined #racket
pera_ has joined #racket
vraid has joined #racket
vraid has quit [Disconnected by services]
vraid has joined #racket
joebobjoe has joined #racket
dustyweb has quit [Ping timeout: 264 seconds]
wingsorc is now known as emacs_apprentice
zmt00 has joined #racket
emacs_apprentice is now known as wingsorc
pera_ has quit [Ping timeout: 244 seconds]
dustyweb has joined #racket
pierpal has quit [Quit: Poof]
pierpal has joined #racket
libertyprime has joined #racket
ZombieChicken has joined #racket
pera_ has joined #racket
dmiles has quit [Ping timeout: 246 seconds]
joebobjoe has quit [Ping timeout: 244 seconds]
libertyprime has quit [Ping timeout: 272 seconds]
dmiles has joined #racket
libertyprime has joined #racket
err0ne has joined #racket
ubLIX has quit [Quit: ubLIX]
libertyprime has quit [Ping timeout: 246 seconds]
ZombieChicken has quit [Remote host closed the connection]
ZombieChicken has joined #racket
pera_ has quit [Ping timeout: 240 seconds]
pera has joined #racket
ubLIX has joined #racket
libertyprime has joined #racket
<lavaflow>
is there much overhead to boxes?
<lavaflow>
I ask because I'm caching some cons cells. some of them need to be ((box ...) . (box ...))
<lavaflow>
some of them could just be a single non-boxed number and a box, but for the sake of consistency I'd like to make those ones boxed too even though it's not necessary, just so I can always return a cons of two boxes and not have to worry about the difference
<lavaflow>
but if boxes have much overhead then I'll avoid using them needlessly.
<lavaflow>
hmm I should probably avoid it I think
<lavaflow>
actually not always boxing introduces another conditional check which could actually be serious since this is going to be pretty hot code... screw it I'll just do both and profile it.
friscosam has joined #racket
joebobjoe has joined #racket
joebobjoe has quit [Ping timeout: 268 seconds]
vraid has quit [Ping timeout: 244 seconds]
<jcowan>
lavaflow: You might want to use mcons instead.
keep_learning has joined #racket
<lavaflow>
that's what I was doing before. the problem is I have a few million cons cells that collectively refer to a only a few thousand different values. I need to update those values quickly.
<lavaflow>
so using cons cells of boxes I can have each of those values only exist once, so I can update it all in a single set
<lavaflow>
well, a single set for the cdr's. Plus a few more for the car's which actually have unique values
<lavaflow>
basically I'm storing a big pile of probabilities as (numerator . denominator). denominators get updated often but there are only a few thousand unique denominators