00:00:00 --- log: started forth/09.12.03 00:01:03 --- quit: dinya (Read error: 104 (Connection reset by peer)) 00:48:04 --- quit: nighty^ ("Disappears in a puff of smoke") 01:35:34 --- join: |dinya_| (n=Denis@188.16.105.99) joined #forth 01:40:35 --- quit: dinya_ (Read error: 60 (Operation timed out)) 02:54:08 --- join: dinya_ (n=Denis@188.17.91.19) joined #forth 03:08:16 --- quit: proteusguy_ (Read error: 110 (Connection timed out)) 03:08:51 --- join: proteusguy_ (n=proteusg@zeppelin.proteus-tech.com) joined #forth 03:11:05 --- quit: |dinya_| (Read error: 110 (Connection timed out)) 03:25:45 --- quit: Al2O3 (Read error: 60 (Operation timed out)) 03:33:25 --- join: Al2O3 (n=Al2O3@c-75-70-11-191.hsd1.co.comcast.net) joined #forth 03:42:23 --- quit: zbrown (farmer.freenode.net irc.freenode.net) 03:42:23 --- quit: dinya_ (farmer.freenode.net irc.freenode.net) 03:42:23 --- quit: nighty_ (farmer.freenode.net irc.freenode.net) 03:42:24 --- quit: tarbo (farmer.freenode.net irc.freenode.net) 03:42:24 --- quit: Quartus__ (farmer.freenode.net irc.freenode.net) 03:42:24 --- quit: mathrick (farmer.freenode.net irc.freenode.net) 03:42:24 --- quit: madgarden (farmer.freenode.net irc.freenode.net) 03:42:25 --- quit: I440r (farmer.freenode.net irc.freenode.net) 03:42:25 --- quit: KipIngram (farmer.freenode.net irc.freenode.net) 03:42:25 --- quit: madwork (farmer.freenode.net irc.freenode.net) 03:42:25 --- quit: probonono (farmer.freenode.net irc.freenode.net) 03:42:25 --- quit: ASau` (farmer.freenode.net irc.freenode.net) 03:42:26 --- quit: uiu_ (farmer.freenode.net irc.freenode.net) 03:42:27 --- quit: gogonkt (farmer.freenode.net irc.freenode.net) 03:42:28 --- quit: hiredman (farmer.freenode.net irc.freenode.net) 03:42:28 --- quit: maht (farmer.freenode.net irc.freenode.net) 03:42:28 --- quit: yiyus (farmer.freenode.net irc.freenode.net) 03:42:28 --- quit: Zarutian (farmer.freenode.net irc.freenode.net) 03:42:29 --- quit: gnomon (farmer.freenode.net irc.freenode.net) 03:42:30 --- quit: Al2O3 (farmer.freenode.net irc.freenode.net) 03:42:30 --- quit: Snoopy_1711 (farmer.freenode.net irc.freenode.net) 03:42:32 --- quit: foxes (farmer.freenode.net irc.freenode.net) 03:42:32 --- quit: uiu (farmer.freenode.net irc.freenode.net) 03:42:32 --- quit: cataska (farmer.freenode.net irc.freenode.net) 03:42:32 --- quit: nottwo (farmer.freenode.net irc.freenode.net) 03:44:00 --- quit: proteusguy_ (farmer.freenode.net irc.freenode.net) 03:44:00 --- quit: segher (farmer.freenode.net irc.freenode.net) 03:44:00 --- quit: crc (farmer.freenode.net irc.freenode.net) 03:44:01 --- quit: aguaithegeek (farmer.freenode.net irc.freenode.net) 03:44:25 --- join: I440r (n=me@c-69-136-171-118.hsd1.in.comcast.net) joined #forth 03:44:25 --- join: nighty^ (n=nighty@x122091.ppp.asahi-net.or.jp) joined #forth 03:44:25 --- join: Al2O3 (n=Al2O3@c-75-70-11-191.hsd1.co.comcast.net) joined #forth 03:44:25 --- join: dinya_ (n=Denis@188.17.91.19) joined #forth 03:44:25 --- join: ASau` (n=user@77.246.231.14) joined #forth 03:44:25 --- join: Snoopy_1711 (i=Snoopy_1@dslb-084-059-099-101.pools.arcor-ip.net) joined #forth 03:44:25 --- join: nighty_ (n=nighty@x122091.ppp.asahi-net.or.jp) joined #forth 03:44:25 --- join: hiredman (n=rattus@67.152.53.125) joined #forth 03:44:25 --- join: tarbo (n=me@unaffiliated/tarbo) joined #forth 03:44:25 --- join: uiu_ (n=ian@schihei.net) joined #forth 03:44:25 --- join: Quartus__ (n=Quartus`@74.198.8.58) joined #forth 03:44:25 --- join: mathrick (n=mathrick@users177.kollegienet.dk) joined #forth 03:44:25 --- join: maht (n=maht__@85.189.31.174.proweb.managedbroadband.co.uk) joined #forth 03:44:25 --- join: yiyus (i=12427124@je.je.je) joined #forth 03:44:25 --- join: Zarutian (n=zarutian@194-144-84-110.du.xdsl.is) joined #forth 03:44:25 --- join: gogonkt (n=info@113.69.187.80) joined #forth 03:44:25 --- join: madgarden (n=madgarde@CPE001d7e527f89-CM00159a65a870.cpe.net.cable.rogers.com) joined #forth 03:44:25 --- join: KipIngram (n=kip@173-11-138-177-houston.txt.hfc.comcastbusiness.net) joined #forth 03:44:25 --- join: gnomon (n=gnomon@CPE0022158a8221-CM000f9f776f96.cpe.net.cable.rogers.com) joined #forth 03:44:25 --- join: foxes (i=flash@221.220.46.126) joined #forth 03:44:25 --- join: madwork (n=madgarde@204.138.110.15) joined #forth 03:44:25 --- join: probonono (n=User@unaffiliated/probonono) joined #forth 03:44:25 --- join: uiu (n=ian@HSI-KBW-078-042-132-111.hsi3.kabel-badenwuerttemberg.de) joined #forth 03:44:25 --- join: cataska (n=cataska@210.64.6.233) joined #forth 03:44:25 --- join: nottwo (n=trannie@designvox-gw.iserv.net) joined #forth 03:44:25 --- join: zbrown (n=suifur@unaffiliated/zbrown) joined #forth 03:44:25 --- mode: irc.freenode.net set +o I440r 03:44:40 --- quit: tarbo (SendQ exceeded) 03:46:42 --- join: crc (n=charlesc@c-68-80-139-0.hsd1.pa.comcast.net) joined #forth 03:46:42 --- join: proteusguy_ (n=proteusg@zeppelin.proteus-tech.com) joined #forth 03:46:42 --- join: segher (n=segher@84-105-60-153.cable.quicknet.nl) joined #forth 03:46:42 --- mode: irc.freenode.net set +o crc 03:46:49 --- quit: nighty^ ("Disappears in a puff of smoke") 03:52:38 --- join: aguaithegeek (n=aguai@98.142.211.123) joined #forth 03:54:32 --- join: nighty^ (n=nighty@x122091.ppp.asahi-net.or.jp) joined #forth 04:16:34 --- quit: gogonkt (Read error: 110 (Connection timed out)) 04:27:44 --- quit: nighty^ ("Disappears in a puff of smoke") 04:28:39 --- join: tarbo (n=me@unaffiliated/tarbo) joined #forth 04:37:29 --- join: DrunkTomato (n=DEDULO@ext-gw.wellcom.tomsk.ru) joined #forth 04:38:43 --- join: gogonkt (n=info@113.69.186.94) joined #forth 05:42:30 Good morning all. 05:42:40 Good evening. 05:46:30 --- nick: ASau` -> ASau 06:02:07 good afternoon 06:11:48 --- join: sunwukong (n=vukung@business-80-99-161-225.business.broadband.hu) joined #forth 06:15:46 --- quit: TreyB ("leaving") 06:15:53 --- join: TreyB (n=trey@adsl-76-254-113-201.dsl.hstntx.sbcglobal.net) joined #forth 06:32:06 --- join: crcx (i=d8012b82@gateway/web/freenode/x-yhjuonzkqgzdypcc) joined #forth 06:53:41 --- join: TreyB__ (n=trey@76.254.113.201) joined #forth 06:58:19 --- quit: TreyB (Read error: 110 (Connection timed out)) 07:37:32 --- quit: TreyB__ ("leaving") 07:37:36 --- join: TreyB (n=trey@76.254.113.201) joined #forth 07:43:13 --- quit: sunwukong ("bye") 07:48:13 --- quit: ASau ("off") 08:08:30 --- join: crc_ (n=charlesc@c-68-80-139-0.hsd1.pa.comcast.net) joined #forth 08:08:30 --- mode: ChanServ set +o crc_ 08:09:21 --- quit: crc (Read error: 60 (Operation timed out)) 08:09:40 --- nick: crc_ -> crc 08:36:30 --- join: PoppaVic (n=pops@adsl-99-150-133-146.dsl.sfldmi.sbcglobal.net) joined #forth 09:10:50 --- quit: crcx ("Page closed") 09:10:50 --- join: Snoopy_1611 (i=Snoopy_1@dslb-084-059-197-145.pools.arcor-ip.net) joined #forth 09:20:39 --- join: GeDaMo (n=gedamo@dyn-62-56-89-110.dslaccess.co.uk) joined #forth 09:39:20 --- quit: Snoopy_1711 (Read error: 110 (Connection timed out)) 09:39:26 --- join: ASau (n=user@83.69.227.32) joined #forth 10:30:25 --- join: dinya (i=Denis@94.180.67.97) joined #forth 10:35:12 --- quit: I440r (Client Quit) 10:50:57 --- join: kar8nga (n=kar8nga@jol13-1-82-66-176-74.fbx.proxad.net) joined #forth 11:22:35 --- join: crcx (i=d8012b82@gateway/web/freenode/x-yebiaqbozucwshnj) joined #forth 11:24:28 --- join: Maki (n=Maki@dynamic-78-30-155-247.adsl.eunet.rs) joined #forth 11:54:37 --- join: dgou (n=dgou@c-24-23-119-48.hsd1.pa.comcast.net) joined #forth 12:05:17 --- quit: mathrick (Read error: 104 (Connection reset by peer)) 12:06:06 --- join: mathrick (n=mathrick@users177.kollegienet.dk) joined #forth 12:07:57 --- part: dgou left #forth 12:52:09 --- join: Snoopy_1711 (i=Snoopy_1@dslb-088-068-207-151.pools.arcor-ip.net) joined #forth 13:02:17 --- quit: Snoopy_1611 (Read error: 110 (Connection timed out)) 13:02:22 --- join: Snoopy_1611 (i=Snoopy_1@dslb-084-059-201-040.pools.arcor-ip.net) joined #forth 13:17:09 --- quit: Snoopy_1711 (Read error: 110 (Connection timed out)) 13:24:04 Does anyone have an opinion on branch predication in Forth-like architectures? 13:25:56 That's hard to tell. 13:27:28 That corresponds to AGAIN/UNTIL/REPEAT. 13:28:32 Well, my cursory survey of the technique indicates that while it's possible for branch predication to slow overall execution (because you execute both the "then" part and the "else part" but keep only one set of results), you generally lose speed only when the alternate sections are lengthy. The benefit is no disruption of the instruction pipeline. 13:28:57 In good Forth you don't have long blocks of code, so that would seem to be a good situation for branch predication. 13:29:26 Provided that *all* code can conditionally execute (both primitives and subroutine calls). 13:29:32 So if you structured your code like this: 13:30:42 if true_word else false_word then you would only pass over a single "throw away" word. You wouldn't drill into it. 13:34:02 KipIngram: I don't much care for "prediction" - sounds like a compiler job; subclass "optimizing". 13:34:52 Not prediction - "predication". 13:35:19 ah, my bad 13:35:35 It's a term used for making instructions conditionally executable. So instead of jumping around a bunch of instructions that shouldn't execute you execute them anyway, they just don't "do" anything. 13:35:48 yes. 13:36:00 like qbranch "conditional" 13:36:57 they do not necessarily "execute" either 13:37:26 well, someone decides - somewhere. 13:37:29 Well, sorry about that phrasing. I guess all they really do is pass through the pipeline. 13:37:33 sure 13:37:40 I was sloppy in my nomenclature. 13:38:02 no, that's my point -- they _do not_ pass through the pipeline, if the issue or fetch stages throw it out already 13:38:19 I've been wasting hours rolling thru youtube, for all my fav Bob & Tom stuff ;-) 13:38:31 Ok, they get fetched. They go throught he part of the pipeline that gets fouled up by branches. 13:38:58 no 13:38:58 code, microcode; piplelines, caches - humbug 13:39:04 I'm picturing a "conditional execution stack" and the following operations: 13:39:14 the problem with badly predictable branches is the big bubbles 13:39:28 which you only get if the problem is _deep_ in the pipeline 13:39:33 1) pop data stack and push that value onto conditional stack 13:40:08 if the top of the conditional stack is false at the time that word occurs, then the conditional stack top just gets dup'd instead. 13:40:27 2) pop conditional stack (this always executes, no matter what). 13:40:35 sounds like doubling/tripling complications. 13:40:54 All other operations (all other primitives and all calls) "execute" only if the top of the conditional stack is true. 13:41:05 ugh 13:41:09 Do you mean by that that you pass through all code? 13:41:26 Why ugh? That's a pretty simple operational definition. 13:41:49 yeah, and it's fuck slow 13:42:10 I'd rather step right along thru my damned nybbles, bytes, etc 13:42:46 Well, how slow it is depends on how much code you have in the alternate sections. If they're short then I think the idea is that you more than make up for the overhead by having a really tight, predictable pipeline that never gets "branch blown". 13:43:01 damned nybbles and bytes? 6 and 66? :-) 13:43:46 Anyway, very preliminary thoughts. 13:44:01 it helps a lot more to have a good counted loop insn :-P 13:44:03 segher: yes, I need my 0666 ;-> 13:44:18 --- quit: proteusguy_ (Read error: 60 (Operation timed out)) 13:46:31 Ah, my day is being made - allow me to share: http://www.youtube.com/watch?v=f6VfWh1atU4&NR=1 13:46:36 * PoppaVic chortles 13:47:56 --- join: Snoopy_1711 (i=Snoopy_1@dslb-088-068-197-029.pools.arcor-ip.net) joined #forth 13:48:55 It's a Bob & Tom day here at home.. All the whackiness one can stand and not choke laughing ;-) 13:49:40 --- quit: Snoopy_1611 (Read error: 60 (Operation timed out)) 13:55:46 --- join: Snoopy_1611 (i=Snoopy_1@dslb-084-059-105-048.pools.arcor-ip.net) joined #forth 13:59:26 --- join: proteusguy_ (n=proteusg@zeppelin.proteus-tech.com) joined #forth 14:00:20 --- quit: GeDaMo ("Leaving.") 14:02:43 --- quit: crcx ("Page closed") 14:08:19 --- join: TR2N (i=email@89-180-135-135.net.novis.pt) joined #forth 14:12:19 --- quit: kar8nga (Remote closed the connection) 14:13:37 --- quit: Snoopy_1711 (Read error: 110 (Connection timed out)) 14:15:41 --- quit: schmx (Read error: 113 (No route to host)) 14:23:30 --- join: schme (n=marcus@c83-249-82-26.bredband.comhem.se) joined #forth 15:12:48 --- quit: PoppaVic (Client Quit) 15:13:08 --- join: PoppaVic (n=pops@adsl-99-150-133-146.dsl.sfldmi.sbcglobal.net) joined #forth 15:28:25 --- quit: Maki ("Leaving") 16:03:27 --- quit: segher (Read error: 60 (Operation timed out)) 17:03:05 --- quit: dinya (Read error: 104 (Connection reset by peer)) 17:36:14 --- join: segher (n=segher@84-105-60-153.cable.quicknet.nl) joined #forth 19:34:33 --- quit: Al2O3 (Connection timed out) 20:06:49 --- quit: gogonkt (Read error: 110 (Connection timed out)) 20:11:30 --- join: gogonkt (n=info@116.5.49.21) joined #forth 21:17:25 --- quit: PoppaVic (Client Quit) 21:27:45 --- join: PoppaVic (n=pops@adsl-99-150-133-146.dsl.sfldmi.sbcglobal.net) joined #forth 22:44:20 --- quit: PoppaVic (Client Quit) 22:52:54 --- part: TR2N left #forth 22:55:56 --- join: Al2O3 (n=Al2O3@c-75-70-11-191.hsd1.co.comcast.net) joined #forth 23:59:59 --- log: ended forth/09.12.03