| --- Log opened Sun Jan 06 00:00:38 2008 |
| 00:04 | <@AnnoDomini> | Okay. I seem to have fixed the OMFGBBQ thing, 'cause it ran properly for 0 now. |
| 00:05 | | You're now known as TheWatcher[zZzZ] |
| 00:07 | | * AnnoDomini facepalms. |
| 00:07 | <@AnnoDomini> | I now know why it bugs out at temp>25. |
| 00:07 | <@AnnoDomini> | I seem to have forgotten an entire term. |
| 00:09 | <@AnnoDomini> | Wait... I haven't forgotten. It's right there, in the code. |
| 00:09 | <@AnnoDomini> | WTF doesn't it show up on the plot? |
| 00:10 | <@AnnoDomini> | Aha. |
| 00:10 | <@AnnoDomini> | I really hate these function formulas. It's really hard to detect mistakes. |
| 00:10 | < Attilla> | How close to absolute zero can it reach? |
| 00:10 | < Attilla> | Or is temp meaning some sort of temporary variable? |
| 00:11 | < Attilla> | (or constant, whatever) |
| 00:11 | <@AnnoDomini> | What I'm doing is a sprinkler system controller, the input variables are temperature(0,35) and humidity(0,100). |
| 00:11 | <@AnnoDomini> | The output is intensity(0,100). |
| 00:11 | < Attilla> | Ah, though wouldn't sprinklers want go off at much higher temperatures? |
| 00:12 | <@AnnoDomini> | At higher temperatures, I think it would be, "call insurance agent, your crops just FAILED due extreme heat". |
| 00:13 | < Attilla> | Oh. I thought you meant an emergency sprinkler system. |
| 00:13 | < Attilla> | Like in case of fires. |
| 00:13 | <@AnnoDomini> | Naw. An agricultural one. |
| 00:13 | <@AnnoDomini> | Thar. It works correctly now. |
| 00:15 | <@AnnoDomini> | WinXP has this nifty function of giving off a ripple centered on the cursor when you push Ctrl+nothing. I find it amusing. |
| 00:17 | < Attilla> | What do you mean "nothing"? |
| 00:17 | <@AnnoDomini> | Nothing. If you press something in conjunction with Ctrl, there's no ripple. |
| 00:20 | <@AnnoDomini> | God damnit. How the hell do I switch to a different plot? The thing just prints on the same one over and over. Which is not to my liking. |
| 00:24 | <@AnnoDomini> | The bugginess of SciLab never ceases to astound me. Right now it went into another crash/infinite loop, FOR NO REASON. |
| 00:25 | | * AnnoDomini mumbles how they're supposed to use this crap now, instead of MatLab, because the Uni is too cheap to renew the license. |
| 00:27 | | * C_tiger idly pimps Sage which is free. |
| 00:36 | <@AnnoDomini> | I don't get a say in what program I'm going to be attempting to pass the exam on. |
| 00:42 | < C_tiger> | True. |
| 00:44 | < Attilla> | There needs to be a #science channel, dangit. I'm seriously lacking anyone to talk with over electrochemical subjects and it's getting me down. Plus the net isn't most helpful and ATHENS access gets me papers which are too advanced to tell me what I want to know, and the net so far seems a bit lacklustre in specificing exactly how what I want to do is done (but it does tell me what a wonderful fin |
| 00:44 | < Attilla> | ish it does) |
| 00:44 | | * AnnoDomini is not amused at the speed of this script. It's only circa 3600 iterations, damnit. mIRC scripts are faster than this! |
| 00:45 | < Attilla> | Ouch. |
| 00:45 | <@AnnoDomini> | Gah. It error'd. |
| 00:46 | <@AnnoDomini> | Fortunately, it doesn't seem to have hung. |
| 00:48 | <@AnnoDomini> | Actually, I tell a lie. It's circa 360000 iterations. Forgot the inner loop. |
| 00:48 | <@AnnoDomini> | But I can't skip this! |
| 00:49 | <@AnnoDomini> | Need values for 3600 points, and each of those points requires a 100 iteration loop. |
| 00:53 | | * AnnoDomini facepalms. |
| 00:54 | <@AnnoDomini> | The help entry for mesh() contains an example which is totally ungrokkable without further examples for the usage of functions called in the example. Naturally, there are no examples for those under their entries. |
| 01:04 | <@AnnoDomini> | Oh, great. This help document, while being more verbose, is partially written in French. These variable names mean nothing to me. |
| 01:10 | < C_tiger> | Atilla, I can talk scienec. |
| 01:10 | < C_tiger> | science. |
| 01:10 | < C_tiger> | It's what I was trained to do. |
| 01:11 | < Attilla> | I know there are people on this network who know Science, but I doubt any know that much about Electrochemistry. Per chance do you? |
| 01:12 | < C_tiger> | Chemical engineer by training. |
| 01:12 | < C_tiger> | I don't know everything but I can think through stuff. |
| 01:12 | < Attilla> | Ah. Okay well what exactly is Cathodic Polarisation, and how is it performed per se? |
| 01:16 | < C_tiger> | Ok, a quick refresher says that it's a way of using different metals so that the potential at the cathode is insufficient to corrode your cathode, thus protecting it longer. |
| 01:17 | < C_tiger> | What exactly do you want to know about it? |
| 01:18 | < C_tiger> | If you write all the half equations out, you'll get a good idea of what kind of metal is needed to protect at a given voltage. |
| 01:18 | < Attilla> | Well I'm trying to replicate an experiment, and one of the preparatory steps is to Cathodically Polarise onto some previously anodised plates for a brief moment. |
| 01:19 | < C_tiger> | Ah. |
| 01:19 | < Attilla> | *-onto |
| 01:20 | < C_tiger> | Ok, what exactly is the experiment, just out of curiosity. |
| 01:20 | < Attilla> | An electroplating experiment. For electroplating iron-nickle alloy onto copper plates. |
| 01:20 | < Attilla> | (And then using analysis on said plates afterwards0 |
| 01:21 | < C_tiger> | I remember that one from high school. |
| 01:21 | < Attilla> | *nickel |
| 01:23 | < C_tiger> | Just from memory, I think you need to set up the system with your plate at the anode so the surface film is eaten off. |
| 01:23 | < C_tiger> | But that doesn't sound right. |
| 01:24 | < C_tiger> | For cathodic polarization, I think you actually do the reverse, you put your anode on the + side of your cell so some of the ions in solution deposit thus preventing the loss of your anode during the course of the experiment. |
| 01:26 | < C_tiger> | The terminology is odd because in this case, you'd put the anode on the cathode side of the cell. |
| 01:26 | < Attilla> | Doesn't anodisation involve putting the plates on the + terminal? |
| 01:26 | < C_tiger> | I hate words like that. |
| 01:26 | < C_tiger> | Um the + side is the cathode. |
| 01:28 | < Attilla> | You sure? Doesn't the cathode *emit* the electrons by repelling them? |
| 01:28 | < C_tiger> | I think. It may depend on if we're talking battery or cell. |
| 01:29 | < C_tiger> | I'm good with half equations. |
| 01:29 | < C_tiger> | I suck at words. |
| 01:30 | < C_tiger> | Ok, you're right, the "anode" is the + terminal in the electrolytic cell. |
| 01:30 | < C_tiger> | It's the - in a galvanic cell. |
| 01:32 | < Attilla> | Ah. Right. Uh. Need to remember the differences. |
| 01:35 | < C_tiger> | Sorry, I'm also manning the LiveChat helpdesk right now so I disappear randomly when there is a question. |
| 01:36 | < Attilla> | (I'm telling myself that) |
| 01:37 | < Attilla> | I'm dealing with galvanic because this is forming layers or stripping, not breaking apart, right? |
| 01:37 | < C_tiger> | No, you're dealing with electrolytic because you're pumping electricity in. |
| 01:37 | < C_tiger> | Hence my "you were right". |
| 01:38 | < C_tiger> | If you stuck a zinc strip and a copper strip into an orange, that's a galvanic cell. |
| 01:38 | < Attilla> | Oh right. Yeah. |
| 01:38 | < Attilla> | Galvanic is batteries or similar. |
| 01:38 | < C_tiger> | Right. |
| 01:38 | < C_tiger> | This has nothing to do with the words "galvanized" or "anodized" which have other meanings. |
| 01:39 | < Attilla> | Yeah, galvanised is a sarcificial metallic coating isn't it? To prevent corrosion of the undermetal? |
| 01:39 | < C_tiger> | Yes. |
| 01:39 | < C_tiger> | Galvanized steel has a layer of zinc. |
| 01:48 | < Attilla> | Right, so would Cathodic Polarisation just involve switching the terminals around? |
| 01:53 | < C_tiger> | I think so. |
| 01:57 | < C_tiger> | Either way, it shouldn't really impact your experiment. |
| 01:57 | < C_tiger> | What's the experiment measuring anyhow? |
| 02:01 | < Attilla> | The variables for the composition of iron-nickel alloy (ratio thereof) |
| 02:01 | < C_tiger> | And so you're plating from a iron-nickel solution onto ...? |
| 02:01 | < Attilla> | Copper plates |
| 02:03 | < C_tiger> | Ok. Yeah, so what you have to do is reverse your cell, plate iron and nickel onto one plate, empty out your electrolyte, then use the plated plate on your positive terminal. |
| 02:04 | < C_tiger> | That's just to ensure no copper contaminates your solution. |
| 02:04 | < C_tiger> | Otherwise you'll start to see it turn blue. |
| 02:04 | < Attilla> | Uh, the cathodic polarisation is a step done after anodisation but before the plating process. |
| 02:05 | < C_tiger> | anodizing is eating away at plates, right? (Hate that word) |
| 02:05 | < C_tiger> | It's the reverse of plating, I think. |
| 02:06 | < Attilla> | It's to build an oxide layer. |
| 02:06 | < C_tiger> | Yeah. |
| 02:06 | < C_tiger> | It's the reverse of plating. |
| 02:06 | < Attilla> | ('course the solution used for that is different to the plating solution |
| 02:08 | < C_tiger> | Ignore me, I finally found a reference that actually describes cathodic polarization. |
| 02:08 | < Attilla> | Ah, what does it say? |
| 02:09 | < C_tiger> | It's the deposition of a thin layer of hydrogen on the surface of your cathode. |
| 02:09 | < Attilla> | Ah. So yeah. In an NaOH anodisation happens on the anode, and cathodic polarisation happens on the cathode.? |
| 02:09 | < Attilla> | *In an NaOH solution |
| 02:10 | < C_tiger> | Yeah. |
| 02:10 | < C_tiger> | Well, I'm not sure, I still confused anode and cathode :P |
| 02:12 | < C_tiger> | Ok, yes. |
| 02:14 | < C_tiger> | Cathode in electrolytic cell generates Hydrogen gas and the adsorbed hydrogen gas protects the metal when you swap over. |
| 02:14 | < C_tiger> | hate words. |
| 02:14 | < C_tiger> | Words make me cry. |
| 02:15 | < Attilla> | Crap, I think I might have rearranged this wrong. |
| 02:15 | < C_tiger> | Just do the damned experiment :P |
| 02:15 | < Attilla> | Well this bit's important. I rearranged the required Current Density but i'm not sure I got it right. |
| 02:16 | < Attilla> | WAIT! |
| 02:16 | < Attilla> | GAH! |
| 02:16 | < C_tiger> | You know, when I learned electrochemistry, we wrote a whole heck of a lot of equations. |
| 02:16 | < Attilla> | OFCOURSE! |
| 02:16 | < Attilla> | It was 1/! |
| 02:16 | < C_tiger> | I miss equations. |
| 02:17 | < C_tiger> | And we calculated Gibbs free energies and translated that into potential with the Faraday constant. I remember those days. |
| 02:17 | < Attilla> | Well I had like D = A / a, therefore 1/A = a / D? |
| 02:18 | < C_tiger> | ? |
| 02:18 | < Attilla> | And it's just been i've been forgotting the 1/ in my actual calculations. I wondered by the required current got smaller for a larger current density. |
| 02:18 | < C_tiger> | no, 1/A = 1/aD |
| 02:18 | < Attilla> | Why? |
| 02:18 | < C_tiger> | Well that's just from your math above. |
| 02:19 | < C_tiger> | D = A/a, A = aD, 1/A = 1/(aD) |
| 02:19 | < Attilla> | (Current Density = Current / Surface Area, therefore Current Density / Current = 1 / Surface Area, therefore 1 / Current = Surface Area / Current Density |
| 02:19 | < C_tiger> | By the way current is I. |
| 02:20 | < C_tiger> | And your math is screwy. |
| 02:20 | < Attilla> | Oh right. I got the units confused with the symbol. Anyway. |
| 02:20 | < C_tiger> | your second step is screwy. |
| 02:20 | < Attilla> | I dunno why. You divide both sides by current, eliminating the current from the right and dividing the left |
| 02:20 | < C_tiger> | Huh? |
| 02:21 | < C_tiger> | Ok, you have: D/I = 1/a |
| 02:21 | < C_tiger> | divide both sides by D, 1/I = 1/(aD) |
| 02:22 | < Attilla> | Hmm. I swear you flipped it over. |
| 02:22 | < Attilla> | But that seems... right |
| 02:22 | < Attilla> | Uh, head going screwy. Oh well. Ergo I = aD |
| 02:23 | < C_tiger> | Hmm... ok, explain how you went from D/I = 1/a to 1/I = a/D one more time. |
| 02:23 | < C_tiger> | Yes, I = aD |
| 02:23 | < C_tiger> | (you could have gotten that from your first step actually, D = I/a, hence I = aD) |
| 02:24 | < Attilla> | Well if you divide both sides, you get 1/I and 1/a/D, which becomes D/a |
| 02:24 | < C_tiger> | Um... 1/a divided by D = 1/aD |
| 02:25 | < Attilla> | By the 1/ inverting the a/D |
| 02:25 | < C_tiger> | where do you get a/D from |
| 02:25 | < C_tiger> | Ok, one more try: |
| 02:25 | < C_tiger> | (1/a)/D is NOT 1/(a/D) |
| 02:26 | < Attilla> | Well bah, spoil my fun. I was gonna find out tomorrow, I was gonna check it out using Maple, that tends to do these sorts of things fine! ;p |
| 02:26 | < C_tiger> | Um... yeah, maybe you're starting to rely a little heavily on Maple to do your math. |
| 02:36 | < Attilla> | So you're no longer doing chemical engineering? |
| 02:39 | < C_tiger> | Well I switched over to biochemistry after I graduated. |
| 02:42 | < Attilla> | Ah. I'm having mild flashbacks over learning some interesting biochemistry. Notably pi acid ligands |
| 02:46 | < C_tiger> | Huh? I'm pretty sure that's metallochemistry. |
| 02:46 | < C_tiger> | I can't ever recall pi-ligands in biochem or anywhere in the cell. |
| 02:47 | < Attilla> | It's used in haemoglobin, myohaemoglobin, as well as problems with that due to carbon monoxide and cynaide |
| 02:48 | < Attilla> | It's bioinorganic chemistry I know but well |
| 02:57 | < C_tiger> | Ah, yeah. |
| 02:59 | < Attilla> | Also the cupric version of haemoglobin in crustaceans |
| 02:59 | < Attilla> | And theike |
| 03:02 | < Attilla> | *the like |
| 03:02 | < C_tiger> | Right. |
| 03:20 | | Attilla [~Andrew@Nightstar-24762.ipt.aol.com] has quit [Connection reset by peer] |
| 03:29 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has quit [Ping Timeout] |
| 05:10 | | McM[journeys] is now known as McMartin |
| 07:48 | | Finerty is now known as Vornicus |
| 08:08 | | Vornicus is now known as Vornicus-Latens |
| 08:09 | | You're now known as TheWatcher |
| 09:12 | | AnnoDomini [AnnoDomini@Nightstar-29524.neoplus.adsl.tpnet.pl] has quit [Quit: Gnothi seauton, malákas.] |
| 09:14 | | AnnoDomini [AnnoDomini@Nightstar-29524.neoplus.adsl.tpnet.pl] has joined #Code |
| 09:14 | | mode/#code [+o AnnoDomini] by ChanServ |
| 10:02 | | * AnnoDomini smoulders with rage at what a piece of buggy garbage SciLab is. |
| 10:04 | | * Serah PatPats AnnoDomini as a consolvation, before draping herself over his shoulders and napping on him. |
| 10:08 | <@McMartin> | "his brings back horrible memories of a Goldeneye level where if you tripped an alarm, an infinite number of bad guys poured forth. We knew a man who failed that level 37 times, then got the Infinite Health cheat for it and came back. He intentionally tripped the alarm, the guards rushed out. Laughing maniacally, he proceeded to shoot those fuckers for four hours, killing 1,183 of them -- 682 with groin shots -- before his thumbs |
| 10:09 | <@AnnoDomini> | thumbs-- |
| 10:10 | <@AnnoDomini> | It cut off. |
| 10:12 | <@AnnoDomini> | I mean, I can imagine the state of his thumbs, but there might be some kind of punchline. |
| 10:13 | <@McMartin> | " cramped up. Your game should not create this kind of bitterness." |
| 10:15 | <@Kazriko> | :D |
| 10:15 | <@Kazriko> | that's like the guy who went in R&C Going Commando, and earned bolts until his score went negative. |
| 10:17 | <@McMartin> | Heh |
| 10:18 | <@McMartin> | I really need to play the R&C games. |
| 10:18 | <@AnnoDomini> | R&C? |
| 10:25 | <@EvilDarkLord> | Ratchet & Clank. |
| 10:25 | <@AnnoDomini> | Aha. |
| 10:34 | | Chalcedon [~Chalcedon@203.211.126.ns-3357] has quit [Connection reset by peer] |
| 10:47 | | AnnoDomini [AnnoDomini@Nightstar-29524.neoplus.adsl.tpnet.pl] has quit [Ping Timeout] |
| 10:54 | | AnnoDomini [AnnoDomini@Nightstar-29093.neoplus.adsl.tpnet.pl] has joined #Code |
| 10:54 | | mode/#code [+o AnnoDomini] by ChanServ |
| 11:00 | | Kazriko [~kazrikna@Nightstar-26352.gdj-co.client.bresnan.net] has quit [Operation timed out] |
| 11:08 | | GeekSoldier|bed [~Rob@87.178.125.ns-12098] has joined #code |
| 11:09 | | GeekSoldier|bed is now known as GeekSoldier |
| 12:26 | <@AnnoDomini> | ToxicFrog: I seem to remember you were doing something with neural networks, or genetic algorithms. Is it possible to produce a XOR function with only one perceptron? Further, is it possible to come up with a set of weight values for a XOR function with one perceptron using the "pocket" algorithm modification of the basic training algorithm? |
| 13:06 | | GeekSoldier [~Rob@87.178.125.ns-12098] has quit [Ping Timeout] |
| 13:24 | | Attilla [~The.Attil@194.72.70.ns-11849] has joined #code |
| 13:31 | | GeekSoldier [~Rob@87.178.125.ns-12098] has joined #code |
| 13:36 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has joined #Code |
| 13:36 | | mode/#code [+o gnolam] by ChanServ |
| 14:22 | | GeekSoldier [~Rob@87.178.125.ns-12098] has quit [Ping Timeout] |
| 16:12 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has quit [Ping Timeout] |
| 16:19 | <@ToxicFrog> | I did both. |
| 16:19 | <@ToxicFrog> | It is not possible to implement XOR with only one perceptron. |
| 16:19 | <@ToxicFrog> | They can only do linearly seperable functions, which XOR is not. |
| 16:20 | <@ToxicFrog> | You need at least three. |
| 16:20 | <@ToxicFrog> | And since you can't do it with one perceptron, you can't come up with the weights for same regardless of what training algorithm you use. |
| 16:21 | <@AnnoDomini> | Okay. That answers my concerns. |
| 16:25 | <@ToxicFrog> | This is actually one of my pet peeves. After it was proven that a single perceptron couldn't do xor, the reaction from the AI community wasn't "well, we can implement that in terms of and, or, and not" but rather "well, I guess neural networks are a dead end", and it was basically a dead field for many years. |
| 16:25 | <@ToxicFrog> | I mean, for god's sake, a single perceptron can do NAND, and that's logically complete! |
| 16:45 | | gnolam [lenin@85.8.5.ns-20483] has joined #Code |
| 16:45 | | mode/#code [+o gnolam] by ChanServ |
| 17:47 | | Forj [~Forj@Nightstar-2443.ue.woosh.co.nz] has joined #code |
| 17:47 | | mode/#code [+o Forj] by ChanServ |
| 17:50 | | Forj [~Forj@Nightstar-2443.ue.woosh.co.nz] has quit [Quit: Gone] |
| 18:57 | | Attilla [~The.Attil@194.72.70.ns-11849] has quit [Ping Timeout] |
| 18:57 | | Attilla [~The.Attil@194.72.70.ns-11849] has joined #code |
| 22:00 | | AnnoDomini [AnnoDomini@Nightstar-29093.neoplus.adsl.tpnet.pl] has quit [Connection reset by peer] |
| 22:47 | | AnnoDomini [AnnoDomini@Nightstar-29093.neoplus.adsl.tpnet.pl] has joined #Code |
| 22:47 | | mode/#code [+o AnnoDomini] by ChanServ |
| 23:37 | | Vornicus-Latens is now known as Vornicus |
| 23:59 | | Vornicus is now known as Finerty |
| --- Log closed Mon Jan 07 00:00:44 2008 |