00:00:00 --- log: started forth/03.07.31
00:02:28 --- join: a7r (~a7r@206.72.82.135) joined #forth
00:30:51 --- quit: skylan (Connection timed out)
01:13:26 --- join: Serg_Penguin (Serg_Pengu@212.34.52.140) joined #forth
01:13:31 --- quit: Serg_Penguin (Client Quit)
01:41:45 --- join: Serg_Penguin (Serg_Pengu@212.34.52.140) joined #forth
01:41:54 --- quit: Serg_Penguin (Client Quit)
02:14:26 --- join: skylan (sjh@vickesh01-4724.tbaytel.net) joined #forth
03:12:04 --- join: njd (melons@njd.paradise.net.nz) joined #forth
03:20:17 --- part: njd left #forth
03:30:23 --- quit: a7r (Read error: 110 (Connection timed out))
05:05:53 --- join: w1k1 (~w1k1@pD9545C06.dip.t-dialin.net) joined #forth
05:25:37 --- join: rO| (~rO|@pD9545C06.dip.t-dialin.net) joined #forth
05:26:11 <rO|> lo
06:53:56 --- join: Serg_Penguin (Serg_Pengu@212.34.52.140) joined #forth
06:54:02 --- quit: Serg_Penguin (Client Quit)
06:58:21 <gilbertdeb> .
07:12:50 --- join: rk (~rk@ca-cmrilo-docsis-cmtsj-b-36.vnnyca.adelphia.net) joined #forth
07:15:45 --- join: mur (murr@baana-62-165-189-186.phnet.fi) joined #forth
07:16:12 <mur> terve
07:16:19 <gilbertdeb> terve mur.
07:16:45 <mur> 33c here
07:16:47 <mur> deadly hot
07:21:56 <rk> mur: ack! evil! the world should be 15c all year every year everywhere
07:23:19 <mur> 21 is fine
07:23:25 <mur> (usually)
07:28:50 <rk> mur: between 15 and 20 :P
07:29:03 <rk> mur: where do you reside??
07:29:10 <mur> finland
07:29:22 <rk> aah :)
07:30:48 <mur> 33c :(
07:31:52 <rk> but europe. europe is cool and nice.
07:32:03 <rk> (i wish i was in geeeermaaaanyyyyy)
07:32:05 <gilbertdeb> riiiight.
07:32:41 <rk> gilbertdeb: if you've lived 12 years in germany, then wander out to the US, you'll want to live in europe again, too
07:32:50 <gilbertdeb> I suppose.
07:32:52 <mur> rk people have common sense in europe too. it's very useful to have...
07:33:06 <gilbertdeb> people have that here too :)
07:33:35 <rk> naah, or at least not the majority of the population
07:34:11 <gilbertdeb> rk: in an anthill you want a lot of worker ants. not more queens or soldiers etc.
07:34:20 <gilbertdeb> a few people are enough to do their thinking for them.
07:34:36 <rk> gilbertdeb: ...
07:34:44 <rk> gilbertdeb: its supposed to be democracy
07:35:15 <gilbertdeb> rk: which means, you create an issue and ask market it for people to popularize with their votes.
07:35:18 <gilbertdeb> thats all it is.
07:35:33 <gilbertdeb> s/ask//
07:35:50 <gilbertdeb> to popularize over an other issue.
07:36:05 <rk> to be honest, i like how germany works. what i dont like about germany is that everything got so damn expensive (i had a vacation there until a few days ago)
07:36:14 <rk> gilbertdeb: i know what it means :P
07:37:10 <gilbertdeb> rk: so a democracy shouldn't need too many thinkers, just a few to think and persuade as many people to vote for their ideas as possible.
07:37:28 <gilbertdeb> if too many people thought, things would go bad!
07:37:39 <rk> i think everyody should think for himself
07:37:47 <gilbertdeb> rk: goodness no!
07:38:09 <rk> nature already chose a halfways balanced system of stupid people and smart people, so its alright actually. 
07:38:25 <gilbertdeb> it is better to divert the funds into purchasing the latest gigahertz computers and the fatest pipes than it is to employ teachers.
07:38:40 <gilbertdeb> the economy benefits in the short run you see?
07:39:31 <mur> oh, btw, Usa is not democracy but republic 
07:39:52 <gilbertdeb> mur: shhhh. we're trying to sell democracy to others be quiet.
07:40:27 <mur> gilbertdeb and that's silly. democracy is BASED on the fact that everyone know what they are voting and the causes of it
07:40:42 <mur> in voting voters shoudl know what voting which means and causes
07:41:13 <gilbertdeb> mur: it is all a matter of marketing. you just want 'voters', which means you do what you can to get more people to agree with you than to disagree with you.
07:41:29 <mur> but that is not democracy. sorry.
07:41:40 <mur> in democracy voters do know the other alternatives too
07:41:59 <mur> openess belongs to democracy. and once again, usa political system is not democracy but republic. :P
07:42:11 <gilbertdeb> mur: sure they do, which is why you get as many people to agree with you as possible.
07:43:12 --- join: hovil (~matt@CommSecureAustPtyLtd.sb1.optus.net.au) joined #forth
07:43:17 --- nick: hovil -> fridge
07:43:18 <rk> gilbertdeb: as far as i know, republic is a kind of democracy
07:43:35 <mur> it's preceeding level of democracy
07:43:42 <rk> mur: as far as i know, republic is a kind of democracy
07:43:45 <mur> it's representative democracy. 
07:43:54 <mur> yes but it is not pure democracy
07:44:00 <mur> it's only partial 
07:44:19 <fridge> is pure democracy where every person votes on every single decision?
07:44:30 <mur> not necessary
07:46:12 <gilbertdeb> who wants pure democracy when you can have impure democracy which sort of works well enough to convince others to use?
07:46:40 <fridge> and what a nasty place it would be if every single person was a politician!
07:47:04 <gilbertdeb> you can't have too many of those, which is why they're voted for to begin with :)
07:47:14 <gilbertdeb> only the most convincing get to be politicians!
07:47:20 <mur> duh
07:47:25 <mur> you got it wrong
07:47:55 <mur> isn't there only one senate member in usa from every area, not many from one larger area, or?
07:48:35 <gilbertdeb> it is better that way, so they can earn more while graciously serving the public.
07:48:53 * rk did not intend to start a conversation about politics
07:49:19 <mur> no that is not good
07:49:33 <gilbertdeb> lets talk about APL
07:49:34 <mur> if there were many from larger area, then there were more choises and larger scope of views
07:49:53 <mur> in such system usually half or so are minority
07:50:00 <mur> that is not very good representative system
07:50:14 <gilbertdeb> it works. there are two different levels.
07:50:17 <rk> alright, chango of topic.......
07:50:28 <gilbertdeb> in one, you have 1 rep per area, in the other, you have more.
07:50:33 <mur> gilbertdeb yeah right, there IS no political parties in USA
07:50:36 <fridge> rk: so about that language forth
07:50:37 <mur> only pressure groups 
07:50:48 <gilbertdeb> mur: now you really understand!!!
07:50:54 <rk> fridge: it is a very intriguing and powerful language.....
07:51:00 <mur> but that's not the idea of politics, gilbertdeb
07:51:09 <mur> because politicians shoudl represent the people
07:51:26 <mur> there shoudl not be external groups to achieve even most basic things
07:51:35 <gilbertdeb> this democracy is controlled by SIGs and the rich/powerful. mur that is an old fashioned idea, and it is one way of interpreting what politicians do ;)
07:51:42 <gilbertdeb> rk:  heard of colorforth?
07:52:38 <rk> gilbertdeb: no. ive only messed a little with gforth.
07:53:09 <gilbertdeb> bah. forth is obsolete, colorforth is the way to go :)
07:53:31 <gilbertdeb> I still don't understand it fully, but those who do get very excited about it.
07:53:41 <rk> gilbertdeb: link?
07:53:56 <gilbertdeb> http://colorforth.com/
07:56:49 <rk> aah
07:59:36 <rk> this seems ... weird ...
07:59:51 <gilbertdeb> pssst. you can't say that about that.
07:59:58 <gilbertdeb> it was created by the creator of forth.
08:03:55 <fridge> I'm not sure how color is superior to punctuation
08:04:36 <gilbertdeb> you can see at a glance what each word's function is for instance.
08:08:25 <TreyB> fridge: colorforth doesn't so much get rid of the punctuation as change its representation.
08:09:41 <gilbertdeb> TreyB: so in other words it is possible to translate from colorforth to punctuated forth and vice versa no?
08:10:10 <TreyB> The division of labor between the compiler and editor has changed in colorforth.  The editor now does all of the parsing, and the compiler essentially does only code generation.
08:10:29 <gilbertdeb> how does the editor do the parsing?
08:13:55 <TreyB> It tokenizes words, converts numbers, etc.
08:14:29 <gilbertdeb> so you have to use coloforth's editor in order to write colorforth?
08:14:34 <TreyB> Yes.
08:14:41 * rk has a broken floppy drive, so tries to use bochs :P
08:19:23 <gilbertdeb> TreyB: aside the editor doing all the work, does the language have other ideas which are relatively uncommon or completely unique?
08:22:25 <TreyB> You can't really call colorforth (at least as Chuck uses it) "just" a language.  You have to look at it as an environment.
08:23:13 <gilbertdeb> s/language/environment/g
08:24:49 <TreyB> I can't think of any that I'd call unique to colorforth.  I don't know of too many modern systems where all of the various ideas live together, though.
08:26:07 <TreyB> You should play with it, if you can.
08:27:30 <TreyB> Have you ever used one of the older PCs (IBM PC, Apple II, TRS-80 Model I) with BASIC in ROM?
08:28:51 <gilbertdeb> emmm I think so.
08:28:57 <gilbertdeb> an old Toshiba.
08:29:24 <gilbertdeb> TreyB: I can never type anything right in it, thats why I'd been hesitating.
08:30:07 <TreyB> colorforth really harkens back to that style environment: instant boot, complete access to the hardware, tokenized source, built-in editor, etc.
08:31:01 <TreyB> Look around for one of the hacks that use a regular keyboard map, or the port to XWindows or Win2K/XP.
08:33:15 <gilbertdeb> what keyboard mapping does it use now? dvorak?
08:34:50 <TreyB> I don't know what to call it.  It uses an input-mode-specific mapping depending on the task.
08:38:09 <rk> can somebody help me? my floppy drive is broken, and im trying to get colorforth working with bochs. i need to make an image. how do i do it?
08:42:30 <gilbertdeb> rk: where are you finding your image?
08:42:39 <gilbertdeb> http://www.oakland.edu/~maslicke/colorforth/vesa-chuck-8bit.tgz 
08:42:57 <rk> gilbertdeb: no, i meant i need to get color.com on a floppy image
08:43:39 <gilbertdeb> have you checked out that url?
08:44:04 <rk> erm
08:44:25 <rk> gilbertdeb: thats not what i need
08:44:41 <gilbertdeb> I'm not of much help.
08:45:56 <rk> TreyB: how are you using it?
08:46:12 <TreyB> colorforth?  I don't, currently.
08:47:06 <rk> TreyB: oh, but when you do? floppy? (which i cant)
08:47:38 <TreyB> I've previously booted from a floppy.  I don't know that I have a usable image, though.
08:49:05 <TreyB> I couldn't find one in the place I normally keep such things.  Apologies.
08:51:19 <gilbertdeb> I can't get the damned thing compiling or booting.
08:51:27 <gilbertdeb> is the language component itself like forth's?
08:54:16 <rk> well, if somebody manages to make a image, send it to me please :P
08:54:18 <rk> bbl
09:34:57 --- quit: mur (Read error: 54 (Connection reset by peer))
09:35:21 --- quit: Fractal (Read error: 60 (Operation timed out))
09:35:57 --- join: mur (murr@baana-62-165-189-186.phnet.fi) joined #forth
09:48:20 --- quit: mur (Read error: 54 (Connection reset by peer))
09:55:36 --- join: mur (murr@baana-62-165-189-186.phnet.fi) joined #forth
10:22:12 <rO|> to try out colorforth *instantly* w/ image etc. needs, Xcolorforth or wincolorforth, worked here adhoc :-)
10:23:29 <rO|> erm, meant "w/o image etc needs"
10:24:50 <rO|> also, there's an alive colorforth chan #c4th on ircnet
10:25:16 --- quit: gilbertdeb ("<--- off to play in the dirt.")
10:25:39 <rO|> tim's cf tftpd just finished
10:26:06 <rk> rO|: where do i find any of that?
10:26:18 --- join: gilbertdeb (gilbert@24.53.192.39) joined #forth
10:26:43 <rO|> afk brb 5 min
10:28:02 <rk> :)
10:30:17 <rO|> rk: on tim neitz's home page
10:31:28 <rO|> http://www.dnd.utwente.nl/~tim/colorforth/
10:32:52 <rO|> the qwerty colorforth is on www.thelma-louise.net
10:33:28 <rO|> i think the xcolorforth is from peter appelmann too
10:37:36 --- join: CrowKilr (~cacaboudi@HSE-Windsor-ppp250496.sympatico.ca) joined #forth
10:37:41 --- quit: CrowKilr (Client Quit)
10:42:04 --- nick: mur -> mur__Bbbl
10:54:51 --- quit: rk ("Client Exiting")
10:58:25 --- nick: mur__Bbbl -> mur
10:58:40 --- join: rk (~rk@ca-cmrilo-docsis-cmtsj-b-36.vnnyca.adelphia.net) joined #forth
11:32:20 --- join: XeF4 (xef4@lowfidelity.org) joined #forth
11:33:05 --- part: gilbertdeb left #forth
13:19:11 --- join: onetom (~tom@novtan.bio.u-szeged.hu) joined #forth
13:31:03 <rk> anybody familiar with colorforth
13:32:06 <XeF4> somewhat
13:32:17 <rk> XeF4: well, i know nothing :)
13:32:31 <rk> i got into the editor, and im starting to understand the keys (somewhat)
13:32:36 <rk> i know nothing else :P
13:33:25 <fridge> load up the hexagons
13:33:32 <fridge> they're very pretty!
13:34:12 <rk> how ?
13:34:40 <fridge> load the block, I can't remember which number it is
13:34:52 <fridge> you'll have to dig through them
13:34:53 <fridge> =(
13:35:34 * rk doesnt even know how to do that
13:35:54 <fridge> can you see the little legend in the corner?
13:36:08 <fridge> of the keymap
13:46:37 <rk> i see the legend, yes
13:46:41 <rk> im in the editor right no
13:46:43 <rk> now*
13:48:02 <fridge> is there a + sign and a - sign?
13:48:15 <fridge> those keys will scroll through the blocks
13:48:52 <rk> he
13:48:55 --- quit: XeF4 ("jalkasieni world domination")
13:49:21 <rk> when i press enter, it shows the + and -, but it just enters them on the current line, no scrolling
13:49:38 <rk> keyboard +\- dont do anything either
13:50:27 <fridge> its hard to describe, but try and find +/- in a different mode if you can
13:51:22 <rk> i found a keypad legend, underneath is "-a9"
14:02:34 <rk> thats not it :( i cant find it
14:36:47 --- quit: rO| (orwell.freenode.net irc.freenode.net)
14:36:47 --- quit: w1k1 (orwell.freenode.net irc.freenode.net)
14:36:47 --- quit: GnuVince (orwell.freenode.net irc.freenode.net)
14:36:48 --- quit: TreyB (orwell.freenode.net irc.freenode.net)
14:36:48 --- join: rO|_ (~rO|@pD9545C06.dip.t-dialin.net) joined #forth
14:38:13 --- join: GnuVince (~vince@24.200.188.158) joined #forth
14:47:08 --- quit: rO|_ (Read error: 104 (Connection reset by peer))
14:57:41 --- quit: fridge ("http://lice.codehack.com")
14:58:49 --- join: w1k1 (~w1k1@pD9545C06.dip.t-dialin.net) joined #forth
15:08:25 --- join: rO| (~rO|@pD9545C06.dip.t-dialin.net) joined #forth
15:39:41 --- join: rO|_ (~rO|@pD954504F.dip.t-dialin.net) joined #forth
15:40:19 --- join: w1k1_ (~w1k1@pD954504F.dip.t-dialin.net) joined #forth
15:42:37 --- quit: rO| (Killed (NickServ (ghosted: rO|_!~rO|@pD954504F.dip.t-dialin.net)))
15:42:56 --- nick: rO|_ -> rO|
15:44:03 --- quit: w1k1 (Killed (NickServ (ghosted: rO|!~rO|@pD954504F.dip.t-dialin.net)))
15:44:05 --- nick: w1k1_ -> w1k1
16:13:25 --- quit: mur ("warped to dream time")
17:04:20 --- join: suprdupr (CrowKilr@HSE-Windsor-ppp251574.sympatico.ca) joined #forth
18:11:53 --- join: kc5tja (~kc5tja@ip68-8-127-122.sd.sd.cox.net) joined #forth
18:12:18 <kc5tja> Wow.  Did I just walk in on a netsplit?
18:12:58 --- quit: suprdupr (orwell.freenode.net irc.freenode.net)
18:12:58 --- quit: w1k1 (orwell.freenode.net irc.freenode.net)
18:12:58 --- quit: rO| (orwell.freenode.net irc.freenode.net)
18:12:58 --- quit: skylan (orwell.freenode.net irc.freenode.net)
18:12:58 --- quit: ian_p (orwell.freenode.net irc.freenode.net)
18:12:59 --- quit: rk (orwell.freenode.net irc.freenode.net)
18:12:59 --- quit: ChanServ (orwell.freenode.net irc.freenode.net)
18:12:59 --- quit: onetom (orwell.freenode.net irc.freenode.net)
18:13:00 --- quit: GnuVince (orwell.freenode.net irc.freenode.net)
18:13:19 --- quit: kc5tja (Client Quit)
18:14:07 --- join: ChanServ (ChanServ@services.) joined #forth
18:14:07 --- join: kc5tja (~kc5tja@ip68-8-127-122.sd.sd.cox.net) joined #forth
18:14:07 --- join: suprdupr (CrowKilr@HSE-Windsor-ppp251574.sympatico.ca) joined #forth
18:14:07 --- join: w1k1 (~w1k1@pD954504F.dip.t-dialin.net) joined #forth
18:14:07 --- join: rO| (~rO|@pD954504F.dip.t-dialin.net) joined #forth
18:14:07 --- join: GnuVince (~vince@24.200.188.158) joined #forth
18:14:07 --- join: onetom (~tom@novtan.bio.u-szeged.hu) joined #forth
18:14:07 --- join: rk (~rk@ca-cmrilo-docsis-cmtsj-b-36.vnnyca.adelphia.net) joined #forth
18:14:07 --- join: skylan (sjh@vickesh01-4724.tbaytel.net) joined #forth
18:14:07 --- join: ian_p (ian@inpuj.net) joined #forth
18:14:07 --- mode: orwell.freenode.net set +o ChanServ
18:14:08 --- mode: ChanServ set +o kc5tja
18:17:22 <suprdupr> hi there
18:17:42 <kc5tja> re
18:17:56 <suprdupr> i got a better semicolon, that compiles short jumps if possible
18:18:08 <suprdupr> its posted at the usual place
18:19:13 <kc5tja> I thought semicolon only compiled a return.
18:19:21 <kc5tja> At any rate, that's the least of my worries at the moment.
18:19:25 <kc5tja> I need cash, and I need it now.
18:19:39 <kc5tja> My fscking roommate just dumped a $500 past-due bill on me.
18:19:49 * kc5tja has all of $82 in my account right now.
18:20:05 <kc5tja> Especially after footing the majority of the rent myself.
18:20:35 <kc5tja> And what sucks, is I can't find a suitable Linux distribution to download and install.
18:20:51 * kc5tja likes Debian, but they've got their problems, and are actually behind the curve on a number of packages.
18:21:17 <kc5tja> RH9 is getting a LOT of good reviews recently, but I can't download an ISO without spending $60 (which I don't have) to join their RH Network.
18:26:09 <suprdupr> maybe you could use damn small linux ;o)
18:26:23 <kc5tja> Not an option.
18:26:35 <suprdupr> that bill thing is not cool, even at work we passed an inch away of loosing our electricty
18:27:01 <kc5tja> Well, I passed -><- that close to getting my Internet cut off, which means I'd have no means of checking my e-mail for orders.
18:27:02 <suprdupr> the bill was lost, no one knows how, a check of 281$ was made to the guy supposed to cut the juice
18:27:28 <suprdupr> internet is essential nowadays ;p
18:30:01 <kc5tja> Since I'm in business and relying on my website for my advertising, e-mail for business communications, etc., yes, it IS important to me.
18:32:58 --- join: rO|_ (~rO|@pD954504F.dip.t-dialin.net) joined #forth
18:33:31 --- quit: w1k1 (Read error: 54 (Connection reset by peer))
18:33:52 --- join: w1k1 (~w1k1@pD954504F.dip.t-dialin.net) joined #forth
18:33:59 --- quit: rO| (Read error: 54 (Connection reset by peer))
18:40:32 * kc5tja isn't convinced that Slackware is what I'm looking for.
18:40:47 <kc5tja> Although that's what I "grew up" on, it also seems like it'd be more trouble than it's worth...
18:41:47 --- join: Herkamire (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth
18:44:06 * suprdupr was initiated by a slackware using guy
18:44:50 <suprdupr> live cds are great, but none is configurable enough easly enough for me
18:45:18 <Herkamire> configureable to do what?
18:45:55 <suprdupr> okok they are most of the time configurable enough through rc.d scripts
18:46:08 <suprdupr> and knoppix got his own knoppix.sh script i think
18:46:35 <suprdupr> the only one I started to like was trinux
18:46:43 <suprdupr> and i cant wait to see the result of the new spinoff
18:46:47 <suprdupr> toward embedded system
18:46:51 <suprdupr> dev.
18:46:51 <Herkamire> suprdupr: what do you want to do with a live CD?
18:48:27 * suprdupr is thinking...
18:49:23 <suprdupr> work on the platform I have to reverse-engineer drivers to work on the PCI bus, using NASM with confortable editor
18:49:42 <suprdupr> to develop forth code ;p
18:57:47 <Herkamire> suprdupr: what on earth are you talking about?
19:00:56 <suprdupr> nothing really important
19:02:13 <Herkamire> if you want to look at UNIX source code for drivers I've heard a bunch of people say that many BSD sources are easier to read than linux
19:02:43 <Herkamire> netBSD was reccomended to me for PPC sources
19:02:50 <kc5tja> I think Slackware is the distribution I'll end up using.
19:03:00 <kc5tja> Nothing else seems to be as rooted in reality.
19:03:23 <suprdupr> herk: thanks for the tip
19:10:46 <suprdupr> slackware-live.org might be what I want
19:14:32 <suprdupr> in about an hour I'll know ;p
19:14:41 <suprdupr> 180mb slackware based live cd
19:15:06 * kc5tja is attempting to download slackware 9's ISO now.
19:17:56 <suprdupr> i got a cd from a magazine with slacware 9 but the install crashed...
19:34:31 <kc5tja> Where did it crash?
19:34:41 * kc5tja has never had a Slack install ever crash on me in the past.
19:34:49 <kc5tja> You do, however, have to pick the correct kernel to boot with.
19:42:58 <kc5tja> One of the things I absolutely adore about Slackware is its package management system.
19:43:11 <kc5tja> It is damn simple.
19:43:15 <suprdupr> the cd was bad
19:43:24 <kc5tja> And it leaves me in total control.
19:43:30 <kc5tja> It can use some improvements, though.
19:44:07 <kc5tja> But such improvements are quite possible without changing a whole lot.
19:44:28 <suprdupr> what are they?
19:46:01 <kc5tja> What are what?
19:47:17 <suprdupr> your enhancements
19:47:30 <kc5tja> Slackware's packages are .tar.gz files, basically.
19:47:44 <kc5tja> Installer meta-data are stored in a specially recognized directory: /install
19:47:51 <kc5tja> Any shell scripts are executed there.
19:47:54 <suprdupr> tar.bz2??
19:48:16 <kc5tja> However, to be generally useful, most installers require the ability to execute pre- and post-install scripts, as well as pre- and post-uninstall scripts.
19:48:52 <kc5tja> The way Slackware treats packages is it will untar them directly on top of / , so there is no chance of executing a pre-install script.
19:49:02 <suprdupr> why not two simple install and uninstall scripts
19:49:06 --- join: Herkamir1 (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth
19:49:21 <kc5tja> All that's needed is to install them somewhere off of the root directory first, so that the installer has a chance to run pre-install scripts.
19:49:30 <kc5tja> suprdupr: You're not thinking clearly.
19:49:43 <kc5tja> Where do these install scripts come from?
19:49:54 <suprdupr> the install could just run install from install and uninstall from a certain package to make the change in a system
19:50:03 <suprdupr> whooo
19:50:14 <kc5tja> And how does it do that?
19:50:28 <kc5tja> Isn't that a bit of a chicken-and-egg problem?
19:50:56 <suprdupr> I don't see any
19:51:06 <suprdupr> just unpack in temp dir
19:51:09 <suprdupr> then run install
19:51:20 <kc5tja> And is this NOT what I just got done saying Slackware doesn't do?  ;)
19:51:41 <suprdupr> then i argue that one dont need preinstall scripts
19:51:43 <suprdupr> ;p
19:51:55 <kc5tja> You are easily proven quite and utterly wrong.
19:52:05 --- quit: Herkamire (Read error: 54 (Connection reset by peer))
19:52:06 <kc5tja> Pre-install scripts are exceptionally handy.
19:52:21 <suprdupr> previous to what exactly?
19:52:30 <kc5tja> They prepare configuration files, verify dependencies, and more.
19:52:52 <suprdupr> that's the job of the install script
19:52:57 * kc5tja sighs!!!
19:53:12 <kc5tja> It can't DO that if the files are already copied over critical already-in-use files!!!
19:53:52 <suprdupr> didnt i said that files could be extracted in a temp directory and then the install could enforce dir structure? ;p
19:54:03 <kc5tja> SLACKWARE DOESN'T DO THIS!!!
19:54:07 <kc5tja> Good grief!!
19:54:11 <kc5tja> How many times must I say this!?!?!
19:54:19 <kc5tja> Forget I ever mentioned anything.
19:54:34 <suprdupr> i looked in a tar ball before... youre the one making me saying things
19:54:46 <suprdupr> I didnt state that what i described was slackware's way
19:54:52 <suprdupr> its all in your head ;p
19:54:52 <kc5tja> No, sorry, if you look on http://www.slackware.com, you'll find my statements are 100% factual and to the point.
19:55:14 <kc5tja> I'm not compelling you to say anything.
19:55:20 <kc5tja> You take responsibility for your own actions.
19:55:37 <suprdupr> [22:50] <suprdupr> I don't see any
19:55:37 <suprdupr> [22:51] <suprdupr> just unpack in temp dir
19:55:37 <suprdupr> [22:51] <suprdupr> then run install
19:55:37 <suprdupr> [22:51] <kc5tja> And is this NOT what I just got done saying Slackware doesn't do?  ;)
19:55:45 <suprdupr> and then
19:55:54 <suprdupr> [22:53] <kc5tja> It can't DO that if the files are already copied over critical already-in-use files!!!
19:55:54 <suprdupr> [22:53] <suprdupr> didnt i said that files could be extracted in a temp directory and then the install could enforce dir structure? ;p
19:55:54 <suprdupr> [22:54] <kc5tja> SLACKWARE DOESN'T DO THIS!!!
19:56:02 <suprdupr> you're hard to follow
19:56:08 <kc5tja> I'm hard to follow?!
19:56:21 <kc5tja> I don't think so.
19:56:50 <kc5tja> I mentioned very succinctly and clearly the method Slackware uses to extract and install package files.
19:57:16 <kc5tja> I then went on to say, "I would have it install in a temporary directory, so that it can run pre-installer scripts." (paraphrased)
19:57:33 <suprdupr> and i stated a solution to the bloated pre/post concept
19:57:43 <kc5tja> You did absolutely nothing of the sort.
19:57:43 <suprdupr> ha ok
19:57:54 <kc5tja> You just said, "OK, we don't need pre-install scripts then."
19:57:55 <suprdupr> i just call that script
19:58:01 <suprdupr> an install script
19:58:05 <suprdupr> its not e install
19:58:08 <suprdupr> it DOES the mv
19:58:22 <suprdupr> moves, indeed
19:58:45 <suprdupr> there's no need for a pre-prefix to the name of that script ;p
19:59:17 <kc5tja> For the purposes of preparing the system as a whole for installing a package, I think there should be.
20:00:34 <kc5tja> Every successful installer and package manager has them.  Every single one.  There's a reason for it.
20:00:42 <kc5tja> Likewise, pre-uninstall and post-uninstall.
20:01:04 <suprdupr> why preuninstall and post uninstall if the script can do the pre, the uninstall and the post?
20:01:32 <suprdupr> giving total control to the script makes the package management easier
20:02:08 <kc5tja> Total control is given to the scripts.
20:02:33 <kc5tja> Asking that question is like asking why we break programs up into subroutines.
20:02:56 <suprdupr> pre install ----??????????????????????????????? post-install
20:03:05 <suprdupr> i dont see the point of doing this
20:03:17 <kc5tja> We have a pre-uninstall script to prepare the system for package removal.  What if we're removing an FTP server, and the server daemon is still running?  We need to kill it.
20:03:32 <suprdupr> it can be made in the uninstall script
20:03:36 <kc5tja> /etc/passwd
20:04:08 <kc5tja> How do you handle the situation where you're installing a package that overwrites /etc/passwd?
20:04:19 <kc5tja> Or, for that matter, removing the package that contains /etc/passwd?
20:04:49 <suprdupr> i dont know at all, i would give contl to the uninstall procedure script written by the guy who knows
20:06:09 <kc5tja> Nice try, but no cigar.  You are the author.
20:09:07 <suprdupr> then i write the damn script without caling it fancy no use names like pre and post
20:09:14 <suprdupr> ?!?
20:09:27 <Herkamir1> suprdupr: shut up.
20:09:31 <suprdupr> i dont understand whats not understandable in my idea
20:09:42 <Herkamir1> suprdupr: you obviously have no idea what your talking about.
20:10:04 --- nick: Herkamir1 -> Herkamire
20:10:04 <suprdupr> k then tell me what happens between pre and post
20:10:47 <kc5tja> For installing, what happens between pre and post install is copying of file data to their proper locations, including any decompression that needs to be done.
20:10:53 <kc5tja> For removing, well, deletion.
20:11:38 <kc5tja> But if you're installing a new package that has its own /etc/passwd, you certainly do NOT want your existing /etc/passwd to be overwritten.
20:11:46 <suprdupr> and THAT cant be coded inside a script? and THAT cant be the answer to this non-problem of package managment?
20:12:10 <kc5tja> Of course it can be coded inside a script.  But it'll probably be handled in a pre- or post-install script.
20:14:20 <kc5tja> Like I said, never mind though.  Forget I mentioned anything.
20:14:40 <suprdupr> I dont see the need for this; thats precisely what im going to do ;p
20:17:57 <suprdupr> It just surprises me how you're inside the "installer: MUST AT ALL COST COPY FILE TO BE AN INSTALLER" illusion. SImple package management system good for every distribution, simple scripts doing their jobs, namely using.. run configure make make install and report any errors, with the most thoughtful curse of action printed on screen
20:18:17 <suprdupr> s/curse/course
20:19:54 <suprdupr> if there's some extras, well they're written in install and uninstal scripts, with comments indicating why they're there, the two scripts are in pre-install-post phase since theyr are what their name is: actual actions
20:20:38 <kc5tja> The problem is, that's been tried for years, and it just doesn't work.
20:20:54 <suprdupr> why?
20:21:02 <suprdupr> why oh why?
20:21:14 <kc5tja> Because it doesn't handle overwriting critical files gracefully.
20:21:28 <kc5tja> Because it doesn't gracefully handle the case of dealing with already running instances of programs.
20:21:44 <suprdupr> kill?
20:21:53 <kc5tja> Because it doesn't gracefully handle the case of updating existing configuration files.
20:22:08 <kc5tja> Last I checked, GNU Configure runs on AmigaOS as well as Windows NT systems, neither of which have 'kill.'
20:22:54 <kc5tja> But, whatever.
20:22:58 <kc5tja> It's not important.
20:24:15 <kc5tja> My point is, and what you got all upset about, hence getting ME upset, is that Slackware does not install its package files to a temporary directory
20:24:32 <suprdupr> well it should, and it would be better
20:24:39 <kc5tja> But it doesn't.
20:24:45 <suprdupr> so slack sucks
20:24:57 <suprdupr> as every other linux around
20:24:57 <kc5tja> No, it just doesn't have a cool package manager.
20:25:06 <kc5tja> Slack as a distribution doesn't suck any more than Debian or RH.
20:25:23 <suprdupr> the more and more I learn the more and more those systems become ugly and bloated
20:25:34 <kc5tja> Slack is one of the least bloated.
20:25:38 <suprdupr> i know
20:25:42 <kc5tja> That's why I'm using it.
20:25:47 <kc5tja> err...going to use it.
20:25:58 <kc5tja> It'll be another 10 hours to grab the ISO.
20:26:11 <suprdupr> ill try my 180mb live iso right away
20:26:44 <kc5tja> One of the reasons I am going for Slack, ironically, precisely is its package management system.
20:26:44 <suprdupr> if i come back it will be in linux, if not well it will wait tommorow
20:27:24 <kc5tja> It lets me create my own proprietary solutions and distribute them to my customer in a convenient package that they can install.
20:28:47 <suprdupr> you mean compile?
20:29:03 <suprdupr> or plain copying install?
20:29:04 <kc5tja> No, I mean install.
20:29:21 <kc5tja> I don't want my customers owning proprietary source code.
20:29:47 <kc5tja> Not for something I spent hours and hours writing on their behalf.
20:30:46 <suprdupr> the truth shall set you free, and others too
20:30:53 <suprdupr> transparency is a must
20:31:00 <suprdupr> you will get free publicity
20:31:07 <kc5tja> I don't want free publicity.
20:31:10 <kc5tja> I want MONEY>
20:31:15 <kc5tja> Last I checked, I had $82 to my name.
20:31:20 <kc5tja> Across ALL accounts that I have.
20:31:21 <suprdupr> give and you will receve
20:31:29 <kc5tja> bull.
20:31:36 <kc5tja> Capitalist economies don't work that way.
20:31:51 <kc5tja> And until I have enough to make rent for next month, my first and foremost fear is losing my house.
20:33:36 <suprdupr> sell your house and live in a van
20:33:53 <kc5tja> Umm..I live in an apartment.
20:33:56 <kc5tja> It's not mine to sell.
20:33:57 <suprdupr> put a vaporizing carburettor in it, or make it run on compressed air
20:34:05 * kc5tja sighs
20:34:11 <kc5tja> Whatever.
20:34:16 <kc5tja> You don't know, you don't care.
20:34:20 <kc5tja> So, never mind.
20:34:21 <suprdupr> build your MSIG generator, power it with a big wheel and a pedal, like old sewing machines
20:34:23 --- nick: kc5tja -> kc-food
20:34:58 <suprdupr> I care, im just pulling out stuff
20:35:49 --- quit: suprdupr ("opensource shall set you free")
20:37:14 --- join: gilbertdeb (gilbert@fl-nken-u2-c3b-50.miamfl.adelphia.net) joined #forth
20:48:22 --- quit: gilbertdeb (Remote closed the connection)
20:59:16 --- join: gilbertdeb (gilbert@fl-nken-u2-c3b-50.miamfl.adelphia.net) joined #forth
20:59:17 --- join: a7r (~a7r@206.72.82.135) joined #forth
21:00:20 --- join: TreyB (~trey@cpe-66-87-192-27.tx.sprintbbd.net) joined #forth
21:05:07 --- quit: kc-food (orwell.freenode.net irc.freenode.net)
21:05:32 --- join: kc-food (~kc5tja@ip68-8-127-122.sd.sd.cox.net) joined #forth
21:05:32 --- mode: orwell.freenode.net set +o kc-food
21:05:36 --- mode: ChanServ set -o kc-food
21:11:40 --- quit: kc-food (orwell.freenode.net irc.freenode.net)
21:12:37 --- join: kc-food (~kc5tja@ip68-8-127-122.sd.sd.cox.net) joined #forth
21:12:39 --- quit: kc-food (orwell.freenode.net irc.freenode.net)
21:12:40 --- join: kc-food (~kc5tja@ip68-8-127-122.sd.sd.cox.net) joined #forth
21:40:22 --- nick: kc-food -> kc5tja
21:40:27 --- mode: ChanServ set +o kc5tja
21:40:32 <kc5tja> Back
21:42:44 <a7r> sup 
21:45:37 * kc5tja is attempting to grab the Slackware 9.0 ISOs.
21:45:42 <kc5tja> Not going too well at the moment though.
21:45:48 <kc5tja> Looks like many hours of download time.
21:46:45 <gilbertdeb> ack. I'm trying at this moment to get the openbsd.iso
21:49:21 * kc5tja is going for Slackware Linux because it's simple, to the point, no nonsense, and I'm already familiar with Linux.
21:50:32 <kc5tja> Plus, I really like Slackware's package management system.
21:50:52 <gilbertdeb> I prefer debian and freebsd's.
21:51:06 <gilbertdeb> they both accept urls.
21:51:11 <a7r> kc5tja: check it http://www.aereon.net/a7r/outgoing/tronics/IMG_0493.jpg 
21:51:18 <kc5tja> BSD's ports system is the best I've seen so far, but it's not conducive for my requirements.
21:51:43 <kc5tja> Debian's system is a mess now, and that's the most polite way I can think of putting it.
21:52:15 <a7r> it's up and running.
21:52:22 * kc5tja would rather let wget handle the URLs -- as a user of Debian since 1.3, I can safely say that Debian's package management system is NOT all flowers.  In fact, in recent years, it's come to rather suck eggs.
21:52:32 <kc5tja> a7r: Checking
21:53:15 <kc5tja> Nothing's coming up.
21:56:17 <kc5tja> Operation timed out.
22:02:35 --- join: Herkamir1 (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth
22:02:38 --- quit: Herkamire (Read error: 104 (Connection reset by peer))
22:03:08 <kc5tja> wb
22:12:27 --- join: Serg_Penguin (Serg_Pengu@212.34.52.140) joined #forth
22:12:31 <Serg_Penguin> hi
22:12:44 <kc5tja> re
22:15:37 <Serg_Penguin> did you ever progged gfx in Forth ?
22:15:45 <kc5tja> No
22:15:56 <Serg_Penguin> gfx at all ?
22:15:58 <kc5tja> FS/Forth will be my first attempt, and even then, it'll be direct framebuffer.
22:15:59 <kc5tja> No.
22:16:07 <kc5tja> Oh, graphics at all, yes.
22:16:16 * kc5tja grew up on the Commodore 64 and Amiga, after all.  :)
22:16:53 <Serg_Penguin> now i mess w/ ASCII editor and wanna make an 'underdraw' option
22:16:57 <Serg_Penguin> imagine:
22:17:11 <Serg_Penguin> you xerox a picture on plastic,
22:17:20 <Serg_Penguin> pot it over monitor
22:17:23 <Serg_Penguin> pUt
22:17:33 <Serg_Penguin> and draw ASCC pic under it
22:17:39 <gilbertdeb> ?
22:17:48 --- nick: Herkamir1 -> Herkamire
22:18:01 <Serg_Penguin> so, i wanna do this in software - ASCII picture under image at same time
22:18:08 <kc5tja> That process of overlaying like that is called 'onion skinning' in the animation industry.
22:18:37 <gilbertdeb> weird. 
22:18:49 <gilbertdeb> the moment you said 'overlaying' someone in the #irix room said 'overlay'.
22:19:24 <kc5tja> Heh
22:19:38 * Serg_Penguin regrets what PC doesn't have GFX and hardware chargen at same time ;)
22:19:52 <kc5tja> Nothing except the Atari 7800 does.
22:19:57 <gilbertdeb> what is GFX?
22:20:12 <kc5tja> gilbertdeb: Gra(f)phiX.
22:20:30 <gilbertdeb> oohhhh.
22:21:31 <Serg_Penguin> i gonna do it in VGA hi-res mode, by messing w/ palette
22:21:42 <Serg_Penguin> or maybe VESA 640x480
22:22:20 <Serg_Penguin> but the Forth what has VESA lib is thunk by wrong bodypart ;((
22:23:00 --- quit: a7r (Read error: 110 (Connection timed out))
22:23:14 <kc5tja> One more reason to appreciate bitplanes -- they make onionskinning super easy.
22:23:27 <kc5tja> They also make CAD/technical representations with multiple layers pretty easy too.
22:23:31 <kc5tja> (easier, I should say)
22:23:43 * kc5tja really misses bitplanes.  :/
22:23:44 <gilbertdeb> who uses bitplaning?
22:23:48 <Serg_Penguin> so, VGA hires
22:24:36 <kc5tja> gilbertdeb: Amiga, VGA (not necessarily SVGA), early Macintoshes, and the Atari ST (in a really bizarre, highly bastardized way).
22:25:25 <gilbertdeb> kc5tja: how old?
22:25:41 <kc5tja> Amigas and STs were using bitplanes will into the 90s.
22:25:55 <kc5tja> Bitplanes aren't a new concept though.  The concept dates back at least to the 70s.
22:26:24 <gilbertdeb> what about the macs, how old?
22:26:33 <kc5tja> Mac II series.
22:26:41 <gilbertdeb> se/30 ?
22:26:41 <kc5tja> Prior to the fx at least.
22:26:49 <kc5tja> Mac SE was monochrome.
22:26:56 <kc5tja> I suppose it was a bitplane.
22:26:57 <kc5tja> :)
22:27:12 <kc5tja> s/was/had/
22:27:13 <gilbertdeb> ah so the mac color classic then?
22:27:39 <kc5tja> Some SGI machines had bitplanes too -- up to 10 bitplanes deep (1024 onscreen colors)
22:27:48 * kc5tja forgets which models though.
22:30:26 <gilbertdeb> kc5tja: I'm thinking hard about getting a ye olde mac and making that my real toy box.
22:30:50 <gilbertdeb> ah it appears all the sgi's have bitplaning.
22:31:00 * kc5tja doesn't know much about the older Mac hardware.  I never played with them extensively.
22:31:26 <kc5tja> Bitplanes are, bar none, the most versatile method of storing and manipulating graphics for a computer.
22:31:34 <kc5tja> It's also the most compact method too.
22:31:45 <gilbertdeb> so why isn't it more popular?
22:31:57 <kc5tja> Beats me.
22:32:04 <kc5tja> I think it's laziness on the part of the computer programmers.
22:32:51 <Herkamire> perhaps bit planes aren't what I was thinking of.
22:33:16 <Herkamire> in one plane would you have say, the most significant bit of the green amount in each pixel?
22:33:16 <kc5tja> A bitplane is just a monochrome bitmap.
22:33:23 <kc5tja> Yes.
22:33:28 <Herkamire> neat
22:33:33 <gilbertdeb> kc5tja: laziness? its not a hardware thing?
22:33:35 <kc5tja> So in a 256 color image, you'd have 8 monochrome bitmaps.
22:33:46 <kc5tja> gilbertdeb: No.  Hardware actually prefers bitplanes.
22:33:47 <Herkamire> cool
22:34:03 <Herkamire> bitplanes sounded horrible when I first heard of them
22:34:03 <kc5tja> In a 24-bit-per-pixel image, you'd have 24 monochrome bitmaps, etc.
22:34:17 <kc5tja> From a software standpoint, they can be somewhat of a pain to deal with.
22:34:30 <kc5tja> I say "somewhat," because I think their advantages greatly overshadow their disadvantages.
22:34:39 <kc5tja> For example, much more compact representation of images.
22:34:43 <gilbertdeb> ah so do all video cards come with bitplaning abilities or something?
22:34:44 <kc5tja> (each plane can be independently compressed)
22:34:45 <Herkamire> but now I'm realizing that you don't have to deal with all the planes if you don't want to
22:35:19 <kc5tja> gilbertdeb: No.  Nowadays, hardware is optimized for chunky display methods.
22:35:25 <kc5tja> Herkamire: Precisely.  :)
22:35:51 <Herkamire> it would be a bit of a pain for drawing lines in bright colors
22:36:00 <kc5tja> Herkamire: Why?
22:36:20 <Herkamire> to get full brightness you have to set the bit for that pixel in each plane right?
22:36:28 <kc5tja> Yes
22:36:43 <Herkamire> you'd have a routine for that of course, but it just seems complex
22:36:46 <kc5tja> You have to do that regardless of its brightness though (unless you already knew that the required bits were already set or not set)
22:37:22 <kc5tja> Well, now-a-days, you'd probably have the hardware draw the line for you.
22:37:48 <Herkamire> :)
22:38:15 * kc5tja notes that anything that can be done with a chunky display, can be done in exactly the same amount of time with a planar display
22:38:25 <kc5tja> Assuming, of course, all other things are equal.
22:39:00 <Herkamire> mmmm. I still don't like it.
22:39:14 <Serg_Penguin> heh, if i wanna PUTPIXEL in planar 24 truecolor, i gotta mess w/ 24 bytes by bitmask !
22:39:16 <Herkamire> I like the idea of being able to change a pixel with a single memory store
22:39:33 <kc5tja> Well, don't PUTPIXEL unless you have to.
22:39:44 <kc5tja> You don't want to PUTPIXEL in chunky modes usually either.
22:39:55 <Serg_Penguin> and w/ chunky truecolor, only "! C!"
22:40:03 <Herkamire> kc5tja: what do you mean by "chunky" exactly?
22:40:22 <kc5tja> Single plane, 8, 16, 24, or 32-bits per pixel.
22:40:34 <kc5tja> Not so much a bitmap as it is a bytemap or wordmap.
22:41:20 <Serg_Penguin> how can i don't PUTPIXEL if i wanna draw a line ?
22:41:34 <kc5tja> Serg_Penguin: You draw multiple pixels.
22:41:45 <Serg_Penguin> ??????
22:42:03 <kc5tja> With a chunky display, you MUST use the equivalent of PUTPIXEL while drawing a line, so it's actually rather slow.
22:42:29 <kc5tja> With a planar display, each write to memory can change a multitude of bits at a time, so it's actually makes up for all the bitmasking.
22:42:43 <kc5tja> For lines with slopes less than 0.5, it's actually faster.
22:42:50 <Serg_Penguin> if i draw slanted line, i gotta find out what bytes i wanna hit, and mock up bitmask !
22:43:11 <Serg_Penguin> it thrash slow !
22:43:18 <kc5tja> Actually, you typically would (a) alternate between two bitmasks, and (b) ROTATE the bitmasks, not compute them from scratch every time.
22:43:34 <kc5tja> Look up the line-slice algorithm for drawing lines.
22:44:01 <kc5tja> It's a refinement of Bresenham's algorithm that takes into consideration the SIMD-nature of bitmaps.
22:44:12 <Serg_Penguin> anyway, i would prefer to have BOTH modes
22:44:33 * Herkamire is still struggling to see the usefullnes/convenience of bitplanes
22:44:59 <Serg_Penguin> planar if different info per plane (like 2 pics to be matched)
22:45:17 <Serg_Penguin> and chunky - for usual gfx, runtime-computed
22:46:02 <kc5tja> Herkamire: From a final output point of view, it doesn't matter.
22:46:27 <kc5tja> From a programmer's perspective, planes makes vastly more sense because it compresses better, and because it becomes very easy to implement blitter hardware.
22:46:49 <Serg_Penguin> IMHO, softwarely simulating plananar from chunky is easier than v-v
22:47:06 <kc5tja> What is v-v?
22:47:22 <Serg_Penguin> visa-versa, if my Latin is not rotten ;)
22:47:40 <kc5tja> What difference does it make?
22:47:41 <Herkamire> the only place I can see planar beeing more convenient is with some transparency stuff where you don't care about exactly how bright stuff is
22:47:52 <kc5tja> Both require the same amount of CPU bandwidth when converting one form to the other.
22:48:15 <kc5tja> Herkamire: Don't forget color palette manipulation too.
22:48:25 <Herkamire> most situations I can think of you know what color you want some pixels to be, and it seems a major pain to have to change memory in 24 places for each of those pixels.
22:48:26 <kc5tja> You can manipulate palette registers to effectively render unused plane bits ... useless.
22:48:42 <Serg_Penguin> but for my mind, chunky is more convenient ;))
22:48:43 <Herkamire> oh
22:48:44 * kc5tja sighs
22:48:52 <kc5tja> Whatever.
22:48:53 <kc5tja> Never mind.
22:49:00 <kc5tja> This seems to be happening to me a lot.
22:49:04 <kc5tja> Maybe I should just go.
22:49:09 <gilbertdeb> whats that?
22:49:32 <Herkamire> so you could set one blue bitplane to make it 100% or 0% blue
22:49:42 <kc5tja> Right.
22:49:45 <kc5tja> I've done it plenty of times.
22:49:49 <Herkamire> ok, now I'm with you
22:49:55 <kc5tja> Although, in a palette system, you don't have a blue plane.
22:50:01 <kc5tja> Unless you set the palette that way.
22:50:19 <Serg_Penguin> palette is cool ^)
22:50:21 <kc5tja> E.g., in a 10-plane system, you have 1024 colors at your disposal (assuming no hardware decompression techniques like Amiga's HAM modes)
22:50:35 <Herkamire> you just have 24 planes that can add any amount to any of R,G,B
22:51:09 <Herkamire> can you have a pane affect R,G and B?
22:51:10 <kc5tja> But you're arguing against that.
22:51:30 <kc5tja> If I'm going to use 24 bitplanes, why would I support a palette?
22:52:06 <Herkamire> ok 10 bit planes I don't care
22:52:27 <Herkamire> 24 bit planes seems silly
22:52:43 * kc5tja shrugs
22:52:48 <kc5tja> Personally, I don't think so.
22:53:00 <kc5tja> But no further progress is being attained in this discussion.
22:53:09 <Herkamire> if you don't have a pallete and you want to make something white, you have to change 24 places in memory
22:53:28 <kc5tja> You have to do that anyway, regardless of the existance of a palette.
22:53:36 <kc5tja> (to set a pixels significant planes)
22:54:11 <kc5tja> But, remember, you're also capable of changing up to 8, 16, or 32 pixels at a time per write.
22:54:43 <Serg_Penguin> but be4 write, u gotta compute that word, messing w/ bitmask, SHIFT etc
22:54:43 --- part: gilbertdeb left #forth
22:54:46 <kc5tja> So it really will only show when you are changing a single pixel.
22:55:07 <kc5tja> Either don't change single pixels often, or use hardware acceleration.
22:55:11 <Serg_Penguin> SIMD-nature IMHO helps only mass-blitting
22:55:38 <kc5tja> Serg_Penguin: That's why AmigaOS's graphics for an 7MHz machine out-performed 25MHz processors of the day in graphics.
22:55:52 <Serg_Penguin> how the hell can it help me do fire effect ?
22:56:11 <Serg_Penguin> function grath ?
22:56:14 <kc5tja> I dunno -- that is not my problem.  :)
22:56:40 <kc5tja> Even on the PC, most graphics-related software keeps a separate, off-screen pixmap anyway, so why not also for a planar system?
22:56:52 * Herkamire is envisioning some cool stuff you could do whith bitplanes and changing the palette
22:57:04 <Serg_Penguin> sprite games on blit - good
22:57:15 <Serg_Penguin> 3D and sci/ CAD - poor
22:57:18 * kc5tja misses good sprite hardware.
22:57:42 <kc5tja> ??!!  AmigaOS had perfectly good CAD programs for it.  I'm *really* not sure where you've derived that comment from.
22:58:08 <Herkamire> ok, time to sleep on it
22:58:12 <kc5tja> As far as 3D is concerned, that's inconsequential -- that's a factor of the hardware accelerator, which as I indicated earlier, couldn't care less about whether it's using chunky or planar.
22:58:28 <Serg_Penguin> i derived it from what CAD uses lotsa PUTPIXEL and not much BLIT
22:58:34 <kc5tja> Repeat please?  I /clear'ed.
22:58:37 <Herkamire> I was damn impressed by the amiga's graphics
22:58:49 <Serg_Penguin> i derived it from what CAD uses lotsa PUTPIXEL and not much BLIT
22:59:02 <kc5tja> One nice advantage of the planar system is this: Drawing to arbitrary color depths is trivial.
22:59:19 <kc5tja> As it is today, the PC needs something like 8 to 10 different "PUTPIXEL" routines, depending on the drawing mode.
22:59:28 <Herkamire> kc5tja: good point
22:59:45 <kc5tja> Bitmap, planar, chunky palette, RGB555, RGB565, RGB888, RGBA8888 come immediately to mind.
23:00:20 <Serg_Penguin> stick to 1 mode per proggie, and you need only 1 toolbox of routines ;)
23:00:31 <Serg_Penguin> it also saves CRT hardware
23:00:58 <Serg_Penguin> my old 17" CRT _hates_ switching gfx mode between game screens
23:01:15 <Herkamire> I'm sure many SDL programs do graphics slower than they should because the putpixel example code checks on the fly each time what putpixel routine it should use.
23:01:17 <kc5tja> And this solves things how?
23:02:02 <kc5tja> Serg_Penguin: Let me remind you that I run an average of 4 to 8 programs concurrently pretty much at all times.
23:02:20 --- quit: Herkamire ("goodnight all")
23:02:30 <Serg_Penguin> under X or win GDI, it neverminds
23:02:36 <kc5tja> Serg_Penguin: And even if I did set only one video mode for the whole global system (which I do anyway), that doesn't change the fact that I can arbitrarily change my video mode at any time.
23:02:52 <kc5tja> Serg_Penguin: Yes, because X and GDI have those 10 different drawing routines in them.
23:03:14 <kc5tja> And I'm not going to use X11 or GDI for FS/Forth.  That's just asking too much for me.
23:03:20 <kc5tja> Thankfully, SDL does bitmap format conversion for me.
23:04:07 <Serg_Penguin> IMHO, ideal Forth kernel must be capable only of extension
23:04:33 <Serg_Penguin> compile new words should be it's _only_ capability
23:04:46 <Serg_Penguin> all features - in libs loaded on demand
23:06:04 <kc5tja> I agree.
23:06:20 <kc5tja> but I'm still not supporting X11 with FS/Forth.
23:06:25 <kc5tja> It's just too damn hard.
23:06:32 <Serg_Penguin> and no one demands ;)
23:06:40 <kc5tja> Demands what?
23:06:48 <Serg_Penguin> X11 in Forth
23:06:57 <kc5tja> Well, I admit it would be kind of nice to have.
23:07:08 <kc5tja> But then I have "portability" issues to resolve.  
23:07:12 <kc5tja> And I'm not about to get into that.
23:07:26 <kc5tja> SDL is so very much simpler, and even permits FS/Forth to be used with fbcon, directFB, and more.
23:07:39 <kc5tja> The only disadvantage is I lose standard drawing libraries and such.
23:08:15 <Serg_Penguin> CAD in Forth may be interesting, but too big proj ;)
23:08:56 <Serg_Penguin> IMHO, the only reason to do X-able Forth - is build-in Forth in naturely GFX app
23:09:26 <Serg_Penguin> like yet another Photoshop or Autocad killer ;)
23:10:29 <kc5tja> Depends on the type of CAD software.
23:10:47 <kc5tja> Such a CAD program does exist.
23:10:54 <Serg_Penguin> in Forth ?
23:10:56 <kc5tja> Yes
23:11:09 <Serg_Penguin> maybe, only Moore's chip design soft...
23:11:16 <Serg_Penguin> but it's way too special
23:15:06 <kc5tja> No
23:15:16 <kc5tja> I'm talking something on the same level as CADKey or earlier versions of AutoCad
23:15:20 <kc5tja> But I can't find the site now.
23:15:26 * kc5tja isn't using the right Google search terms
23:18:00 <kc5tja> OK, I can't find it now.
23:18:02 * kc5tja sighs
23:39:00 <kc5tja> OK, I'm going to bed.
23:39:08 --- quit: kc5tja ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK")
23:50:25 --- quit: Serg_Penguin ()
23:59:59 --- log: ended forth/03.07.31