00:00:00 --- log: started forth/04.01.13
00:01:37 * scope is still away[auto away after 60 minutes idle] (1hr 59mins 59secs)
00:03:49 <fridge> righto
00:04:02 <fridge> thanks for the informative updates on your status scope
00:47:24 --- join: qFox (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth
00:48:16 --- quit: qFox (Read error: 104 (Connection reset by peer))
00:49:34 --- join: qFox (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth
01:01:37 * scope is still away[auto away after 60 minutes idle] (2hrs 59mins 59secs)
01:08:52 --- join: Serg (~z@212.34.52.140) joined #forth
02:01:37 * scope is still away[auto away after 60 minutes idle] (3hrs 59mins 59secs)
02:01:37 --- quit: onetom (Read error: 104 (Connection reset by peer))
02:28:46 --- quit: scope ("fractal2 mirc script (ver 1.0betar1) · http://fractal2.net")
02:34:08 --- quit: fridge (Remote closed the connection)
02:54:45 --- quit: Serg ()
03:42:42 --- join: scope (~junk@njd.paradise.net.nz) joined #forth
04:34:17 --- join: rO|afk (rO_@pD9E5935E.dip.t-dialin.net) joined #forth
04:34:36 --- nick: rO|afk -> rO|
04:46:38 * scope is away[auto away after 60 minutes idle] [log- off, page- off] @ 01:42
05:48:29 --- join: fridge (~hovil@CommSecureAustPtyLtd.sb1.optus.net.au) joined #forth
05:48:52 <rO|> hi fridge
05:49:17 <fridge> hi!
05:50:06 <rO|> 16bit protected mode in 80286
05:50:29 * rO| is looking for a good pointer on that topic
06:35:42 <Robert> Why would anyone want to deal with that?
06:36:40 <fridge> http://ros.rubyforge.org/wiki/wiki.pl?BestFeatures
06:36:48 <fridge> someone want to fill in the forth entry?
06:44:13 --- join: Herkamire (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth
06:48:47 --- join: schihei (~schihei@p5085DAA0.dip.t-dialin.net) joined #forth
07:11:55 * arke is away: schooool
07:22:39 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth
07:52:45 --- join: downix (~downix@adsl-153-121-56.mia.bellsouth.net) joined #forth
08:03:43 --- quit: rO| (Read error: 60 (Operation timed out))
08:08:44 --- quit: schihei (Client Quit)
08:30:47 --- join: schihei (~schihei@pD9E5C324.dip.t-dialin.net) joined #forth
08:47:51 --- join: kc5tja (~kc5tja@66-91-231-74.san.rr.com) joined #forth
08:47:51 --- mode: ChanServ set +o kc5tja
08:55:30 --- join: Robert_ (~snofs@c-705a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth
08:58:12 --- quit: Robert (Read error: 60 (Operation timed out))
08:59:39 --- join: mur (~mur@smtp.uiah.fi) joined #forth
09:14:16 --- nick: Robert_ -> Robert
09:49:45 --- join: suprdupr (~Vapo_Rule@cnq7-86.cablevision.qc.ca) joined #forth
09:49:45 --- quit: suprdupr (Remote closed the connection)
09:53:35 --- join: rO| (rO_@pD9545AD6.dip.t-dialin.net) joined #forth
11:03:12 --- join: aktnot (ident@233.80-202-65.nextgentel.com) joined #forth
11:03:31 --- quit: aktnot (Client Quit)
11:34:19 --- quit: arke (Read error: 60 (Operation timed out))
12:12:24 --- join: gilbertdeb (~knoppix@fl-nked-ubr2-c3a-37.miamfl.adelphia.net) joined #forth
12:12:29 <gilbertdeb> hello #forth :)
12:12:36 <kc5tja> re
12:12:38 <Robert> Hey gilbert
12:12:47 <gilbertdeb> hi robert, kc.
12:12:49 <gilbertdeb> what's news?
12:13:10 <kc5tja> Kind of multitasking between FS/Forth and my ForthBox computer concept.
12:13:16 <downix> hey Sam
12:13:16 <Robert> I fried my little transciever.
12:13:24 <gilbertdeb> forthbox?
12:13:29 <kc5tja> brb
12:14:07 <mur> terve
12:15:15 <gilbertdeb> terve mur.
12:15:22 <gilbertdeb> what's your transceiver, robert?
12:15:29 <gilbertdeb> a HAM transceiver?
12:16:45 <kc5tja> back
12:16:58 <kc5tja> gilbertdeb: Yes.  It's my 65816-based computer that I've been thinking of making for some time now.
12:17:32 <kc5tja> But I'm kind of see-sawing on the video architecture.  I just realized today that the video display architecture I wanted to use with it would take 1.3 seconds to clear the screen.  :D
12:17:33 <kc5tja> Oops.  :)
12:17:45 <gilbertdeb> heheh. slow wipes ;)
12:18:00 <kc5tja> Yeah.
12:18:08 <kc5tja> Well, it's all graphical mode, 640x480, 256 colors.
12:18:19 <kc5tja> So it takes 300KB of a frame buffer to display the screen.
12:18:49 <kc5tja> The problem is, the CPU is only running at 6.25MHz, over an 8-bit data bus.
12:18:51 <kc5tja> :)
12:19:13 <gilbertdeb> bare metal forth :-?
12:19:17 <kc5tja> I also assumed 6502 timings.  It might be *slightly* faster if I used 16-bit instructions.  It ought to be roughly 0.8 seconds in 16-bit code.
12:19:35 <kc5tja> gilbertdeb: I'm assuming not even Forth; just raw, purest assembly language.
12:19:37 <Robert> gilbertdeb: Yes
12:20:03 <kc5tja> Still, at 6MHz, 0.8 seconds to clear a screen is intolerable to me.
12:21:09 <kc5tja> So, needless to say, I'm rethinking everything.
12:21:19 <gilbertdeb> you're not going to fabricate your own 6502 are you? :P
12:21:25 <kc5tja> gilbertdeb: ummmm
12:21:34 <kc5tja> no.
12:21:39 <kc5tja> That would be retarded.
12:21:42 <gilbertdeb> heheh.
12:21:57 <kc5tja> But making my own MISC is not outside of the realm of possibility or profitability.
12:22:25 <gilbertdeb> in fpga right?
12:22:31 <kc5tja> gilbertdeb: Or two CPLDs.
12:23:11 <gilbertdeb> which is cheaper?
12:23:24 <kc5tja> In terms of expense on my part, two CPLDs.
12:23:33 <kc5tja> $15 to $30 for the CPU, but $200 for the programming environment.
12:23:51 <gilbertdeb> eeek!
12:23:52 <kc5tja> versus $7 for a single FPGA, versus $2K for an FPGA's programming environment capable of handling the job.
12:24:53 <kc5tja> $230 is a lot cheaper than $2007.
12:25:00 <gilbertdeb> by most accounts :)
12:25:14 <kc5tja> (and, frankly, Xilinx quoted $2500, so the price total for me would be $2507).
12:26:17 <kc5tja> And that's not counting the cost of the EEPROM needed to boot it with in a production circuit either.  That'll cost about $5 too.
12:28:09 <kc5tja> I wonder, if I were to build a CPU in discrete components (a hobby I rather miss, actually), if I made it with asynchronous logic, how fast could I get it to run?
12:28:14 <Herkamire> what's a CPLD?
12:28:51 <gilbertdeb> programmable logic device thingamagic.
12:28:58 <kc5tja> Herkamire: Complex Programmable Logic Device -- a PLA chip on steroids.  Not nearly as cool as an FPGA, but generally much faster and cheaper.
12:30:56 <gilbertdeb> faster?
12:31:10 <gilbertdeb> kc which vendor'd you get for this?
12:31:34 <kc5tja> FPGAs are fast only when transfering data between CLBs.  Timing is a very real constraint, however, when routing signals beyond adjacent CLBs or on/off chip.
12:31:49 <kc5tja> gilbertdeb: I haven't.  I'm in no position to start buying or building anything right now.
12:31:53 <kc5tja> I'm poor, remember?
12:32:24 <gilbertdeb> yeah me too. i remember.
12:32:49 <gilbertdeb> but i think i might save up for something like this.
12:33:04 <gilbertdeb> admittedly i hadn't heard of cplds until a few lines back.
12:33:09 <kc5tja> But if I get a PLD programmer, I'm thinking I could resell services for custom chips, especially to the 6502/65816 and 8051/Z-80 communities.
12:33:19 * downix nods
12:33:30 <downix> FPGA's timing *sucks*
12:33:36 <downix> it is difficult to get timing accurate
12:34:03 <kc5tja> Like, I know that folks on 6502.org are looking for 6551 replacements.  I can't provide a pin-compatible replacement, but I could offer a chip or hybrid module that is substantially better.
12:35:01 <kc5tja> Well, its propegation delays are rather nice, but realize that each CLB represents about 2 to 3ns of propegation delay.  Then there are the fractional ns delays incurred by the interconnects.  The longer the interconnect, the longer the delay.
12:35:17 <kc5tja> (which makes sense when you think about it; long wires are good approximations for large capacitors, as far as a chip is concerned).
12:35:50 <downix> that reminds me, I've almost finished a 6581 replacement
12:35:59 <downix> I also have a CIA replacement
12:36:31 <kc5tja> That's why I'm half-way interested in building an asychronous, discrete component CPU.  I'm positive I can get it running at least into the 35 to 50 MIPS range, even with discrete components.
12:36:49 <downix> *nods*
12:36:57 <downix> If you want the CIA clone, let me know
12:37:13 <kc5tja> Nahh
12:37:21 <kc5tja> That's OK.
12:37:23 <gilbertdeb> downix how'd you make a replacement?
12:37:24 <gilbertdeb> on paper?
12:37:33 <kc5tja> CIA is nice, but I think one can do better.
12:38:24 <kc5tja> downix: Maybe you should post it to OpenCores.org.  :D
12:38:26 <downix> gilbertdeb:  in PLD
12:38:34 <gilbertdeb> heheh. downix how come you're not in #flux ?
12:38:36 <downix> kc5tja:  I'd need to clean it up a bit first.
12:38:42 <downix> what is #flux?
12:38:50 <gilbertdeb> florida linux users exchange :)
12:39:01 <downix> I'm not much of a Linux user
12:39:06 <gilbertdeb> yeah, still!
12:39:07 <downix> More a PUG guy
12:39:12 <gilbertdeb> PUG?
12:39:17 <downix> Pegasos User Group
12:39:31 <gilbertdeb> ok. but you're in miami right?
12:39:45 <downix> West Palm Beach, yes
12:39:55 <gilbertdeb> ah i see.
12:40:18 <gilbertdeb> what's pegasOS?
12:40:33 <downix> the Pegasos is a PowerPC based motherboard produced by Genesi SARL
12:40:58 <gilbertdeb> do you work for them or something?
12:41:28 <downix> Yup
12:41:40 <downix> Run Gentoo and MorphOS on mine
12:42:11 <gilbertdeb> i have a mac 6100/66 i don't suppose pegasOS can run on there, or can it?
12:42:23 <downix> Um...
12:42:26 <downix> dude
12:42:27 <kc5tja> gilbertdeb: Pegasos is the name of the motherboard.
12:42:30 <downix> it's a MOTHERBOARD
12:42:35 <kc5tja> gilbertdeb: MorphOS is the operating system.  :)
12:42:38 <gilbertdeb> aaah!
12:42:43 <gilbertdeb> <--- does the duh thing.
12:42:48 <gilbertdeb> i read it as PegasOS.
12:42:59 <downix> it's the Pegasos
12:43:05 <downix> http://www.pegasosppc.com
12:43:15 <kc5tja> I still do too.  I think it's a stupid name.  They could have done a better job with its spelling.  :)
12:43:15 <gilbertdeb> i get it now. pegasos not pegas OweEss
12:43:22 <downix> We're in the second revision of the motherboard
12:43:34 <downix> kc5tja:  the spelling is intentional, actually.  It's greek.
12:43:41 <gilbertdeb> pegasus?
12:43:43 <downix> same with Genesi itself
12:43:46 <gilbertdeb> isn't that a horsey?
12:43:54 <downix> Pegasus is the Romanized form of Pegasos
12:44:22 <kc5tja> gilbertdeb: With wings even.
12:45:05 <kc5tja> downix: Are these boards actually shipping, right now?
12:45:09 <downix> Yup
12:45:09 <gilbertdeb> but given that you're in the computer industry, and OS has a very specific meaning ...
12:45:12 <kc5tja> If so, how much?
12:45:15 <gilbertdeb> n e way. this looks cool.
12:45:23 <downix> 299 euro
12:45:35 <downix> (which means it sucks in the US due to the weak dollar)
12:45:55 <gilbertdeb> downix is MorphOS free?
12:46:04 <downix> it's free with the Pegasos
12:46:40 <kc5tja> downix: I might still consider one.  It's **MUCH** cheaper than an AmigaOne board.
12:46:44 <kc5tja> Does it have on-board video?
12:46:51 <gilbertdeb> hmmm how much is it?
12:46:52 <downix> No on-board video
12:46:59 <downix> gilbert:  299 euro
12:47:03 <gilbertdeb> ah yes. i saw that.
12:47:22 <kc5tja> NEAT!  STAC 9766 codec chip!  Hifn is a STAC spinoff, and we had quite a line of 97xx series chips.  :D
12:47:25 <gilbertdeb> and this is being sold/made in wpb?
12:47:47 <downix> No, it's being sold/made in Germany
12:47:49 <downix> I telecommute
12:48:04 <kc5tja> downix: Is this the telecommute job you were asking about the other day?
12:48:04 <downix> kc5tja:  I love the sound chip other than 1 thing, you need an amp for it
12:48:20 <kc5tja> I listen to sound via headphones anyway.
12:48:27 <downix> Not quite, one of our partners needs a coder to update/port their software.
12:49:00 <kc5tja> To Linux, or to MorphOS?
12:49:09 <kc5tja> Oh wait, you did say to Linux.
12:49:13 <downix> *nods*
12:49:16 <kc5tja> But I refuse to do Windows coding.  Sorry.
12:49:22 <downix> Understandable
12:49:26 <kc5tja> I'll work at In-N-Out before I code for XP.
12:49:30 <downix> heh
12:49:53 * kc5tja notes Pegasos has Open Firmware on it.  :)
12:50:02 <downix> It's 99% CHRP sped
12:50:05 <downix> erm, spec
12:50:06 <gilbertdeb> downix so you're their forthy guy?
12:50:10 <downix> it's missing 1 component
12:50:19 <downix> gilbert:  Actually I'm in marketing.
12:50:26 <gilbertdeb> :O
12:50:30 <gilbertdeb> cpld/marketing
12:50:30 <kc5tja> So I have a question: I don't see any AGP ports.  Are folks still making PCI video cards?
12:50:32 <gilbertdeb> hmmm. i see.
12:50:40 <downix> It has AGP 2.0
12:50:57 <kc5tja> That's only 2x, right?
12:51:08 <downix> no, 2.0 covers 1x, 2x, and 4x
12:51:19 <downix> and 8x cards work on it, but won't run at-speed
12:51:34 <kc5tja> What kind of power consumption does it have?
12:51:37 <downix> (actually, some 8x cards perform better with the Pegasos than they do on "real" 8x ports)
12:51:57 <downix> The G3 consumes 11W's total, G4's closer to 17 total
12:52:11 <kc5tja> What clock speeds?
12:52:52 <downix> right now it's G3-600 G4-1Ghz, but we support up to 2Ghz for either
12:52:57 <downix> (pity noone makes them yet)
12:53:07 <downix> Also, allows dual-CPU
12:53:13 <downix> (for G3 or G4)
12:53:19 <kc5tja> Any plans for G5s?
12:53:42 <downix> Yes, that'll require a new motherboard tho
12:53:51 <downix> rev 3 is slated for 970 support
12:53:56 <kc5tja> That's fine.  I don't swap CPUs without swapping a mobo as well.
12:54:10 <kc5tja> Especially if the mobo/CPU combination is cheap.
12:54:22 <kc5tja> And this is, bar none, the single cheapest PPC solution I've *ever* seen.
12:54:22 <downix> *nods*
12:54:26 <kc5tja> It's almost too good to be true.
12:54:31 <kc5tja> How are they making their money?
12:54:42 <downix> We make good profit from them actually
12:54:48 <downix> tells you how much Apple gouges
12:55:04 <gilbertdeb> downix, how come this isn't /. ed yet?
12:55:06 <gilbertdeb> :)
12:55:14 <downix> it's been on /. several times
12:55:20 <gilbertdeb> i must have missed it.
12:55:27 <gilbertdeb> so can you install osX on it as well?
12:55:35 <downix> Not directly
12:55:43 <downix> you can run it using Mac on Linux, however
12:55:49 <downix> kind of a VMWare for PowerPC
12:55:55 <gilbertdeb> i see.
12:56:09 <downix> Apple has a dongle, can't by-pass it legally
12:56:22 <gilbertdeb> dongle? where?
12:56:26 <downix> in the BIOS
12:56:30 <gilbertdeb> oh i see.
12:56:41 <gilbertdeb> do you know anything about the 6100/66 bios es?
12:56:45 <gilbertdeb> the ancient machine thingies?
12:56:55 <downix> They're too ancient for much use
12:56:59 <downix> they're the NuBus boxes, right?
12:57:05 <gilbertdeb> yeah .
12:57:16 <gilbertdeb> that's 10 bucks gone down the drain.
12:57:21 <gilbertdeb> a whole pizza with nothing.
12:57:32 <downix> Actually the 6100 is not a bad machine
12:57:41 <downix> I've given them as xmas gifts before
12:57:48 <gilbertdeb> heheh. how recently?
12:57:53 <downix> last xmas
12:58:02 <gilbertdeb> to who?
12:58:09 <downix> my wifes cousin
12:58:09 <gilbertdeb> did they throw it back at you?
12:58:15 <downix> and no, she loved it
12:58:20 <downix> her own computer
12:58:23 <gilbertdeb> i see.
12:58:26 <downix> (she's also only 14)
12:58:46 <gilbertdeb> i hope she takes it apart.
12:59:16 <downix> kc5tja:  I'll be honest, we rate our AGP at 1x speed, but that's because we rated it by constant, reliable speed over peak performance.
12:59:24 <gilbertdeb> speaking of which ... why on earth is the ibook's harddisk so HARD to get to?
13:00:36 <fridge> because they want most people to put it in the 'too hard' category
13:00:42 <fridge> and pay apple technicians to access it
13:00:49 <gilbertdeb> yeah. that.
13:01:05 <gilbertdeb> it'll cost too much for me.
13:01:17 <kc5tja> PC architecture, however, prefers wide-open hardware access, but impossible-to-manage software.
13:01:22 <downix> *nods*
13:01:22 <gilbertdeb> my middle name is: Impoverished.
13:01:23 <kc5tja> Thus creating an equally false economy.
13:01:53 <downix> We've been working on bugs in the Peg 2, but nothing as bad as those of the Peg 1
13:02:09 <downix> we had to pull the first revision pegs off of the market due to problems with the northbridge
13:02:31 <downix> the issues with the peg 2 have been little more than firmware bugs
13:02:39 <downix> (so a quick reflash and viola)
13:04:38 <kc5tja> Peg2 being the G4 boards?
13:05:08 <kc5tja> Here's another question for you: any damned CPU fans?
13:05:20 <downix> G3 or G4, Peg 2 just means revision of the motherboard
13:05:37 <downix> We have a CPU fan on the G4, but not because it really needs one
13:05:51 <kc5tja> It must be there for a reason.  
13:05:57 <downix> Cheaper
13:06:03 <kc5tja> All I know is that my current computer, an 800MHz Athlon, sounds like it's about to take off.
13:06:04 --- nick: proteusguy -> proteus-away
13:06:11 <downix> the heatsink for it would run $8, the fan ran $4
13:06:29 <kc5tja> I would personally pay the extra $4 for the silence of a good quality heatsink.
13:06:49 <downix> A VGA or northbridge cooler is a drop-in replacement
13:07:04 <kc5tja> I don't think I'd mind a fan if they didn't make so much damned noise.  >:(
13:07:21 <downix> I did not notice any noise
13:07:26 <downix> the PSU was louder
13:07:27 <downix> 8)
13:07:32 --- quit: fridge ("Leaving")
13:07:36 <kc5tja> I'd rather have a huge heatsink and a slow fan than a small heatsink and a fast-spinning fan.
13:07:40 <gilbertdeb> i'll be back in 10.
13:07:41 --- quit: gilbertdeb ("Leaving")
13:07:59 <kc5tja> Yes, well, that's another thing that simply needs to go.
13:08:06 <downix> *nods*
13:08:08 <kc5tja> My ham radio's power supply is nice.
13:08:11 <kc5tja> VERY quiet.
13:08:18 <downix> only on the G4's, and it is rapid to replace
13:08:19 <kc5tja> And its fan kicks in only when it's absolutely needed, not all the time.
13:08:27 <downix> not held on by complex clips and paste
13:08:37 <downix> just a pair of mounting points at the corner
13:08:41 <downix> literally, a northbridge cooler
13:10:57 <kc5tja> Was that C= 299 for the G4 1GHz board?
13:11:06 <downix> No, the G3-600Mhz
13:11:10 <downix> it's 499 for the G4
13:11:27 <kc5tja> In US money that's about $600.
13:11:41 <downix> I know, I hate how the US$'s gone in the crapper
13:11:47 <kc5tja> I think it's great.
13:11:51 <downix> hrm?
13:12:01 <downix> paying higher prices for everything is great?
13:12:03 <kc5tja> It's about time this country gets a swift kick in the ass, that we're not the only folks on the face of the planet.
13:12:09 <downix> ah, good point
13:12:20 <kc5tja> And maybe bring HOME more business instead of going OUT all the time.
13:12:25 <downix> *nods*
13:12:27 <kc5tja> But I digress.
13:12:42 <kc5tja> So, $600 for the mobo, less RAM, but with CPU.
13:12:54 <downix> Right
13:12:55 <kc5tja> $100 for RAM brings it to $700.
13:13:00 <kc5tja> Maybe $750.
13:13:13 <downix> uses any PC2100 RAM, most PC2700 works as well
13:13:17 <kc5tja> A good harddrive will cost about $100 too, so let's say $850.
13:13:32 <downix> can use any ATX or MicroATX case
13:13:37 <kc5tja> A decent video card will also cost around $100, so $950.
13:13:39 <downix> don't need a "P4 ready" PSU
13:14:18 <kc5tja> Except for the clock rate, it's competitive with PCs.
13:14:23 <downix> *nods*
13:14:33 <downix> Actually for thuroughput and latency, it beats PC's
13:14:45 <kc5tja> They need to get that clock rate up to make sales drones notice though.
13:14:50 <Herkamire> G4 1MHz you're talking about right?
13:14:52 <kc5tja> sure, RISC is like that.
13:14:55 * downix nods
13:14:57 <kc5tja> Herkamire: Yeah.
13:15:13 <kc5tja> G4 1GHz, BTW.  :)
13:15:19 <downix> I ment the northbridge, it's got a crossbar capable of the hundreds of billions of bits per second.
13:15:41 <downix> which, in the dual-CPU config is *very* nice
13:15:42 <downix> 8)
13:15:42 <kc5tja> downix: I care more about CPU power than NorthBridge power at this point.
13:15:52 <downix> *nods*
13:15:59 <kc5tja> I don't intend on doing anything that would even remotely tax the NorthBridge.
13:16:13 <downix> basically means the CPU power is being stolen due to a crappy northbridge design like you'll find on a Mac.
13:16:21 <Herkamire> downix: how much more us$ for a dual G4 setup?
13:16:25 <downix> erm, s/is/isn't
13:17:00 <downix> Herkamire:  We haven't made the dual G4's for sale yet.  But anticipated price if we do is $200 over the single CPU
13:17:10 <downix> we,, 200 euro
13:17:18 --- quit: schihei (Client Quit)
13:17:27 <Herkamire> sounds like great stuff
13:17:50 <downix> but not much need for the dual if we can instead push development of the G5-capable board instead
13:17:57 <downix> (which also would allow dual)
13:18:04 <Herkamire> could I fit this stuff in a 1RU case?
13:18:11 <downix> 1U no, 2U yes
13:18:18 <downix> the CPU card is just a hair too tall for the 1U
13:18:30 <kc5tja> Herkamire: The motherboards are ATX form-factor compliant.
13:18:41 <Herkamire> what's ATX?
13:18:55 <downix> kc5tja:  You can fit them into a 2U ATX case, however.  I have mine in one.
13:19:01 <kc5tja> Herkamire: The spec that all PC motherboards are made to fit.
13:19:05 <downix> right next to my A3000 rackmount.
13:19:06 <downix> 8)
13:19:12 <kc5tja> Herkamire: Lets you fit any motherboard into any ATX-compliant case.
13:19:21 --- join: gilbertdeb (~knoppix@fl-nked-ubr2-c3a-37.miamfl.adelphia.net) joined #forth
13:20:00 <Herkamire> I see.
13:20:01 <downix> the Peg is a microATX, it's smaller than a normal ATX
13:20:03 <gilbertdeb> kc have you heard of the bootable lispOS thing?
13:20:06 <kc5tja> downix: I no longer have any need or desire to purchase an Amiga 3000 or 4000.  Modern CPUs are so fast that they can actually emulate them faster than the real hardware would permit.
13:20:20 <downix> kc5tja:  Mines got a PPC in it tho.
13:20:22 <kc5tja> downix: It's still an ATX-compliant board though.  :)
13:20:26 <downix> true
13:20:28 <Herkamire> I've never done x86 hardware stuff. (except installing a harddrive once)
13:20:39 <gilbertdeb> http://www.cs.uit.no/~frodef/los0/ <---
13:21:02 <gilbertdeb> Herkamire, would you know how to replace an ibook's harddisk?
13:21:02 <kc5tja> gilbertdeb: No, I've known several Lisp-based OSes have existed in the past, but none have been commercially successful for long.
13:21:05 <Herkamire> (which was a pain in the ass because the IDE cable that the box came with wasn't long enough to reach the HD bay)
13:21:21 <gilbertdeb> kc5tja, this is a bootable one. amateur not commercial.
13:21:24 <downix> kc5tja:  it just runs my flyer tho.
13:22:11 <kc5tja> Herkamire: I no longer use those flat IDE cables.  I now prefer the bundled cables, which are more expensive, but better all-around.
13:22:27 <kc5tja> And, frankly, serial ATA is coming down in price VERY fast now-a-days.
13:22:29 --- join: Nutssh (~Foo@dunwlessnat.rice.edu) joined #forth
13:22:33 * kc5tja was shocked.
13:22:36 <downix> For the Peg 3 we're thinking also a wholesale conversion to SATA
13:22:47 <gilbertdeb> spam alert: 
13:22:50 <gilbertdeb> The Movitz system aspires to be an implementation of ANSI Common Lisp that targets the ubiquitous x86 PC architecture "on the metal". That is, running without any operating system or other form of software environment. Movitz is a development platform for operating system kernels, embedded, and single-purpose applications. There can potentially be several completely different operating systems built using Movitz. 
13:23:16 <downix> we *almost* did it for the Peg 2, but elected against it due to the high cost of SATA drives.
13:24:12 <gilbertdeb> dedicated machines :)
13:24:13 <kc5tja> Well, I think that parallel IDE ports and drives are going to last a long, long time to come.
13:24:19 <downix> *nods*
13:24:41 <kc5tja> I think a wholesale conversion wouldn't be the best idea, especially since SATA gives nothing over SCSI.
13:24:43 <gilbertdeb> downix so who in the US are buying your solutions?
13:24:48 <downix> The Peg 3 would have the two ATA as the Peg 2 does, but also have 4 SATA ports.
13:24:53 <kc5tja> er, serial SCSI rather (which, today, basically means Firewire)
13:25:10 <Herkamire> I was just pissed that the cable that came with it didn't reach the only place in the box where you could fit a HD
13:25:15 <downix> gilbertdeb:  usually developers, but also sold a pile to ShopIP (one of Woz's companies)
13:25:44 <gilbertdeb> downix, do they usually buy it with morphOS?
13:25:53 <gilbertdeb> or are they opting for linux?
13:25:55 <Herkamire> I had to rip some metal out of a devider so I could get the cable to go diagonaly stright for the HD
13:26:02 <downix> gilbert:  you get MorphOS and Debian linux for free
13:26:09 <gilbertdeb> cool!
13:26:15 <downix> some folk also run OpenBSD
13:26:24 <gilbertdeb> hmmm.
13:27:15 <downix> and QNX is *almost* ready for it
13:27:34 <gilbertdeb> there's no chance morphOS would run on the 6100 i take it?
13:28:05 <kc5tja> I doubt it.
13:28:12 <downix> No
13:28:13 <kc5tja> If you don't have OpenFirmware, it's unlikely to run.
13:28:27 <downix> MorphOS runs only on Openfirmware machines or old AMiga's with PPC CPU's
13:28:29 <gilbertdeb> i might as well toss the box out.
13:29:15 --- quit: Nutssh ("Client exiting")
13:29:41 --- nick: proteus-away -> proteusguy
13:29:45 <kc5tja> gilbertdeb: Can't you get Linux running on it?
13:29:46 <downix> I collect old computers, send it my way
13:29:47 <downix> 8)
13:30:05 <gilbertdeb> kc nah, i have enough linux boxes around.
13:30:12 <kc5tja> I really dislike it when people threaten to throw computers out.  Computers are rarely so useless as to just warrent a blatant discard.
13:30:21 <gilbertdeb> heheh. downix what do you have so far?
13:30:28 <downix> let's see
13:30:32 <downix> oldest to youngest
13:30:38 <downix> DEC PDP-8
13:30:42 <downix> Tiger Craco
13:30:48 <downix> Atari 600XL
13:30:49 <gilbertdeb> kc the old machines would be a lot more useful if apple exposed the innards to one and all to see.
13:30:50 <downix> Commodore VIC-20
13:30:55 <gilbertdeb> it makes it a lot more hackable.
13:30:58 <downix> Magnavox Oddyssey 2
13:31:01 <downix> Commodore 64
13:31:03 <downix> TI-99/a
13:31:09 <gilbertdeb> you have a dec pdp-8???
13:31:13 <gilbertdeb> does it still run?
13:31:15 <downix> yup
13:31:20 <downix> and yup
13:31:20 <gilbertdeb> what's on it?
13:31:21 <kc5tja> gilbertdeb: Well, the Linux sources ought to be able to tell you something about how the boot system works, and how to touch the hardware in useful ways...
13:31:36 <downix> for OS, old Multics I believe
13:31:41 <gilbertdeb> really???
13:31:45 <downix> anyways, to continue
13:31:46 <gilbertdeb> does multics come with sources???
13:31:48 <downix> Amiga 1000
13:31:50 <downix> Amiga 500
13:32:02 <downix> Atari Jaguar
13:32:09 <downix> Amiga 1200
13:32:13 <downix> and that's about it so far
13:32:18 <kc5tja> Your Amiga 3000?
13:32:20 <gilbertdeb> downix, does multix come with sources?
13:32:58 <downix> No
13:33:04 <downix> Oh, right, the 3000
13:33:11 <downix> I don't consider machines I run everyday
13:33:24 <gilbertdeb> hahah.
13:33:35 <kc5tja> You did say to the youngest.  :)
13:33:37 <gilbertdeb> the only exotic box i have is an sgi i2 r10k ip28.
13:33:45 <downix> which is why the Alpha and VAX are also not on the list
13:33:47 <downix> oooo
13:33:50 <downix> I want an SGI
13:34:01 <gilbertdeb> all they seem to do is run irix :(
13:34:13 <downix> So?
13:34:16 <gilbertdeb> some run openbsd/netbsd/linux but ... mine defiantly runs nothing but irix.
13:34:18 <downix> I want it for the collection
13:34:28 <gilbertdeb> ebay has lots of them
13:34:37 <downix> yeah, but $80 for shipping?
13:34:46 <gilbertdeb> yeah that could be annoying.
13:34:46 <downix> I buy from local throwouts
13:35:04 <kc5tja> I'm curious to find out who would purchase a ForthBox kit or pre-assembled computer, and for how much...
13:35:28 <gilbertdeb> marketing research, kc5tja ;)
13:35:38 <kc5tja> gilbertdeb: What do you think I'm doing right now?  :)
13:35:39 <gilbertdeb> you need to throw out the word kit, and let in 'solution'
13:35:41 <gilbertdeb> it is magic.
13:36:13 <kc5tja> gilbertdeb: No, because I don't have the production facilities to resell a ton of pre-fabricated boards.
13:36:16 <kc5tja> That is $$$$$$$$.
13:36:22 <kc5tja> And $$$$$ is something I just don't have.
13:36:29 <gilbertdeb> yeah but your local bank does.
13:36:37 <kc5tja> gilbertdeb: I have less than $1000 in the bank.
13:36:38 <kc5tja> No credit.
13:36:40 <gilbertdeb> well, they can be annoying to deal with so scratch that.
13:36:43 <kc5tja> You tell me how I'm going to get a loan?
13:36:48 <kc5tja> And how that loan is going to cover costs?
13:36:58 <gilbertdeb> hmmm. you need investors.
13:37:07 <gilbertdeb> angel investors. people who simply have utter faith in you.
13:37:25 <kc5tja> And who would invest in a lousy 5MHz to 7MHz 8-bit computer with no expansion slots, running FORTH of all things?  :)
13:37:32 <gilbertdeb> hahahaha
13:37:38 <gilbertdeb> c.l.f 
13:37:45 <gilbertdeb> but then again, this is forth we're talking about.
13:37:46 <downix> Actually
13:37:53 <gilbertdeb> they'll go make their own damned forthBox 
13:37:55 <kc5tja> Seriously, I'd get more interest from the CPU than the whole box.
13:38:47 <gilbertdeb> forth's been dying the death of a thousand cuts for so long ...
13:38:59 <kc5tja> Forth can't be commercialized.
13:39:02 <kc5tja> It's too simple.
13:39:11 <gilbertdeb> forth.com 's tried.
13:39:18 <gilbertdeb> so have the openfirmware folks.
13:39:25 <kc5tja> By its very definition, it is so simple that it is a commodity item as soon as you think of it.
13:39:43 <kc5tja> Forth, Inc. has been wildly successful because they target *very* specific niche markets.
13:39:53 <kc5tja> They also have a world-class sales force.
13:39:54 <gilbertdeb> they're profitable?
13:40:05 <kc5tja> gilbertdeb: Dude, they've been in business since 1973.  :)
13:40:13 <gilbertdeb> heheheh. yeah
13:40:18 <gilbertdeb> i needn't have asked.
13:40:21 <kc5tja> :)
13:40:33 <gilbertdeb> they've outlived what? 350,623 .com's?
13:40:37 <gilbertdeb> that's longevity.
13:40:39 <kc5tja> At least.
13:40:53 <kc5tja> But here's the other thing too:
13:40:57 <kc5tja> What would be my business plan?
13:41:04 <tathi> I was under the impression that openfirmware is being fairly successful as well...
13:41:13 <gilbertdeb> you'd want to compete with forth.com i'd presume.
13:41:16 <kc5tja> "My goal is to make a PC that I'm happy with.  It needs to be very quiet, low in power, and capable of text editing and Internet access."
13:41:24 <gilbertdeb> hahahaha
13:41:31 <gilbertdeb> a wind up computer?
13:41:33 <kc5tja> Boohoo.  The bankers and investors would look at that and laugh so hard they'd need to change their underwear.
13:41:48 <kc5tja> tathi: Every MacOS X box has a copy.  :)
13:42:00 <gilbertdeb> every sun box (i think) too.
13:42:15 <kc5tja> Every RS/6000 box since the PowerPC too.
13:42:20 <gilbertdeb> really?
13:42:23 <gilbertdeb> >:)
13:42:37 <kc5tja> The CHRP motherboard basically is an RS/6000 motherboard.
13:44:00 --- nick: gilbertdeb -> debtreblig
13:44:06 <downix> the only difference between CHRP and the Pegasos is the lack of an OpenPIC controller
13:44:09 <downix> noone makes them anymore
13:44:28 <debtreblig> i did impress this one ibook user once by revealing the forth prompt.
13:44:30 <debtreblig> she was a she :D
13:44:43 <debtreblig> but she wasn't sufficiently impressed to stay long enough :(
13:44:47 --- nick: debtreblig -> gilbertdeb
13:45:39 <gilbertdeb> downix has anyone put your mamaboards in a laptop yet?
13:46:05 <downix> No, it's a teensy too big for that.
13:46:18 <downix> but it could fit into a slightly bigger than normal laptop
13:46:24 <gilbertdeb> you're not gonna shrink it a tad?
13:46:35 <gilbertdeb> what? the lunchboxes of old?
13:46:40 <downix> Not that big
13:46:42 <gilbertdeb> an Osborne 1 :)
13:46:46 <downix> heh
13:46:48 <kc5tja> downix: You know, if they could get away from that CPU slot, they could get higher clock rates.  Slots add a metric buttload of capacitance to a CPU bus.
13:46:50 <gilbertdeb> lugger.
13:47:12 <downix> kc5tja:  PPC CPU's don't come in easy-to-use packages, preferring being surface-mounted.
13:47:23 <kc5tja> downix: Yes, I know.
13:47:31 <kc5tja> That's why they're able to run so dang fast.
13:47:32 <kc5tja> :)
13:47:34 <downix> we got less capacitance from the slot than with the ZIF PPC's
13:47:39 * kc5tja notes that CPLDs and FPGAs are the same way.
13:48:29 <downix> *nods*
13:48:39 --- join: fridge (~fridge@dsl-203-33-164-15.NSW.netspace.net.au) joined #forth
13:48:50 <downix> we had a ZIF arrangement awhile back, the slot ended up being the better route
13:48:58 <kc5tja> With the MISC concept, I was thinking of making the ForthBox mainboard itself just a passive backplane for four CPLDs, each on its own plug-in card.
13:49:50 <kc5tja> Damn it, people need to start using more asynchronous logic!!
13:50:01 <downix> lol
13:50:05 <kc5tja> I'm serious.
13:50:11 <downix> I know you are
13:50:17 <kc5tja> Asynchronous buses are the best thing that ever happened to computers.
13:50:25 <gilbertdeb> why? is it easy to work with?
13:50:47 <kc5tja> I love how that, "Asynchronous logic is once again being researched, as synchronous systems are starting to show critical faults as technology improves."
13:50:54 <kc5tja> gilbertdeb: Harder to design, MUCH easier to work with.
13:50:58 <downix> gilbert:  better end performance.
13:51:07 <kc5tja> And frankly, it's not harder to design once you wrap your head around it.
13:51:41 <kc5tja> But it does take more logic to implement.
13:52:08 <downix> Eddas uses asynch, so I know
13:52:49 <kc5tja> But now that we're getting into 1GHz front-side buses and all, we're looking at printed circuit board trace lengths that are so long, even at only a few inches, that you can actually put several bus transactions on a single wire before the first one will respond with data.
13:53:14 * kc5tja notes that the PowerPC has what's called a "split address and data bus" to take advantage of this, which is another reason PowerPCs blow the snot out of x86s.
13:53:42 <fridge> mmm snot
13:54:53 <kc5tja> Asynchronous buses would help mitigate this.
13:54:56 <downix> *nods*
13:55:43 <kc5tja> Moreover, asynchronous logic would be even better -- self-timed circuitry always tends to run at the fastest possible speed, dissipates MMUUCCHH less power (because a chip's clock skew control circuitry and distribution nets aren't needed), and has less RF emissions.
13:56:07 <kc5tja> All this, despite roughly 50% more transistors for the same level of functionality.
13:57:02 <kc5tja> If you want to make things go faster, pack them in tighter.  Oh, and the bus will always run at the right speed, because addressing a fast memory will, well, go fast, while a slow device will run slow.
13:57:41 <kc5tja> Instruction timings won't be measured in nebulous concepts like "cycles" anymore (especially when you get into superscalar architectures.  Case in point: notice how Intel nor AMD document cycle counts for their instructions anymore.  Even THEY have no idea how long it takes to run an instruction!).
13:57:56 <kc5tja> Instead, they'll be measured in nanoseconds, which is a substantially more useful metric.
13:58:31 <gilbertdeb> so who shall bell the cat?
13:58:55 <kc5tja> What?
13:59:04 <gilbertdeb> who is going to implement this stuff ^
13:59:09 <kc5tja> Whoever.
13:59:14 <gilbertdeb> ^
13:59:14 <kc5tja> Sun is actively looking into it.
13:59:21 <gilbertdeb> you should look into it too.
13:59:24 <kc5tja> Of course.
13:59:34 <kc5tja> I've been a proponent of asynchronous lgoic for years and years.
13:59:43 <kc5tja> But in programmable logic, your density goes way down. :(
14:00:42 <kc5tja> Programmable logic chips are really optimized for synchronous systems.
14:00:57 <kc5tja> FPGAs are the best you can use for asynch logic, because they have the most passive logic on them.
14:01:15 <kc5tja> I've been toying with the concept of cross-coupled RAM chips to implement programmable logic too.  :)
14:01:51 <kc5tja> where the data bus of one feeds the address pins of another, the data bus of THAT chip feeds back into the address bus of the original, etc.
14:02:09 --- join: Nutssh (~Foo@dunwlessnat.rice.edu) joined #forth
14:02:39 <kc5tja> That would give me about 15ns propegation delay for any arbitrary logic, depending on the simplicity of the circuit being programmed.
14:03:07 <kc5tja> Well, 30ns if you have two cross-coupled RAM chips.  But the cost goes down as the interconnects go up, because more work is being done at each stage.
14:09:02 <gilbertdeb> Zzzzz
14:09:06 --- quit: gilbertdeb ("Leaving")
14:28:52 --- quit: scope ("fractal2 mirc script (ver 1.0betar1) · http://fractal2.net")
14:34:04 --- part: madgarden left #forth
14:43:26 <kc5tja> I'll be back in an hour I hope...
14:43:31 --- quit: kc5tja ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK")
14:54:25 --- quit: tathi (""removing secondary HD"")
15:01:09 --- join: arke (~chris@wbar8.lax1-4-11-099-104.dsl-verizon.net) joined #forth
15:20:31 --- quit: qFox (Read error: 104 (Connection reset by peer))
15:28:44 * arke is back (gone 00:27:42)
15:51:43 --- join: Sonarman (~matt@adsl-64-160-165-192.dsl.snfc21.pacbell.net) joined #forth
15:56:33 --- join: scope (~junk@njd.paradise.net.nz) joined #forth
16:27:01 --- quit: downix (Read error: 54 (Connection reset by peer))
16:52:48 --- join: kc5tja (~kc5tja@66-91-231-74.san.rr.com) joined #forth
16:52:48 --- mode: ChanServ set +o kc5tja
16:53:15 <kc5tja> HOLY COW.  I might actually have a real, honest to goodness paying gig for FalvoTech.
16:54:16 --- join: dubious (~marc@209.71.234.197) joined #forth
16:58:29 <ayrnieu> kc5 - doing what?
16:58:36 <kc5tja> I can't say precisely.
16:58:49 <kc5tja> It's for a new product that is to be sold in the future.
16:59:02 <kc5tja> But it's embedded C++ coding for linux.
16:59:06 <ayrnieu> kc5 - good idea.  Your employers will never find themselves at a loss to know what you might've said on #Forth
16:59:09 <ayrnieu> oh, nifty.
16:59:18 <dubious> how does forth go about executing what it interprets? does each statement map to symbols in the executable or are statements converted to binary and executed?
16:59:25 <ayrnieu> I mean, the people selling you the contract.
16:59:37 <kc5tja> ayrnieu: Well, it is logged.
16:59:42 <kc5tja> And the logs are publicly available online.
16:59:50 <ayrnieu> dubious - Forth does not demand either method.
17:00:02 <ayrnieu> kc5 - er, I know.
17:00:34 <dubious> ayrnieu: aha, I suspected that. where can I read about the pros and cons of either approach, which are probably the only two approaches (I hope to be wrong on that, would like to learn more about this)
17:00:38 <ayrnieu> kc5 - I think you caught my tone well enough, though =)
17:00:41 <arke> hiya
17:00:44 <arke> kc5tja: hi
17:00:58 <kc5tja> ayrnieu: Actually, I thought you were being facetous.  :)  But true enough.
17:01:02 * kc5tja is actually rather excited.
17:01:05 <arke> kc5tja: I HATE BEANER BITCHES!! THEY CAN'T DRIVE WORTH SHIT ^___^ seriosuly, though...
17:01:05 <kc5tja> It'll start off slow to start with.
17:01:07 <ayrnieu> dubious - why do you hope to have spoken in error?
17:01:24 <arke> kc5tja: Ok, I was driving along this road, speed limit 40
17:01:32 <arke> kc5tja: I was going exactly 40
17:01:36 <dubious> ayrnieu: it would mean there are other ways to interpret code I could learn about.
17:02:11 <arke> kc5tja: this beaner bitch pulls out of some driveway in front of me (taking a left), so both me and a guy coming from the other side had to hit the brakes a little.
17:02:37 <ayrnieu> dubious - Forths generally compile what ever they find to a more convenient internal format and then directly execute that.
17:02:42 <arke> kc5tja: then she saw me, slammed the brakes, and was standing in the middle of the other side of the street while waiting for me to pass.
17:02:43 <ayrnieu> also, whatever.
17:02:44 <kc5tja> arke: OK, I want to know something.  How many people in this channel are Mexican?
17:02:45 <kc5tja> :)
17:02:57 <arke> kc5tja: Erm, dunno?
17:03:22 <dubious> ayrnieu: is the internal format like a parse tree or actual binary?
17:03:26 <ayrnieu> kc5 - please pretend that two of me gave either answer to that question.
17:03:29 <arke> kc5tja: and you do know I'm not racist, right? :)
17:03:56 <ayrnieu> dubious - yes.
17:04:41 <dubious> ayrnieu: that was a "one or the other" question, not a "yes or no" ;)
17:05:01 <dubious> or perhaps I'm just being made fun of and not realising it
17:05:21 <ayrnieu> dubious - if the such mechanisms interest you, you should find out what the following mean: STC ITC DTC 'native-coded Forth'
17:05:26 <ayrnieu> also, if such.
17:05:55 <ayrnieu> dubious - and then you probably want to read up on ColorForth and 'edit time' and partially compiled source.
17:05:59 <arke> Rush kicks ass.. :)
17:08:32 <ayrnieu> dubious - and if you want to ask 'either-or' questions, you may find it easier to begin with contradictory options =)
17:09:13 <ayrnieu> but, err, I don't know of any Forth for which 'parse tree' really fits the internel format.
17:10:12 <dubious> ayrnieu: a C compiler for example could generate a parse tree of the code before outputing binary. however, the parse tree could be interpreted right there and therefore run the given code without ever generating binary code. this is why I thought it was an 'either-or' question.
17:10:31 <ayrnieu> parsing words and the notion of multiple 'times' for code would probably give a parse tree the fitss.
17:10:45 <ayrnieu> dubious - yes, a C implementation can choose either implementation.
17:11:11 <dubious> ayrnieu: though the syntax of forth is very different from C, granted, the same approach could be taken as well.
17:11:12 <kc5tja> Well, I need to get ready for Aikido.
17:11:13 <ayrnieu> er, either method.
17:11:59 <arke> Darwin runs on the PowerPC and x86 platforms. What's more, it is even possible to build a "fat" kernel, containing both platform executables in a single file.   <--- !!!!
17:12:11 <ayrnieu> dubious - not really, because a Forth program doesn't fall into a neat compiletime/runtime contrast.
17:12:23 * ayrnieu experiences terminology failure.  Darn it.
17:12:31 <ayrnieu> kc5 - have fun.
17:12:58 <kc5tja> Forth programs are interpreted outside of colon definitions, compiled inside.  Think of Python, where things inside the def construct are compiled, but outside, they're interpreted.  Same basic idea.
17:13:11 <ayrnieu> and inside colon definitions, too =)
17:13:18 <kc5tja> Only Forth uses this dichotomy much more fluidly -- the [ and ] words can be used to switch from compile-mode to interpret-mode at will, which can be rather handy.
17:13:25 <kc5tja> ayrnieu: I just said that.
17:13:34 <ayrnieu> : +1 1 + ; immediate : foo [ 2 ] +1 literal . ;
17:13:52 <ayrnieu> kc5 - thank you!  My terminology failure happened with 'dichotomy'.
17:14:03 <ayrnieu> kc5 - I lag a bit, sorry.
17:14:06 <kc5tja> :)
17:14:10 <arke> heh cool, Open Firmware for Mac is written in forth :)
17:14:10 <kc5tja> That's OK, I need to jet.
17:14:16 * kc5tja is away: AIKIDO!!!
17:14:37 <ayrnieu> arke - Sparcstations have a similar system.
17:14:55 <arke> ayrnieu: Coolness!
17:14:57 <arke> kc5tja: have fun
17:17:26 <ayrnieu> ( or even ) : printer create , does> @ . ; 1 printer a 2 printer b : foo a b ; foo foo : wait-for ." More foos: " begin foo dup key <> until ; char q wait-for bye
17:19:47 <ayrnieu> but I'll block on that.  Stuff to do.
17:19:49 <slava> ayrnieu, the initial compiler for LSD is working. i have a rather contrieved benchmark where it runs in 30 seconds with the interpreter, 9 with the compiler.
17:20:06 <slava> ayrnieu, same algorithm written in java runs in 25 ms though -- there's some work left to do :)
17:20:21 <ayrnieu> slava - so it seems =)
17:22:55 <dubious> ayrnieu: unless I am mistakenly reading the colorforth website, don't people find it a problem that it runs standalone? what if I happen to like running under an OS?
17:23:38 <ayrnieu> dubious - you probably don't use colorforth, then =)  Or you write a hosted colorforth, since nobody has really done so, yet.
17:24:36 <Sonarman> there are hosted colorforth
17:24:36 <Sonarman> s
17:24:37 <dubious> ayrnieu: indeed, I'm just starting to learn about the virtues of forth
17:25:03 <ayrnieu> Sonar - no interestingly hosted colorforths that I've noted.  To which do you refer?
17:25:32 <ayrnieu> The one 'hosted' colorforth that I played with in dosemu didn't have much truck with the OS =)
17:25:33 <dubious> hosted under an emulator like bochs? :P
17:25:38 <Sonarman> there
17:25:44 <Sonarman> there's Xcolorforth
17:25:55 <ayrnieu> hokay, I'll look for it.
17:26:15 <Sonarman> and two (that i know of ) hosted colorforths for windows
17:26:31 <ayrnieu> I think that I'd rather make a 'colored' outer interpreter for gforth, but not just now.
17:27:13 <dubious> if the point of running colorforth standalone that forth provides greater benefits when run in ring 0, couldn't it have been written as a kernel loadable module?
17:27:42 <dubious> err, make that "shouldn't" instead of "couldn't".. anything is possible, or just about.
17:27:57 <ayrnieu> dubious - maybe, but that doesn't follow as 'the point' of colorforth.
17:28:27 <ayrnieu> and Chuck would probably have more trouble writing a kernel module than a stand-alone system, at this point.
17:29:37 <dubious> ayrnieu: right, I had already missed his point but I understand what he feels.
17:30:19 <ayrnieu> dubious - er, OK.
17:47:51 --- quit: dubious ("Leaving")
17:59:53 --- join: blockhead (default@dialin-847-tnt.nyc.bestweb.net) joined #forth
18:00:17 <arke> Hrm, neat...SDL allows me to create user defined events and push them onto the event queue
18:00:37 --- join: zardon (~zardon@h24-68-59-145.gv.shawcable.net) joined #forth
18:04:41 <zardon> what happened to #ppc, where did it move?
18:05:18 <arke> #hell
18:05:27 <arke> ^_^
18:05:49 <zardon> hmm
18:06:07 <Sonarman> was #ppc a channel on the processor, or was it for ppc linux users?
18:07:05 <zardon> mpu
18:07:11 <Sonarman> ok
18:07:18 <zardon> but it seemed to have alot of ppc64 linux folks
18:07:31 <arke> zardon: check OFTC, lots of people are moving to it
18:08:01 <zardon> politics?
18:08:19 <Sonarman> partially, i guess
18:08:28 <arke> --> You are now talking on #ppc
18:08:28 <arke> --- Topic for #ppc is Debian/GNU Linux, mostly german PowerPC Chan |realname erwünscht | realname has to be set! | We want the coexistance of Linux and OSX, not the flame. | The G5 rocks (prob.)!  |  iBooks & Powerbooks rule
18:08:28 <arke> --- Topic for #ppc set by ChanServ!services@services.oftc.net at Mon Jan 12 07:10:57
18:08:33 <arke> OFTC
18:08:38 <arke> 3 people there, including me :(
18:08:50 <Sonarman> that's the channel i remember
18:08:54 <zardon> hmm
18:09:01 <zardon> there were many more on freenode
18:09:05 <Sonarman> dreamind probably hangfs out in there
18:09:47 <arke> checking EFNet right now.
18:10:15 <arke> 17 people in #ppc there
18:10:26 <Sonarman> what's the topic?
18:10:37 <arke> no topic
18:12:10 <zardon> hmm
18:12:23 <zardon> i don't suppose you have experiance with PSIM?
18:12:53 <arke> er, no :)
18:13:03 <Sonarman> no, this is the first time i've heard about it
18:13:07 <Sonarman> arke: http://sources.redhat.com/psim/
18:13:16 <zardon> PPC simulator
18:13:27 <arke> SWEET!
18:13:35 <Sonarman> is there a difference between a simulator and an emulator?
18:13:42 <zardon> i have a nasty problem, with the IP jumping in an almost logarithmic fashion
18:14:00 <Sonarman> Herkamire could probably help you
18:14:24 <arke> I NEED A OSX ROM SOMEBODY
18:14:29 <zardon> well, PSIM just has special Open Firmware interfaces
18:14:38 <zardon> OS X ROM?
18:14:56 <Sonarman> I NEED A WINDOWS BIOS SOMEBODY
18:15:01 <Sonarman> =)
18:15:10 <arke> zardon: yep :)
18:15:10 <zardon> OS X does not really use ROMs
18:15:23 <arke> zardon: well, a bootable OSX hard drive rom
18:15:41 <zardon> that will take some doing
18:15:41 <Sonarman> you mean an image of a hard drive?
18:15:54 <zardon> PSIM is really used in gdb as a simulator
18:16:50 <arke> eek
18:17:02 <arke> Sonarman: you have OSX, you could..
18:17:05 <arke> Sonarman: oh nevermind.
18:17:09 <arke> WHERES MYSTICONE!?
18:17:50 <Sonarman> just enjoy Linux and forget about OS X :)
18:18:04 <arke> Grr!
18:18:09 <arke> I WANNA TRY OSX THOUGH
18:18:32 <zardon> swing by a apple store
18:18:41 <Sonarman> yeah, or CompUSA
18:18:51 <zardon> that implies living in .us
18:18:58 <Sonarman> i know that he does
18:20:18 --- quit: Nutssh ("Client exiting")
18:23:36 --- join: downix (~downix@adsl-153-121-56.mia.bellsouth.net) joined #forth
18:23:39 <downix> hello everyone
18:23:50 <Sonarman> hi
18:26:52 <downix> what's up?
18:43:35 <downix> other than I'm trying to figure out driver coding in forth
18:52:22 <ayrnieu> I've only ever written a PC soundcard driver in Forth =)
18:53:27 <downix> more than I
18:54:11 <ayrnieu> five lines or so of colored Flux code, IIRC.  Hardly impressive.
18:56:40 <Sonarman> post it to kristopherjohnson.net when you're don e:)
18:56:44 <downix> still more than me
18:56:52 <downix> I'm trying to figure out *how* to code them in forth
19:00:43 <ayrnieu> It depends on the environment around the Forth, generally.
19:01:18 <ayrnieu> if you used a Forth embedded into the kernel, you'd probably use the means it offers you.  For Enth/Flux I used p@ p! and such
19:01:45 <ayrnieu> Sonar - do you do much Quartus Forth programming?
19:01:59 <Sonarman> ayrnieu: none
19:02:04 <ayrnieu> I used to, but my Clie's emulation breaks down into hard results often enough with my little Chinese-English dictionary.
19:02:09 <ayrnieu> also, hard resets.
19:02:28 <Sonarman> what dialect of Chinese are you learning?
19:02:31 <ayrnieu> Mandarin.
19:02:41 <ayrnieu> often called 'Chinese' =)
19:02:46 <Sonarman> cool.  I'm learning Cantonese
19:02:57 <ayrnieu> ah, nifty.  Where?
19:02:58 <proteusguy> Ni Hao Ma
19:03:07 <Sonarman> Nei Hoo Maa
19:03:11 <ayrnieu> proteus - hen3 hao3.  ni3 ne?
19:03:14 <proteusguy> show off. :)
19:03:15 <Sonarman> ayrnieu: just at school
19:03:35 <Sonarman> ayrnieu: Wo hao :)
19:03:40 <ayrnieu> Sonar - do they teach you with that romanization?
19:04:01 <Sonarman> ayrnieu: no, we don't use romanization in class, that's just my version of it
19:04:04 <proteusguy> I've always thought that forth my be attractive to people who have non-phoenetic languages.
19:04:30 <ayrnieu> proteus - it certainly offers more opportunities for sane identifiers.
19:04:36 <blockhead> non-phoenetic?
19:04:48 <proteusguy> the character set is non-pheonetic.
19:05:01 <ayrnieu> blockhead - Chinese does not 'phonetisize' well.  proteus probably really wants to talk about ASCIIzation.
19:05:06 <proteusguy> meaning you can't "sound out" the spelling of a word.
19:05:08 <downix> ayrnieu:  it's for OpenFirmware!
19:05:13 <Sonarman> ayrnieu: how much of the chinese character set do you know?
19:05:36 <ayrnieu> since Koreans would also benefit from a relaxed 'identifier', though their language more easily romanizes.
19:05:45 <proteusguy> actually there's not much in the way of "spelling" at all.
19:05:51 <blockhead> ayrnieu: oh, I think I understand. thank you.
19:05:55 <ayrnieu> Sonar - I know maybe 2000 characters.  I haven't counted.
19:06:52 <proteusguy> 2000 is a pretty good amount! How long have you been learning?
19:07:11 <ayrnieu> proteus - since about March, last year =)
19:07:18 <Sonarman> ayrnieu: that's more than I know, for sure :)
19:07:49 --- join: Nutssh (~Foo@gh-1177.gh.rice.edu) joined #forth
19:07:49 <ayrnieu> proteus - what other languages do you know, or study?
19:07:55 <proteusguy> hmm... Are you learning full time? Have native speakers helping?
19:08:04 <ayrnieu> proteus - yes.
19:08:28 <proteusguy> I used to be very good at reading french, ok at writing it, decent understanding verbally, horrid speaking. :P
19:08:49 <ayrnieu> (j'ne parle pas francaise =)
19:08:55 <ayrnieu> also, 'francais'
19:09:09 <Sonarman> you forgot the squiggly c :p
19:09:19 <proteusguy> Know a smattering of Japanese.
19:09:34 <ayrnieu> proteus - watashi wa!
19:09:43 <ayrnieu> proteus - definitely a *smattering*.
19:09:49 <proteusguy> Just enough to be polite and get some food. Being polite requires more knowledge than one would think! ha ha
19:10:22 <ayrnieu> proteus - aye, I suppose it involves more than ordering in English and then saying 'arigatoo gozaimusu' or such.
19:10:42 <proteusguy> haha indeed.
19:11:20 <proteusguy> My ex-gf is Chinese so I learned some spoken words but almost no characters and a little pin yin.
19:12:05 <ayrnieu> proteus - did she laugh at your tones?
19:13:03 <proteusguy> Yes of course. I can still only casually differentiate three tones, four when I'm careful. Two of them (3 & 5 I think) sound identical to me.
19:15:50 <ayrnieu> proteus - oh, what dialect did she teach you?
19:16:20 <proteusguy> Mandarin primarily but her family also speaks "Hawkien". Not sure of the spelling.
19:16:44 <ayrnieu> huh, OK.  I haven't heard of "Hawkien".
19:18:01 <proteusguy> Its one of the dozens of major dialects. Hard to describe. Very different from Mandarin and Cantonese.
19:19:14 <Sonarman> Hakka?
19:19:44 <proteusguy> well how I spelled it is how they pronounce it. Otherwise can't confirm or deny.
19:19:52 <ayrnieu> Mandarin only has four tones (and 'five' if you count 'no tone'): high, rising (start low and end high, like at the end of an English question), low (start 'regular' or so, dip down, and then rise -- but not quite as high as 'rising'), and 'falling' (start less-than-high and fall rapidly).
19:20:33 <proteusguy> Yes - I count flat as a tone. All the books I've read do as well.
19:21:06 <ayrnieu> It doesn't interest me so much, I suppose.
19:21:57 <ayrnieu> but (with 'no tone' as 5), I listed those in the order that I learned to number them as.  Does your numbering system agree?
19:22:03 <proteusguy> You've been studying since March and it doesn't interest you? ha ha How is that?
19:23:16 <proteusguy> hmm... I was confusing the "bouncing" tone with the one that rises. Don't recall the actual order.
19:23:38 <ayrnieu> proteus - Chinese tends to flatten tones in certain circumstances.  I find it easier to listen to tones than their absense.
19:23:53 <ayrnieu> proteus - the one I called 'low'?  Two tones rise =)
19:24:28 <proteusguy> Its obvious to me that context is more often a differentiator than the absolute tone - but you gotta know enough of the language to get a context!
19:24:41 <proteusguy> So where are you learning this?
19:25:14 <ayrnieu> aye, you have to learn the language =)
19:44:45 <Herkamire> I don't think psim can run Mac OS
19:57:56 <Herkamire> http://sources.redhat.com/psim/faq/faq_5.html#SEC5
19:58:08 <Herkamire> it says it doesn't do graphics yet
19:58:21 <downix> PSIM is very slow
20:00:13 <chandler> it's a cycle-accurate simulator, not an efficient emulator
20:00:20 <chandler> emulating ppc on x86 would be an exercise in PAIN
20:00:31 <downix> yup
20:00:42 <downix> I just have run into so many folk that can't tell the difference
20:00:47 <downix> UAE is an emulator
20:00:51 <downix> it is *not* cycle-accurate
20:00:56 <downix> (for a good example)
20:01:09 <Herkamire> their sample statistics show the emulator runing at 250,731 instructions/second
20:01:09 <chandler> the basilisk II engine even does JIT these days
20:01:22 <chandler> hah. 250khz!
20:01:32 <chandler> OS X will run at blazing speed!
20:01:34 <downix> no 250 instructions/second
20:01:45 <downix> that's actually 75khz
20:01:59 <Nutssh> I wonder if PPC on x86-64 might be a bit less painful than x86-32. At least you have registers.
20:02:19 <chandler> it would be less painful. can x86-64 do pseudo big-endian?
20:02:37 <chandler> of course, FP would still be a major pain
20:02:41 <downix> No chandler
20:02:42 <chandler> and OS X is incredibly heavy on FP
20:02:47 <downix> x86-64 is still LE
20:02:54 <Herkamire> downix: how do you figure 75khz?
20:03:44 <downix> Herkamire:  Well, I don't know which model PPC PSIM does, but some varients can pump out 4 instructions per clock cycle
20:04:01 <chandler> downix: dude, have you looked at the 970? :-)
20:04:03 <downix> not counting an Altivec
20:04:12 <chandler> oh
20:04:13 <Herkamire> downix: ahh. ok. I was just looking at the statistics, and they show more cycles than instructions.
20:04:22 <downix> chandler:  Well, I didn't think PSIM would have been upgraded to 970's.
20:04:31 <chandler> even still, I think the 970 can do two int + two fp + two load/store per cycle, from memory
20:04:32 <downix> Herkamire:  ewww!  8)
20:05:15 <Herkamire> downix: their example code is loaded with contition branches
20:05:42 <downix> gotcha
20:05:49 <Herkamire> about a third of the branch predictions failed
20:06:08 <downix> not good then
20:06:47 <Herkamire> psim looks like it'd be a cool sort of profiler
20:06:55 <downix> *nods*
20:07:02 <chandler> the builtin instruction counters work just as well
20:07:08 <chandler> hence shark on OS X
20:08:07 <Herkamire> you could probably run my forth with psim though :)
20:16:17 <arke> b4ck4g3
20:18:26 <Sonarman> y0
20:20:42 <arke> Herkamire: you got an OSX rom?
20:20:46 <arke> ^__^
20:20:51 <chandler> ROM?
20:20:57 <chandler> you are so confused, child
20:21:02 <arke> heh
20:21:03 <arke> lol
20:21:12 <arke> yeah
20:21:13 <arke> I am
20:21:20 <arke> I meant an OSX HD image
20:21:51 <chandler> no. and you wouldn't be able to run it with psim anyway.
20:21:59 <chandler> psim does not emulate the rest of the hardware
20:22:08 <arke> ugh :(
20:22:13 <arke> no graphics?
20:22:26 <arke> suxx0r!
20:22:35 <chandler> no uninorth bridge, so no memory controller
20:22:39 <chandler> no PCI devices of any sort
20:22:40 --- quit: downix ("Leaving")
20:22:42 <chandler> no ATA controller
20:22:46 <arke> :(
20:22:48 <chandler> no open firmware
20:22:51 <arke> :(
20:23:07 <arke> I wonder if they're ever gonna do bochs-ppc...
20:23:20 <chandler> why would anyone, except as an exercise in pain?
20:23:29 <chandler> in portable C code, it would be so slow as to be useless
20:23:50 <arke> lots of people use bochs.
20:23:57 <chandler> in raw x86 assembly, you /might/ be able to get 10-8:1. maybe. dowhnill with a tailwind
20:24:04 <chandler> yes, because x86 is braindead
20:24:10 <arke> true.
20:24:13 <chandler> it's the universal bytecode these days
20:24:14 <arke> very true.
20:24:22 <arke> x86 _is_ braindead.
20:24:25 <chandler> Virtual PC can do 2:1 emulation on a G4 with L3 cache
20:24:38 <arke> and I guess it's easy to emulate x86 on ppc, but not quite the reverse :)
20:24:41 <arke> chandler: wow!
20:24:49 <chandler> of course, graphics is slower
20:25:26 <arke> VPC = MS VMWare or MS VMWare + hard-coded Windows?
20:25:26 <chandler> but that's because of JIT, which is made possible by the increased number of registers
20:25:38 <chandler> arke: Virtual PC was a Connectix product; it's just a PC emulator
20:25:45 <chandler> there are special drivers for windows but it runs many OS's
20:25:53 <chandler> I've run everything from BeOS to NeXTStep and Linux in it
20:25:57 <arke> ...er?
20:26:09 <chandler> oh. the x86 version is a vm, if that's what you're asking
20:26:10 <arke> Microsoft just came out with this MS Virtual PC for the Macintosh
20:26:15 <chandler> yeah. MS bought Connectix.
20:26:19 <arke> Oh.
20:26:21 <arke> Cool.
20:26:29 <chandler> unfortunately it doesn't yet run on G5
20:26:31 <arke> er, not so.
20:26:37 <arke> Ugh.
20:26:46 <chandler> the next version (later this year, they claim) will
20:26:58 <Sonarman> chandler: what is uninorth?
20:27:17 <chandler> Sonarman: it's just Apple's unified north / south bridge
20:27:32 <Sonarman> ok, evidently i have more reading to do :)
20:28:20 <chandler> memory, ATA, firewire, USB, audio, PCI, AGP, and phy all in one chip
20:32:35 <arke> Wow.
20:32:46 <arke> MS VPC is quite a nice product, from what I see.
20:33:11 <arke> I think it should be more "integrated" though...
20:33:30 <arke> As in, the start menu in the dock, but each "Windows" window being its own "Mac" window
20:33:43 <arke> That way, it would really be transparent (and cool!)
20:34:02 <chandler> yeah. but I bet they'll never do that
20:34:21 <arke> Of course.
20:34:29 --- quit: proteusguy ("Leaving")
20:34:31 <arke> That would make people ditch PC+windows and be mac only
20:34:33 <arke> ^__^
20:35:15 <chandler> well, I did that already :-)
20:35:28 <arke> I would if I could afford it
20:36:33 <chandler> ok...
20:36:58 * chandler finds that people complaining about the affordability of macs are looking at G5s and not the more affordable options
20:37:28 <Sonarman> arke: or you could get a pegasos or amigaone system :)
20:37:46 <chandler> Sonarman: that ends up being just as expensive as a low-end iBook
20:37:48 <arke> Well, G4's are quite expensive too.
20:37:55 <chandler> arke: really?
20:38:02 <arke> From what I see...
20:38:13 <chandler> what are you looking at?
20:39:14 <arke> http://www.bestbuy.com/site/olspage.jsp?id=1062632293080&skuId=6011585&type=product
20:39:36 <chandler> why would you buy from best buy?
20:39:46 <chandler> btw, are you a student?
20:40:49 <chandler> that's a nice computer but you're paying for the monitor mostly
20:41:31 <arke> Yes, I'm a high school student
20:41:43 <arke> high school students get 5% off, i think
20:41:46 <chandler> Apple has educational discounts
20:42:03 <chandler> better than nothing. and I thought they raised the discount recently
20:42:32 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth
20:42:55 <arke> hi tathi
20:43:07 <tathi> hey arke
20:44:04 <chandler> my suggestion is to look at an iBook
20:44:12 <chandler> great machines, quite small
20:44:19 <arke> laptops, right?
20:44:21 <chandler> don't complain about the price when you're looking at a premium machine
20:44:23 <chandler> yeah
20:44:55 <arke> http://www.bestbuy.com/site/olspage.jsp?id=1066095232795&skuId=6153075&type=product
20:46:03 <arke> http://www.macofalltrades.com/product.asp?dept%5Fid=2&pf%5Fid=G4400XX1&mscssid=0WVXVMLJENQT8M5V55NSQ3V3FXW7BRQ1
20:46:11 <chandler> argh. you are /still/ looking at the most expensive in the line
20:46:26 <chandler> the Yikes is called that way for a reason
20:46:29 <chandler> it has no AGP slot
20:46:33 <chandler> thus, no Quartz Extreme
20:46:35 <chandler> and that means slow
20:46:47 <chandler> Apple sells iBooks for $800 on their web site, under the "special deals" section
20:46:57 <ayrnieu> chandler - ooh.
20:47:11 <chandler> they have a G3, but they are quite fast in use
20:47:17 <chandler> spend another $50 on memory and you are set.
20:47:46 <arke> will G3 run OSX?
20:48:16 <chandler> oh yes
20:48:22 <chandler> sufficiently zippy too
20:48:32 <arke> oh, coolness
20:48:37 <arke> any g3 will run it?
20:48:40 <chandler> oh, they have much nicer ones for $900 too - with a DVD reader and CDRW
20:48:50 <chandler> there are one or two really old G3s that won't
20:49:04 <Herkamire> not any g3
20:49:05 <chandler> 900 G3, 40GB HD, combo drive, 12" for $900
20:49:25 <arke> Hrm, hold on
20:49:28 <ayrnieu> sure, but what resolution on that 12"?
20:49:28 <chandler> oh, they have the same one refurbished for $850
20:49:29 * arke balances checkbook
20:49:33 <chandler> ayrnieu: 1024x768
20:49:37 <chandler> http://store.apple.com/1-800-MY-APPLE/WebObjects/AppleStore.woa/70307/wo/Ge45jEeIvZLF2QIClJ625k30TJm/0.0.7.1.0.5.13.0.3.0.0.0.0.3.1.1.0?29,70
20:49:37 <Herkamire> the beige g3s don't run OS X... I think that's the distictcion
20:49:46 <chandler> anything with builtin USB does
20:49:56 <ayrnieu> hm.
20:50:13 <ayrnieu> Then in a few months I'll buy an apple.
20:50:19 <chandler> cool
20:50:44 <Herkamire> just get an apple that is a couple years and you will get a nice price
20:51:12 <arke> 	iBook 700MHz/128MB/20GB/CD-ROM/E/56K/12.1" TFT - Refurbished 		$679.00
20:51:24 <chandler> arke: no checkbox there, they're out
20:51:26 <chandler> and you really want a combo drive
20:51:45 <chandler> you could call and ask about inventory
20:51:49 <arke> 	iBook 800MHz/128MB/30GB/CD-ROM/E/56K/12"TFT - Refurbished 	Add 	$749.00
20:51:51 <chandler> you might get something more up to date than the page
20:51:56 <chandler> that still does not have the combo drive
20:52:01 <arke> Ok, lemme balance my checkbook to see how I'm doing.
20:52:34 <chandler> the 900 has a better graphics card in it too - radeon mobility 9200 IIRC
20:54:04 <chandler> oh, it's a 7500
20:54:08 <chandler> still, that's no slouch
20:54:51 <arke> 231.24
20:54:56 <arke> my current savings account
20:54:57 <arke> :(
20:55:23 <chandler> hm
20:55:29 <chandler> that's a bit low for any computer
20:55:35 <arke> shit, i need to find my bank statement..
20:56:00 <arke> aaH, GOT IT
20:56:04 <arke> er, sorry caps
20:56:14 <ayrnieu> You can get a decent enough (read: not at all decent, by anyone's but my starved standards) laptop off of eBay -- or an OK (read: way better than I've ever touched, by my starved standards) desktop system.
20:56:50 <chandler> there are too many frauds on ebay
20:57:20 <ayrnieu> chandler - none of them that've bit me, but the desktop systems to which I referred I did not find on eBay.
20:57:48 <chandler> I've been bit on smaller stuff on fleaBay
20:57:54 * ayrnieu has bought two laptops off of eBay, three PDAs, and numerous small happy things like "Spirited Away"
20:58:16 <arke> spirited away was an odd movie.
20:58:32 <ayrnieu> I so love it.
20:59:24 <arke> it wasn't bad, i liked it, it just was odd
20:59:49 <ayrnieu> it doesn't strike me as all that odd, but OK.
21:00:20 <ayrnieu> you didn't watch a dubbed version or something, did you?
21:00:22 <chandler> arke: btw, high school discount is same as uni discount - between 5 and 20% depending on the item
21:00:35 <arke> sweet..
21:00:45 <arke> what sort of items would have more discount?
21:00:51 <chandler> more expensive ones :-)
21:01:09 <chandler> still won't bring the price down to refurbed level
21:01:14 <chandler> do you have an apple store near you?
21:01:26 <chandler> sometimes they sell "refreshed" stock, which are returned units which have been reloaded
21:01:43 <ayrnieu> huh, I didn't know that 'apple store's existed.
21:01:53 <chandler> hah. you have been living in a cave, haven't you? :-)
21:02:00 <arke> oh hell yeah, 5 mcdonald's free meal cards
21:02:05 <chandler> http://www.apple.com/retail/
21:02:21 <arke> I would love an apple computer.
21:02:25 <arke> i would loooove it.
21:02:45 <Sonarman> why?
21:02:49 <arke> I wonder if they already got an X windowmanager that behaves a little more apple like.
21:02:54 <arke> BECAUSE X86 IS FUX0RED
21:03:01 <arke> and I love macs.
21:03:08 <chandler> you do know that OS X doesn't use X11, right?
21:03:11 <chandler> anyway there is X11 for OS X
21:03:14 <chandler> and it is mac-like
21:03:22 <ayrnieu> I've never particularly enjoyed macs, but I don't have any dark biases against them.
21:03:33 <chandler> arke: http://www.apple.com/macosx/features/x11/
21:03:41 <chandler> ayrnieu: throw away all your impressions of pre-OS X macs
21:03:49 <chandler> OS X is not really Mac OS; it's NeXTStep
21:03:51 <ayrnieu> chandler - =)  OK.
21:03:52 <arke> chandler: I know about that, i meant a window manager for X that is OSX like
21:04:17 <arke> from what I read, OSX is a maccish nextstep :)
21:04:26 <chandler> yes, it is
21:04:30 <arke> but the interface is similar.
21:04:41 <chandler> in some ways. there are big influences from *step too
21:04:42 <arke> I used to have an ancient mac, and I liked it better than any windows box.
21:04:47 <arke> you can just use it.
21:10:11 <Herkamire> well, I'm going to go play with ppforth on my Palm V in the dark.
21:10:24 <arke> :P
21:10:31 <ayrnieu> Herka - not Quartus Forth?  Or Dragon Forth?
21:14:22 --- part: Nutssh left #forth
21:32:16 --- join: default_ (default@dialin-847-tnt.nyc.bestweb.net) joined #forth
21:35:27 * kc5tja is back (gone 04:21:11)
21:41:53 --- quit: blockhead (Read error: 60 (Operation timed out))
21:43:15 --- nick: default_ -> blockhead
21:44:40 --- quit: blockhead ("Client Exiting")
21:47:03 --- quit: tathi ("leaving")
21:48:21 <arke> kc5tja: how was aikido?
21:48:25 --- join: proteusguy (~proteusgu@216.27.161.121) joined #forth
21:48:36 <zardon> kc5tja: have you used the psim/gdb combo?
21:48:51 <kc5tja> zardon: No.
21:48:55 <kc5tja> arke: Slow, but good.
21:49:07 <kc5tja> We had a lot of new beginners to the class, so we went over the very, very basic stuff.
21:49:27 <kc5tja> Things like the four principles of aikido, and sensei proceeded to demonstrate everything.
21:49:50 <kc5tja> I could physically realize only three of the four principles -- the fourth wasn't possible with my skill level at this time.
21:49:50 <arke> xpde.com <--- GOOD effort, really
21:49:53 <kc5tja> But it was fun.
21:50:00 <arke> :)
21:50:09 <arke> I'm looking for another WM again.
21:50:19 <chandler> tried icewm?
21:50:22 <arke> xfce isn't suiting me anymore.
21:50:43 <arke> chandler: icewm itself is fine, but its ugly and ... well, not enough, in a sense... lol hard to explain
21:50:52 <kc5tja> Oh my God!  XPDE shall never grace my desktop ever again.
21:50:58 <kc5tja> Oh, my virgin eyes!!
21:51:07 <chandler> arke: ok. there's always GNOME. but I like icewm + rox
21:51:18 * kc5tja uses ROX-Filer too.  *Sweet* system.
21:51:20 <arke> kc5tja: :P
21:51:29 <arke> kc5tja: I know, but the effort is amazing.
21:52:00 <kc5tja> True, but to me, that's not nearly as hard as actually innovating a useful user interface, something truely distinctive and compelling.
21:52:22 <arke> I think I'm gonna have to just start my own WM someday.
21:52:30 <arke> and then change it when I don't like it.
21:52:32 <kc5tja> That's why I always hated 90% of X window managers/desktop environments.  They always try to copy Windows.  Windows, Windows, Windows.  Bahh.  If I wanted to use Windows, I'd install Windows.
21:52:34 <arke> but that's AFTER frapiar.
21:52:43 <arke> kc5tja: heh
21:54:05 * proteusguy got banned from redhat for making that windows remark! :P
21:54:28 <kc5tja> proteusguy: You get banned from #redhat for just showing up.  :D
21:54:53 <arke> I wish there was a OSX like WM out there ...i would probably be using that.
21:55:05 <proteusguy> Evidently!
21:55:07 <arke> I'm considering a composite system.
21:55:37 <arke> xfce4 with a KDE Panel, gnome panel, and rox
21:55:39 <arke> lol
21:55:41 * ayrnieu finds ratpoison a happy enough WM =)
21:55:44 * proteusguy glances at ROX-Filer. Interesting...
21:55:49 <zardon> there is a wm called Oberous or somthing that looks like Aqua
21:56:47 <kc5tja> zardon: Google can't find anything on that.
21:56:59 <arke> OroborOSX
21:56:59 <arke> More than just another window manager!
21:56:59 <arke> A complete environment for X11 on Mac OS X
21:57:33 <proteusguy> So what do you like about ROX so much kc5? 
21:57:34 <arke> Maybe I'll switch back to wmaker.
21:57:36 <chandler> yeah. it's useless now that Apple ships X11 with the OS
21:57:36 * arke is away: sleep
21:57:50 <chandler> proteusguy: fast, simple, good UI
21:57:55 <kc5tja> proteusguy: Speed.  And unobtrusiveness.
21:58:19 <kc5tja> proteusguy: It's also incredibly small, and very, very featureful for its size.
21:58:58 --- quit: scope (Read error: 54 (Connection reset by peer))
21:59:01 <kc5tja> It supports panels and the like, and plug-ins for the panel are trivial to write software for.  No need for CORBA (though XML-RPC is used, XML-RPC is very, very much simpler to support than CORBA), and its AppDirs are soooo nice.
22:03:12 <arke> I think I'm gonna go with window maker + rox + customized xfce panel.
22:03:15 <arke> and I sleep.
22:06:19 <proteusguy> I like small and fast. Don't like CORBA. Sounds promising.
22:08:59 --- join: Serg (Serg_Pengu@212.34.52.140) joined #forth
22:13:34 <zardon> hello
22:16:14 <proteusguy> howdy.
22:16:22 <Serg> hi ;)
22:16:57 <zardon> how is the hacking in russia going?
22:17:20 <Sonarman> kc5tja: do you know where ROX applications look for libraries?  i've installed ROX-Clib in ~/lib but the compilation of an application which uses it is failing
22:17:42 <kc5tja> Sonarman: ??  I never had a problem.
22:17:54 <Sonarman> you put them in ~/lib?
22:18:14 <kc5tja> What file, specifically, are you looking for?
22:24:13 <Sonarman> i don't know, i'll try agian later
22:31:34 <Serg> kc5tja: i bought a handful of parts and now think how to construct my first RX ;))
22:31:54 <Serg> how 2 make board, case etc. almost w/o tools ;)
22:38:58 <kc5tja> :)
22:39:04 <kc5tja> Good luck.
22:39:35 <Serg> how do you make it in such situation ?
22:42:58 <kc5tja> Heh, luck.
22:43:07 <kc5tja> I have yet to make a working RX.
22:43:18 <kc5tja> I made a working AF amplifier, but that's about it.  That isn't too hard.
22:43:26 <kc5tja> I made it dead-bug style, without concern for case, etc.
22:43:34 <kc5tja> Just the components and printed circuit board.
22:43:42 <kc5tja> Soldered the components directly to each other.
22:43:49 <kc5tja> I used the PCB as a flat ground-plane.
22:45:32 <Serg> i seen things made on worn out book cardboard ;)) 
22:45:52 --- quit: Sonarman ("leaving")
22:46:08 <Serg> 2-side PCB, one side - solid, other - cut in 10x10 mm squares, tite ?
22:46:20 <Serg> right ?
22:46:57 <kc5tja> Serg: That's more like "Manhatten" style, as we call it.  It's related to dead-bug style, and is also used where suitable PCB tools don't exist.
22:47:12 * Serg needs some support for knobs ;)) kinda like case
22:47:19 * kc5tja nods
22:47:34 <kc5tja> When I start making my PCB CAD tool, I have to consider both the board itself and the case that the board will go in.
22:47:39 <kc5tja> I'm not yet sure how I'm going to do this.
22:47:57 * Serg doesn't want to buy full tool set, drill etc.
22:48:11 <kc5tja> I just buy the things I need, as I need them.
22:48:39 <kc5tja> Like, for some of the things I'm looking to do, I may need a programmable logic programmer.  :(  That's going to be expensive.
22:48:53 <kc5tja> So I'm really, really holding off on that.
22:49:38 <Serg> heh, i seem to need grips, saw, drill ... :( too cumbersome and i doubt i'll need it again
22:49:55 <kc5tja> Yeah, that's one of the reasons I'm holding out.
22:50:22 <kc5tja> Xilinx has already retired a few of their chip lines, so anyone who has a programmer for those chips now finds themselves with a door-stop.  :(
22:50:46 <Serg> at fu... home, i have no even stable table to fix the grips ;(((
22:50:51 <kc5tja> Then there is the issue of soldering 0.5mm pitch pins.  :(
22:51:28 <kc5tja> Serg: You know, all the homebrew hackers I know all have home-built workbenches.  :)
22:51:39 <kc5tja> So I'm not sure the problem is restricted to your country.  :)
22:52:00 <kc5tja> Right now, my tables are both flimsy card tables.
22:52:06 <kc5tja> Definitely not suitable for anything electronic.
22:52:13 <kc5tja> My computer monitor is wobbling as I type this.  :)
22:52:19 <Serg> oops ;((
22:52:31 <kc5tja> Yeah.
22:52:34 <kc5tja> Oh well.
22:52:37 <Serg> no, my issue is utter inner misery of my mom and all relatives
22:52:48 <kc5tja> Someday, I'll be able to afford a real desk.
22:53:05 <Serg> i am the FIRST ONE IN ALL FAMILY who works and has rising income
22:53:25 * kc5tja nods
22:53:33 <Serg> afford ? it's damn cheap !!
22:53:44 <kc5tja> Not here.
22:53:52 <Serg> issue is great mayhem then changing furniture, not price
22:53:55 <kc5tja> I either have to built it myself, or spend a few hundred $ on a table.
22:54:18 <kc5tja> And my take-home pay isn't much.  My rent is $500/month, and my income is only $800/month.
22:54:44 <kc5tja> Considering food and utilities, that leaves me with about $100/month for general spending money.  Take groceries out of that, and that leaves me with, oh, $50.  :(
22:54:57 <Serg> and lack of space - 13m^2, full of scrap and trash what i can't' toss out, coz it's not mine but mom's
22:55:11 * kc5tja nods
22:55:18 * kc5tja wishes he had a 13m^2 room.  :)
22:55:32 <Serg> oops, how little do u have ?
22:55:52 <Serg> i remember myself 4 y old, driving tricicle in that room
22:55:52 <kc5tja> My room is about 10 feet by 10 feet, so ... about 9 m^2 I'd guess.
22:56:21 <kc5tja> (I'm estimating 10 feet at 3.33m)
22:56:33 <Serg> now it's like scrapyard, only narrow path in middle :((
22:57:26 <Serg> i bought 2 warehouse shelves ( 100x50x200 cm ), but it doesn't help much
22:57:38 * kc5tja nods
22:58:44 <Serg> and tossed old dead fridge to make room ;))
22:59:29 <Serg> or 'to free some space', how is more right ?
22:59:41 <kc5tja> Both are correct English phrases.
23:01:00 <Serg> are you now capable of TX, what power/band ?
23:01:18 <kc5tja> Now that I have an antenna tuner, it is at least safe for me to transmit on various bands.
23:01:38 * Serg contacted local hams to buy old RX, and think of QSO (and QSL ;) w/ IRC sked ;))
23:01:39 <kc5tja> I can transmit up to 25W safely at the moment, though the rig is capable of handling 100W.
23:01:50 <kc5tja> Above 25W, things start buzzing quite loudly in my room.
23:02:52 <kc5tja> However, my privileges are still restricted to just morse code on all HF bands except 10m.
23:03:09 <Serg> they have 3-vibrator YAGI for 20m ;))
23:03:30 <kc5tja> Just to let you know: in English, that'd be a 3-element Yagi.  :)
23:03:54 <Serg> 4-storey house + maybe, 10m macht
23:04:00 <kc5tja> Saying something has 3 vibrators is rather erotic in nature.  :)
23:04:31 <Serg> i imagine how large they gotta be to be usable as YAGI elements ;))))
23:04:48 <kc5tja> macht is a mast?  The pole on which the antenna sits?
23:04:52 <Serg> for 20m band ;))
23:04:56 <kc5tja> Yeah, I've seen a 20m antenna before -- it's huge.
23:05:02 <Serg> yeah, mast ;))
23:05:13 <kc5tja> What is incredibly frightening is a 160m band, 4-element yagi.
23:05:15 <kc5tja> Good grief.
23:05:30 <kc5tja> That thing is so large, each element has its own set of guy-wires.  :/
23:07:50 <Serg> ;))
23:10:19 <Serg> why not 5/8 ?
23:10:49 <Serg> or 'phased array' of 4 5/8, what can be switched to 4 ot 8 directions ?
23:10:59 <kc5tja> Because that's even larger.  :)
23:12:24 <Serg> hmmm... here in RU, it's hard to imagine what guy having callsign never built a rig ;))
23:12:49 <Serg> at least, decent RX or/and simple CW TX ;))
23:20:08 <Serg> kc5tja: how do u think, is it possible to make DC reciever for AM, w/ PLL (phase lock loop) ?
23:20:26 * Serg is sick of superhets
23:21:14 <kc5tja> It ought to be easy to do, sure.  AM would excel with a DC receiver, because DC receivers add both upper and lower sideband energy into the same audio signal.
23:21:34 <kc5tja> Wow, my website is apparently moderately popular.
23:21:47 <Serg> Forth site ?
23:21:51 <kc5tja> I even have a few folks from the Czeck republic viewing it.
23:21:59 <kc5tja> Yes, and it's not even that recent.
23:22:01 <kc5tja> err
23:22:02 <kc5tja> that old rather.
23:22:04 * kc5tja is tired.
23:22:25 <kc5tja> I've had 3 search strings for Falvo Web Design
23:22:31 <kc5tja> and Falvo Training
23:22:33 <kc5tja> etc.
23:22:38 <kc5tja> Most interesting.
23:22:40 <Serg> ;) nice
23:22:47 <kc5tja> Too bad I don't get any real customers from it all though.
23:22:52 * Serg wanna try myself as designer
23:22:58 <kc5tja> But, I might have bagged a customer thanks to another friend of mine today.
23:23:07 <kc5tja> For custom software development.
23:23:10 * kc5tja is going to specialize.
23:23:23 <kc5tja> Falvo Technical Solutions is going to concentrate on embedded systems development in C and Forth.
23:31:20 <zardon> nice
23:31:49 <zardon> particular mpus?
23:38:07 --- join: Nutssh (~Foo@gh-1177.gh.rice.edu) joined #forth
23:53:33 <kc5tja> zardon: I try to be open, but I'm partial to x86 and Motorola CPUs.
23:53:46 <kc5tja> I don't have any real-world experience with PowerPCs, but I can't see how they can be THAT much different from MIPS.
23:53:52 <kc5tja> (which I have SOME real-world experience with.)
23:54:23 <kc5tja> Anyway, it's midnight here, and I'm getting tired.
23:54:26 <kc5tja> So, I'm off to bed.
23:54:34 <kc5tja> Night all!
23:54:41 --- quit: kc5tja ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK")
23:59:59 --- log: ended forth/04.01.13