00:00:00 --- log: started retro/08.07.07
01:45:30 --- quit: Raystm2 (Read error: 110 (Connection timed out))
03:36:34 --- quit: nighty__ (Read error: 110 (Connection timed out))
03:50:19 --- quit: virl (Remote closed the connection)
12:22:37 --- join: virl (n=virl__@chello062178085149.1.12.vie.surfer.at) joined #retro
13:20:00 * crc is working on a module which will add an inline class for the words that correspond to single ngaro instructions. It may help performance, and will allow generation of tighter code.
13:42:17 <crc> results are in
13:42:27 <crc> Standard compiler:
13:42:34 <crc>  - Recursive FIB (39)... 208.0144 seconds
13:42:34 <crc>  - Countdown Loop (1,000,000)... 12.228482 seconds
13:42:34 <crc>  - Nest/Unnest (256 million pairs)... 42.299306 seconds
13:42:49 <crc> With Inlining:
13:42:56 <crc>  - Recursive FIB (39)... 85.860101 seconds
13:42:56 <crc>  - Countdown Loop (1,000,000)... 12.38329 seconds
13:42:56 <crc>  - Nest/Unnest (256 million pairs)... 30.796164 seconds
13:43:22 <crc> a noticeable boost in two cases, and a tiny slowdown in one. 
13:43:42 <crc> This is running under my standard VM, not the new threaded one
13:46:11 <crc> Threaded VM with Inlining:
13:46:13 <crc>  - Recursive FIB (39)... 51.989871 seconds
13:46:13 <crc>  - Countdown Loop (1,000,000)... 8.292265 seconds
13:46:13 <crc>  - Nest/Unnest (256 million pairs)... 17.577663 seconds
13:46:35 <crc> I'm running the tests without inlining now...
13:48:50 <crc>  - Recursive FIB (39)... 130.537818 seconds
13:48:50 <crc>  - Countdown Loop (1,000,000)... 8.325312 seconds
13:48:50 <crc>  - Nest/Unnest (256 million pairs)... 27.092447 seconds
13:49:26 <crc> It looks like inlining is a definite win. I may need to make it part of the retro core...
14:45:29 --- join: Raystm2 (n=Ray@unaffiliated/raystm2) joined #retro
14:45:30 --- mode: ChanServ set +o Raystm2
14:48:45 <crc> hi Ray
14:49:19 <crc> http://retroforth.blogspot.com/2008/07/improving-performance.html
15:45:59 <Raystm2> hi Charles. Back from the store. Reading the report. This is going to be good. :)
16:11:02 <crc> I'm pleased with the results
17:48:28 --- join: nighty__ (n=nighty@210.188.173.246) joined #retro
17:57:06 --- join: Raystm2_ (n=Ray@unaffiliated/raystm2) joined #retro
17:57:06 --- mode: ChanServ set +o Raystm2_
17:58:16 --- nick: Raystm2_ -> release
17:58:34 --- nick: release -> Raystm2_
17:59:28 --- nick: Raystm2 -> Guest63962
18:01:12 --- nick: Raystm2_ -> Raystm2
18:02:39 --- kick: Guest63962 was kicked by Raystm2 (pow!)
18:02:44 --- mode: Raystm2 set -o Raystm2
23:58:30 --- join: jamc (i=ly9i0qko@90-231-88-54-no84.business.telia.com) joined #retro
23:59:59 --- log: ended retro/08.07.07