IRC log started Wed Jan 13 00:00:00 1999 [msg(TUNES)] permlog 1999.0113 ωνω abi [nef@bespin.ml.org] has joined #tunes ωνω SignOff ultima: #TUNES (Ping timeout for ultima[user-38lc7fo.dialup.mindspring.com]) ωνω ultima [ultima@user-38lc7fo.dialup.mindspring.com] has joined #tunes ωνω SignOff ultima: #TUNES (Ping timeout for ultima[user-38lc7fo.dialup.mindspring.com]) ωνω Connection closed from irc.us.openprojects.net: Success ωνω Connecting to port 6667 of server irc.us.openprojects.net [refnum 0] ωνω BitchX: For more information about BitchX type /about ωνω Welcome to the Internet Relay Network TUNES (from varley.openprojects.net) ωνω Your host is varley.openprojects.net, running version u2.10.03.lagtime.desynch3.indent8.dead.whox2.zombie (from varley.openprojects.net) ωνω This server was cobbled together Sat May 9 1998 at 12 25:56 EDT(from varley.openprojects.net) ωνω varley.openprojects.net u2.10.03.lagtime.desynch3.indent8.dead.whox2.zombie dioswkfcg biklmnopstv ωνω [local users on irc(24)] 12% ωνω [global users on irc(68)] 33% ωνω [invisible users on irc(139)] 67% ωνω [ircops on irc(11)] 5% ωνω [total users on irc(207)] ωνω [unknown connections(0)] ωνω [total servers on irc(27)] (avg. 7 users per server) ωνω [total channels created(57)] (avg. 3 users per channel) !varley.openprojects.net Highest connection count: 59 (45 clients) ωνω Mode change [+iw] for user TUNES ωνω TUNES [tunes@bespin.ml.org] has joined #tunes ωνω Topic for #TUNES: The TUNES Channel http://www.tunes.org/ for the TUNES OS project ωνω topic set by Tril [Mon Jan 11 19:23:54 1999] ωνω [Users(#TUNES:3)] [ TUNES ] [ Fare ] [ abi ] ωνω Channel #TUNES was created at Wed Dec 30 09:08:43 1998 ωνω BitchX: Join to #tunes was synced in 6.726 secs!! !sterling.openprojects.net!! Remote CONNECT lucas.openprojects.net 8004 from ^lilo !bear.openprojects.net!! Remote CONNECT lucas.openprojects.net 8005 from ^lilo ωνω NetSplit: fontana.openprojects.net split from varley.openprojects.net [12:31pm] ωνω BitchX: Press Ctrl-F to see who left Ctrl-E to change to [fontana.openprojects.net] ωνω Netjoined: fontana.openprojects.net varley.openprojects.net ωνω Fare [rideaufr@esmeralda.enst.fr] has joined #TUNES ωνω binEng [bineng@dialup245-2-1.swipnet.se] has joined #tunes ωνω SignOff binEng: #TUNES (Leaving) ωνω SignOff abi: #TUNES (Ping timeout for abi[bespin.ml.org]) ωνω Connection closed from irc.us.openprojects.net: Success ωνω Use /Server to connect to a server ωνω Connecting to port 6667 of server irc.us.openprojects.net [refnum 0] ωνω BitchX: For more information about BitchX type /about ωνω Welcome to the Internet Relay Network TUNES (from lackey.openprojects.net) ωνω Your host is lackey.openprojects.net, running version u2.10.03.lagtime.desynch3.indent8.dead.whox2.zombie (from lackey.openprojects.net) ωνω This server was cobbled together Sat May 9 1998 at 12 25:56 EDT(from lackey.openprojects.net) ωνω lackey.openprojects.net u2.10.03.lagtime.desynch3.indent8.dead.whox2.zombie dioswkfcg biklmnopstv ωνω [local users on irc(8)] 4% ωνω [global users on irc(81)] 36% ωνω [invisible users on irc(147)] 64% ωνω [ircops on irc(11)] 5% ωνω [total users on irc(228)] ωνω [unknown connections(0)] ωνω [total servers on irc(26)] (avg. 8 users per server) ωνω [total channels created(58)] (avg. 3 users per channel) !lackey.openprojects.net Highest connection count: 17 (16 clients) ωνω Mode change [+iw] for user TUNES ωνω TUNES [tunes@bespin.ml.org] has joined #tunes ωνω Topic for #TUNES: The TUNES Channel http://www.tunes.org/ for the TUNES OS project ωνω topic set by Tril [Mon Jan 11 19:23:57 1999] ωνω [Users(#TUNES:3)] [ TUNES ] [ abi ] [ Fare ] ωνω Channel #TUNES was created at Wed Dec 30 09:08:43 1998 ωνω BitchX: Join to #tunes was synced in 5.899 secs!! ωνω ultima [ultima@user-38lcecm.dialup.mindspring.com] has joined #tunes ωνω Tril [dem@xws004.xtrn03.wwu.edu] has joined #tunes hi twil! hi ooltima heh yo know i just watch the log for people to join to see if I should sign on if everyone did that nobody would show up Hah thats lame IRC wastes time yeah I multitask tho 02:40pm mi reading news too! =) 02:50pm Later... ωνω SignOff ultima: #TUNES (Leaving) 03:00pm ωνω SignOff Tril: #TUNES (Tril has no reason) ωνω _QZ [brand@p0wer.qzx.com] has joined #tunes hum 04:40pm ωνω SignOff Fare: #TUNES (Connection reset by pear) ωνω ^lilo [lilo@varley.openprojects.net] has joined #Tunes π ^lilo/#Tunes looks in <_QZ> greetings 06:00pm ωνω tcn [tcn@cci-209150250106.clarityconnect.net] has joined #tunes heh.. IRC is out of style today :) ωνω SignOff tcn: #TUNES (Leaving) 06:10pm ωνω Beholder [chaldea@ppp-163.m2-3.sub.ican.net] has joined #tunes Anyone seen tril today? Or fare? fare are you suggesting that my a Fairie? You know, Gay? <_QZ> fare? fare are you suggesting that my a Fairie? You know, Gay? <_QZ> hum <_QZ> === SIGNOFF #tunes Fare Connection reset by pear <_QZ> fare died about an hour ago Hey QZ <_QZ> u seen clementine? Clementine? Nope :) <_QZ> core's os <_QZ> he has a working component system with vm and all <_QZ> it just gives a blue screen and lets u use a cheap debugger <_QZ> and it supports smp QZ: Thats pretty awesome. What do you think of Flux? <_QZ> should i think anything of it? <_QZ> its nothing impressive, just another unix clone Not really, just wondering what people think of it. I thought Flux was an OS dev kit? <_QZ> full blown os kit <_QZ> it uses freebsd and linux drivers and has everything to be its own os Ahh. Not interesting then. <_QZ> linux could be an os dev kit How goes BRiX dev? <_QZ> u have all the source <_QZ> well after seeing clementine i decided to work faster on brix <_QZ> :) <_QZ> core is going to release it soon and i must beat him I was going to ask you some stuff about BRiX, if you don't mind? I was wondering how you were doing your clustering technology? <_QZ> clustering? 06:30pm Distributed computing (sorry clustering is new new buzzword term) <_QZ> well its not true distributed computing How so? <_QZ> all code in brix works on an object <_QZ> it uses O-spaces where action threads run around an object <_QZ> each O-space is isolated from others <_QZ> to do distributed computing (dc) u must relocate an object to another machine and send messages to it (run actions) <_QZ> and brix can query machines on the network, in its dc domain, to find idle machines and relocate objects for processing to those machines And it returns a result? <_QZ> no, u relocate the object back to the original machine No raw communications channel to send data back and forth? 06:40pm <_QZ> the actions can send objects back How do you implement your network communications? <_QZ> t what level? <_QZ> at <_QZ> the interface to everything is the same I assume the NIC is objectified, but what do you do to send stuff back and forth between BRiX computers? Is it raw data streams or objects also? <_QZ> the storage fs, networking, serial, parallel, etc is all handled the same way <_QZ> everything is objects Hmm, how would TCP/IP be implemented? <_QZ> u can only send messages in brix object.action(params) <_QZ> object is the data in the o-space, the action is the message that gets threaded in the o-space and uses the params Are you worried about network latency that will (potentially) cause? <_QZ> eh? <_QZ> that what will cause? The fact that objects have to be passed over the network (sorry going back to the BRiX -> BRiX machine communications) <_QZ> well yer not gonna be doing it over a 28.8 modem :) <_QZ> most ppl use 100baseT <_QZ> 10baseT could do it if the objects are small What?! Most ppl use 10 Base T :) You only use 100 BaseT when you are connecting network segments :) <_QZ> no u use 100baseT when u need to do dc Oh, yes for DC :) Not for file transfers and such :) <_QZ> i will be uprading my home network to 100baseT after i get brix doing everything that the OSes/programs i use do <_QZ> and then all the machines will become a single brix dc system 06:50pm <_QZ> Hmm, how would TCP/IP be implemented? <_QZ> now what did u wanna know there? And if you get some developers on your side, you could get some kick ass games going (clustered Half-Life... no more stoppage when explosions happen :) <_QZ> stoppage? <_QZ> i dont get stoppage :) QZ: Just genrally how any network standard will work in BRiX... I mean you can't send objects with IPX, it's just a pure data stream. I think you meant that the protocols are Objects themselves which interact directly with the NIC QZ: P200 MMX, Voodoo2, 64 Megs Ram, Win95 OSR2.1... I get bogged to 10 FPS when huge explosions happen <_QZ> bummer <_QZ> celeron472, voodoo1, 128meg, win98... i get good fps all the time <_QZ> how do u et fps in halflife? It's entirely processor dependent, I suspect. My friend has a AMD K6-2 380 with a TNT card and it bogs to about 20 FPS QZ: You don't... I'm only estimating based on how the controls react <_QZ> what mode are u in? <_QZ> res 800x600x16 bit color, Glide ωνω SignOff Beholder: #TUNES (Read error to Beholder[ppp-163.m2-3.sub.ican.net]: Connection reset by peer) ωνω Beholder [chaldea@ppp-163.m2-3.sub.ican.net] has joined #tunes Sorry QZ, missed your reply <_QZ> i use 640 Thats probably another reason... do you have your Voodoo1 clocked up to 75 yet? <_QZ> i have it on the faster of the 2 clocks 07:00pm My friend was quite psycho into cooling and overclocking his Diamond Monster card... he had it performing near the speed of a voodoo2 card (with the help of 2 486 heat sinks and fans :) <_QZ> how can u overclock it? ωνω ultima [ultima@user-37kbajb.dialup.mindspring.com] has joined #tunes <_QZ> i was told it can only be overclocked by the setting in the config π ultima/#tunes growls lowly It's a setting the the registry <_QZ> ultima: u freak dont run yer spider on me _QZ u freak dont killall httpd on me QZ: I'm talking to the guy in ICQ and he said he had it up to 58 stable... π ultima/#tunes ponders grr <_QZ> well if it can be oc'd then i can do it <_QZ> i have a 12" oscillating fan blowing over the cpu and cards QZ: Guess you have to with the celery oc'ed :) My Diamond Monster II can't go higher than 95 or it craps out ωνω SignOff ultima: #TUNES (Leaving) Gotta go now... sleep ωνω SignOff Beholder: #TUNES (Read error to Beholder[ppp-163.m2-3.sub.ican.net]: Connection reset by peer) 07:10pm ωνω ultima [ultima@user-37kb9va.dialup.mindspring.com] has joined #tunes hullo <_QZ> hi 09:20pm ωνω tcn [tcn@cci-209150250111.clarityconnect.net] has joined #tunes Hey que tal, tcn >>> tcn [tcn@cci-209150250111.clarityconnect.net] requested PING 916292103 from TUNES ωνω GMOL [gmol@h24-64-168-191.mt.wave.shaw.ca] has joined #tunes Hello.. who all is here, anybody whose active on the mailing list? Yeah, I am This is Maneesh, you are? 09:40pm Tom Novelli So you're proposing we set a deadline and write a bunch of papers before actually coding anything.. hmm.. I think it's reasonable That's fine with me, as long as the papers aren't too longwinded and academic, and people don't spend too much time prettying them up I think our biggest problem is lack of commitment nothing will come of nothing... what do you think? yes...exactly acedemics suffer from uneeded complexity We need concise summaries, not ramblings YES Hola welcome back, ultima We need implementation specifics Not academic explanations of why language X and theory Y suck tcn im coding muh butt off Once we have a plan that anyone can follow, the implementation is easy. cool What do you mean by specifics? hehe.. some things sound good in theory, but just don't work. Sometimes you just have to TRY an idea to know.. <_QZ> someone NEEDS to set down some standards fast and then begin coding and changing the standards as u encounter problems <_QZ> working code gives ppl hope and they will work more <_QZ> papers dont mean jack, but code shows progress ωνω SignOff GMOL: #TUNES (Ping timeout for GMOL[h24-64-168-191.mt.wave.shaw.ca]) <_QZ> i have rewritten brix more times than i can remember and each time i learn something new which makes the next rewrite even better There are some ideas floating around (Arrows, Prism, etc..) that need to be summarized in a way that people can understand them. _Short_ papers would help. Yeah, if you try to get everything right the first time, it might end up being like MULTICS.. <_QZ> and u cant kiss everyones ass, fare/tril must tell everyone how its gonna be and it will force some ppl to leave tunes but it will move tunes forward I learned that the 386 protection model is based on MULTICS.. no wonder it sucks <_QZ> there will never be a set of standards that pleases everyone 09:50pm Well if they want Tunes to be bloated, I'm out :) If they want to research forever, I'm out.. <_QZ> im making brix for myself <_QZ> if no one else wants to use it, WHO CARES It might inspire someone, QZ.. if it's a "proof of concept" for something new.. <_QZ> the work of someone doing what they want is always better than someone trying to kiss ass haha, yeah <_QZ> s/than someone/than that of someone/ Hey qz.. what output format are you using with NASM? <_QZ> bin Ok, same here. I might try RDF for modules.. Now that they've got nasm .98 out (sort of) <_QZ> here u wanna see something thats very secret? :) sure 10:00pm ? <_QZ> ? ? wots very secret <_QZ> nothing mr spider <_QZ> :) hey ultima, what OS do you use? tcn: LameIX Freebsd? yeah FBSD OBSD and Linux currently Freebsd? yeah right now im in fbsd coding a game queso is cool :) QZ, I tried queso on a buncha sites, like all the major OS makers.. works great <_QZ> heh queso is ok its not too accurate i hacked the kernel source for linux a bit back tomake it look like windows <_QZ> sure <_QZ> i need some smp docs 10:10pm <_QZ> know of any???? qz: linux-2.0.36.tar.gz :) <_QZ> well code is ok but docs are better :) <_QZ> do smp 486 machines exist? QZ yes <_QZ> wow <_QZ> i did not know that QZ well actually in theory they could exist i dont know if they got manufactured I'm not a big hardware buff I've heard of dual 386's, 10 years ago.. don't know if it was "SMP" though Look at Intel's site.. Hey QZ, you ever look at John Fine's page? good stuff there.. <_QZ> ya 10:20pm <_QZ> how do u do 64bit subtraction on the x86? carry 11:10pm <_QZ> borrow yeah :) <_QZ> ; edx:eax - ecx:ebx <_QZ> sub eax,ebx <_QZ> sbb edx,ecx hehe.. you thought Tunes was disorganized, look at NASM.. <_QZ> nasm works <_QZ> ok i think brix can detect mhz of a cpu now :) cool I guess the only doc's on RDOFF2 are the source.. rdoff sucks u really need to write your own binformat rdoff is more of the authors experimenting than anything else Yeah, I should prolly add a new output format to nasm <_QZ> core wrote his own heh shit oh yeah, xcom <_QZ> he has patches to binutils and nasm 11:20pm hoowah! well nasm should not be used for binaries its just not designed that way besides you'd be braindead if u used it huh? Linux binaries? Sooner or later I gotta write my own assembler, might as well start now ;) no system binaries woo Botz 0.1 works ult: what would you use instead? hehe.. get this: NASM 0.98 should be available around October 15 3 months later.. Heh dont ulove it <_QZ> ok here goes my mhz test 11:30pm did it work? <_QZ> no bummer.. what'd it do? <_QZ> it said 0mhz <_QZ> UGH <_QZ> found the bug RDF looks allright for now.. simple to load & link.. hahahah RDF is crappy though use ELF Isn't ELF too damn complicated? hmm, lesse <_QZ> YES yES YES <_QZ> oops jesus christ! 640 bytes for an 8 byte "program"!! I don't need all that shit, ultima btw, what's all this coding you're working on like mad? <_QZ> my mhz detection works kick ass How do you do it, with a "inc eax" loop and a timer? 11:40pm hmm I would do a nop loop or how about "mov ecx, -1" then "loop $" :) tcn: The coding im workin on mad? ultima: yeah hehehe what is it? its a school project a Brix clone? :) pretty lame phear it pretty good for a couple of hours work yeah, programming courses are so lame, I don't take 'em <_QZ> ; wait for the timer to tick <_QZ> ; RDTSC _QZ: Send a copy of brix on over <_QZ> ; wait for timer to tick again <_QZ> ; RDTSC <_QZ> ; mhz=(t2-t1)/54945 Whoa <_QZ> where t2 is the second RDTSC uhm <_QZ> and it gives a perfect mhz reading RDTSC == what? <_QZ> read time stamp counter <_QZ> its a pentium thing how does that work? I mean, how would it to MHZ detection <_QZ> the tsc is a 64bit register that incs every clock tick <_QZ> cpu clock tick that is cpu clock tick...u mean the timer that happens 18.2 times a second tick? <_QZ> so t2-t1 is the number of clock cycles in 1/18.2 seconds ah ok send it here <_QZ> on a 200mhz cpu u have 200 million cpu clock ticks per second =) I'll try it on my 386 =) <_QZ> in 1/18.2 seconds u have 10989010.989 cpu cycles <_QZ> 10989010.989/54945 = 200 what's the Model-Specific Register for? <_QZ> msr is not one but many regs <_QZ> k6 has 7 or 8 <_QZ> i assume they are the same as the pentiums <_QZ> ok im gonna crash now later <_QZ> tomorrow i will encrypt the brix image and let u gys have it :) <_QZ> or ill just turn everything off <_QZ> so ucan see the mhz <_QZ> :) why dont u just give us sourcE? u said it'd be open source <_QZ> ya WAY AFTER i get it all done <_QZ> cant take the chance of M$ cloning brix before my brix is done Yeah, I'm thinking about doing the same thing Polish it up good before I release it <_QZ> :) 11:50pm Except I jsut want it for my own use erm ur not nic nice even Yeah I like OSDev to learn <_QZ> this is my baby and i will protect it I kinda want to make a commercial Firewall OS though Maybe I'd better just stay in Tunes, or some OS project, so I don't have to write everything for it :) free source but commercial support facilities <_QZ> and if i release source then i will have a thousand morons telling me what needs to be changed in the kernel <_QZ> when i dont give a damn no one says u have to listen to em think It would be worse than spam though what would Linux be if Linus listened to AST? who/what the hell is AST? <_QZ> hehe the company? Andrew S. Tannenbaum oh.. hahaha He's a professor in .no, he like taught linus or osmething <_QZ> no It would be a curiosity like Minix and Amoeba He said publically that Linus was a moron That a monolithic kernel was a 'stupid' idea <_QZ> ya and his was better <_QZ> i read that news article too I saw the email transcript of Linus and Andy's big argument.. well microkernels and unix dont mix Hrmmmm Amoeba is also unix-like, right? I thought I saw X in with the sources How fast could an interpreter be? if I made a basic compiled LISP interpreter Forth interpreters are wicked fast reverse polish LISP, almost as fast.. π ultima/#tunes is thinking like ForthLisp Lisp syntax but designed as a general purpose language... Forth interpreters are so fast, nobody ever bothered to save compiled code as with other languages :) tcn: Well, I'm thinking of compiling the lisp into a forth-like bytecode why are they so fast? <_QZ> i just gotta tune my 386/486 mhz code now brb hehe, the syntax is so simple. Just read a word, execute it (or compile it), and repeat :) My interpreter is prolly under 100 lines.. in ASM! heh thats what I was thinking basically like interpreted assembler... frick im hitting the sack goodnight guys ωνω SignOff ultima: #TUNES (Leaving) heh, he crashed quick damn, it's 3 already [msg(TUNES)] newlog 1999.0114 IRC log ended Thu Jan 14 00:00:01 1999