00:00:00 --- log: started forth/02.08.06
00:06:17 <kc5tja> back
00:08:23 <kc5tja> GOING TO BED
00:08:24 <kc5tja> wow
00:08:30 <kc5tja> Going to bed -- that's better. :)
00:08:43 --- quit: kc5tja ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK")
00:09:55 --- quit: proteusguy (Read error: 110 (Connection timed out))
01:51:52 --- join: proteusguy (~irc@24-197-147-197.charterga.net) joined #forth
02:20:40 --- join: Serg_penguin (~Z@nat-ch1.nat.comex.ru) joined #forth
02:21:01 <Serg_penguin> hi
03:11:02 --- quit: proteusguy (Read error: 110 (Connection timed out))
03:30:39 --- quit: Serg_penguin (Read error: 110 (Connection timed out))
04:21:59 --- join: Serg_penguin (~Z@nat-ch1.nat.comex.ru) joined #forth
04:22:05 <Serg_penguin> hi
04:22:27 --- quit: Serg_penguin (Client Quit)
04:50:46 --- join: proteusguy (~irc@24-197-147-197.charterga.net) joined #forth
05:11:43 --- quit: proteusguy (Read error: 110 (Connection timed out))
05:38:21 --- join: Serg_penguin (~Z@217.10.35.226) joined #forth
05:38:27 <Serg_penguin> hi
05:38:44 --- quit: Serg_penguin (Client Quit)
06:50:59 --- join: proteusguy (~irc@24-197-147-197.charterga.net) joined #forth
07:09:41 --- quit: proteusguy (Read error: 110 (Connection timed out))
07:24:59 --- join: proteusguy (~irc@24-197-147-197.charterga.net) joined #forth
08:09:40 --- quit: proteusguy (Success)
08:13:37 --- join: kc5tja (~kc5tja@ip68-8-206-226.sd.sd.cox.net) joined #forth
08:27:27 --- join: Kitanin (~clark@h-209-91-66-234.gen.cadvision.com) joined #forth
08:51:50 --- nick: kc5tja -> kc-shower
08:51:50 --- quit: Kitanin (Read error: 104 (Connection reset by peer))
08:52:38 --- join: proteusguy (~irc@24-197-147-197.charterga.net) joined #forth
09:32:47 --- nick: kc-shower -> kc5tja
09:47:19 --- quit: kc5tja ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK")
10:03:25 --- join: Speuler (l@62.206.56.189) joined #forth
10:03:38 <Speuler> salve
10:04:47 --- join: I440r (~mark4@67.241.61.155) joined #forth
10:05:08 <Speuler> 'day m4
10:05:33 <I440r> hi
10:06:09 <I440r> doing anything these days dood ?
10:06:20 <Speuler> yeah. 
10:06:25 <Speuler> work you mean ?
10:06:30 <I440r> ya
10:06:35 <Speuler> sort'f
10:06:47 <Speuler> all linux-related
10:07:09 <Speuler> misc stuff
10:07:09 <I440r> cool
10:07:30 <I440r> anything forth related ? :)
10:07:34 <Speuler> edu, promotional and radio stuff
10:07:56 <Speuler> 'll be using forth when shell scripts won't do
10:08:12 <I440r> :)
10:08:14 <Speuler> which didn't happen yet
10:08:37 <Speuler> the office i'm in right now needs presentation systems
10:08:55 <Speuler> beamers, 60" plasma screens, to show presentations on
10:09:07 <I440r> presentation ?
10:09:12 <I440r> aha
10:09:12 <I440r> cool
10:09:18 <Speuler> commercials, tv, and the like
10:09:28 <Speuler> server-driven
10:09:31 --- quit: proteusguy (Success)
10:09:44 <Speuler> for easy change of sequence
10:10:20 <Speuler> everything, including tv control (channel, full-screen v windowed), done with bash :)
10:10:33 <I440r> lol
10:10:36 <Speuler> interfaced to cookies :)
10:10:56 <Speuler> client and server
10:11:16 <I440r> why bash ?
10:11:33 <Speuler> happened to be my default shell
10:12:06 <Speuler> why program when shell scripts do ?
10:12:28 <I440r> because i cant bash :)
10:12:40 <Speuler> simple stuff
10:13:33 <Speuler> my 5 year old sister could do that
10:13:44 <Speuler> if i'd had a five years old sister
10:14:15 <I440r> lol
10:14:36 <I440r> bash never interested me realy
10:16:49 <Speuler> t'is not my intention to become a bash script professional.
10:17:20 <Speuler> for simple things it can be quite suited though
10:18:33 <Speuler> phood grab
10:18:44 --- nick: Speuler -> PhoodPhrenzy
10:19:02 <PhoodPhrenzy> oh...
10:19:10 <PhoodPhrenzy> had a job on a bsd machine...
10:19:17 <I440r> cool 
10:19:24 <PhoodPhrenzy> had to break in :)
10:19:46 <PhoodPhrenzy> gateway/firewall
10:19:51 <PhoodPhrenzy> unknown root pwd
10:20:13 <PhoodPhrenzy> first bsd xperience
10:20:38 <PhoodPhrenzy> got in :)
10:22:41 <I440r> heh
10:22:42 <I440r> brb
10:22:44 <I440r> neeed coffee
10:23:13 --- quit: PhoodPhrenzy ("klasikahl: well, my suggestion is sleep, and reinstall when you're fresh.  you are, however, free to do whatever you choose.")
10:27:37 --- join: Kitanin (~clark@h-209-91-66-234.gen.cadvision.com) joined #forth
10:33:50 <I440r> hi kit
11:07:06 --- join: proteusguy (~irc@24-197-147-197.charterga.net) joined #forth
11:07:07 --- quit: Kitanin (Read error: 104 (Connection reset by peer))
11:12:01 <futhin> i440r: what's news with isforth?
11:22:13 <I440r> not much
11:22:26 <I440r> i still didnt release 1.09b yet
11:22:32 <futhin> okay
11:22:34 <futhin> but what is it coming with?
11:22:44 <futhin> the tui windowing stuff? the socket stuff? 
11:22:46 <I440r> 1.09b has dynamic memory allocation
11:22:48 <futhin> graphics stuff?
11:22:54 <I440r> and the windowing stuff yes
11:22:59 <I440r> not the sockets yet
11:23:08 <I440r> that still needs to have DNS 
11:23:17 <futhin> what the hell is it doing with dynamic memory allocation?
11:23:35 <I440r> isforth doesnt use any but user apps would need it
11:23:35 <futhin> heh
11:23:43 <futhin> hm?
11:23:45 <I440r> being able to allocate and free memory is very important
11:23:52 <I440r> your mud will use it alot for sure
11:23:59 <futhin> since when do REAL Forth Coders use dynamic memory allocation?
11:24:29 <I440r> isforth allows you to grab memory from dictionary space but you cant allocate it and then FREE it up
11:24:42 <futhin> does chuck moore use dynamic memory allocation in his mud? :P
11:24:52 <I440r> does chuck moore have a mud ?
11:24:55 <I440r> chuckmud :)
11:24:57 <futhin> heh
11:25:01 <I440r> shipmud ?
11:25:02 <I440r> erm
11:25:05 <I440r> chipmud even
11:25:12 <futhin> i was sort of joking
11:25:19 <futhin> but "chipmud" sounds like a good idea
11:25:21 <I440r> i never would have guessed :P
11:25:43 <futhin> well i'm saying "would chuck moore use dynamic memory allocation if he were to code a mud?"
11:25:52 <I440r> of corse
11:26:13 <futhin> are you absolutely 102% positive?!
11:26:22 <I440r> 103.5%
11:26:24 <futhin> heh
11:26:35 <I440r> :)
11:26:51 <futhin> yeah but he would implement the dynamic memory allocation code at the begining of the mud code
11:27:03 <futhin> instead of using the implementation's dynamic mem alloc code? :P
11:28:28 <futhin> humhumhumhum
11:28:33 <I440r> actually, the meta compiler will prolly use dynamic memory allocation
11:28:47 <I440r> allocate a bunch of memory to target compile the kernel into or something
11:29:01 <I440r> that is if i ever get an assembler i can use
11:29:27 <futhin> hey, question
11:29:39 <futhin> is it harder to code a forth in asm in linux or in dos?
11:35:33 <I440r> hang on.. brb
11:36:43 <I440r> back
11:37:00 <I440r> i would say its harder in linux because i dont know linux as wel las i know dos
11:37:11 <I440r> and linux places arbitary restrictions on what you can and cant do
11:37:19 <futhin> i'm talking about assembler-level
11:37:25 <I440r> fucking protected mode not letting me write directly to the display :P
11:37:29 <I440r> so am i
11:37:33 <futhin> k
11:37:46 <I440r> theres alot of info you need to be ablt to code asm in linux
11:37:49 <I440r> and its all HIDDEN
11:37:54 <I440r> they dont WANT you coding in assembler
11:38:21 <futhin> yeah
11:38:25 <futhin> bunch of C losers
11:38:30 <I440r> xaktlee
11:38:52 <futhin> mmm, win32 asm coding is soo much fun ;)
11:41:23 <I440r> what i want to know is where the hell GOOGLE got its name :P
11:47:22 <futhin> why?
11:48:52 <I440r> its important we know these things :)
11:52:21 --- join: proteus (~irc@24-197-147-197.charterga.net) joined #forth
11:53:09 <I440r> i need a defered does> word
11:53:39 <I440r> a non creating does> word where teh does> part is defered till later heh
11:53:58 <futhin> dieeee DEFER!
11:54:05 <I440r> hrm 
11:54:12 <I440r> no defer is good unless its abused
11:54:25 <I440r> defered words arent so you can do forward references
11:55:00 <I440r> isforth however needs some rethinking on its defered initializaiton chain
11:55:08 <I440r> its too complex but i cant see a way out of it
12:09:44 --- quit: proteusguy (Connection timed out)
12:09:54 <futhin> DIEEE DEFER!
12:10:21 <futhin> colorforth has two dictionaries or something?   it eliminates the need for DEFER doesn't it?
12:27:47 --- quit: proteus ()
12:30:31 --- join: proteusguy (~irc@24-197-147-197.charterga.net) joined #forth
12:34:51 --- quit: proteusguy (Read error: 104 (Connection reset by peer))
12:35:53 <I440r> no
12:36:11 <I440r> defer allows you to revector things like KEY and EMIT 
12:36:30 <futhin> wasn't there another word for revectoring things?
12:36:45 <I440r> defer foo
12:36:48 <I440r> ' bar is foo
12:36:55 <futhin> there is:  DEFER, IMMEDIATE, FORWARD..  what words are there??
12:37:10 <I440r> forward references are EVIL
12:37:18 <futhin> i recall some forth implementation that does not use DEFER
12:37:20 <futhin> what does it use?
12:37:27 <I440r> dunno
12:37:35 <I440r> brb
12:37:39 <I440r> upgrading mirc :P
12:37:43 --- quit: I440r ()
12:43:06 --- join: I440r (~mark4@1Cust155.tnt3.bloomington.in.da.uu.net) joined #forth
12:43:30 <futhin> come on, there's forths with no DEFER
12:43:33 <futhin> they use a different word
12:43:34 <futhin> what is it
12:43:39 <I440r> i dunno
12:43:50 <futhin> you lie! you told me before! ;P
12:46:04 --- join: Kitanin (~clark@h-209-91-66-234.gen.cadvision.com) joined #forth
12:47:27 <I440r> hi kit
12:47:35 <I440r> u there this time ??? :)
12:49:23 <I440r> guess not
12:52:12 <I440r> ok, this is the only channel he is in and his idle time keeps resetting but he isnt talking here
14:04:04 --- quit: Kitanin (Read error: 104 (Connection reset by peer))
14:07:18 --- join: kc5tja (~kc5tja@user-24-214-86-42.knology.net) joined #forth
14:11:03 * kc5tja pollutes the mind of yet another person with Forth....bwahahahaha!!
14:11:31 <kc5tja> (actually, he's trying to learn C, but he's having a hard time.  I recommended he try Forth instead; it's different enough that he might get it.)
14:12:06 <I440r> :)
14:12:15 <I440r> have him start with asm if he doesnt knnow it already
14:19:09 --- join: Kitanin (~clark@h-209-91-66-234.gen.cadvision.com) joined #forth
14:34:12 --- join: proteusguy (~scherrey@24-197-147-197.charterga.net) joined #forth
14:57:45 <I440r> bbl, got alot of rtfm to do :)
14:57:50 --- quit: I440r ()
15:02:06 --- nick: kc5tja -> kc-meeting
15:07:59 <futhin> hmmmmmmmmmm
15:08:00 * Kitanin looks around...
15:08:05 <Kitanin> Is he gone?
15:08:07 <futhin> yeah
15:08:18 <futhin> you didn't talk to him and so he left
15:08:21 <futhin> you meanie! :P
15:08:24 * Kitanin dances...
15:08:33 <Kitanin> I have _no_ idea who he is.
15:08:54 <futhin> i440r ?
15:09:05 <Kitanin> Yeah.
15:09:09 --- join: proteus (~scherrey@24-197-147-197.charterga.net) joined #forth
15:09:15 <futhin> he's the guy who's coding isforth
15:09:29 <futhin> he's the original founder of this channel
15:09:41 <futhin> (i stole the channel from him, bwhahaha)
15:10:17 --- quit: proteusguy (Read error: 110 (Connection timed out))
15:10:22 <Kitanin> Oh, okay.  :-)  Well, now I feel bad.
15:10:25 <futhin> kitanin: maybe he just wanted to chat, asshole :P
15:11:02 <futhin> (asshole only if you deliberately ignored him when he was trying to say hi and stuff to you)
15:11:15 <futhin> hm
15:11:22 <futhin> kitanin: so what do you do?
15:11:30 <futhin> with forth
15:11:42 <Kitanin> No, I wasn't.  I started up IRC, then wandered away for six hours.  :-(
15:12:13 <Kitanin> Right now, not much.  <sigh>
15:12:38 <futhin> awesome, chuck moore finally replied to my email
15:12:52 <futhin> kitanin: how familiar are you with forth?
15:13:19 <Kitanin> It's been about 15 years since I did anything serious.
15:14:07 * Kitanin eeps.
15:14:23 <Kitanin> Time to fly to Milwaukee.  Yay!
15:14:46 <futhin> anything at milwaukee?
15:16:02 <futhin> ahh, i remember, the annual forth coding convention is at milwaukee
15:16:16 <Kitanin> The GenCon gaming convention.  :-)  And now, I must take my leave of you.  Tell I440r I'm sorry about not responding, but I've spent about twenty minutes at my desk today.
15:16:49 --- part: Kitanin left #forth
15:18:17 --- quit: futhin (Read error: 104 (Connection reset by peer))
15:18:20 --- join: futhin_ (~thin@h24-64-175-61.cg.shawcable.net) joined #forth
15:19:57 --- nick: futhin_ -> futhin
15:30:26 --- nick: kc-meeting -> kc5tja
15:30:59 <kc5tja> re
15:31:01 <kc5tja> Cool.
16:08:54 --- quit: proteus (Read error: 110 (Connection timed out))
16:10:57 --- join: cleverdra (jfondren@0-1pool36-45.nas2.florence1.sc.us.da.qwest.net) joined #forth
16:12:52 <futhin> hi cleverdra :)
16:13:18 <futhin> what are you up to these days?
16:13:25 <futhin> you are never on irc nearly enough :P
16:13:50 <cleverdra> Trying to get a few good lisp systems working.
16:14:12 <cleverdra> (It's not easy.)
16:14:34 <cleverdra> Yeah, but soon I'll be in Basic and I won't be on at all for a while =)
16:15:08 <futhin> hmm
16:15:21 <futhin> do you have an interest in forthchips?
16:17:02 --- join: I440r (~mark4@1Cust28.tnt2.bloomington.in.da.uu.net) joined #forth
16:17:34 * cleverdra notes that Basic was Air Force Basic training.
16:17:56 <cleverdra> I do have an interest, futhin.
16:21:09 <futhin> how mnay forthchips would you buy if they were 3 bucks a piece (assume F21)
16:22:03 <cleverdra> well, one, at least to start.
16:22:38 <futhin> well it costs $60 grand for 20,000 forthchips
16:22:49 <futhin> and we'd need to get a whole bunch of forth coders to put in money
16:23:03 * cleverdra nods.
16:23:05 <futhin> unfortunately, this cost does not include packaging :(
16:23:14 <futhin> it could go up to 50 bucks a chip or 100 bucks
16:23:15 <kc5tja> That we know of.
16:23:49 <futhin> kc5tja: why can't we use a smaller pin package that is more common and thus a lot cheaper
16:23:53 <futhin> like 40 pin ?
16:24:06 <kc5tja> Why can't we?  I don't recall saying we couldn't.
16:24:13 <futhin> is there any _huge_ disadvantages going from 68 pins to 40 pins
16:24:47 <kc5tja> That depends on your application.
16:25:17 <kc5tja> If you're using it as the main CPU, probably not.  If you're using it as a DMA controller or bus controller, yes, absolutely.
16:25:21 <futhin> my application is to sell 20,000 forth chips ;)
16:25:36 * kc5tja sighs
16:25:53 <futhin> heh
16:33:59 <I440r> i guess mrreach doesnt come in here no more
16:34:50 <futhin> naw
16:34:59 <futhin> comes on opn for #rei
16:35:04 <futhin> real estate investing is l33t
16:36:07 <futhin> forth distracts me
16:36:10 <futhin> but not mrreach ;)
16:37:16 <kc5tja> Heheh
16:37:33 <kc5tja> Actually, MrReach came on fairly late at night the past couple of nights, didn't he?
16:37:42 <futhin> yeah
16:37:43 <kc5tja> I remember staying up to 2AM talking about engines with him once.
16:37:46 <futhin> yup
16:37:49 <futhin> that was a few days ago
16:37:52 * kc5tja nods
16:38:22 <kc5tja> Man, I can't wait to finish my Forth environment.  I really can't.  It's going to be a lot of fun getting it up and working.
16:38:49 <futhin> hurry up and wait!
16:39:04 <kc5tja> I hope I can do it before college starts up though, because I have a feeling I'll be needing it.  :-)
16:39:31 <futhin> you are taking some classes this fall?
16:41:05 * kc5tja nods
16:41:12 <kc5tja> Going back to school to get my physics degree.
16:41:31 <kc5tja> Taking English 100 and Math 135 (composition and comprehension, and pre-calc, respectively)
16:43:08 <futhin> fun fun
16:44:37 <kc5tja> Well, I don't want to be doing this stuff for the rest of my life.  I like Physics, and a degree is virtually necessary to get any kind of reasonably paying job these days, so....
16:56:16 <I440r> is mrreach pissed off or something ?
16:56:26 <I440r> or just avoiding #forth because its a distraction
17:00:37 <kc5tja> Only he knows at this time.
17:04:38 --- join: Speuler (~l@pD9502572.dip.t-dialin.net) joined #forth
17:04:45 <Speuler> salve
17:07:27 <kc5tja> No thanks; I'm not injured.
17:07:40 <Speuler> :)
17:08:06 <Speuler> you can get hurt with comments like that
17:08:19 <kc5tja> Not when you're in a totally different country.  :D
17:09:46 <Speuler> people never do believe what dangers they run
17:10:08 <Speuler> such as the bloke who wouldn't want to believe that one can get killed by 6 volt
17:10:35 <Speuler> until he got killed by a car accumulator which dropped on his head
17:11:01 <kc5tja> Car accumulator?
17:11:15 * kc5tja imagines a mass of cars falling on this guy's head.
17:11:19 <Speuler> car battery ... the lead-acid thing
17:11:25 <kc5tja> That's 12V 
17:11:26 <kc5tja> :)
17:11:33 <Speuler> now they are
17:11:35 <kc5tja> But at any rate, it's current that kills, not voltage.
17:11:57 * kc5tja has successfully lived through experiencing 120V first-hand.
17:12:00 <kc5tja> And I do mean hand.  :D
17:12:10 <kc5tja> Fortunately, the current was quite small.
17:12:24 <kc5tja> It only takes 0.01A of current to stop the heart.
17:12:29 <Speuler> i=u/r
17:12:36 <kc5tja> I = E / R
17:13:00 <Speuler> given constant R, i gets bigger with higher U
17:13:10 <kc5tja> Why do you use U to represent voltage?
17:13:22 <kc5tja> At any rate, that bloke must have been soaking wet for that to happen.
17:13:31 <Speuler> same reason why i represent current as I
17:13:35 <Speuler> instead of A
17:13:48 <kc5tja> We use E or V to represent voltage here.
17:14:11 <kc5tja> I and R are pretty universal, it seems, but U?  To me, U is the internal energy contained in a gas.
17:14:26 <Speuler> why do you represent current as I ?
17:14:33 <kc5tja> Intensity.
17:14:55 <Speuler> R seems to be obvious
17:14:59 <I440r> the current through a conductor is proportional to the potential difference between its ends
17:15:00 <kc5tja> A is the unit of measurement for current, not its representation.
17:15:17 <I440r> erm 'through a GIVEN conductor' even
17:15:18 <kc5tja> E is for electromotive force, and of course, V is for voltage.
17:15:20 <Speuler> thought U is international for voltage
17:15:49 * kc5tja came within two weeks of getting his NIST certification in metrology, and I've never once seen U used for voltage, anywhere.
17:16:26 <Speuler> is used everywhere here
17:16:58 <kc5tja> What country are you in?
17:17:11 <Speuler> germany
17:17:27 <kc5tja> How would you say voltage or electromotive force in German?
17:17:45 <kc5tja> (more accurately, how would you spell it?)
17:17:51 <Speuler> spannung = voltage
17:18:15 <Speuler> translates to "tension"
17:18:30 <kc5tja> Hmm...  
17:18:46 <kc5tja> Well, anyway...
17:19:19 <kc5tja> At 6V, he'd need to drop his resistance down to only 600 ohms.  Most humans are in the hundreds of kilohms at the least.
17:19:38 <kc5tja> He had to have had wet hands or something.
17:19:46 <Speuler> a.c is more dangerous than d.c
17:19:59 <kc5tja> An urban legend.
17:20:04 <I440r> speuler no its not :)
17:20:08 <kc5tja> Both will equally contract muscles the same way.
17:20:23 <I440r> eddison tried to show that - went arround electrocuting cows to sho it
17:20:38 <kc5tja> In fact, DC will do additional damage in that it'll electrolyze(!!) bodily fluids, while AC will not.
17:20:45 <I440r> tesla went arround making arks of light jump from one had to the other to prove it wasnt :)
17:21:18 <kc5tja> That's why AC is used to drive LCD screens -- to prevent the electrolysis of the liquid crystal compound.  :)
17:21:23 <Speuler> a.c causes the heart to sync with the a.c frequency
17:21:37 <I440r> kct u mean my tft display on my laptop uses ac ?
17:21:37 <kc5tja> Speuler: When you've exceeded 0.01A, who cares?
17:21:43 <kc5tja> I440r: Yes.
17:21:49 <kc5tja> 32kHz AC usually.
17:22:25 <Speuler> heart suffers damage quickly with some frequencies
17:22:32 * kc5tja sighs
17:22:36 <Speuler> even if the current wouldn't be lethal as such
17:22:37 <kc5tja> CURRENT kills.
17:22:41 <kc5tja> Frequency has little affect.
17:22:57 <Speuler> everything kills, given an o.d.
17:23:13 <kc5tja> I mean, I've lit up at 60Hz, and nothing happened.
17:23:22 <kc5tja> 120V, 60Hz -- wall current.
17:23:43 <kc5tja> (well, nothing permanent.  A few bristles of hair stood on end for while... :) )
17:23:46 <Speuler> for how long ? left arm through leg ?
17:24:34 <kc5tja> Right arm through left arm.
17:24:38 <Speuler> most of europe runs on 220 v. not comfortable to touch it :(
17:24:53 <kc5tja> Well, no, I wouldn't want to touch it either.
17:24:58 <kc5tja> You think I WANTED to experience that?
17:25:21 <Speuler> touch wire with the back of your hand if you need to do so.
17:25:21 <kc5tja> "Tony, is this TV set ready to be repaired?"  "Yes, go ahead."  **ZOT!!**  
17:25:43 <kc5tja> Yeah, let's touch a wire with the most sensitive part of the hand.  That's real smart.  :)
17:26:03 <kc5tja> I have a better idea...  Use an RF detector to detect AC, or a capacitance meter to detect DC.  :D
17:26:07 <Speuler> if muscles contract, your finger won't grip the wire...
17:26:30 <kc5tja> Well, when you're working with components on a PC board in a device with a hot chassis, that's kind of hard to do.
17:26:37 <kc5tja> In fact, it's utterly impossible.
17:26:41 <Speuler> you break the circuit that way
17:26:51 <Speuler> (back of you hand)
17:27:34 <kc5tja> I don't know if I can believe that, since current will cause all muscles to contract, not just the ones in contact with the wire
17:27:56 <kc5tja> But it's more believable than some of the other stories I've heard.
17:28:07 <kc5tja> I still prefer to use meters instead of fingers.  :)
17:28:22 <Speuler> if your finger muscles contract, you'd put them around the wire
17:28:37 <Speuler> if touching it with the inside of your hand
17:31:52 <kc5tja> So you could use something like GForth to learn with, for example.
17:32:00 <kc5tja> GAHHY!H!H!
17:34:06 <cleverdra> (all muscles contract, but some are stronger than others.)
17:35:02 <futhin> blahblahblahblahblahblahblahblahblah
17:35:14 <futhin> lookie at what i typed
17:35:17 <futhin> forget about everythign else
17:35:24 <Speuler> kc5tja: you're adressing me when proposing to learn to use gforth ?
17:35:39 <kc5tja> Speuler: No, addressing someone in a different channel all-together.
17:35:49 <futhin> ah
17:35:51 <Speuler> ah
17:35:52 <futhin> stupid client of yours
17:36:03 <futhin> you keep making that mistake heh :P
17:36:06 <kc5tja> It should be smart enough to read my mind about who I'm talking to.
17:36:12 <futhin> heh
17:36:25 <kc5tja> actually, I've been thinking about an alternative interface for IRC the other day.
17:36:37 <futhin> that woudl be tricky, how to code the irc client to be smart enough to do that ?
17:37:13 <kc5tja> Imagine having a state-smart IRC client that remembers who you were last talking to.  So, for example, to talk in #Forth next, I'd put it in angle-brackets, like XML.  <#Forth>This is to everyone here.<Futhin>This is just to Futhin<#Forth>
17:37:53 <kc5tja> Of course, it makes using < and > more difficult.  I haven't quite figured that one out yet.  :-)
17:38:12 <futhin> i have an idea
17:38:24 <futhin> 1:hi futhin!
17:38:27 <futhin> 2:hi #forth
17:38:29 <cleverdra> /#Forth now I'm talking to #Forth
17:38:47 <futhin> where 1: is the most recent person/channel you were talking to
17:38:49 <cleverdra> If you're going to use typical IRC command syntax.
17:38:50 <futhin> and 2: is the 2nd most recent
17:38:53 <kc5tja> cleverdra: Too hard to parse correctly.
17:39:06 <kc5tja> Actually, I have a better idea.
17:39:08 <cleverdra> kc5 - oh, with users.
17:39:18 <kc5tja> Keep a configuration that maps control keys to users or channels.
17:39:28 <cleverdra> futhin - I'd prefer to discriminate between persons and channels.
17:39:42 <kc5tja> CTRL-A for #Forth, CTRL-B for Speuler, CTRL-C for #hamradio, etc.
17:39:59 <cleverdra> kc5 - and maybe ALT-chars to assign?
17:40:09 <kc5tja> cleverdra: I don't discriminate between them.  IRC lets me treat them both equally anyway, so there's no reason I shouldn't.
17:40:39 <kc5tja> cleverdra: I'd probably use a /-command for that.  /assign ^a #forth or something.
17:40:46 <kc5tja> ALT-keys would work of course, too.
17:41:06 <futhin> no, you could hit ALT-a to remember channel/person you are currently talking to
17:41:15 <futhin> and ctrl-a to go back to that channel/person
17:41:37 <cleverdra> err
17:41:42 <cleverdra> /msg futhin hello
17:41:46 <cleverdra> ALT-a
17:42:13 <futhin> yeah
17:42:30 <cleverdra> kc5tja with that, after a /msg, would unqualified input go to futhin?
17:45:36 <futhin> you have current channel/person
17:45:53 <futhin> and alt-a remembers channel/person
17:46:11 <futhin> and ctrl-a switches current channel/person to remembered channel/person
17:50:18 --- join: Fractal (ajta@h24-77-171-228.ok.shawcable.net) joined #forth
17:57:56 --- quit: Speuler (Read error: 60 (Operation timed out))
17:59:13 --- join: Speuler (~l@pD9502572.dip.t-dialin.net) joined #forth
18:18:20 --- join: ayrnieu (jfondren@0-1pool36-45.nas2.florence1.sc.us.da.qwest.net) joined #forth
18:19:37 --- quit: cleverdra (Read error: 104 (Connection reset by peer))
18:19:43 --- nick: ayrnieu -> cleverdra
18:22:09 --- quit: cleverdra (Client Quit)
18:30:51 --- join: CrowKiller (Vapo_Rulez@24.212.5.233) joined #forth
18:33:13 <CrowKiller> look at this: http://www.bytecraft.com/touchsw.html AKA the recipe for the perfect (cheapest) homemade keyboard interfaced via parallel port
18:33:50 <CrowKiller> one component! one wire per key! truly amazing
18:53:19 <futhin> one wire per key, interesting :)
18:55:14 <CrowKiller> using Chuck's timers, it could be one wire and that all lol
18:55:35 <kc5tja> Chuck's timers aren't designed for that modus operandi.
18:55:43 <kc5tja> His timer is a time-domain reflectometer.
18:55:44 <CrowKiller> or maybe one wire with resistances along it, like a dac
18:55:58 <kc5tja> Just use a normal count-down timer (F21 has 3 of them) for this.
18:56:12 <kc5tja> There's no reason to go exotic for this.
18:59:58 <CrowKiller> maybe there's a manner of having a full keyboard made using only this method, not only 8 keys
19:00:48 <CrowKiller> using an old cd, or aluminium foil with paper, or pcb with heatshrink wrap like the guy used
19:01:18 <CrowKiller> making our own keypads and keybaord is dead esay knowing that little electrical trick
19:12:03 --- join: ictx (jcomeau@sdn-ar-001flflauP300.dialsprint.net) joined #forth
19:19:58 <I440r> hi ictx
19:20:31 <ictx> hi, yes i code forth sometimes, not professionally
19:20:45 <I440r> on what platform ?
19:20:48 <kc5tja> Boy, he sure knows the routine, doesn't he?  ;D
19:20:49 <ictx> x86
19:20:53 <I440r> cool
19:21:00 <I440r> what os ??? :)
19:21:15 <Fractal> Hey I440r, when's isforth's integrated mp3 support coming out?
19:21:24 <ictx> linux; i've tried both bigforth and gforth, besides my own postforth
19:21:24 <I440r> soon as you write it :)
19:21:25 <Fractal> That'd be l33t.
19:21:44 <I440r> ictx you did postforth ????
19:21:47 <I440r> neat 
19:21:58 <ictx> such as it is; yes. i'm working on a usable version now
19:22:04 <I440r> i dont like bigforth or gforth but bigforth is the better of teh two
19:22:14 <kc5tja> (I wonder if I can replace PalmOS on my Palm IIIx with a 68K version of FS/Forth...)
19:22:24 <I440r> im writing isforth - 32 bit direct thread4ed forth using syscalls for all i/o
19:22:31 <I440r> kernel is assembled with nasm
19:22:48 <I440r> i need an assembler extension for it tho
19:22:57 <ictx> i coded postforth in gas
19:23:15 <I440r> yea heh - i cant handle gas, at&t will rot your brain heh
19:23:19 <ictx> yes, the assembler will be the first thing in the new postforth, if and when i ever complete it
19:23:49 <I440r> my BIGGEST problem regarding the assembler is that i cannot accept 5 # ax mov
19:23:52 <I440r> etc etc
19:24:21 <I440r> what i want is an assembler extension that can take the existing nasm sources with some minor edits (insert spaces etc)
19:24:33 <kc5tja> Can you accept move.l #$DEADBEEF,d0?
19:25:14 <I440r> no
19:25:16 <Fractal> kc5tja : You can run uClinux on the palm, btw.
19:25:24 <I440r> move.l is NOT an x86 instruction :P
19:25:36 <kc5tja> I440r: That's not the point.  :)
19:26:06 <kc5tja> Fractal: Yeah, I remember someone at 3Com porting it for demonstration purposes.  I thought that was cool.  :)  But it's just that I wonder if the flash ROM file format is adequately documented.
19:26:17 <I440r> kc5 when i code 68k i code 68k - at&t decided in their infinite stupidity that x86 assembler should LOOK like 68k assembler
19:26:25 <I440r> they were wrong :P
19:26:28 <kc5tja> I don't want to have to reverse engineer uCLinux to figure it out.
19:26:55 <kc5tja> I440r: That's an interesting theory, considering AT&T used that syntax about 10 years before the 68000 was ever dreamed of.
19:26:58 <I440r> ictx u working on an assembler extension ?
19:27:10 <ictx> just getting started really
19:27:15 <Fractal> Yes, that could be a problem. You could always set up a SLIP link through the serial port.
19:27:31 <I440r> ivev downloaded gigs of dox and so far all it does is tie my head in knots :P
19:27:35 <ictx> i'll probably think it over for another month or two before writing any code
19:27:35 <kc5tja> Fractal: ??  What exactly would that do for me?
19:27:37 <Fractal> Course you would still need to use the flash.
19:28:02 <I440r> ictx what syntax will your assembler use ?
19:28:09 <I440r> gas?
19:28:16 <I440r> or standard x86?
19:28:17 <ictx> src dest opcode
19:28:23 <Fractal> kc5tja : Well, you could talk to your computer...
19:28:27 * Speuler smiles
19:28:54 <I440r> ictx i cant get my head arround that either. first and foremost im an assembler coder. when i code asm i THINK asm. having to think backwards in asm just doesnt work for me
19:28:59 <kc5tja> fractal: Yes, of course.  I thought you were talking about using SLIP to flash it.  
19:29:01 <I440r> when im coding forth i think forth
19:29:30 <ictx> i440r: why is src dest backwards?
19:29:49 <I440r> it should be opcode destination, source
19:29:50 <Fractal> kc5tja : Oh, I see. I guess I switched topics rather abruptly..
19:29:58 <kc5tja> Whatever, I440r.
19:30:30 <kc5tja> Fractal: :)
19:30:44 <I440r> dont get bitchy at me kc5 dood, i said 'I cant handle it when its backwards';
19:31:06 <kc5tja> You're the one getting bitchy.  It's not our problems that you can't wrap your mind around it.
19:31:09 <I440r> isforth may or may not support a forth style assembler but if it doesnt support what i want it isnt going to have one 
19:31:11 <I440r> PERIOD
19:31:21 <ictx> i440r: that's an x86 anomaly; pdp-11 macro assembler doesn't work that way 8^)
19:31:27 <kc5tja> You don't see me using isforth, now do you?  :)
19:31:36 <I440r> kc5 your loss :)
19:31:37 <I440r> and mine
19:31:58 <ictx> i440r: what's important is you code YOUR forth the way YOU want it, anyway
19:32:18 <I440r> ictx yes but isforth isnt JUST for me
19:32:36 <ictx> guys, please don't message me, i'm using a crappy X-based irc client that pops open a new window each time you do
19:32:47 <I440r> its prime reason for existing is so that i can produce linux applications without even THINKING about c
19:33:02 <I440r> ictx what client ?
19:33:12 <ictx> zircon
19:33:14 <kc5tja> fractal: Actually, I was planning on using SLIP for TCP/IP on my laptop (since I'm coding it for that platform first).
19:33:18 <I440r> aha. i used that once heh
19:33:25 <I440r> try xchat - much nicer
19:33:45 <kc5tja> TCP/IP should be easy to implement given that channel.
19:33:52 <Fractal> kc5tja : Cool. You ever try PLIP? It's good if you have a laplink.
19:34:24 <I440r> so what brings you to #forth? - most of the well known forthers dont seem to want to hang out here :)
19:34:30 <kc5tja> No, I don't have a laplink cable.  I suppose I could make one though.  Is it a 4-bit wide path?  Both of my parallel ports are EPP, though, I *think*.  I'm pretty sure of it.  I could be wrong on the laptop though.
19:34:31 <Fractal> BitchX-1.0c17+ by panasync - OpenBSD 2.9
19:34:40 <Fractal> ^ There's a client.
19:34:59 <ictx> i was invited here by futhin, he saw me hanging out in the deadsville efnet #forth
19:35:01 <Fractal> Ya, I have no idea of the specs.
19:35:07 <I440r> aha
19:35:18 <I440r> yea. futhin owns this channel now - but its STILL mine "{
19:35:20 <I440r> :P even
19:35:25 <I440r> heh
19:35:40 <I440r> jeff fox comes by occasionally
19:35:45 <I440r> chuck moore has been in here
19:35:54 <Fractal> Yeah, I guess it would have to be 4 bits wide.
19:35:55 <kc5tja> How often does Jeff stop by now-a-days?
19:35:59 <I440r> THAT alone makes this channel worth it for me heh
19:36:05 <I440r> ive not seen him in weeks
19:36:06 <Fractal> So you'd have to break up the bytes...
19:36:26 <kc5tja> Fractal: That's cool; at roughly 2MHz strobe rate, that could give a potential speed of 8Mbps.  :)
19:36:28 <I440r> anyone ever in efnet #forth ?
19:36:34 <kc5tja> Hey, that's *ALMOST* 10-base-2 speed... :D
19:37:01 * kc5tja didn't even know efnet HAD a #forth channel.  :)
19:37:05 <Fractal> Yeah, well it's faster than a serial link, in any case.
19:37:15 * kc5tja nods
19:37:26 <kc5tja> Purchasing a long-enough cable is the only concern I have.  :)
19:37:39 <Fractal> Heh. Ya. Buying stuff sucks.
19:37:47 <I440r> kc5 i have a nice wireless serial connection here (not in use)
19:37:50 <Fractal> I just sunk $50 canadian into a sound card.
19:38:33 <Fractal> It's great, though. I can play 2 things at once.
19:38:45 <Fractal> Quake gun shots and my MP3s.
19:38:53 <I440r> ictx ur forth have a url ?
19:39:58 <ictx> whois me
19:40:38 <ictx> never mind, i forgot to set it: http://risp.org/members/jcomeau/
19:40:56 <I440r> hehe
19:41:45 <Fractal> ictx : So you're a DEC man, eh? I've got a complaint about VAX assembly.
19:42:17 <ictx> fractal: don't blame me, i didn't design it 8^)
19:42:25 <Fractal> Heh. Ok.
19:42:54 <kc5tja> I440r: No; I'd like to reduce the RF sources around my ham radio, thanks.  :)
19:43:31 <I440r> kc5 ??? eh ???
19:44:03 <I440r> ictx check out isforth.clss.net :)
19:44:09 <I440r> im readking your sources
19:44:58 <kc5tja> I440r: Unless your wireless serial port uses optical, I'm not particularly interested in using wireless technology at this time.
19:45:04 <kc5tja> In the future, I might reconsider.
19:45:42 <I440r> oh that heh yea
19:45:52 <I440r> no - its not optical, i think its uhf or something
19:46:04 <I440r> me either - thats why its not in use heh
19:48:21 <kc5tja> Hmm..UHF might be usable too...but I like the idea of using parallel port interfaces.  It's easier and more convenient for me.
19:48:58 <I440r> prolly faster too
19:49:11 * kc5tja can also make a simple parallel to serial to optical link for it quite cheaply.  With the readily available 33MHz and 50MHz clock oscillators, I should have no problems with aliasing.
19:49:27 <kc5tja> Oh it definitely is faster.
19:49:36 <kc5tja> Maximum theoretical throghput is approaching 10-base-2 speeds.
19:49:50 <kc5tja> (8Mbps ideal; probably more like 6Mpbs or so real-world)
19:55:52 --- quit: ictx ("bbl")
20:09:57 <kc5tja> Hmmm...looks like the average transmission rate on a mode 0 PLIP connection is only 300kbps or so.
20:10:13 <kc5tja> That's double serial's speed, but....I'm not sure it's worth the additional hardware expense.
20:10:31 <Fractal> Understandable...
20:11:42 <CrowKiller> buy a parallel straight trough cable and code your own drivers if the ports is epp, should be doable ant you get like 4mbps of speed
20:13:00 <Fractal> For PC to PC communication, you need a laplink, otherwise you'd have the input pins talking to the other input pins.
20:13:45 <kc5tja> Fractal: If the port is EPP, 8-bit bidirectional is supported (in fact, this is PLIP Mode 1).
20:14:19 <Fractal> Oh really? Even with just a printer cable?
20:16:19 <kc5tja> No, you need an IEEE-1248 capable cable for that mode (e.g., a real parallel cable, and not just a printer cable)
20:16:57 <kc5tja> But even so, I'm not sure if my laptop supports EPP, and I can't get into its BIOS to configure it (there is literally no keyboard combo to do this).
20:25:56 <Fractal> Compaq?
20:27:01 <kc5tja> No
20:27:10 <kc5tja> Toshiba Satellite TS-210.
20:27:23 <kc5tja> 486-era 
20:27:37 <Fractal> Oh. They may do the same thing as compaq: They put the BIOS configuration program on a partition on the HDD...
20:27:39 <kc5tja> So the Phoenix BIOS that it has is woefully inadequate.  :(
20:28:14 <kc5tja> No, I got into it once before using the keyboard, but I can't for the life of me figure out what it is, and I expired every idea I had.
20:29:37 <Fractal> Heh. Harsh. ctrl-f1? delete?
20:36:13 <kc5tja> Nothing works.
20:36:27 <Fractal> :(
20:37:44 <Soap`> Try searching for tsetup.exe on the Toshiba support site.
20:38:15 <kc5tja> Soap`: Thanks.  I'll try that when I get home.
20:38:24 <kc5tja> In the mean time, I need to hit the road.  :)
20:38:35 --- quit: kc5tja ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK")
20:57:47 <Speuler> 'nite
20:57:52 <futhin> 'nite
20:57:57 --- part: Speuler left #forth
21:46:26 --- join: kc5tja (~kc5tja@ip68-8-206-226.sd.sd.cox.net) joined #forth
21:48:04 <kc5tja> rehi
21:48:25 --- quit: Soap` (Read error: 104 (Connection reset by peer))
21:48:53 <futhin> hello kc5tja
21:50:24 <futhin> home built optical links using 33 mhz or 50 mhz clock oscillators?
21:52:05 <CrowKiller> depends on the laptop pport speed rating and/or the speed of the program that use it
21:52:53 * kc5tja knows the hardware specs for Centronics -- it can transfer up to 2MBps using the Centronics minimum possible timings.
21:53:10 <kc5tja> So I know that up to 2MBps (20Mbps) can be *theoretically* possible using full 8-bit transfers.
21:53:17 <kc5tja> (16Mbps -- sorry)
22:01:17 <CrowKiller> the appnote talks about up to 4Mbit per second IIRC
22:01:39 <CrowKiller> with a 4MHz micro running at +-4 MIPS
22:02:19 <CrowKiller> i dont remember the exact number but i've found it quite fast for such a little MHZ device
22:09:34 <kc5tja> futhin: converting the parallel port's signals into and back from a serial data stream fast enough that the 2MHz parallel data rate is maintained.
22:09:46 <kc5tja> WIth enough fudge factor to eliminate aliasing problems.
22:10:26 <futhin> 2mhz sounds slow ?
22:13:01 <kc5tja> Not when you're transferring 8 bit clumps at a time.
22:13:15 <kc5tja> 2MHz is the same as 2MBps on an 8-bit wide data path.
22:17:23 <CrowKiller> 2Mhz/8bits <=> 16MHz/1 bit (serially)
22:18:14 <kc5tja> Exactly.
22:18:32 <kc5tja> Times two to meet the Nyquist rate, so 32MHz (33MHz crystal).
22:18:50 <futhin> g'night
22:19:00 <kc5tja> Oh, and then there's the fact that there's two-way communications going on too, so really, it should be 64MHz.  :)
22:19:02 <futhin> bedtime and all
22:19:03 <kc5tja> 73 futhin
22:19:17 <kc5tja> But that can be obtained easily enough with the 32MHz crystal.
22:23:33 <CrowKiller> i just resolved the matter with dictionaries
22:23:59 <CrowKiller> i completely eleminate them, sicne all source for the system is in memory in my system, i'll search that instead!
22:24:17 <CrowKiller> LOL I wanted to build a dictionary but culd not get an elegant enough solution
22:25:14 <CrowKiller> now I have it I discarded it! Forth should be thaught by preachers acting as programmer, doing an optimization like only forth can do and tell forth is the best way to go aloud with a backing of a dancing gospel choir lol
22:25:47 <CrowKiller> Big forth partys, hurray! Rejoy becuase Forth is here! bla bla bla ;ppp
22:25:57 <futhin> good idea, a big forth party
22:26:08 <I440r> nonono
22:26:13 <I440r> not a bigforth party!
22:26:16 <futhin> lol
22:26:17 <CrowKiller> would be really a funny way to introduce forth in a videotape presentation
22:26:18 <I440r> and ISFORTH party!!!
22:26:22 <CrowKiller> the joys of forth programming
22:26:31 <kc5tja> Can I ask you something?  How is searching the source any faster than searching a linked list?
22:27:02 <CrowKiller> at least I dont have to allocate memory, and tahts waht i wanted to do
22:27:17 <kc5tja> OOOooooo Kay... :D
22:27:20 <futhin> there's a forth party in calgary!!  october 8, come to calgary and lets partay!
22:27:23 <kc5tja> : ALLOT   HERE + H ! ;
22:27:26 <kc5tja> That's all you need.  :)
22:27:33 <kc5tja> : ,   HERE ! 1 CELLS ALLOT ;
22:27:40 <kc5tja> : C,   HERE C!   1 CHARS ALLOT ;
22:27:52 * kc5tja doesn't allocate any memory in his Forth system.  :)
22:27:52 <CrowKiller> chuck stated on the cforth mailing list that its better soemtimes to use memory for speeding programs though ;p
22:28:07 <I440r> kc5 should be : , here 1 cells alot ! ;
22:28:21 <kc5tja> I440r: It's the same.
22:28:24 <I440r> tho yours is more readable
22:28:37 <I440r> it is and it isnt, you realy should allocate before you use it heh
22:28:45 <I440r> tho- thats splitting hairs realy
22:28:52 <CrowKiller> i code in asm me too
22:29:18 <kc5tja> CrowKiller: But how does searching the source code help speed up searching for words?
22:29:41 <kc5tja> You're going to be looking for words on a byte-by-byte basis.  That's SLOWER than searching a linked list.
22:30:45 <CrowKiller> with MHZ computers and this process occuring only in edit time, who cares? ;o)
22:30:57 <kc5tja> You should be caring.
22:31:12 <kc5tja> Because how can you find the compiled form of a word by searching its source?
22:31:18 <futhin> before you can gain entrance to the forth party, you will have to answer a series of trick questions that will quickly determine several things:  does the "forther" like C or OOP or FP or ANS Forth?  if true, he can't get in :(
22:31:19 <kc5tja> The source contains the *SOURCE*, not the compiled result.
22:31:35 <kc5tja> futhin: Not true.
22:31:53 <kc5tja> FP can solve problems that scaled arithmetic cannot (not without a lot more bits that is).
22:31:53 <CrowKiller> everyone should come
22:31:55 <CrowKiller> even kids
22:32:15 <kc5tja> OOP is also not strictly scorned against.  It's only its abuse that is scorned.
22:32:15 <CrowKiller> introduce them to the beauty and flamboyance of the perfect software, for everyone ;p
22:32:23 <CrowKiller> little thing, but magnified to the extreme
22:32:42 <futhin> kc5tja: are you talking specific to x86 architecture or non-specific? (about FP)
22:32:53 <CrowKiller> even knowit all gurus would enjoy, to some extent
22:32:58 <kc5tja> Non-specific.
22:33:09 <kc5tja> Floating point was designed and implemented specifically to overcome the issues of rational and scaled arithmetic.
22:33:25 <kc5tja> Namely, it's inability to simultaneously handle extremely large and extremely small numbers.
22:33:42 <kc5tja> I'm talking about plus or minus several orders of magnitudes.
22:33:56 <futhin> what's wrong with the fixed/integer floating point ?
22:34:02 <kc5tja> It's range is limited.
22:34:42 <kc5tja> There's a tradeoff involved.
22:34:53 <kc5tja> Fixed point numbers are extremely accurate, but imprecise.
22:35:02 <kc5tja> Floating point numbers are inaccurate, but very precise.
22:35:13 <kc5tja> In scientific computing, precision counts more than accuracy.
22:36:34 <futhin> hmm
22:36:55 <futhin> but floating point numbers can be simulated with fixed point computing?
22:37:17 <futhin> um
22:37:17 <futhin> nevermind
22:37:24 <futhin> besides, i really should be going to bed
22:37:24 <kc5tja> We use fixed point to implement floating point.
22:37:28 <futhin> goodnight
22:37:36 <CrowKiller> gn
22:37:38 <kc5tja> Not many people realize thata.
22:39:50 <futhin> hm?
22:39:56 <futhin> fixed point with forth
22:39:59 <futhin> or with the cpu & fpu?
22:40:23 <kc5tja> Both.
22:40:26 <kc5tja> Does it matter?
22:40:32 <kc5tja> Floating point works by realizing a simple truth:
22:40:47 <kc5tja> 123.45 == 1.2345 x 10^2
22:41:03 <kc5tja> The number is stored in integer form as 12345, along with an exponent of 2
22:41:13 <kc5tja> (the decimal point is implicit, just like it is in fixed point)
22:42:00 <kc5tja> But because the exponent is separate from the "mantissa", *ALL* of the bits in the mantissa contribute to its precision, while the exponent determines the number's order of magnitude.
22:42:45 <kc5tja> With a 32-bit mantissa and a 16-bit exponent, you can express numbers as small as -1e-65536 to +2,400,000,000e+65535
22:43:02 <kc5tja> Those are *IMMENSELY* large numbers.  They'd require a minimum of 512-bit integers to represent.
22:43:09 <kc5tja> Yet we're doing it in 48 bits.
22:44:21 <I440r> so we can take 512 bits of program data and convert them to their floating point equiv and store them in 48 bits ?
22:44:31 <kc5tja> With loss of precision yes.
22:44:44 <I440r> i.e. it would be a lossy compression method
22:44:47 <kc5tja> (BTW: I got the attributes of fixed-point and floating-point backwards; my mistake)
22:44:50 <kc5tja> Yes
22:44:53 <I440r> ok
22:45:03 <kc5tja> Floating point numbers is nothing more complicated than scientific notation in binary.
22:45:05 <I440r> would it be a good lossy compression method? heh
22:45:25 <kc5tja> That depends on how much precision you want to lose.
22:45:37 <kc5tja> If you want ultimate in precision and accuracy, use fixed point.
22:45:46 <kc5tja> Otherwise, floating point will let you express larger and smaller numbers.
22:45:49 <I440r> no compression hehehe
22:46:18 <kc5tja> No, no compression.
22:46:44 <kc5tja> Not unless you want to quantize numbers (this is why JPEG and MPEG are lossy compression; they quantize the frequency components created by the DCT algorithm).
22:46:44 <CrowKiller> I think the whole edit process is simplified by using only one kind of memory: video screens. With color acting as a token for the function of each char (a char is the minimal word) you can search consecutive entries of chars easly. If the cellwidth for the video screen was 32 bit, chuck would have made it textmode compatible, but he doesn't have.
22:47:35 <CrowKiller> so searching tokens in colorforth and completely eliminate the dictionary could be fast enough if it,s not amde at compile time
22:48:25 <CrowKiller> the editor could only "Compile" each screen individually, to improve factoring
22:49:07 <CrowKiller> my compiler is only limited by 128 userdefined words and a definite set of functions taht can reside inside 256 bytes of storage
22:49:31 <CrowKiller> so it fits perfectly for this: would compiling program ona  screen basic would be better?
22:49:40 <CrowKiller> basis*
22:53:39 <kc5tja> I don't know.  I just don't think it'll work, at least not as you expected.  But if it does, I'd be interested in seeing/reading the details of it.
22:54:06 <CrowKiller> i'm almost finish fixing the littel things that kept me from writing code
22:55:11 <I440r> ok wtf is the word r)
22:55:43 <I440r> i think its a typo and should be r>
23:00:34 <kc5tja> Just because of that, I'll have to rename R> and >R to ]R and r[ (case sensitive of course!) ... :)
23:02:57 <I440r> :)
23:03:16 <I440r> why not }r and r{ ???
23:11:18 <CrowKiller> because that's his Forth ;p
23:13:37 <kc5tja> :)
23:13:42 <kc5tja> Nahh, I'll stick with R> and >R
23:16:50 --- join: Soap` (~flop@202-0-42-22.cable.paradise.net.nz) joined #forth
23:19:50 <I440r> good idea heh
23:24:19 <I440r> nite nite
23:24:24 --- quit: I440r ()
23:55:40 --- quit: CrowKiller ("User pushed the X - because it's Xtra, baby")
23:59:59 --- log: ended forth/02.08.06