00:00:00 --- log: started forth/04.06.25 00:20:39 --- join: solar_angel (~jenni@Toronto-HSE-ppp3685160.sympatico.ca) joined #forth 00:28:31 --- join: lalalim_ (~lalalim@p508ABAC5.dip.t-dialin.net) joined #forth 00:41:34 --- quit: lalalim (Read error: 110 (Connection timed out)) 00:46:40 --- quit: zardon_ (Read error: 110 (Connection timed out)) 00:55:20 --- join: doublec (~doublec@gprs.vodafone.net.nz) joined #forth 01:15:04 --- quit: doublec () 01:37:23 --- quit: topher ("Client Exiting") 02:06:52 --- quit: Herkamire ("bed") 02:44:39 --- join: qFox (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth 03:12:23 --- join: aum (~aum@port-204-54-210.fastadsl.net.nz) joined #forth 03:32:18 --- join: Topaz (~top@spc1-horn1-6-0-cust217.cosh.broadband.ntl.com) joined #forth 03:57:06 sweet, getting wxForth working 03:57:16 wxForht! 03:57:22 that sounds cool 03:57:33 can you give me a url to it? 03:57:45 hey, i've only just got a test prog working 03:57:51 will be a while b4 i publish 03:58:03 oh, you're writing it! even cooler! 03:58:27 i've been building 'forpy' - a fork of FICL-forth that interfaces with the Python API 03:58:53 and am working on some Forth words for building, instantiating, subclassing, accessing Python classes 03:59:22 so a kinda roundabout journey - forth -> python -> wxPython -> wxWindows 04:00:46 wow, it interfaces with python? 04:00:49 that's... interesting. 04:01:16 well, i love python, it's got a really cushy 'as-you-like-it' oop model, and a great api 04:01:17 my FORTH environment stands alone. 04:01:28 solar_angel: no it doesn't 04:01:36 no it doesn't? 04:01:42 solar_angel: a REAL forth environment doesn't even need the hardware! 04:02:01 aum - you're talking to a chip designer who's in the process of designing chips to run her FORTH environment on 04:02:09 like i said, my FORTH environment stands alone 04:02:17 solar_angel: faaarrrrrrrrrrk!!! :)) 04:02:32 ? 04:02:36 solar_angel: the Forth community will hail you as a Godess, and celebrate your birthday 04:02:41 lol 04:03:11 solar_angel: will you get manufacturer support to tool up and scale up a good prod run? 04:03:31 i'm not rich... looking at FPGAs for implementation at present. 04:03:47 * aum would be delighted to have a Forth PCI card 04:04:38 solar_angel: have you been doing Forth for long? 04:04:57 depends what you mean by FORTH. 04:05:04 FORTH itself, not really... under a year, probably. 04:05:10 stack-based languages... that's another story 04:05:59 i invented my first one back in... '95 or 96, i think. 04:06:33 solar_angel: that's impressive 04:06:48 i suppose. 04:06:49 so how did you feel switching from your own thang to the world of Forth? 04:07:07 it's not all that much different, honestly... especially since my FORTH isn't anywhere remotely near ansi 04:07:15 what forth is ?!? 04:07:21 point.' 04:07:39 well, several Forths try to do the ANS Core+Ext 04:07:56 mine doesn't even try to be ans. 04:08:11 I hacked FICL-forth, so now in addition to the usual param/return/float stacks, it's got a separate stack for python objects 04:08:14 it uses nul-terminated strings, for one thing :P 04:08:20 now that's odd :) 04:08:29 have you played with python? 04:08:34 i wrote a forth-like calculator once, as a test for the C++ standard library i was writing.... 04:08:48 it was odd, because it had arbitrary-precision integers on the stack 04:08:57 well done! 04:08:59 yes, i've used python. i used to do a fair bit in it 04:09:11 i massively like Python's OO paradigm 04:09:23 it's kinda cool. 04:09:33 i've been working on an OO extension in my forth. 04:09:52 i feel there's a slight impedance mismatch between forth and oo 04:10:00 hard to blend them in a way that feels good 04:10:04 not necessarily. 04:10:22 see, i already have a dictionary/vocabulary system that approximates OO already. 04:10:31 there's an OO extension in FICL-forth, but it's painful 04:10:33 it's just a matter of hooking it in. 04:10:58 i've done a callback wrapper that allows Python methods to be implemented in Forth 04:11:05 cool. 04:11:32 i'm trying to rid myself of my last external dependancy 04:11:35 haven't implemented python locals yet, so the stack manipulation is frying my brain 04:11:38 i'm still reliant on nasm. 04:11:43 ahh, ok 04:12:02 so you want your Forth system to spit the raw binary? 04:12:26 i want to be able to reassemble my core from inside itself. 04:12:46 but for a given platform, you've got to 'boot' your forth onto it somehow 04:13:00 sure. that's where you cross-assemble. 04:13:04 i'm not quite to that point yet. 04:13:06 but it's close 04:13:29 any chance of a linux kernel replacement, with drivers, by Thursday next week? 04:13:37 * aum ducks 04:14:00 i don't see a point in replacing the linux kernel... 04:14:21 slightly more seriously - are you building a graphical toolkit 04:14:21 ? 04:14:33 but Cy/VOS boots quite nicely on most hardware, and supports memory management, floppy drive, pic, pit, dma, pc speaker, rs232, keyboard, psaux, vgacon, vga... 04:14:49 nic? 04:14:54 tcp stack? 04:15:07 not quite yet... i'm adding interrupt support to my serial driver at the moment... 04:15:19 i have my dev box cabled into a Cisco 2502 through both serial ports 04:15:27 right 04:15:31 RFC1055 is on my immediate todo list 04:15:43 do you find you're getting comfortable leverage? 04:16:05 leverage? 04:16:24 well, layers of abstraction which eliminate menialities 04:16:51 it's not too bad. i spend less time actually implenting code than i do designing. 04:17:01 or reading up on bizarre PC hardware 04:17:04 that's a good place to be 04:17:14 exactly. 04:17:33 once in a while an odd hardware peculiarity snags me for a bit 04:17:37 trivial question - do you think local variables in Forth suck? 04:17:40 psaux is a right royal mess. 04:17:49 local variables? like, the LOCALS keyword? 04:17:55 i don't have them implemented. 04:18:15 so within your words, you do everything purely on the stacks? 04:18:27 pretty much 04:18:37 if i can't, i usually break it into a couple words 04:18:55 guess it would be comfy for you, since you've spent most of a decade in stack-based programming 04:19:26 mostly just toying around, except for the time i wrote a compiler that used a stack-based intermediate language... 04:19:44 the optimizer was a cool project... i still haven't added a comparable thing to my current FORTH system 04:20:01 most stack manipulation words compiled to zero assembly instructions. it was quite cool. 04:20:13 nice one 04:20:36 i rather liked it. 04:20:45 i've spent most of that time in C++, honestly. 04:20:57 i wrote an extensive standard library, starting in a time before stl existed. 04:21:04 totally different paradigm from forth 04:21:09 that's for sure. 04:21:20 my FORTH system doesn't have built-in templates for genetic algorithms :P 04:21:45 that brings up a thought - AI in Forth, or at least, neural nets 04:21:46 i go from writing a massive standard library to writing an OS that weighs in at 60k. 04:22:03 hehehe, my first neural net program was written in qbasic :P 04:22:15 it was odd, though, had a full mouse-driven GUI and everything 04:22:19 60k - that's good 04:22:33 takes me back to the 8-bit era 04:22:55 hehehe 04:23:03 the 8-bit era was a trip 04:23:21 i remember feeling in heaven after expanding from 4k RAM to 16k 04:23:39 yikers. i'm not that old, i started on a 128K machine. 04:23:58 still a 6502-based machine, though 04:24:05 my first machine - SC/MP processor, didn't even have CALL/PUSH/POP instructions 04:24:06 (for the most part... it also had a Z80, but it hardly used it...) 04:24:14 which machine was that; 04:24:16 ? 04:24:21 * solar_angel started on the C128 04:24:28 National Semiconductor SC/MP - 8-bit 04:24:38 took 3 machine instructions to emulate a CALL 04:24:43 oh. cute. 04:26:00 i guess the whole spirit of Forth is about breaking free from all dependencies on other software, right down to the bare metal 04:26:11 * solar_angel thinks she probably learned to read from the Commodore 128 Reference Manual. 04:26:20 lol 04:26:21 exactly :) 04:26:33 well, i'm actually being half-serious. 04:26:36 i started programming when i was 5. 04:26:42 eek 04:26:51 i thought i was an early starter at 13 04:27:00 i had already designed my first computer by then. 04:27:14 although it wouldn't have worked in the designed state, and would have costed a fortune... 04:27:29 i also had built my first computer-controlled robot (after reprogramming the 128's joystick port as an output device) 04:27:37 * aum wonders if DNA is (even partially) stack-based 04:27:48 and... um... probably built my boolean calculator as well. 04:28:24 not really.... DNA is a 3-unit tetranary code (base 64) 04:28:36 you must have been Alan Turing or similar in a previous life 04:28:40 it's more like massive parallel processing. 04:28:44 lol, why do you say that? 04:29:01 you've taken to programming like Mozart took to music 04:29:52 * solar_angel blushes. 04:29:59 well that has to be the nicest thing anyone's ever said to me. 04:30:06 surely not 04:31:01 do you get a hard time in life? 04:31:31 yes 04:31:47 how so, if you don't mind me asking? 04:31:55 a bunch of ways. 04:32:15 amongst other things, i'm autistic. and not exactly mildly. it makes functioning in this world a huge challenge. 04:32:41 i have Aspergers, so I (partially) know where you're coming from 04:33:16 *nods* lots of the brightest geeks are somewhere on the autistic spectrum. 04:33:28 i'm almost an extreme example of that, i guess. 04:33:56 must be tough relating to people in the physical world 04:34:24 it is. 04:35:04 do you feel connected to your emotions, or does that whole area feel a bit dead? 04:35:24 i'm quite well connected to my own emotions. 04:35:47 from what i can tell, it's mostly a misconception that autistic people *lack* emotion. it's more like we just have a totally different set of them. 04:36:02 i can't relate to neurotypicals at all. my partner is also autistic, though, and we relate perfectly. 04:36:18 different set of 'protocols' of emotional interaction? 04:36:47 that's a good way of putting it, yes. 04:37:07 i've never had much trouble relating to other autistics, and the internet has made finding others to interact with a lot easier 04:37:23 that'd be a god-send 04:37:29 it is. 04:37:57 could it be that autistics wire their brains, and write their neural microcode differently? 04:38:23 absolutely. 04:38:33 there are many things that autistics do a *lot* better than neurotypicals. 04:38:38 out of the ones i know, most are in either math or computers. 04:38:39 that's for sure 04:38:57 analytical disciplines, with objective parameters 04:39:00 exactly. 04:39:45 have you wondered what might cause one to structure one's brain in a way which neurotypicals label as 'autistic'? 04:40:02 well, that part seems like a birth difference. probably something genetic. 04:40:06 or, when in life this happens? in-utero? infancy? 04:40:58 a good example (and an odd one) is that autistics are born with smaller heads than neurotypicals, but have way more development of the brain after birth, ending up with substantially larger cranial diameters after 4 or 5 years. 04:41:31 it's just that i practise an alternative health therapy called 'Rebirthing', which seeks to promote better/deeper self-understanding 04:41:43 it goes along with my current theory that we have fewer instincts (less prejudice, less irrationality, etc, amonst other things) and more learned-behaviors than neurotypicals. 04:42:22 i've seen countless cases where people can set up major aspects of their existence from single experiences, even in the womb 04:42:34 (i've never met a prejudiced autistic who wasn't thoroughly brainwashed by propaganda, and even those would listen to reason...) 04:43:16 must be weird to look out at the mainstream world - seeing people who'd rather die than change a fundamental irrational belief 04:43:20 So. Whan will the sane take over? 04:43:37 aum: It isn't for you? 04:43:38 aum - *very* weird. 04:44:11 solar_angel: i feel that to a large extent, but i can certainly imagine the feeling would be greater for someone from a more autistic viewpoint 04:44:38 Robert - referring to what? 04:44:50 * aum thinks the sane will take over when people learn en masse to buy into a 'plus-sum game' 04:45:01 solar_angel: What is? 04:45:17 aum: What is that? 04:45:42 a 'plus-sum game' is a mind-set of abundance, the paradigm of 'there's more than enough for everybody' 04:45:57 * solar_angel smiles at aum. 04:46:03 a plus-sum game.. that would be quite something. 04:46:12 most of the world plays a 'zero-sum game' - the belief that if anyone has something, then one is automatically being deprived 04:46:33 or, a belief that the size of the pie is unavoidably limited 04:47:31 i've come to believe that the 'the root of all evil' is not the 'love of money', it's the fear of being without money 04:47:58 that is an astute observation. 04:48:10 although sadly, i don't think it's entirely correct. 04:48:20 the problem with society at the moment is an increasing concentration of wealth. 04:48:25 some of my clients are millionaires, with holes in their socks 04:48:53 some of my clients are on the bones of their asses, dirt-poor, but always find money for what they need, and manage to save and live well 04:49:59 tis one thing i'd like to ask actually, does anybody ever 'feel rich'?, or do people always think they need at least twice what they actually have? 04:50:18 i guess autistics might be less entrapped by the reptilian aspects of the mating drive - the need to find/control the most 'genetically viable' mates at all costs 04:50:26 (when compared to the rest of the (global) population, they're extremely wealthy) 04:50:45 Topaz: Nigerians are the world's happiest people 04:50:54 * Robert looks at aum 04:50:58 and I'm not referring to the 419 spammer/scammers 04:51:43 one measure of wealth - how long you can cease work for before worrying about money 04:51:51 exactly. 04:51:56 i feel comfy now. 04:52:00 i don't work. i like it this way. 04:52:12 i don't work outside of writing a FORTH-based OS, anyway. 04:52:15 but that's fun. 04:52:18 aum: Then everybody could we wealthy - just not at the same time. 04:52:21 be* 04:52:29 what will this OS run on? 04:52:37 Topaz - it already runs on x86 PC's. 04:52:43 i'm in the process of a couple other ports 04:52:47 ooh, fun fun 04:52:48 mostly waiting on development systems. 04:53:24 Topaz: Do you ever play strategy games? 04:53:32 a forth-based pxt cellphone would be leet 04:53:52 pxt? 04:54:06 camera cellphones 04:54:09 Robert: occasionally, though i haven't played any computer games in ages 04:54:20 (for a start, most won't run on linux ;) 04:54:30 i'm overdue to walk the dog - back approx 20 mins... 04:54:33 i like red alert, though i'm no good at it 04:55:17 strategy games are fun. 04:56:10 solar_angel: if you want to keep in touch, my email is david@freenet.org.nz, feel welcome to drop me a line 04:57:03 i'm constantly in this channel anyway. 05:04:18 Topaz: Me neither. But you can never have too much defense for your base, right? No matter how much you have, you don't feel really secure. 05:08:12 well, depends :D 05:08:17 but i see what you mean 05:24:09 :) 06:28:00 aum: wxForth url? 06:29:54 --- join: madwork_ (~madgarden@derby.metrics.com) joined #forth 06:59:57 --- quit: Topaz (Remote closed the connection) 07:00:03 --- join: segher (~segher@blueice4n1.de.ibm.com) joined #forth 07:07:23 --- quit: solar_angel ("later") 07:15:45 --- join: squire (~vikas@202.142.105.254) joined #forth 07:16:46 after learning lisp I wrote better C code, after learning a bit of forth I write better Lisp code 07:31:24 --- join: wossname (wossname@HSE-MTL-ppp60177.qc.sympatico.ca) joined #forth 07:44:43 --- quit: squire (Read error: 104 (Connection reset by peer)) 07:57:36 :) 07:57:44 erm he left 07:58:21 I found one hit on google for wxForth, and it was a log from this channel in 2002, asking about wxForth. 08:01:34 hehe 08:07:39 ;) 08:09:40 first disk done, yay 08:20:36 hi 08:20:52 hi 08:20:56 * arke is just using glue and tape and paper 08:20:59 Hi slava 08:21:00 * arke is building a tesla turbine from paper 08:23:26 I can't remember what the inner hub diameter is on a 5.25" HD, but the media surface would be great. And being aluminum, cutting the exhaust ports shouldn't be too hard. 08:24:01 ;) 08:24:09 * arke has 2 disks cut out 08:24:15 * arke starts on the third 08:26:53 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 08:42:22 --- quit: segher ("Leaving") 08:47:12 --- join: solar_angel (~jenni@Toronto-HSE-ppp3685160.sympatico.ca) joined #forth 08:53:39 --- join: topher (~chris@lsanca1-ar42-4-61-175-184.lsanca1.dsl-verizon.net) joined #forth 09:02:17 --- quit: tathi ("leaving") 09:07:36 sometimes, vim *really* pisses me off. I'm toggling between two files, comparing some differences. And half the time this brain dead program doesn't remember the screen position of where it was. And there's no consistency to it, that I can find. Well, other than not working right when I really need it. 09:07:57 why not use vimdiff? 09:08:05 or screen? 09:08:16 Because yank buffers don't work across screen. 09:08:24 use vimdiff then 09:08:26 I dunno about vimdiff. Never heard of it. 09:08:33 man vimdiff 09:08:36 it works :) 09:10:04 So it does. Now, how the hell to actually move differences back and forth... 09:10:29 i forget, but there's a couple commands for it 09:10:34 try :help diff 09:11:19 --- quit: topher ("Client Exiting") 09:53:33 --- quit: solar_angel ("*later*") 11:36:03 --- quit: fridge (Remote closed the connection) 11:36:04 --- quit: madwork_ (Read error: 54 (Connection reset by peer)) 11:55:00 --- join: proteusguy (~proteusgu@69.79.24.31) joined #forth 12:35:16 --- join: ASau (~asau@158.250.48.204) joined #forth 12:35:23 Dobryjj vecher! 12:35:40 --- join: Topaz (~top@spc1-horn1-6-0-cust217.cosh.broadband.ntl.com) joined #forth 12:35:56 Dobryjj vecher, Topaz! 12:37:39 er, hi :) 12:37:51 Privet, ASau :) 12:42:23 God kvael! 12:43:47 kvaell* :) 12:44:36 OK. 12:44:49 Anything new? 12:45:02 I've made my diploma. 12:45:32 Now I'm (relatively) free. 12:46:14 I may afford one or two free months. 12:47:22 Cool! 12:47:30 Congratulations, ASau :) 12:47:57 How is it called in English? 12:48:00 Ph. D.? 12:48:20 Not sure, that's one title... 12:49:31 Anything new from crc? 12:49:49 Has he seen hacked c-forth? 12:50:47 Don't know 12:51:11 Well, I can wait for him. 13:26:49 --- quit: wossname (Read error: 104 (Connection reset by peer)) 13:27:21 --- nick: lalalim_ -> lalalim 13:54:32 --- join: Herkamire (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth 14:03:50 God kvaell, Herkamire! 14:05:33 hi ASau :) 14:06:23 What's new? 14:06:42 Where's crc? 14:51:03 --- quit: cmeme (Read error: 54 (Connection reset by peer)) 14:51:23 --- join: cmeme (~cmeme@216.184.11.30.unused.swcp.com) joined #forth 14:51:23 --- join: cmeme2 (~cmeme@216.184.11.30.unused.swcp.com) joined #forth 14:53:17 --- quit: cmeme (Remote closed the connection) 14:53:17 --- quit: cmeme2 (Read error: 104 (Connection reset by peer)) 14:53:58 --- join: cmeme (~cmeme@216.184.11.30.unused.swcp.com) joined #forth 15:09:07 --- quit: cmeme (Read error: 54 (Connection reset by peer)) 15:20:23 --- join: cmeme (~cmeme@216.184.11.30.unused.swcp.com) joined #forth 15:21:02 --- quit: cmeme (Read error: 54 (Connection reset by peer)) 15:25:11 --- join: cmeme (~cmeme@216.184.11.30.unused.swcp.com) joined #forth 15:47:29 --- join: solar_angel (~jenni@Toronto-HSE-ppp3685160.sympatico.ca) joined #forth 15:52:40 --- join: SDO (~SDO@68.170.20.201) joined #forth 16:02:39 Good evening, solar_angel and SDO! 16:02:54 hi asau 16:03:21 Anything new? 16:03:51 I've been away for a week. 16:04:05 not sure... 16:04:10 some progress on my OS project 16:04:19 That's good. 16:04:28 solar_angel: what sort of os? 16:04:39 I have little progress in my optimizer. 16:04:42 Herkamire - FORTH-based OS for x86, and eventually for PPC. 16:05:18 i've been thinking about my optimizer a bit, and looking through the sources of my old one 16:05:36 I've almost coded Hooke-Jeeves' alg. in Scheme, and partially in Forth. 16:05:39 definately going to have to build one in soon. 16:05:48 hooke-jeeves? 16:07:47 R. Hooke, T. A. Jeeves, 'Direct search solution of numerical and statistical problems', J. Assn. Comp. Mach., 8, 212-219, 1061. 16:08:03 alright, what does it do? 16:08:13 :) 16:08:38 i implemented Brandenberg-Sporer a while ago too... 16:08:39 It optimizes function of several arguments. 16:08:45 i bet that means just as much to you :P 16:08:57 Hm. 16:09:08 Reference? 16:09:20 what do you need an optomizer for? 16:09:53 [3] Sporer, Th., et al, The use of multirate filter banks for coding 16:09:54 of high quality digital audio, 6th European Signal Processing Conference 16:09:54 v1 pp211-214, Amsterdam, 1992 (http://www.lte.e-technik.uni-erlangen.de/~spo/eusipco.corrected.ps) 16:09:56 Nonlinear fit. 16:10:31 amsterdam! nl! 16:11:07 And main thermodyn. problem. 16:11:42 Link does not work. 16:11:49 probably not. *shrugs* 16:11:52 it was just a reference i found 16:11:58 i think i've got the ps file around somewhere... 16:12:24 May I ask you to send it by e-mail? 16:12:36 asau at hotbox dot ru 16:12:44 i'd have to actually find it. 16:12:58 it's basically an efficient implementation of MDCT 16:14:10 Sorry, I don't know. 16:14:14 What's MDCT? 16:14:17 Simplex? 16:14:22 Modified Discrete Cosine Transform 16:15:28 How does it converge? 16:15:42 converge? 16:16:02 it's totally deterministic 16:16:09 there's only ever one solution to an MDCT or an IMDCT 16:16:12 Well. 16:16:39 Does it require differenciation? 16:17:22 Probably not. 16:17:37 But it requires integration, I think. 16:17:52 This is not good too. 16:17:56 no, actually, it doesn't. 16:17:59 Though it's better. 16:18:13 it's a recursively-expanded series expansion 16:18:16 Maybe, reference? 16:19:12 google has a few. or for simple explanation, it's very much like an FFT 16:19:36 with a few quirks that get around the edge problems and windowing issues in continuous streams. 16:19:59 I can't understand this language. 16:20:18 My problems are of quite another class. 16:20:25 I'm not EE. 16:20:38 *nods* you're doing numeric solving and optimization 16:20:45 which is a totally different set of problems 16:20:54 you're not converting to and from frequency domains 16:21:05 Right. 16:21:23 I don't have frequencies at all. 16:22:25 Thanks for the idea to try to google. 16:22:25 exactly. 16:22:32 sure 16:22:53 (trust me, the google links are *way* more approachable than the paper itself.) 16:26:25 That depends on your nearest library. 16:29:35 no, i don't mean findable. i mean approachable. 16:29:49 i mean you can sit down with the paper in front of you and be left less clueful than before you started 16:29:52 took me hours to decypher it 16:30:01 and i'm relatively skilled in this area 16:30:09 (no expert, to be sure, but not new either) 16:30:18 C gives you enough rope to hang yourself, think of Subversion as a sort of rope storage facility.? ?Brian W. Fitzpatrick 16:30:19 In the world of open-source softw 16:30:31 Forth gies you a way to hang yourself more efficiently with LESS ROPE! 16:30:52 lol!! 16:30:52 Dobryjj vecher, I440r! 16:30:54 LESS ROPE! 16:30:56 that was supposed to be IF C... 16:30:57 now that's brilliant 16:31:25 solar_angel, I can't agree. 16:31:34 ASau - can't agree with what? 16:32:02 the paper i pointed out didn't really explain the concepts of spectral density or what an MDCT is, just how to do one fast. 16:32:07 I've met more papers that were more clear than anything found with google. 16:32:12 sure 16:32:14 just not that one 16:33:46 Well, I can't imagine anything analogous to reference implementation in chemistry. 16:35:24 cool, you're in chemistry? 16:35:55 I've made diploma in chem. thermodyn. 16:36:02 cool 16:36:31 Why I need optimization is this. 16:36:53 Non-linear fit. Weighted least squares. 16:37:23 Solving main thermodyn. problem is minimizing Gibbs energy function. 16:37:40 Well, constrained actually. 16:37:45 *nods* 16:38:00 But you can go with Lagrange function. 16:38:02 wait... Gibbs energy function? 16:38:25 "How the stuff works." 16:38:26 oh ok, nevermind. 16:38:37 i knew the equation but not the name. 16:38:40 it's been a while 16:38:56 You measure thermodynamic properties (or thermochemical). 16:40:02 *nods* you're trying to make G small so that you can ensure the reaction is forward-biased. 16:40:27 Then you apply nonlinear fit: measured values to model function G(T, p, x_i). 16:40:28 yes? 16:40:39 No. 16:40:42 ohhhh 16:40:45 now that's interesting 16:40:48 i misunderstood :) 16:42:22 When you fix T and p, you have x (molar ratio of components) for G to be minimal. 16:42:45 ahhh 16:43:05 you're trying to find the optimal temperature, pressure, and concentration for a reaction? 16:43:15 So when you minimize G(x) you find equilibrial concentrations. 16:43:22 *nods* 16:43:40 ohhhhh, so then you know your K. 16:43:40 interesting 16:43:54 * solar_angel is really rusty on chemistry. 16:44:20 hi all 16:44:28 Hi slava 16:44:33 Privet, slava! 16:44:56 how goes? 16:45:32 ASau - i now think i see what you're using the optimization algorithm for. fascinating :) 16:48:28 hi all 16:48:37 Actually the final result is either phase diagram or stability diagram. 16:49:03 Good evening, aum! 16:49:11 morning here 16:49:35