00:00:00 --- log: started forth/02.05.27
00:25:44 --- quit: kc5tja ("[BX] Do you... BitchX?")
00:43:28 --- join: Serg_penguin (~snaga_NOI@nat-ch0.nat.comex.ru) joined #forth
00:45:05 --- quit: Serg_penguin (Client Quit)
01:46:35 --- quit: Soap` (Read error: 104 (Connection reset by peer))
02:56:03 --- join: goshawk` (goshawk@panix1.panix.com) joined #forth
03:04:40 --- join: mur (ammu@baana-62-165-189-112.phnet.fi) joined #forth
03:53:34 --- quit: onetom ("ircII EPIC4pre2.508 -- Accept no limitations")
04:39:29 --- nick: mur -> murBBL
05:29:37 --- nick: murBBL -> mur
05:41:10 --- quit: mur ("MURR! end of file reached. continuing filling logs some other time.")
06:11:52 --- join: Serg_penguin (~snaga_NOI@nat-ch0.nat.comex.ru) joined #forth
06:24:27 * Oscarian is away: sleep time
06:33:25 <Serg_penguin> :)
06:40:24 --- join: jim2 (~jim@12.240.101.86) joined #forth
06:42:12 <jim2> Hi. I'd like to find out where the log of the second conversation with Mr. Moore is... Thanks. Please msg me? tnx agn... bye :)
06:42:15 --- part: jim2 left #forth
07:11:23 --- quit: Serg_penguin ()
07:21:14 --- join: Serg_penguin (~snaga_NOI@nat-ch0.nat.comex.ru) joined #forth
07:23:04 --- quit: Serg_penguin (Client Quit)
07:29:37 --- join: sif (~siforth@ip68-9-58-81.ri.ri.cox.net) joined #forth
07:29:37 <sif> Type sif: <forth code here> (or /msg sif <forth code here> to play in private)
07:29:49 <rob_ert> sif: .( :))
07:29:51 <sif> rob_ert: :  Word not found: )
07:29:54 <rob_ert> Bah
07:30:09 <rob_ert> sif: : x ." Hug me!" ; x
07:30:10 <sif> rob_ert: Hug me!
07:32:00 --- join: mur (ammu@baana-62-165-189-112.phnet.fi) joined #forth
08:46:36 --- join: Herkamire (~jason@ip68-9-58-81.ri.ri.cox.net) joined #forth
08:46:43 --- quit: Herkamire (Client Quit)
08:46:59 --- join: Herkamire (~jason@ip68-9-58-81.ri.ri.cox.net) joined #forth
08:55:14 <Etaoin> hooray
08:56:51 <mur> whooo!
08:57:24 <Etaoin> a toybot!
09:00:28 --- quit: Etaoin ("raise KeyError, "I'll bet that's the last time you lock them in the car..."")
09:06:12 --- join: Etaoin (~david@ljk2-22.sat.net) joined #forth
09:11:46 <mur> ?'
09:17:02 <Etaoin> ?
09:26:30 <mur> a toybot?
09:27:06 <Etaoin> sif
09:27:26 <mur> ???
09:27:52 <Etaoin> sif: : saysomething ." something" ; saysomething
09:27:54 <sif> Etaoin: something
09:32:51 --- quit: Herkamire ("leaving")
10:22:10 <mur> oh fofrth bot
10:22:12 <mur> `?
10:30:54 <Etaoin> looks like it, doesn't it?
10:32:17 --- join: I440r (~mark4@1Cust115.tnt3.bloomington.in.da.uu.net) joined #forth
10:32:22 <rob_ert> Hey.
10:32:46 <I440r> hi
10:32:48 <I440r> has Oscarian said anything lately ? heh
10:32:50 <I440r> ive not talked to him yet :)
10:32:56 <I440r> btw ill be working on that dns stuff later today
10:33:24 <I440r> have you added your stuff to my sockets code ?
10:33:41 <I440r> the telnet should be in telnet.f i think
10:33:59 <I440r> because http/ftp/irc etc all basically use telnet connections
10:34:23 <rob_ert> He's sleeping now :)
10:34:32 <I440r> heh i guess:)
10:34:32 --- nick: mur -> murFOod
10:34:34 <rob_ert> Aussies are always on the wrong side of the world:/
10:34:42 <I440r> heh
10:34:49 <I440r> no they are just upside down
10:34:54 <rob_ert> :)
10:35:49 <I440r> ive given isforth the ability to create headerless words
10:36:15 <I440r> also added forget and empty.
10:36:23 <I440r> empty forgets all user defined words
10:39:23 <I440r> brb - making coffee
10:57:22 --- nick: murFOod -> mur
10:57:26 --- join: kc5tja (~kc5tja@stampede.org) joined #forth
10:59:21 <I440r> hi kc5
10:59:28 <kc5tja> re
10:59:38 <I440r> i updated my 8051 disassembler to disassemble bit addresses and direct addresses symbolically
10:59:50 <I440r> its on the isforth.clss web page
10:59:56 <kc5tja> Cool
10:59:57 <I440r> thers a link to the source file that is
11:00:13 <I440r> im going to add the ability to create user defined symbols too
11:01:10 * kc5tja is daydreaming of building my own home computer again, entirely from TTL discrete logic.
11:01:18 <kc5tja> (including the CPU and video display controllers)
11:03:32 <I440r> lol
11:03:35 <I440r> your sic:)
11:03:38 <I440r> you need help :)
11:03:46 <kc5tja> No, it's a hobby.
11:03:46 <I440r> i cant do hardware...
11:03:49 <I440r> ya
11:03:50 <I440r> i know
11:03:56 <I440r> me and hardware -->
11:03:58 <kc5tja> I've built CPUs in the past before.
11:04:06 <kc5tja> I just never built a whole, complete system before.
11:04:10 <I440r> oh look whats... GADZZZTS.... oopts
11:04:21 <kc5tja> Video hardware can be incredibly easy to implement though.
11:04:39 <I440r> ive got code to do vidio output with an 8051
11:04:49 <I440r> piddly little 4k 12 mhz thingie
11:05:04 <kc5tja> You mean the 8051 generates the video waveform directly?
11:05:13 <kc5tja> For NTSC or VGA?
11:05:17 <I440r> not sure
11:05:23 <I440r> its in the book upstairs
11:05:25 <I440r> prolly ntsc
11:05:35 <kc5tja> I really, really, really doubt the 8051 is doing it on its own.
11:05:51 <I440r> dunno - it might be a circuit with an 8051 controlling it...
11:05:57 <kc5tja> I'm willing to bet it has an external video chip it connects to.
11:05:59 <I440r> i never realy looked at it
11:06:57 <kc5tja> I don't know what Intel was thinking when they came up with the whole 8048-series of microcontrollers (of which 8051 is a member of); I mean, a 12 clock execute cycle is wholly unjustified.
11:07:04 --- join: Herkamire (~jason@ip68-9-58-81.ri.ri.cox.net) joined #forth
11:07:06 <kc5tja> Especially for a Harvard architecture chip! :(
11:07:18 <kc5tja> re Herkamire
11:07:24 <Herkamire> hey :)
11:07:38 <I440r> hi herk
11:07:56 <I440r> kc5 i LOVE the 8051
11:08:02 <kc5tja> I abhore it.
11:08:11 <I440r> :)
11:08:31 <kc5tja> It's got an instruction set that is even worse than the original 8008.
11:08:38 <I440r> bullshit!
11:08:43 <I440r> its instruction set is perfect
11:08:46 <kc5tja> HAHAHAH!
11:08:50 <I440r> :)
11:09:05 <I440r> and you dont NEED to stick with harvard if you dont want to
11:09:06 <kc5tja> I programmed the 8008, 8080, and the 8048/8051, and I can tell you the 8008 is vastly superior in performance and flexibility to the 8051.
11:09:12 <Herkamire> I like the instruction set on the X18...
11:09:27 <kc5tja> I440r: *I* don't select Harvard or not; Intel does.
11:09:31 <I440r> my 8051 forth (unfinished) requires the select lines for external data and external code be or'd
11:09:39 <I440r> kct nope YOU do
11:09:54 <I440r> wire or (or gate) the select lines for external memory
11:09:56 <kc5tja> That's a new feature then; Intel's original 8051 design was strictly Harvard.
11:10:03 <I440r> kct NOO
11:10:19 <kc5tja> Whatever
11:10:30 <I440r> there are TWO select lines coming out of the 8051. tie them together and put them onto the select line of your external ram
11:10:41 <I440r> code fetches and data fetches BOTH go to the same memory
11:10:53 <I440r> my 8051 forth relies on this :)
11:11:03 <I440r> but i never finished that forth
11:11:13 <kc5tja> For your generation of 8051, maybe that's true.
11:11:26 <kc5tja> That wasn't the case for the 8051 I was working on, since it could access data RAM and code space concurrently.
11:12:10 <kc5tja> At any rate, it still doesn't excuse Intel for choosing the most god-awful instruction set in the known universe, and it definately doesn't excuse them for using a 12 clock bus cycle!
11:12:15 <I440r> kct no it couldnt. not unless it could execute two instructions at the same time. one fetching from data the other fetching from code,....
11:12:31 <I440r> both of which would require a seperate DPTR
11:12:41 <I440r> unless the code fetch was an instruction fetch @pc
11:13:01 <kc5tja> That's what it was.  That is the only way the 8051 can keep up with its competitors.
11:13:02 <I440r> or a movc a,@a+pc
11:13:25 <kc5tja> An instruction can be executing while it's fetching another instruction, you know...
11:13:35 <I440r> and thers still nothing to stop you wire oring the select lines heh
11:13:41 <I440r> of corse
11:13:59 <kc5tja> Like I said, maybe on your specific 8051, but not on the unit I worked on.
11:14:09 <I440r> but you cannot fetch from external data memory and external code memory at the same time
11:14:21 <kc5tja> I440r: Then it's not a Harvard architecture.
11:14:24 <I440r> becayse port 2 is used to select the upper 8 bits of the address
11:14:35 <kc5tja> The one I was using was.
11:14:51 <I440r> get me the part number on that
11:15:05 <I440r> and i bet the datasheet will tell you you can wire or the select lines
11:15:07 --- quit: jim (Remote closed the connection)
11:15:09 <kc5tja> I can't now -- I'm not working at that company anymore.
11:15:20 <I440r> you cannot read from two different external memory locations
11:15:33 <I440r> you can execute internal rom while fetching external data
11:15:50 <I440r> harvard architecture is just seperation of code and data.
11:15:52 <kc5tja> I440r: IF THE CPU HAS HARVARD ARCHITECTURE, IT CAN FETCH AN INSTRUCTION AT THE SAME TIME AS EXECUTING ANOTHER; This means that it *CAN* fetch from two different spaces concurrently.
11:15:56 <I440r> internally that is enforced
11:15:59 <kc5tja> Why must I repeat myself six hundred times?
11:16:14 <I440r> kc5 dood that has nothing to do with harvard architecture
11:16:19 <I440r> err
11:16:28 <kc5tja> I440r: That is the very definition of a Harvard architecture!!!!!
11:17:03 <kc5tja> http://www3.sk.sympatico.ca/jbayko/cpuAppendC.html
11:17:07 <I440r> harvard architecure is simply a seperation of code space from data space. internally the chip has two memory areas, each with its own select line
11:17:13 <I440r> you also have external address space
11:17:19 <I440r> for BOTH code and data
11:17:27 <I440r> which means you have to have two external select lines
11:17:31 <I440r> which you can TIED
11:17:33 <I440r> tie
11:17:41 * kc5tja sighs
11:17:41 <I440r> you cannot however tie internal code and data
11:17:51 <I440r> if you are executing INTERNAL code space
11:18:00 <I440r> you can fetch external code or data at the same time
11:18:27 <I440r> you CANNOT ever on any 8051 read from two different areas of external memory
11:18:31 <I440r> its flat impossible
11:18:43 <I440r> because port 2 is used for the upper 8 bits of the address
11:18:50 <I440r> you cant have more than one value on p2
11:19:34 <I440r> well - you SORT of can... but not at the same time :)
11:19:50 <kc5tja> I440r: In case you haven't noticed, I gave up on this conversation a while ago.
11:20:03 <kc5tja> I am not repeating myself again.
11:20:04 <I440r> thers a latch - and some instructions use the latch to select the addres... others put the address directly onto P2 for the duration of the instruction
11:20:17 <I440r> well you were wrong dood - i know the 8051 inside out
11:21:01 <kc5tja> Yeah, apparently I was wrong.
11:21:07 <kc5tja> And so was the company I worked for.
11:21:13 <kc5tja> And the datasheets for it.
11:21:19 <kc5tja> But, that's a non-issue.
11:21:27 <kc5tja> Whatever.  I'm through with arguing.
11:21:37 <I440r> kc5 dont get bitchy with me dood. your memory of how you did what ever it was you did is FAULTY
11:21:54 <I440r> you CANNOT address two different memory address spaces externally at the same time
11:22:07 <I440r> not unless you dedicated port B to select addresses
11:22:22 <I440r> erm p1 even
11:22:32 * Etaoin waits for insults to start flying
11:22:34 <I440r> and i dont even know if that is pissible
11:22:47 <I440r> possible
11:23:48 <I440r> anyway time to code
11:23:51 --- quit: I440r ("Reality Strikes Again")
11:24:07 --- join: kors (~kors@nbc-2.ukr.net) joined #forth
11:24:51 <kors> evening.
11:25:22 <Etaoin> depending on where you are...
11:25:53 <kors> Etaoin: A, really, sorry :). Good <whatever you have> :).
11:26:55 <kc5tja> Etaoin: There will not be any insults from me.  They're not needed.
11:29:33 <Etaoin> kc5tja: eh. it seemed like the next step.
11:29:45 <Etaoin> kors: same to you
11:30:27 <kors> Etaoin: :)
11:30:30 --- nick: mur -> mur_away
11:32:21 * kors wonders if anyone was able to launch Moore's color forth under vmware or (even better) bochs? Or, at list, could point me to ANY FM on it...
11:32:36 <kors> s/list/least/
11:34:40 <Etaoin> I thought about trying that, but didn't
11:37:33 <kors> Etaoin: Seems like it works directly with AGP board. And I wouldn't mess with sources to the new dialect of language I barely remember. Was just very interested what's Chuck up to. Had a very positive experience very long ago with forth on 8-bit machine.
11:38:33 <kors> ok, another one - anyone seen an implementation of CF for standard VGA, or even better *nix console?
11:52:36 <kc5tja> kors: Nope.
11:52:46 <kc5tja> But it should be relatively easy to implement, I'd imagine.
11:52:57 <kc5tja> I believe CrowKiller is working on something of that nature.
11:53:29 <kors> kc5tja: Ok, thanks.
12:03:18 <rob_ert> Kors i taket!
12:06:42 <kors> rob_ert: ???
12:06:59 <rob_ert> Hehe
12:07:06 <rob_ert> That's a swedish expression.
12:07:25 <rob_ert> Used to express one being suprised.
12:07:48 <kors> rob_ert: :) Wow :). My nick is of other origin (russian :).
12:08:19 <rob_ert> It means "Cross in the roof".
12:08:28 <rob_ert> I see. Privet.
12:08:30 <rob_ert> ;)
12:08:45 <kors> rob_ert: Privet :)
12:09:19 * rob_ert thinks it's good with ukranian IRC friends.
12:09:56 <kors> :)
12:21:01 --- join: Stepan (~stepan@p50847D99.dip.t-dialin.net) joined #forth
12:21:24 <Stepan> ha.. this got pretty full since i was here last time
12:21:53 <kors> :(. The sources for CF seem to be using masm syntax. And I haven't seen it for ages.
12:21:55 <rob_ert> Hi :)
12:22:12 <rob_ert> Hehe... I prefer NASM's syntax.
12:22:13 <Stepan> Hey rob_ert :)
12:22:33 * kors desperately rummages through the backups in search of the assembler from unnamed redmond horror :)
12:22:42 <kc5tja> The original ColorForth was written using MASM, yes.
12:22:44 <kors> Stepan: hi.
12:22:49 <kors> rob_ert: me too :)
12:23:14 <rob_ert> Portable, easier to read, more logical...
12:23:44 --- nick: mur_away -> mur
12:24:06 <kors> rob_ert: feels so strange - I'll have to use the DOS emulation to make a bootable floppy for standalone forth...
12:24:59 <Stepan> I finally started writing some basic IEEE 1275 forth words today... http://openbios.ph-freiburg.de/cgi-dom/viewcvs.cgi/fcode/base/
12:25:15 <rob_ert> kors: Haha, yeah :)
12:25:31 <rob_ert> And may I ask what IEEE 1275 is?
12:27:48 <kors> rob_ert: afair, It has something to do with Open Firmware (boot loader used on Sparc worstations)
12:29:00 <rob_ert> I see...
12:29:06 <rob_ert> Never used one of those.
12:35:45 <Stepan> Its really nice - far more than a boot loader. Its a firmware implementation that abstracts device drivers (ie on expansion cards like gfx adapters) from the used instruction set (ie CPU)
12:36:25 <Stepan> by using forth. This makes Open Firmware compliant hardware work on any open firmware system, nut like broken x86 compliant hardware with intXX bios functions etc
12:36:40 <rob_ert> Sounds very nice. I'd really like to get a compiler with something cleaner than a BIOS that's backwards compatible to the 70s.
12:36:44 <Stepan> it comes with a nice abstraction layer and drivers/packages
12:36:48 <rob_ert> s/compiler/computer/
12:37:07 <Stepan> rob_ert: exactly :)
12:37:25 <Stepan> openbios is a free open firmware implementation (well, going to be)
12:37:48 <Stepan> thus making it easy for everybody to enhance it or port it to a new platform
12:38:08 <rob_ert> Sounds nice. But I think the whole PC arch. should be replaced :)
12:38:30 <rob_ert> Why not get 2 computers if you want to run PONG from 1978?
12:38:37 * kors has a forth bootloader too. Though on i386, level 2, and serving different task. :). Open Firmware is really cool.
12:42:48 <kors> I mean the one used to load the freebsd kernel and modules.
12:43:04 <kc5tja> kors: That's FICL.
12:43:17 <kc5tja> http://ficl.sourceforge.net
12:45:28 <kors> kc5tja: Indeed.
12:47:07 <Stepan> rob_ert: the hammer architecture is really nice thoughj
12:47:25 <rob_ert> What is that?
12:48:10 <Stepan> x86-64 ... the upcoming amd 64bit arch
12:48:33 <Stepan> it's based on x86/ia32 but it enhanced it in a really nice way... more registers etc
12:48:51 <kors> Stepan: the one from AMD?
12:48:53 <Stepan> its so much nicer than this itanium crap
12:48:57 <kc5tja> 64-bit mode is incompatible with 32-bit modes, though.
12:48:58 <Stepan> kors: jupp.
12:49:17 <Stepan> kc5tja: you can run all 32bit apps transparently... shouldnt be any problem
12:49:19 <rob_ert> Does it support older modes?
12:49:24 * kors really likes AMD stones.
12:49:25 <kc5tja> Stepan: Nope.
12:49:53 <kc5tja> Stepan: The 64-bit instruction prefixes come from the single-byte INC instruction forms; hence, the single-byte INC EAX (or other register) will *break* when running in 64-bit mode.
12:50:39 <kc5tja> The ability to run 32-bit and 16-bit software comes from the operating system that's running on the processor.  But it definately isn't transparent.
12:51:08 <kc5tja> And switching from 64-bit to IA-32 mode will be relatively expensive (but with today's clock speeds, few will notice).
12:51:09 <Stepan> kc5tja: sure,.. the OS has to support it. but thats not a problem, is it?
12:51:24 <Stepan> 32bit apps will die out anyways
12:51:28 <kc5tja> Stepan: I'm just saying it's not "transparent."
12:51:39 <Stepan> it's transparent in "user space
12:51:40 <Stepan> "
12:51:49 <kc5tja> Stepan: I disagree.  99.99% of the people that use computers have no need for 64-bit software at all.
12:51:52 <Stepan> thats what i meant... on os level you have to do lots more stuff
12:52:10 <Stepan> like the syscall interface is completely different in 32bit and 64bit as well
12:52:15 <kc5tja> Our CPUs still have 8086 support, after all.  :)
12:52:20 <rob_ert> Hmm...
12:52:50 <rob_ert> I hope to continue like this, "old" computer being sold for for almost nothing.
12:52:58 <Stepan> kc5tja: agreed. but this is why i prefer hammer a lot over itanium
12:53:16 <Stepan> itanium is a nice philosophical implementation, but no real world use
12:53:20 <kc5tja> I like them both, frankly.  Each is nice in their own ways.
12:53:31 <kc5tja> Well, we'll see when McKinnley hits the market.
12:53:34 <kc5tja> If it ever does.
12:55:59 <Stepan> it will. i heard some birds whistling that they are delivered to some companies... not that i would be allowed to tell you we have one, if we did ;)
13:09:12 --- join: XeF4 (~XeF4@dsl-XIV-08.kotikaista.weppi.fi) joined #forth
13:09:26 <rob_ert> Hello.
13:09:48 <XeF4> olleh
13:11:29 <kors> XeF4: Hi.
13:23:31 --- join: TheBlueWizard (TheBlueWiz@ip-216-25-205-204.vienna.va.fcc.net) joined #forth
13:23:31 --- mode: ChanServ set +o TheBlueWizard
13:23:34 <TheBlueWizard> hiya all
13:23:51 <rob_ert> Hi :)
13:23:56 <TheBlueWizard> hiya rob_ert
13:25:32 <kors> TheBlueWizard: Hi,
13:26:55 <TheBlueWizard> hiya kors
13:41:11 --- quit: XeF4 ("pois")
13:53:19 --- join: Soap` (~flop@202-0-42-22.cable.paradise.net.nz) joined #forth
13:53:47 <TheBlueWizard> hiya Soap`
13:54:22 --- join: rob__ert (~robert@h237n2fls31o965.telia.com) joined #forth
13:55:24 <Soap`> 'ello
13:56:31 --- quit: rob_ert (Killed (NickServ (Ghost: rob__ert!~robert@h237n2fls31o965.telia.com)))
13:57:01 --- nick: rob__ert -> rob_ert
14:16:48 --- join: Fare (fare@samaris.tunes.org) joined #forth
14:17:16 <rob_ert> Hey Fare.
14:17:21 <Fare> gakuk
14:17:26 * Fare is sad
14:20:16 <rob_ert> Why? :(
14:21:02 <rob_ert> Fare?
14:21:50 <TheBlueWizard> hiya Fare
14:24:38 * Fare needs sentimental support
14:24:44 <rob_ert> Huh?
14:24:49 <rob_ert> What's up?
14:25:12 <TheBlueWizard> what's wrong?
14:25:22 <Fare> I'd rather discuss it on #blues
14:25:50 <rob_ert> I have to go to bed :(
14:25:56 <rob_ert> No support for you today :(
14:26:00 <rob_ert> Night all.
14:26:16 --- quit: rob_ert ("Nothing is real. I think.")
14:32:07 --- join: davidw (~davidw@adsl-ull-217-108.42-151.net24.it) joined #forth
14:49:20 --- nick: kc5tja -> kc-food
15:10:21 --- quit: Herkamire (Ping timeout: 14400 seconds)
15:11:49 --- nick: kc-food -> kc5tja
15:44:33 --- quit: Fare ("3053")
15:52:12 --- nick: air|cabin -> air
16:02:36 --- quit: davidw (Read error: 113 (No route to host))
16:12:13 --- quit: kors (Read error: 110 (Connection timed out))
16:16:07 <TheBlueWizard> gotta go...gonna work on my lil script :)
16:16:10 <TheBlueWizard> bye all
16:16:16 --- part: TheBlueWizard left #forth
16:20:01 --- join: Herkamire (~jason@ip68-9-58-81.ri.ri.cox.net) joined #forth
16:21:33 * Oscarian is back (gone 09:57:08)
16:22:15 * mur never left
16:22:34 <mur> blah thebluewizard was here 3 hours and i missed him wih 3 minutes
16:22:36 <mur> baah
16:22:45 <mur> by 3 min
16:53:00 --- quit: mur ("MURR! end of file reached. continuing filling logs some other time.")
17:26:37 --- quit: Fracta| (Read error: 110 (Connection timed out))
18:27:10 --- join: I440r (~mark4@1Cust219.tnt3.bloomington.in.da.uu.net) joined #forth
18:43:48 --- join: CrowKiller (Vapo_Rulez@cnq5-233.cablevision.qc.ca) joined #forth
18:44:01 <CrowKiller> hi
18:44:36 <kc5tja> re
18:44:52 * kc5tja got AHEAD, IF, THEN, and ELSE working in my target compiler.
18:53:23 --- join: Shrdlu (~david@ljk2-22.sat.net) joined #forth
18:53:28 <I440r> hi Shrdlu
18:53:36 <I440r> davidw ?
18:53:49 <Shrdlu> hello
18:53:59 <Shrdlu> oops.
18:54:02 <Shrdlu> I'm already on
18:54:05 --- quit: Shrdlu (Client Quit)
18:54:06 <I440r> lol
18:54:09 <kc5tja> re
18:54:20 <kc5tja> Got the target compiler to properly handle IF THEN and ELSE.
18:54:27 <I440r> cool
18:54:29 <kc5tja> Working on BEGIN AGAIN UNTIL WHILE and REPEAT now.
18:54:37 <I440r> i never used forward myself...
18:55:02 <I440r> for nxt and do loops too :)
18:55:18 <I440r> i called 'next' nxt because i think the word NEXT has special meaning in forth
18:55:37 <I440r> i personally dont seperate out "system programming" from "application programming"
18:55:40 <I440r> they are one and the same
18:59:55 * kc5tja uses AHEAD, not FORWARD...I'm not sure what you're getting at.
19:00:16 <kc5tja> Also, since my Forth is now subroutine threaded with inlined primitives, I no longer even have a need for NEXT either.
19:03:52 <I440r> i meant ahead
19:03:53 <I440r> lol
19:03:56 <I440r> never used it
19:04:18 <kc5tja> Well, there's no law that says you have to.
19:04:25 <kc5tja> But for me, it's just the tool I need.
19:05:17 <I440r> heh its just not something i ever used, im not even sure how to implement it :)
19:05:33 <kc5tja> It's just a forward branch...  :)
19:05:41 <kc5tja> It's like IF, except it's unconditional.
19:07:59 <I440r> of
19:08:01 <I440r> oh
19:08:31 <I440r> compile branch >mark
19:08:32 <CrowKiller> kc5tja: did you watched all the aha.rm file?
19:09:13 <CrowKiller> or anybody on this chan, if you listened to that it would be great to have you thoughts on this
19:12:50 <kc5tja> CrowKiller: No, I couldn't.
19:12:56 <kc5tja> It crashed halfway through the file, repeatably.
19:13:09 <kc5tja> The file is corrupted.
19:13:16 <CrowKiller> it was doing that when the file was incomplete
19:13:17 <kc5tja> I440r: Basically, yes.
19:13:33 <kc5tja> CrowKiller: Like I already said, I have the whole file -- all 140MB of it.
19:13:37 <kc5tja> The file is not incomplete.
19:13:55 <CrowKiller> i know i can read you know ;p
19:13:58 <kc5tja> And repositioning the progress bar, or using the fast forward/rewind buttons caused it to hang hard.
19:14:58 <CrowKiller> I was experiencing the very same problems but in my case it was because the file missed a few dozen megabytes ;p
19:15:34 <CrowKiller> its frustrating that only a small "notch" in the stream can cause these hangups
19:16:49 <kc5tja> What I don't like is how just barely tapping the forward button will cause the program to rebuffer it (or so it says), even at the very beginning of the file.  I just click, as fast as I can, the button while it's at the very beginning fo the file, and it hangs.
19:17:02 <kc5tja> I literally let it sit there for an hour, and it still didn't rebuffer.
19:17:32 <CrowKiller> i can only agrre with that it did the same thing to me ;p
19:17:56 <Soap`> I think the lesson here is that realmedia is the spawn of the devil.
19:18:42 <kc5tja> Soap`: No.  Devilspawn at least is interactive while it ruins your life.  Real doesn't even do that.  It's too fuckin' stupid to be devilspawn.
19:19:40 <CrowKiller> i saw a compression scheme that was a lot better than that, its at www.kttech.com IIRC
19:20:37 --- join: CrowKilr (Vapo_Rulez@cnq5-233.cablevision.qc.ca) joined #forth
19:20:37 --- quit: CrowKiller (Read error: 104 (Connection reset by peer))
19:20:45 --- nick: CrowKilr -> CrowKiller
19:20:51 <CrowKiller> www.kttech.com
19:21:19 <CrowKiller> maube with some reverse engineering, we could get a forth version of this algorythm
19:25:11 <CrowKiller> these algorythms are the most efficient and well done I know
19:28:13 <I440r> the ceo worked for nasa - she probably knows something of forth
19:30:34 <kc5tja> And now, BEGIN AGAIN WHILE REPEAT and UNTIL are done.
19:31:06 <kc5tja> CrowKiller: Reverse engineering is illegal in the USA now (DMCA); you'll have to do it over in Europe somewhere.
19:32:52 <I440r> what is dmca ?
19:32:55 <CrowKiller> dmca... what a shame
19:33:02 <CrowKiller> digital millenium copyright act
19:33:22 <CrowKiller> the stupidiest law ever in the information technology era
19:33:34 <CrowKiller> never heard of that?
19:33:49 <CrowKiller> there so much cases in court about the DMCA
19:34:27 <CrowKiller> any person can send a "DMCA" letter to your ISP forcing him to close your website or to face legal charges
19:35:13 <I440r> lol
19:35:16 <CrowKiller> Europe is preparing an even tighter incarnation of the DMCA IIRC
19:35:28 <kc5tja> A man from Russia had posted some information about how to decrypt Adobe Acrobat files, which is just ROT13 (basically).
19:35:32 <CrowKiller> the scientology mivement used such letters to close sites that were agianst them
19:36:07 <kc5tja> That man gave a conference in the USA about something (totally unrelated), and was arrested right there, on the spot, as soon as he got off the plane.  The charge?  Violating DMCA.
19:36:17 <CrowKiller> this particular case is the most famous, he was incarcerated lol
19:36:18 <kc5tja> He was in prison for almost two years.
19:37:04 <I440r> fucking democrats
19:37:06 <I440r> fucking up our country
19:37:37 <kc5tja> Democrats?  No....democrats fought violently against it.  Republicans, the great lovers of monopolies and big industry, are the ones who pushed it through.
19:37:57 <kc5tja> But since Microsoft paid everyone off in Congress, it got passed.
19:38:58 <CrowKiller> I live in canada, so maybe the DMCA isnt applicable here
19:39:18 <kc5tja> Don't know.
19:39:52 <I440r> hey can i get in trouble for reverse engineering libc??? lol
19:40:02 <I440r> its open source :)
19:40:46 <kc5tja> No; it's open source.
19:40:59 <kc5tja> But you can get in trouble for, say, probing Windows NT to see if a security hole might be present.
19:41:21 <kc5tja> That's half the reason Microsoft pushed the law through.  Notice how, after DMCA was passed, the security violation alerts for Windows NT all but disappeared?
19:41:37 <I440r> lol
19:41:54 <I440r> and samba ?
19:41:54 <I440r> they trying to kill taht too
19:42:38 <Oscarian> howdy I440r...i missed you yesterday to return your greeting :)
19:42:49 <I440r> hi Oscarian heh
19:42:57 <I440r> thats ok - irc is like that :)
19:42:59 <Oscarian> i was off pretending to be doing something productive...a bit like i am now
19:43:00 <CrowKiller> now everybody got their perosnnal exploits to toy with without being detected ;p
19:43:01 <deltab> glibc is open source; other libcs are usually proprietary
19:43:10 <I440r> plus im told your on the wrong side of the world ;)
19:43:26 <Oscarian> well, "other" side yes ;)
19:43:40 <Oscarian> i'm a dinky di aussie ;)
19:43:40 <I440r> the UNDER side :)
19:44:00 <I440r> well your in #forth so you cant be ALL bad :)
19:44:01 <I440r> lol
19:44:07 <Oscarian> (that means ridgey didge, if you didn't know that) :P
19:44:11 <I440r> so you code forth ?
19:44:12 * kc5tja wants to visit Australia some day
19:44:32 <Oscarian> no, i'm here lurking after rob_ert recommended forth to me as a language to use
19:45:07 <I440r> he recommended good
19:45:51 <I440r> its a little bit difficult to get your head arround at first but once you get past the basics is a piece of cake
19:46:01 <I440r> hell... if i do it it has to be :)
19:47:02 <CrowKiller> Oscarian, the best place to learn about forth is htpp://www.ultratechnology.com
19:47:24 <CrowKiller> all you need to know on Forth is there, the rest is up to you ;p
19:47:28 <Oscarian> the first thing he said was that forth is all about stacks...
19:47:38 <Oscarian> which helped me understand stuff already
19:47:54 <I440r> it is
19:48:00 <I440r> there are 2 stacks
19:48:13 <Oscarian> 2 stacks...?
19:48:14 <I440r> to begin with you need only bother yourself with the parameter stack
19:48:20 <Oscarian> ah, okay
19:48:24 <I440r> to pass parametrers you put them on the parameter stack
19:48:25 <kc5tja> Actually, http://www.taygeta.com is somewhat better I think; ultratechnology.com assumes some prior knowledge of Forth.  taygeta.com (or .org or whatever it happens to be this week) has tutorials and introductions on it.
19:48:26 <Oscarian> what's the second stack?
19:48:30 <I440r> 2 or more. but 2 usually
19:48:41 <Oscarian> thanks kc5tja
19:48:45 <I440r> the second stack is called the return stack
19:48:48 <kc5tja> The second stack is the return stack.
19:49:03 <I440r> do you by any chance code c ?
19:49:27 <Oscarian> i know a little C/C++
19:49:33 <I440r> c only used one stack so the return address for a function call is in the way. forth doesnt have this problem
19:49:41 <I440r> 1 2 [enter]
19:49:47 <I440r> you now have 1 and 2 on the stack
19:49:54 <I440r> do-something
19:50:03 <I440r> ill explain do-something
19:50:11 <I440r> in forth everything you type is one of 3 things
19:50:17 <I440r> 1. its a word in the dictionary
19:50:20 <I440r> 2. its a number
19:50:24 <Oscarian> so, each word is a list of things to manipulate (in a group) on the stack?
19:50:30 <I440r> 3. its a mistake
19:50:58 <I440r> no. but words are sometimes lists ill get to taht :)
19:51:02 <I440r> but they arent on the stack
19:51:07 <Oscarian> oh...
19:51:09 <I440r> when you type something on the keyboard
19:51:22 <I440r> forth first searches its entire dictionary to see if there is a word with that name
19:51:28 <I440r> if there is it executes that word
19:51:49 <I440r> if there isnt it passes the string thats not a word to the word number
19:52:17 <I440r> number tryies to interpret the string as a number. if it works it leaves that nubmers value on.....
19:52:20 <I440r> the parameter stack
19:52:29 <I440r> 1 2 3 foo 4 5 6
19:52:32 <I440r> sort of thing
19:52:46 <I440r> words in forth can use ANY characters in their names
19:52:47 <Oscarian> hmm, okay...
19:52:48 <I440r> +
19:52:50 <I440r> is a word
19:53:00 <I440r> you pronounce the name as "plus"
19:53:01 <Oscarian> oh right
19:53:06 <I440r> plus takes 2 parameter
19:53:16 <I440r> it removes both parameters from the stack
19:53:23 <I440r> adds them and pushes the result
19:53:23 <Oscarian> so if you type "1 2 +" you've got two numbers and a word?
19:53:33 <I440r> .  (pronounced DOT is another word
19:53:47 <I440r> yes heres what forth does whenyou type that
19:53:56 <kc5tja> Quite often, you'll find that certain constants like 0 1 2 -1 and -2 are words too, but that optimization is falling by the wayside as more advanced native-code compilers are coming around.
19:53:56 <I440r> first. forth accepts your line of input
19:54:28 <I440r> all input in forth is SPACE delimited
19:54:32 <kc5tja> Oscarian: No, after 1 2 +, you have only one number -- the sum of 1 and 2.
19:54:36 <I440r> you cannot have a space in a word name
19:55:04 <I440r> forth first parses your input for the first space delimted string. the "1"
19:55:05 * Oscarian takes all this in and tries to stop it from escaping his brain
19:55:12 <I440r> this is a number so the number 1 gets put on the stack
19:55:25 <I440r> then forht parses the 2 which also goes on the stack
19:55:26 <kc5tja> Follow I440r -- he's explaining the process in detail.
19:55:44 <I440r> when forth parses the + it finds it in the dictionary and EXECUTES that word
19:55:59 <I440r> so when plus executes ALL of its parameters are ready for it on the stack
19:56:17 <I440r> it removes its parameters, adds them and pushes the result onto the stack
19:56:30 <Oscarian> hmm, sounds simple...and nice
19:56:35 <I440r> it is simple
19:56:50 <I440r> now.
19:57:00 <I440r> there are quite a few words in the dictionary
19:57:04 <Oscarian> rob_ert pointed me to an IDE controller written in forth on the 'web...it was suprisingly short
19:57:04 <I440r> + - * /
19:57:14 <I440r> those add, subtract multiply and divide
19:57:28 <CrowKiller> forth is all able to being the nicest (to your brain and to the computer) and the simplest you can be
19:57:31 <I440r> their names are usually pronounced... plus, minus, star, slash
19:58:12 <I440r> if you have an item on the stack that you need to multiply by 3 and add 5 to...
19:58:15 <I440r> 3 * 5 +
19:58:30 <Oscarian> do forth compilers also store "plus", etc, as a word?
19:58:41 <I440r> well
19:58:58 <I440r> the forth has a whole slew of pre-existing words in its dictionary
19:59:04 <I440r> PLUS is always one of them heh
19:59:11 <Oscarian> ah, okies
19:59:18 * Oscarian wonders off for a bit...just getting a drink
19:59:25 <I440r> k
20:00:00 <Etaoin> if you're gone, we won't know it unless you tell us L)
20:00:08 <Etaoin> s/L/:/
20:00:36 <I440r> he did tell us heh
20:01:12 <Etaoin> I know
20:01:22 <I440r> :)
20:01:56 <Etaoin> my point is he didn't have to
20:02:14 <I440r> oh heheh
20:02:14 <I440r> lol
20:02:23 <I440r> he could sneek out and we wouldnt know :)
20:02:30 <Etaoin> exactly
20:03:49 <I440r> mental note: Etaoin is sneaky
20:04:53 <CrowKiller> i do it all the time, it explains my latency when IRCing ;p
20:05:27 <I440r> heh
20:06:37 <Etaoin> heck, it's no different from leaving your irc client connecting while intently hacking on a new piece of code, ignoring anything anyone says
20:06:53 <kc5tja> Like me... :)
20:07:30 <I440r> lol i cant do thgat
20:07:33 <I440r> that
20:07:40 <I440r> if im online i gotta be chattin or whatever
20:07:44 <I440r> i can never code
20:08:55 <Oscarian> haha, i'm back!
20:09:04 <Oscarian> i was away for longer than i thought
20:09:05 <I440r> ok
20:09:12 <Oscarian> i got sidetracked by the couch
20:09:32 <I440r> you also need to learn some basic stack manipulation
20:09:32 <I440r> the words that do those are
20:09:43 <I440r> dup drop nip rot
20:09:45 <Etaoin> it's easy to ignore irc if no one is saying anything interesting
20:09:59 <I440r> there are others but those are the basic ones
20:10:09 <Oscarian> duplicate, drop...what are the other two?
20:10:57 <Etaoin> understand stack notation?
20:11:05 <Oscarian> not really...
20:11:16 <CrowKiller> nip removes the TOS (Top of stack) and place the second item of the stack into it
20:11:24 <CrowKiller> or
20:11:25 <kc5tja> Bzzz...
20:11:26 <I440r> nip is a drop of the second stack item
20:11:29 <kc5tja> Just the opposite.
20:11:30 <I440r> without dropping the top one
20:11:34 <kc5tja> NIP ( n1 n2 -- n2 )
20:11:36 <CrowKiller> it only remove the seond item on stacl
20:11:36 <I440r> rot rotates the third item out to the too
20:11:38 <I440r> top
20:11:44 <CrowKiller> replacing it by the third
20:11:53 <Oscarian> ah okay, so nip is like push
20:12:03 <Oscarian> or is that pop
20:12:06 <I440r> CrowKiller: nip doesnt delete the top item. it deletes the second item
20:12:12 <Oscarian> oh
20:12:16 <kc5tja> No, NIP is like DROP, except that it drops the second top of stack, not the topmost.
20:12:24 <Oscarian> ah
20:13:10 <Oscarian> so drop removes the top item, nip removes the second item, dup would create another item on the stack the same as what is currently there?
20:13:19 <kc5tja> Yes
20:13:21 <kc5tja> DUP ( n1 -- n1 n1 )
20:13:24 <I440r> forth has a special sort of comment that tells you what a word does with the stack
20:13:28 <Oscarian> what does run do?
20:13:30 <Oscarian> rot
20:13:39 <I440r> ( n1 n2 --- ..... )
20:13:44 <Oscarian> (where did 'run' come from =-/ )
20:13:45 <I440r> the .... part im leaving blank for now
20:13:47 <kc5tja> ROT ( n1 n2 n3 -- n2 n3 n1 )
20:13:57 <I440r> rot rotetes the third itep out to the top
20:14:09 <I440r> everything on the left of the --- or -- or what ever is
20:14:10 <Oscarian> ah
20:14:21 <I440r> whats on the stack BEFORE you execute a word
20:14:35 <I440r> everything to the right of the --- is whats on the stack AFTER you execute the word
20:14:41 <I440r> the stack comment for + might be
20:14:49 <I440r> ( n1 n2 --- n3 )
20:15:06 <I440r> the TOP of stack is always to the right
20:15:16 <I440r> swaps stack comment is
20:15:21 <I440r> ( n1 n2 --- n2 n1 )
20:15:24 <CrowKiller> actually a rot is very ugly IMHO, unless its faster to do it the other way, wich is quite rare; better to push it on the return stack to use it later
20:15:48 <CrowKiller> read: unless its faster to do it THIS way
20:16:19 <CrowKiller> the return stack can be used as a second temporary data stack
20:16:21 <kc5tja> CrowKiller: Or you can't because the return stack will get cruft put onto it from other processing steps.
20:17:07 <CrowKiller> hmmm in ugly forths, yes, but its wise to take that into account ;p
20:17:20 <kc5tja> CrowKiller: you can't always do that.
20:17:41 <Oscarian> could i use forth to program a PIC?
20:17:44 <kc5tja> CrowKiller: I cite the mere existance of ROT as proof of that.  :)
20:17:47 <CrowKiller> i know thats why i say its waise to take that into consideration when working on a "foreign" forth
20:17:49 <Oscarian> or do i have to stick with asm for that?
20:18:12 <CrowKiller> oscarian, the forth methodology is great to code PICS
20:18:19 <Oscarian> ah, good stuff
20:18:23 <CrowKiller> you can write the forth you need when you need it
20:18:33 <CrowKiller> the way you want it
20:18:52 * kc5tja isn't so sure about that, CrowKiller ... PICs don't have writable program storage...
20:18:58 <kc5tja> Not from the PIC instruction set anyway.
20:18:58 <Oscarian> i saw on a web page yesterday someone who had written PONG for a PIC
20:19:02 <I440r> there are some PIC's that would be very difficult to code a forht for heh
20:19:11 <I440r> oscar codes asm?
20:19:20 <Oscarian> not yet, but i'm learning...
20:19:24 <Oscarian> slowly...
20:19:33 <CrowKiller> kc5tja: pic17 can read code from exteranl ram, and otehr rom based Ucontrollers can serve as bytecode interpreters
20:19:36 <Oscarian> i'm out of work at the moment, so i'm trying to find things to do with my spare time
20:20:18 <I440r> me 2
20:20:26 <I440r> u code embedded ?
20:20:30 <Oscarian> nope
20:20:35 <Oscarian> just fiddling with stuff in linux
20:20:40 <CrowKiller> I worked with SXs and AVRs and they can do Forth very well, even from external ram or serial stream
20:20:42 <kc5tja> CrowKiller: I suppose so; a bit of a slow way to do it, but it would work, I guess.
20:21:04 <I440r> ok :)
20:21:11 * kc5tja needs to re-organize his target compiler's source code... :/
20:21:19 <Oscarian> i don't know very much at all, i need to read through that online asm coding book...
20:21:34 <Oscarian> which, i should go find again while i'm thinking of it...
20:22:04 <I440r> :)
20:22:12 <CrowKiller> asm is the basic skill
20:22:26 <CrowKiller> after you learn forth and thats all dont need to learn any other language
20:22:27 <CrowKiller> ;p
20:23:37 <I440r> crow asm and forth
20:23:45 <I440r> i think all coders. ALL CODERS should be required to serve an apprenticeship
20:24:10 <Etaoin> there's some neat stuff other than forth and asm. I don't mean c
20:24:14 <I440r> where they are not allowed to code in ANY language other than PURE ASM for the first 10 years
20:24:14 <I440r> after that they can learn forth
20:24:56 <CrowKiller> i started to code on PICs like devices and it served me well, taught me binary arithmetics and all
20:26:25 <Oscarian> i'm also contemplating learning either LISP or Haskell
20:26:44 <Oscarian> but i need to actually know something useful before i try to learn either one
20:29:06 <Etaoin> lisp is awsome
20:29:14 <Etaoin> haven't tried haskell
20:30:03 <Oscarian> well, i have to go out...thanks for the small guide I440r, i'll find it useful i'm sure...
20:30:17 <Oscarian> i might try out some forth this afternoon after i get back
20:30:44 <Oscarian> catchas all!
20:30:53 * Oscarian is away: gone out for a while
20:31:12 <I440r> oscar if you forget it all its ok
20:31:28 <I440r> cuz if you stick with it youll remember what i said
20:37:15 --- quit: I440r ("well its time to go code some more :)")
20:44:11 --- quit: Herkamire ("goodnight")
20:56:02 <kc5tja> brb; rebooting.
21:08:48 <kc5tja> back
21:10:26 <CrowKiller> re
21:10:40 * CrowKiller is in the design phase of his aha-like system
21:11:51 * kc5tja is re-organizing his target compiler before he adds more features to it.
21:11:57 <kc5tja> Cleaning up the source code.
21:19:44 <CrowKiller> my aha system is based on 8 bits token, so it can be used in microcontrollers
21:20:23 <CrowKiller> the draft is almost finished, ill polish things up and will post it on a local webserver for this chan's viewing pleasure ;p
21:21:38 <kc5tja> Cool
21:30:57 <CrowKiller> its 60% finished, ill post it, wait a bit ill start the server
21:32:41 <CrowKiller> http://24.212.5.233
21:32:58 <CrowKiller> I cut and pasted it into Netscape Composer
21:33:08 <CrowKiller> dont look at the html source lol
21:37:09 <CrowKiller> please press reload i changed some stuff in the first paragraph
21:37:23 <CrowKiller> just in acse your viewing it lol
21:37:40 <CrowKiller> here its 1 a.m so i must get to bed
21:38:36 <CrowKiller> since I sleep beside it, I must shut down my PC, sorry for the inconvenience, it will be back tommorow, maybe even done better
21:38:41 <CrowKiller> good night all
21:38:56 --- quit: CrowKiller ("Bibi veut des fruits")
21:49:43 --- quit: Soap` ()
22:11:11 --- quit: Etaoin ("raise OverFlowError, "Ewwww, now it's all over the floor."")
22:12:57 --- join: rob_ert (~robert@h237n2fls31o965.telia.com) joined #forth
22:24:54 <kc5tja> What is a good naming convention for words that embed things into the current definition?
22:25:13 <kc5tja> I'm already using XXX, to indicate words that compile things into the current definition.
22:25:41 <kc5tja> My problem in particular is that I'm writing code to embed a string into a colon definition, for use with S" and ." words.
22:25:57 <kc5tja> S, is already defined for strings, and is used by S" and .".
22:26:08 <kc5tja> hmm...
22:26:09 * kc5tja thinks
22:32:58 <kc5tja> Never mind -- I don't need to rename the word.  It'll never be in the target compiled dictionary anyway.
22:36:36 --- join: Fracta| (zopqss@h24-77-171-228.ok.shawcable.net) joined #forth
22:38:10 <rob_ert> Hi.
22:39:35 <kc5tja> re
22:39:41 * kc5tja is cleaning up the source code to his target compiler.
22:48:28 --- join: Serg_penguin (~snaga_NOI@nat-ch0.nat.comex.ru) joined #forth
22:48:46 <Serg_penguin> hi
22:48:59 <Serg_penguin> i have a stupid question on FORGET
22:49:36 <Fracta|> FORGET it.
22:50:01 <Serg_penguin> really, i need :
22:50:25 <Serg_penguin> to FORGET by cfa adress, like EXECUTE, not by name
22:50:48 <Fracta|> I dunno. Go ask in #forth.
22:50:59 <Serg_penguin> to forget the word defined after the specified word
22:51:01 <kc5tja> This IS #forth
22:51:09 <Fracta|> No it isn't.
22:51:15 <Serg_penguin> like : USER-XTNS ;
22:51:22 <kc5tja> My BitchX bar reads #Forth.  What channel am I in?
22:51:34 <Serg_penguin> FORGET USER-XTNS
22:51:37 <Fracta|> #forth_sucks_ass
22:52:04 <Serg_penguin> all beyond USER-XTNS should go hell, but USER-XTNS itself should stay
22:52:11 <Serg_penguin> how can i do it ?
22:52:27 <kc5tja> Fracta|: if you don't like Forth, why are you here?
22:52:52 <Fracta|> When you create USER-XTNS, create USER-XTNS2 immediatley after, then FORGET that?
22:52:59 <kc5tja> Serg_penguin: No; Forth provides no standard way to forget based on an xt or cfa.
22:53:38 <Serg_penguin> i know..
22:53:45 <Fracta|> Ok. I'm lost. That's why I'm not in #forth, I guess...
22:54:07 <Serg_penguin> but how can i deefine it ?
22:54:21 <Serg_penguin> manually setting HERE and LATEST vars ?
22:54:27 <Fracta|> Uh... : USER-EXTNS2 ;  ??
22:54:37 <kc5tja> That would be highly compiler specific.
22:54:46 <kc5tja> Fracta| has a better solution; or you can use MARKER.
22:55:02 <Fracta|> MARKER, eh? I'll look into that...
22:55:13 <Serg_penguin> but does MARKER go after use, or stay ?
22:55:36 <kc5tja> It goes after it's used.  But it's most often used like this:
22:55:42 <kc5tja> FOO MARKER FOO blah blah,
22:55:58 <kc5tja> So although FOO disappears after it's used, MARKER FOO immediately re-creates it.
22:56:25 <Serg_penguin> i wanna make 2 things:
22:56:45 <Serg_penguin> fdisk highly programmable&customizable
22:57:08 <Serg_penguin> and a game w/ executing source pieces from world/levels files
22:57:20 <Serg_penguin> for both i need what i iasked
22:57:39 <Serg_penguin> ??
22:57:53 <Serg_penguin> 1) i urgently need
22:58:01 <kc5tja> If the sources are loaded in via INCLUDE or LOAD or THRU (the former being for files, and the latter two being for blocks), then MARKER solution would work for your needs.
22:58:42 <Serg_penguin> 2) maybe will do just as educational example
22:59:07 <Serg_penguin> what if user will FORGET ;) to define marker foo at start of his script ?
22:59:10 <kc5tja> s/INCLUDE/INCLUDED/
22:59:41 <Serg_penguin> what if levels will grow unnecessary because of extra src line ?
23:00:03 <kc5tja> You can create a load chain.  INCLUDED takes a string on the data stack, so you can create a file called "marker.fs" for example, and do this:  S" myExtensions.fs" INCLUDE marker.fs
23:00:14 <kc5tja> Where marker.fs contains the following one line:  FOO MARKER FOO INCLUDED
23:01:19 <kc5tja> Anything beyond this requires intimate knowledge of the specific Forth compiler you're writing for, unfortunately.
23:02:06 <Serg_penguin> u sure dont know it - i use 2 russian forths, one dos 64k, 2nd - full dos dpmi
23:03:05 <kc5tja> I'm not sure how to take the last message...what do you mean?
23:05:40 <Serg_penguin> i am Russian... and these forths have all docs in rus, so u hardly can learn their internals
23:06:20 <kc5tja> Hmm..that brings to mind that they're quite likely not ANSI compliant either.
23:06:24 <Serg_penguin> second is no-source ;(, 1st is full-src, lots of rus comments ;)
23:06:39 <kc5tja> In that case, instead of FOO MARKER FOO, you might want something like FORGET FOO : FOO ; to get the same effect.
23:07:09 <Serg_penguin> no, both have marker
23:07:20 <Serg_penguin> both claim to obey ansi
23:08:39 <kc5tja> OK, hopefully that will work for your needs.
23:09:45 * kc5tja has a coworker from Russia where I work.  It's interesting to see how Forth has a fairly high level of interest in Russia.  (At least when he was still living there, back when the USSR still existed)
23:10:09 <kc5tja> versus America, where Forth has almost zero interest.
23:18:19 * Oscarian is away: Frazzled, I'm sure
23:18:30 * Oscarian is back (gone 00:00:03)
23:18:44 <Oscarian> oops
23:22:54 --- quit: sif (Read error: 110 (Connection timed out))
23:26:32 * Oscarian is away: de-smelling
23:27:20 --- join: Soap` (~flop@202-0-42-22.cable.paradise.net.nz) joined #forth
23:29:55 <Serg_penguin> maybe, ist because poor comuter power ?
23:30:13 <Serg_penguin> and 70's USA designs copiet in 90's
23:30:28 <Serg_penguin> PDP-11 in home computer, 16 k ram...
23:30:59 <Serg_penguin> it was 1st time i run a BASIC proggie....
23:31:34 <Serg_penguin> it was a big keyboard , not huge boxes of real pdp-11
23:32:45 <Serg_penguin> and costed 550 rub while avg monthly income was 70-110 rub for 1 working man
23:34:21 <Serg_penguin> it was be4 rub (and ussr) crashed...
23:34:39 <Serg_penguin> hey !!!!
23:35:02 <Serg_penguin> afk..
23:36:40 <kc5tja> Ahh
23:36:50 <kc5tja> Sorry; I was busy hacking away on my own Forth compiler environment.
23:37:10 * Oscarian is back (gone 00:10:38)
23:37:44 <kc5tja> I'm really pleased to see that it's coming along; however, I'm just a tad upset at the code quality it produces.  But, so far, it's only 31% larger code, so I guess I shouldn't be too terribly worried yet.
23:38:09 <kc5tja> It actually falls in line with what's to be expected between direct threaded and subroutine threaded code.
23:40:12 <kc5tja> Most of the code bloat seems to come from the fact that 16-bit x86 code (IA-16) isn't orthogonal, and thus, many CPU instructions do the job of one Forth VM instruction.
23:40:37 <kc5tja> The constant adding and subtracting to BP is also a problem too.
23:40:59 <kc5tja> Those represent wasted cycles more than anything else.
23:41:25 <kc5tja> But that's OK -- this version of FS/Forth is the very first time I've ever done any native code compilation techniques of any kind.  I'm really pleased with the results overall.
23:42:11 <kc5tja> as far as PDP-11s, I can see that happening.  In America, computer power has grown so incredibly fast, that the need for good programming techniques and languages has been obscured by technological progress.
23:46:54 <Serg_penguin> i just don't grasp it
23:47:17 <Serg_penguin> u definitely may get more from ANY cpu by good optimisation
23:47:28 <Serg_penguin> from Z80 to P4
23:48:11 --- join: davidw (~davidw@adsl-ull-217-108.42-151.net24.it) joined #forth
23:48:12 <kc5tja> Correct; however, the Z-80 definately has its limitations.  :)
23:48:57 <kc5tja> The 6502 was easily twice as fast as the Z-80 at everything except block memory moves (and that's only because the Z-80 had a LDIR and LDDR instructions)
23:49:03 <kc5tja> for the same clock speed.
23:49:22 <Serg_penguin> demo-makers did a realtime raytracing  working well on my p166 (320x200, closed room w/5-10 yextured primitives)
23:49:39 <Serg_penguin> why industry did not do it even or p4 ?
23:50:07 <kc5tja> Well, the only real difference, from a software perspective, between the P4 and the p166 is clock speed.
23:50:13 <kc5tja> So the same code would still work on the P4.
23:50:30 <kc5tja> Today's most advanced Pentium-class machines are still just glorified 80386 processors.  :)
23:51:28 <kc5tja> What I don't understand is why the software industry adopts a source distribution method, and let the processor the software runs on be free to change as needed.
23:52:14 <kc5tja> Forth *excels* at this like no other language in the world (I have plans of porting my FS/Forth System I program to the Commodore 64 just as a demonstration of retargetability of the target compiler).
23:52:42 <kc5tja> And the one-pass compiler makes compilation almost unnoticable.
23:53:02 <kc5tja> Compared to waiting for the disk to deliver the next chunk of source text, compilation IS zero overhead.
23:53:40 <kc5tja> s/software industry adopts/software industry didn't adopt/
23:53:47 <kc5tja> sorry -- I didn't catch that error until just now.
23:54:30 <Serg_penguin> i wanna add in-place vocabulary edition to my forth compiler..
23:54:52 <kc5tja> How do you mean?  Retroactive word redefinitions?
23:55:09 <Serg_penguin> w/ ability to even toss out words and shrink/grow wokabulary, insert in middle (correcting all calls/branches)
23:55:21 <kc5tja> Ahh, I see.
23:55:44 <kc5tja> That can come in handy for if you load two or three modules, but you want to remove only the middle modules.
23:56:17 <Serg_penguin> but words doing wthis must be first...
23:59:23 <kc5tja> Right.
23:59:59 --- log: ended forth/02.05.27