00:47
<
yopp >
kalman filter is awesome, but it's kinda brain breaking
01:35
dimday has joined #rubinius
02:37
houhoulis has quit [Remote host closed the connection]
03:00
dimday has quit [Ping timeout: 256 seconds]
03:01
dimday has joined #rubinius
05:40
GitHub28 has joined #rubinius
05:40
<
GitHub28 >
rubinius/master 19e3174 Brian Shirai: More locking fixes.
05:40
GitHub28 has left #rubinius [#rubinius]
05:57
travis-ci has joined #rubinius
05:57
travis-ci has left #rubinius [#rubinius]
07:39
heroux_ has joined #rubinius
07:41
heroux has quit [Read error: Connection reset by peer]
07:41
heroux_ is now known as heroux
07:56
mister_solo has joined #rubinius
08:06
amclain has quit [Quit: Leaving]
08:29
mister_solo has quit [Ping timeout: 255 seconds]
08:51
mister_solo has joined #rubinius
09:09
dimday has quit [Ping timeout: 246 seconds]
09:24
mister_solo has quit [Ping timeout: 256 seconds]
09:57
mister_solo has joined #rubinius
10:02
mister_solo has quit [Ping timeout: 265 seconds]
10:26
mister_solo has joined #rubinius
10:43
guilleiguaran__ has quit [*.net *.split]
10:43
halorgium has quit [*.net *.split]
10:43
lbianc has joined #rubinius
10:43
halorgium has joined #rubinius
10:50
guilleiguaran__ has joined #rubinius
11:18
djellemah has quit [Ping timeout: 256 seconds]
11:22
meh` has joined #rubinius
11:34
<
yopp >
there still no jit on arm, right?
11:46
<
yopp >
Runtime: 0.09 (Ruby 2.2.2)
11:46
<
yopp >
Runtime: 0.37 (rubinius 2.5.5)
11:52
<
yorickpeterse >
Correct
11:56
kagaro has joined #rubinius
11:57
nirvdrum has joined #rubinius
12:07
<
yopp >
thats on x86
12:07
<
yopp >
I'm running agains daily well data, but seems like it's still much worse than mri
12:08
<
yorickpeterse >
without any repro/whatever there's little we can help with
12:10
<
yopp >
working on it, just cleaning stuff a bit
12:19
<
yopp >
; Runtime: 4.17
12:19
<
yopp >
; Runtime: 10.51
12:19
<
yopp >
first mri, second rbx
12:19
<
yopp >
; Runtime: 14.60
12:20
<
yorickpeterse >
well, at least we're doing better than JRuby :P
12:20
<
yorickpeterse >
does this take JIT warmups into account?
12:20
<
yorickpeterse >
hm, doesn't seem like it
12:20
<
yopp >
it's plain runtime on small dataset
12:20
<
yorickpeterse >
can you turn this into a benchmark-ips benchmark?
12:22
<
yopp >
yep, gimme a sec
12:32
<
yorickpeterse >
Hmmm
12:32
<
yorickpeterse >
now could you report this as an issue? :P
12:33
<
yopp >
I've removed "benchmark" block, but it made no diffirence
12:33
<
yopp >
I remember something about blocks and JIT
12:36
<
yopp >
ah, okay, jruby actually faster
12:37
<
yopp >
I've tested against jruby 9.0.0.0-pre
12:37
<
yopp >
What you would like to see in the issue?
12:37
<
yopp >
Just "not fast enough?" :)
12:37
<
yorickpeterse >
Well, a description of the end use case would be useful
12:37
<
yorickpeterse >
to give some idea what this would be used for
12:39
<
yopp >
The first gist is pretty much it: I'm filtering data from ultrasonic range sensor, to get a water level in my well :)
12:41
mister_solo has quit [Ping timeout: 256 seconds]
12:42
<
yopp >
; Runtime: 10.82
12:42
<
yopp >
; jruby 1.7.20.1 (1.9.3p551) 2015-06-10 d7c8c27 on Java HotSpot(TM) 64-Bit Server VM 1.7.0_67-b01 +jit [darwin-x86_64]
12:42
<
yopp >
; Runtime: 9.24
12:42
<
yopp >
; rubinius 2.5.5 (2.1.0 433d9b8c 2015-05-15 3.5.1 JI) [x86_64-darwin14.4.0]
12:57
<
yopp >
hum, jruby is actually better than mri
12:58
<
yopp >
on the bench/ips
12:58
<
yopp >
but on the real data it's actually slower
13:04
<
yorickpeterse >
that's probably the JIT warming up
13:04
<
yorickpeterse >
also I think pre2 is much faster than pre1
13:15
cremes has joined #rubinius
13:38
<
yopp >
in /usr/local/Cellar/crystal-lang/0.7.3/src/matrix.cr:220: undefined method '*' for Nil
13:39
<
yopp >
(╯°□°)╯︵ ┻━┻
13:44
djellemah has joined #rubinius
13:49
<
yopp >
; Runtime: 0.85
13:49
<
yopp >
crystal rocks!
14:06
mister_solo has joined #rubinius
14:12
RageLtMan has quit [Ping timeout: 265 seconds]
14:47
RageLtMan has joined #rubinius
15:12
|jemc| has joined #rubinius
15:19
rakaur_ has quit [Read error: Connection reset by peer]
15:19
rakaur has joined #rubinius
15:46
heroux has quit [Ping timeout: 244 seconds]
15:56
|jemc| has quit [Ping timeout: 252 seconds]
15:57
|jemc| has joined #rubinius
16:03
mister_s_ has joined #rubinius
16:07
mister_solo has quit [Ping timeout: 248 seconds]
16:46
mister_s_ has quit [Quit: So long, and thanks for all the fish!]
16:47
heroux has joined #rubinius
17:04
pietr0 has joined #rubinius
18:26
pietr0_ has joined #rubinius
18:27
pietr0 has quit [Ping timeout: 250 seconds]
18:27
pietr0_ is now known as pietr0
18:54
GitHub44 has joined #rubinius
18:54
<
GitHub44 >
rubinius/1.8.7 3f00924 Brian Shirai: Merge remote-tracking branch 'origin' into 1.8.7
18:54
<
GitHub44 >
rubinius/1.8.7 854648a Brian Shirai: Fixed locking for Thread#critical= lock.
18:54
GitHub44 has left #rubinius [#rubinius]
19:23
travis-ci has joined #rubinius
19:23
travis-ci has left #rubinius [#rubinius]
19:56
pipework is now known as Spaceghostc2c
20:10
Spaceghostc2c is now known as |work
21:44
unreal_ has quit [Ping timeout: 264 seconds]
21:47
unreal has joined #rubinius
22:12
GitHub123 has joined #rubinius
22:12
<
GitHub123 >
[rubinius] nirvdrum opened pull request #3448: Fixed return value for Rubinius::Mirror::Range#step_float_iterations_… (master...fix_step_float_iterations_size)
http://git.io/vLFkl
22:12
GitHub123 has left #rubinius [#rubinius]
22:32
GitHub160 has joined #rubinius
22:32
<
GitHub160 >
[rubinius] YorickPeterse closed pull request #3448: Fixed return value for Rubinius::Mirror::Range#step_float_iterations_… (master...fix_step_float_iterations_size)
http://git.io/vLFkl
22:32
GitHub160 has left #rubinius [#rubinius]
22:32
GitHub138 has joined #rubinius
22:32
<
GitHub138 >
rubinius/master 2ceaa67 Yorick Peterse: Merge pull request #3448 from nirvdrum/fix_step_float_iterations_size...
22:32
GitHub138 has left #rubinius [#rubinius]
22:32
<
GitHub138 >
rubinius/master 03b3154 Kevin Menard: Fixed return value for Rubinius::Mirror::Range#step_float_iterations_size.
22:51
travis-ci has joined #rubinius
22:51
travis-ci has left #rubinius [#rubinius]
23:19
coffeeju1 has quit [Ping timeout: 245 seconds]
23:30
meh` has quit [Ping timeout: 255 seconds]
23:32
coffeeju1 has joined #rubinius
23:37
pietr0 has quit [Quit: pietr0]