00:00:00 --- log: started forth/01.12.05
04:45:07 --- join: edrx (edrx@200.240.18.84) joined #forth
05:54:58 --- quit: edrx ("[x]chat")
09:07:07 --- join: Speuler (akhandel@banane.icafe.spacenet.de) joined #forth
09:07:11 <Speuler> hihi
09:16:30 --- join: edrx (edrx@copacabana-ttyS8.inx.com.br) joined #forth
09:30:56 --- quit: edrx ("[x]chat")
09:41:27 <Speuler> brb
10:28:44 <Speuler> back
11:12:03 --- nick: Speuler -> PhoodPhrenzy
11:42:21 --- nick: PhoodPhrenzy -> BloodDrain
11:58:56 --- quit: BloodDrain (Remote closed the connection)
13:56:50 --- join: Speuler (l@passionsfrucht.icafe.spacenet.de) joined #forth
13:56:55 <Speuler> hihi
13:57:02 <MrReach> greetings!
13:57:06 <MrReach> how are you?
13:57:12 <Speuler> just fine
13:57:15 <Speuler> good day
13:57:25 <MrReach> what's new?
13:57:32 <Speuler> clients
13:57:44 <MrReach> for the cafe'?
13:57:50 <Speuler> nope. myself
13:57:56 <MrReach> awesome!
13:58:04 <MrReach> what are you doing this time?
13:58:08 <Speuler> contacted in the cafe ...
13:58:13 <Speuler> programming
13:58:38 <Speuler> possibly forth ...
13:59:12 <Speuler> will get clearer what to use next week when we meet again
13:59:19 * MrReach nods.
13:59:32 <Speuler> those people do know forth
13:59:32 <MrReach> what can yoiu tell me about the nature of the prog?
13:59:43 <MrReach> what platform?
14:01:00 <Speuler> microcontroller
14:01:02 <Speuler> no os
14:01:22 <MrReach> ah! ok, embedded then
14:01:22 <Speuler> people are electronics
14:01:37 <Speuler> that's whz they know forth
14:01:40 <Speuler> a bit that is
14:01:50 <MrReach> what do they usually use?
14:01:58 <Speuler> vietnamese ...
14:02:18 <Speuler> c
14:02:21 <Speuler> assembler
14:02:23 <MrReach> Forth is good for that, but so few people know how to use it that other languages are often chosen, like C, Java, or Smalltalk
14:03:05 <MrReach> funny, Forth often ends up smaller than pure assembler
14:03:16 <Speuler> thez were jsut looking for spec s of good scopes
14:03:24 <MrReach> takes some extra clocks, though
14:03:25 <Speuler> stopped at lecroy
14:03:39 <MrReach> "lecroy"?
14:03:43 <Speuler> but run into problems viewing pdf files ...
14:03:49 <MrReach> OH! that reminds me ...
14:03:53 <Speuler> lecroy makes oscilloscopes
14:03:57 <Speuler> good ones
14:04:00 <Speuler> expensive too
14:04:12 <Speuler> range 10k ... 30 k$
14:04:16 <MrReach> on IA32 arch, do you think we could get docolon and unnest to 1 clock?
14:04:33 <Speuler> i don't think so
14:04:50 <MrReach> sounds like you've tried
14:04:56 <Speuler> maybe if different threading considered 
14:05:00 <Speuler> return thrading ?
14:05:36 <Speuler> thrading
14:05:36 <MrReach> heh
14:05:36 <Speuler> threading
14:05:36 <Speuler> what's wring with the 
14:05:36 <Speuler>  
14:05:36 <MrReach> do you mean STC?
14:05:36 <Speuler> e
14:05:36 <Speuler> nope.
14:05:41 <Speuler> ESP is ip
14:05:44 <Speuler> return is next
14:05:55 <Speuler> brb
14:05:58 <MrReach> oh, I get it
14:06:03 <MrReach> I think
14:06:49 <MrReach> do you mean, docol saves return adr, then CALLS token parser, when it returns, docol resets the return adr?
14:11:37 <Speuler> back
14:11:50 <MrReach> wb
14:12:33 <Speuler> i haven't grokked  retunr thrading totally yet ... brb ... 
14:12:45 <MrReach> ok
14:18:16 <Speuler> back
14:18:22 <Speuler> interesting problems today
14:18:23 <MrReach> wb
14:18:27 <Speuler> very different
14:18:35 <MrReach> oh?
14:19:04 <Speuler> not just the standard problems of how to choose a channel in irc ...
14:19:19 <Speuler> more challenging things
14:19:32 <Speuler> (this one was easy oo)
14:19:34 <Speuler> too
14:19:49 <Speuler> find a potcode of a certain town
14:19:52 <Speuler> post
14:19:58 <Speuler> sticky kyboard
14:20:00 <MrReach> heh
14:21:23 <Speuler> well, the vietnamese electronicians went for scope specs
14:21:34 <Speuler> and couldn|t view the pdf file
14:21:36 * MrReach nods.
14:21:48 <MrReach> you are programming the guts of an o-scope, then?
14:22:00 <Speuler> i applies some wizardry on the keyboard
14:22:11 <Speuler> they seem to have been impressed
14:22:47 <Speuler> when, while chatting, they found that i|m an electronician myself
14:22:56 <Speuler> knowing some programming too
14:23:02 <Speuler> they got kind of excited
14:23:11 <Speuler> swapped business cards
14:24:23 <MrReach> the new digital o-scopes are hella neat
14:24:28 <Speuler> and talked me into an appointment
14:24:43 <Speuler> the scope is just meant as tool
14:24:45 <MrReach> managing memory bandwidth is a nightmare, though
14:25:03 <Speuler> they|re into pos terminals adnd the like
14:25:11 <MrReach> oh, they don't manufacture them, were just thinking of purshasing one?
14:25:14 <MrReach> ok
14:25:20 <Speuler> multi-bank, interleaved ram helps
14:25:25 <Speuler> yes
14:25:33 <MrReach> yes, that was the solution I came up with
14:25:46 <Speuler> that's what manufacturers do
14:25:57 <MrReach> also, fifos so can use the chip's burst modes
14:26:13 <Speuler> with 2 gigasamples per second ?
14:26:36 <MrReach> yes, why not?
14:27:08 <Speuler> trigger event detection is difficult to do quickly
14:27:31 <Speuler> got to detect condition before trace is gone from memory
14:27:49 <Speuler> especially with complex triggering events
14:28:02 <MrReach> that's true, the comparator usually has a capacitor in it somewhere, which decreases bandwidth
14:28:17 <MrReach> oh, I see what you're saying
14:28:34 <Speuler> lecroy is very good with complex trigger events
14:28:43 <MrReach> at 2Ghz ... you simply can't ... you can only do simple level triggering
14:28:57 <Speuler> i've been working with a 20k$ scope, lecroy, a while ago
14:29:10 <Speuler> well, more like 30$
14:29:36 <MrReach> it's wierd to think that there are scopes that will let you look at microwave waveforms
14:29:42 <Speuler> with memory, big enough, you post-trigger
14:29:43 <MrReach> bet it's a big box
14:29:47 <Speuler> yes
14:30:26 <MrReach> the processor can't keep up with the data stream from the ADCs
14:30:48 <MrReach> oh, unless it looks at every 4th or 8th sample, or something like that
14:30:57 <Speuler> but, what it couldn't do well, was displaying the output of my poor-man "logic-analyser"
14:31:12 <Speuler> brb, new problem ...
14:31:12 <MrReach> ok
14:37:41 <Speuler> back
14:37:52 <MrReach> i'm here
14:37:55 <Speuler> this client wa looking for a job :)
14:38:07 <Speuler> logic analyser ...
14:38:13 <Speuler> 16 bit bus,
14:38:15 <MrReach> a job at the cafe'?
14:38:24 <Speuler> nope
14:38:38 <Speuler> how to look for a job on internet
14:38:45 <MrReach> ah!
14:38:56 <MrReach> a "smart" logic analyser?
14:38:56 <Speuler> using two 8-bit dacs,
14:39:12 <Speuler> one one lower 8 addresslines, other on upper 8 lines,
14:39:26 <MrReach> ok
14:39:28 <Speuler> feeding output on x/y analog scope
14:39:31 <Speuler> voila
14:39:43 <Speuler> monitor for cpu activity
14:39:49 <MrReach> lasatious figures?
14:39:58 <MrReach> ah! ok
14:40:00 <Speuler> roughly comparable
14:40:13 <Speuler> you see address space as 256 lines , 256 dtos each
14:40:27 <Speuler> to each referenced address, the beam jumps
14:40:43 <MrReach> yes, still point means no activity
14:40:59 <MrReach> moving point means proc is clocking
14:41:11 <MrReach> (probably can't see point at all when clocking)
14:41:30 <Speuler> with a little bit of practice, you can see with a glimpse waht the cpu is doing
14:41:52 <Speuler> with cheap scopes, you can see even single events
14:41:56 <Speuler> like , an interruot
14:42:11 <MrReach> heh
14:42:26 <Speuler> sequential coded draws a horiyontal line
14:42:39 <Speuler> jumping around causes some more complex image
14:42:47 <Speuler> as most code is repetitive,
14:42:50 <MrReach> yes, wouldn't THAT be an interesting plot
14:42:57 <Speuler> more or less constant shapes form
14:43:06 <MrReach> a seemingly randome set of short horizontal lines
14:43:20 <Speuler> when i finished that thing (R/2R resitor network ...)
14:43:29 <Speuler> and put it into operation,
14:44:02 <Speuler> the electronicians where very surprised to find out how much info you could extract from the display
14:44:15 <MrReach> indeed
14:44:19 <Speuler> 10 minutes later, the camera was present too
14:44:47 <Speuler> but, this thing worked only well with cheap anolog scopes
14:44:49 <MrReach> but the fancy scope didn't handle it well
14:45:03 <MrReach> it insisted on a uniform horizontal sweep?
14:45:21 <Speuler> the modern digital ones couldn't cope with the speed, cause they were sampling inputs asynchronously
14:45:34 <Speuler> where against the analog scopes displayed continously
14:45:45 <Speuler> no sweep
14:45:48 <Speuler> x/y
14:45:53 <MrReach> right
14:46:25 <Speuler> then i went into writing scope-plotting routines :)
14:46:35 <Speuler> zero and one were easy ...
14:46:35 <MrReach> heh
14:47:12 <Speuler> that gave us a minimal output device
14:47:52 <Speuler> cpu crashs looked interesting too
14:48:09 <Speuler> and, nice thing about, you could see right away when the cpu has crashed
14:48:20 <MrReach> I'll have to remember that trick
14:48:28 <MrReach> a very simple, very powerful tool
14:48:35 <Speuler> zou also could see how efficient a forth program works ...
14:49:02 <Speuler> that it was. but something one must have seen to believe it
14:49:19 <Speuler> the techs didn't expect much from it
14:49:29 <Speuler> until they saw it in operation
14:49:43 <Speuler> fromon then, they wouldn't work without
14:49:55 * MrReach nods.
14:50:00 <Speuler> for that company, i was actually programming
14:50:13 <MrReach> too cheap and easy to be w/o one
14:50:50 <Speuler> i was observing the attempts of the hardware folks to get a new baord running for a few weeks ...
14:51:01 <Speuler> until i thought i need to do something about it
14:51:10 <Speuler> took me 2 hours to build it
14:51:32 <MrReach> did you have to slow the clock on the board a bit?
14:51:37 <Speuler> nope
14:51:40 <Speuler> full speed
14:51:46 <MrReach> which was?
14:51:46 <Speuler> real-time analyser
14:52:04 <Speuler> i believe it was 10 mhy, maybe 20
14:52:14 <MrReach> ok
14:52:18 <Speuler> 20
14:52:59 <MrReach> where did you learn to do this?
14:53:21 <Speuler> i love it when i can contribute outside of my main qualification
14:53:31 <MrReach> i bet
14:53:33 <Speuler> i just thought it up
14:54:03 <Speuler> problem was: what is the cpu busy with ?
14:54:15 <Speuler> how to know ? watch the address bus
14:54:39 <Speuler> as no digital instrument for that many lines was around,
14:54:48 <Speuler> and all we had was the 2 channel analog scope,
14:54:55 <Speuler> output had to be analog ...
14:55:03 <Speuler> one leads to the next
14:55:51 <Speuler> that was for the company making the braille terminals
14:56:04 <Speuler> and my first H8 experience
14:56:15 <Speuler> which i can reuse now
14:56:39 <MrReach> you should write an article for QST or Electronic Design
14:56:53 <Speuler> i had status of wizard-of-all-trades there
14:57:19 <MrReach> you might even get it as a featured project in the ARRL Radio Amateur's Handbook
14:57:34 <Speuler> was the company i contacted by walking in with a floppy disk, containing cforth,
14:57:44 <Speuler> and showed what i could do for them
14:57:51 <MrReach> heh
14:58:26 <Speuler> 15 minutes later, i was adjunct developer for a project
14:58:37 <Speuler> task: support the developer with tools
14:58:49 <Speuler> 2 months later, the developer walked out
14:58:56 <Speuler> i took over the project
14:59:20 <Speuler> was the company too my most challenging accomplishent was realised
14:59:45 <Speuler> still looks good on my c.v.
15:00:29 <Speuler> one project was outsourced...
15:00:45 <Speuler> cpu, programmable logic, pc card
15:00:57 <Speuler> 6 motnhs later, that company dropped out
15:01:02 <Speuler> couldn't make it
15:01:32 <Speuler> returned non=functioning prototype, undocumented ...
15:01:51 <Speuler> i got in another developer ...
15:02:08 <Speuler> and us two, got it running in another 4 months
15:02:13 <MrReach> yes, I remember the episode where the company went under
15:03:07 <Speuler> i quit that company when they hired a coordinator
15:03:16 <Speuler> many people did ...
15:03:42 <Speuler> killed productivity
15:03:50 * MrReach nods
15:04:11 <Speuler> people were performing bacause it was fun before
15:04:19 <MrReach> tried to install new communications systems, when all that was needed was a bit of tweeking with current system?
15:05:11 <Speuler> the point i dropped out was when the coordinator left me in the believe that i had damaged a prototype board, while he knew he did it ...
15:05:34 <MrReach> i remember that
15:06:00 <Speuler> felt like i couldn't work with that person
15:06:17 <Speuler> he felt he couldn't work with me :)
15:06:27 * MrReach nods
15:06:36 <MrReach> how is the company doing today?
15:06:48 <Speuler> he was lacking basic understanding of technics
15:07:13 <Speuler> but tried to have it done his way
15:07:36 <Speuler> first, we only made fun from him
15:07:41 <MrReach> yes, coordinator should NOT have to know about electronics
15:08:04 <Speuler> but then they should shut up about the proper dealing with design
15:08:05 <MrReach> at the same time, coordinator should know when s/he is outside their area of expertises
15:08:31 <Speuler> think he felt like a kind of mini-god
15:08:42 <MrReach> heh, yes
15:09:05 <Speuler> pray to him, the omnipotent coordinator
15:09:16 <MrReach> is not the job of the coordinator to make sure that proper communication between groups happens, and to streamline current processes as much as possible?
15:09:35 <Speuler> that should have been his job
15:09:55 <MrReach> if I were in a position like that, I wouldn't feel very powerful at all
15:10:08 <Speuler> but he presented himself as version 2 of the allmighty
15:10:11 <MrReach> it is as like that any change I make will break what already exists
15:10:19 <MrReach> and I would be intimidated with that
15:10:34 <Speuler> maybe he overcompensated
15:10:58 <Speuler> play down his uncertainty
15:11:05 <MrReach> well, he might have had to show justification for his job
15:11:21 <Speuler> he's not with the company anymore btw
15:11:22 <MrReach> and needed to produce measurable results in the first month or something
15:11:47 <Speuler> i managed to coexist for about 8..9 months
15:11:58 <Speuler> i was not the first one to quit
15:12:13 <Speuler> but might have done better to leave earlier
15:12:49 <Speuler> the atmosphere didn't improve by waiting it out
15:12:56 <MrReach> heh, I doubt my employer would have been too happy with me the first month
15:13:22 <Speuler> people who have left have been asked by the company to return ...
15:13:30 <MrReach> I probably would have imply wandered around (seemingly aimlessly) watching people work
15:13:40 <Speuler> company has informed about me too
15:14:01 <Speuler> well, i'm too busy nowadays to consider
15:14:06 * MrReach nods.
15:14:36 <Speuler> the nick name he had was giftzwerg
15:14:51 <Speuler> means, "poisonous dwarf"
15:15:09 <MrReach> HAHAHA!
15:15:15 * MrReach laughs out loud.
15:16:20 <Speuler> i put a node on the blackboard once,
15:16:25 <Speuler> saying about:
15:17:14 <Speuler> "there are people, who know what they can't manage, and people, who manage, what they don't know"
15:18:16 <MrReach> heh
15:18:18 <Speuler> i think he guessed it was me who had put it onto the board
15:18:38 <Speuler> i'm not sure whether he understood
15:18:42 <MrReach> that doesn't translate well, but I bet it's quite elegant in German
15:18:59 <Speuler> was a dutch company
15:19:37 <MrReach> Swedish, then
15:19:56 <MrReach> wifey's home
15:20:10 <Speuler> just returned ?
15:20:14 <MrReach> yep
15:20:19 <MrReach> brb
15:20:23 <Speuler> go greet her then
15:21:41 <MrReach> back
15:21:48 <MrReach> I've got to drive to the next state
15:21:55 <MrReach> be back in a couple of hours
15:22:05 <Speuler> i'll be gone by then
15:22:19 <MrReach> yes, probably
15:22:20 <Speuler> back on friday, i guess
15:22:23 <MrReach> see you tomorrow?
15:22:26 <MrReach> ok
15:22:26 <Speuler> nope
15:22:32 <Speuler> probably not
15:22:36 <MrReach> very well
15:22:42 <MrReach> best of luck to you
15:22:48 <Speuler> you too
15:22:55 <Speuler> don"t rush
15:23:01 <Speuler> rushing kills
15:23:05 <MrReach> heh, lots of snow here
15:23:17 --- nick: MrReach -> MrGone
16:08:54 --- part: Speuler left #forth
21:45:21 --- join: edrx (edrx@200.240.18.59) joined #forth
23:36:41 --- quit: edrx ("[x]chat")
23:59:59 --- log: ended forth/01.12.05