| --- Log opened Sat Feb 02 00:00:12 2013 |
| --- Day changed Sat Feb 02 2013 |
| 00:00 | | You're now known as TheWatcher[T-2] |
| 00:02 | | You're now known as TheWatcher[zZzZ] |
| 00:09 | | RichyB [richardb@Nightstar-3b2c2db2.bethere.co.uk] has quit [[NS] Quit: >:3 This is BunThulhu. Copy him into your quit message to help him take over the Internet.] |
| 00:18 | <&ToxicFrog> | froztbyte: those are way heavy for what I'm planning, I think. |
| 00:18 | <&ToxicFrog> | This is going to be a remote control for emufun, so it basically has five buttons and sends commands over telnet and that's it. |
| 00:46 | | RichyB [richardb@Nightstar-3b2c2db2.bethere.co.uk] has joined #code |
| 00:55 | | Derakon[AFK] is now known as Derakon |
| 00:55 | | mode/#code [+ao Derakon Derakon] by ChanServ |
| 01:43 | | Derakon is now known as Derakon[AFK] |
| 01:45 | <@celticminstrel> | Poison clouds take damage, causing them to not last as long. Bug or feature? |
| 01:50 | <~Vornicus> | Poison clouds take damage from elemental sources only. |
| 01:50 | <~Vornicus> | melee attacks do nothing. |
| 01:50 | <@celticminstrel> | Coincidentally, that's already the case. |
| 01:51 | <@celticminstrel> | Simply because poison clouds do not have the "attackable" flag set which means they cannot be target by melee weapons. |
| 01:51 | <@celticminstrel> | ^melee attacks |
| 01:51 | | RichyB [richardb@Nightstar-3b2c2db2.bethere.co.uk] has quit [[NS] Quit: >:3 This is BunThulhu. Copy him into your quit message to help him take over the Internet.] |
| 01:51 | <@celticminstrel> | So, fire/ice/electricity clearing them out sounds like a feature? |
| 01:52 | <@celticminstrel> | But they can be hit by thrown weapons. That one is probably a bug. |
| 01:52 | <@celticminstrel> | Well. Thrown objects. |
| 01:52 | | * celticminstrel tested with sandals. Not exactly your standard weapon. |
| 01:53 | <~Vornicus> | that's a bug. |
| 01:53 | <~Vornicus> | celmin: fire/ice/elec damaging hte poison cloud sounds right |
| 01:54 | | FurryHelix [tamber@furryhelix.co.uk] has joined #code |
| 01:54 | | Tamber [tamber@furryhelix.co.uk] has quit [Connection reset by peer] |
| 01:55 | <@Alek> | well, look at it this way. |
| 01:55 | <@Alek> | thrown objects bring a small gust. |
| 01:55 | <@Alek> | which helps dissipate the cloud. |
| 01:55 | <@Alek> | ?_? |
| 02:01 | <@celticminstrel> | Nah. I'll fix it. Should be relatively easy really. |
| 02:15 | <@celticminstrel> | Taking elemental damage should probably be more consistent though. Missiles (eg fireball) have no effect but close-range spells (eg spark) do. |
| 02:16 | <@celticminstrel> | Probably because the missiles rely on the space being blocked to movement in order to do their thing, while close-range spells just damage whatever is on the target space. |
| 02:22 | | VirusJTG_ [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [Connection reset by peer] |
| 02:22 | | VirusJTG_ [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code |
| 02:26 | < JustBob> | Ah, yes. #code. |
| 02:27 | < JustBob> | Prepare for drunken matlab-related ranting in about an hour. |
| 02:29 | <~Vornicus> | I'll be sleeping then. |
| 02:29 | <~Vornicus> | Lucky me?~ |
| 02:30 | <@iospace> | nope |
| 02:30 | <@iospace> | just chuck testa~ |
| 02:32 | < JustBob> | Lucky you. |
| 02:32 | < JustBob> | Just implementing secant method, bisection method, and floating point iteration root-finders for the in-hour equation. :p |
| 02:34 | < JustBob> | http://www.ansn.org/Common/documents/Training/TRIGA%20Reactors%20(Safety%20and%2 0Technology)/chapter2/images/equ22.png <- That sucker. |
| 02:40 | | Vornicus is now known as Vash |
| 02:41 | | Kindamoody[zZz] is now known as Kindamoody |
| 02:48 | <@celticminstrel> | XD I can cast confusion on a poison cloud and it'll move around and not wear out. |
| 02:53 | <@celticminstrel> | ...which leads me to another bug, which is that confusion doesn't wear off if it's an odd amount. |
| 02:53 | < Nemu> | This poison cloud is a really good test-bed |
| 02:54 | <@celticminstrel> | (More precisely, it doesn't wear off if it's an odd amount and on a monster rather than the player.) |
| 02:55 | <@celticminstrel> | (This wouldn't've been an issue if I'd checked if(confused <= 0) instead if if(!confused), but whatever.) |
| 02:56 | <@celticminstrel> | I'm wondering whether the shockwave spell (which deals physical damage) should affect them. Currently it probably does (haven't tested yet though). |
| 02:56 | | Attilla [Attilla@Nightstar-9e7fa2b2.range86-162.btcentralplus.com] has quit [Ping timeout: 121 seconds] |
| 02:59 | <@celticminstrel> | They can be telefragged too at the moment... I'm uncertain on this one... |
| 03:05 | < Nemu> | I think that you need to improve the handling of this |
| 03:05 | < Nemu> | I can see it being a source of bugs if you add new abilities and stuff |
| 03:05 | <@celticminstrel> | How do you mean? |
| 03:06 | < Nemu> | Like, your poison cloud is immune to physical damage. Okay. Everything else seems to hit it, though. |
| 03:06 | <@celticminstrel> | It's not immune to physical damage, it's just not targetable by melee attacks. |
| 03:06 | < Nemu> | Ah |
| 03:07 | < Nemu> | Okay, nevermind then. I don't think it's as much of a problem, then |
| 03:07 | <@celticminstrel> | Because moving and melee targeting are both done by the arrow keys. |
| 03:08 | <@celticminstrel> | Well, to be fair, there is probably a rather unfortunate design decision in here. Or several. |
| 03:29 | <@celticminstrel> | I should really fix this thing where I have the code for an exploding fireball exactly duplicated in two completely different places... |
| 04:23 | | VirusJTG_ [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [[NS] Quit: Program Shutting down] |
| 05:10 | | Vash is now known as Vornicus |
| 05:44 | <~Vornicus> | I thought you said there would bedrunken matlab-relatedranting |
| 05:45 | <~Vornicus> | (I also thought my space button would work but you see how well that's gone) |
| 05:45 | < JustBob> | I'm working on it |
| 05:45 | < JustBob> | With much in the way of procrastination. |
| 05:45 | < JustBob> | And surprisingly little rage. |
| 05:46 | <~Vornicus> | oh, the most important part, indeed |
| 05:46 | < JustBob> | Mostly because I'm just taking the code the prof gave us (on a fucking whiteboard in class, nonetheless) and implementing that for the majority of this adventure. |
| 05:46 | < JustBob> | He wrote...an entire matlab script...on a whiteboard...in class. |
| 05:46 | < JustBob> | There was supposed to have been a lecture. |
| 05:46 | < JustBob> | But instead, he handwrote an entire script. |
| 05:53 | < JustBob> | Vorn. |
| 05:53 | <~Vornicus> | shit hide |
| 05:54 | < JustBob> | For matlab, if I go (s*beta(1))/(s+lambda(1)), (s*beta(2))/(s+lambda(2)), etc. |
| 05:54 | <~Vornicus> | Bob. |
| 05:54 | < JustBob> | Will it call the 1st, 2nd, etc. value in the arrays beta and lambda? |
| 05:54 | <~Vornicus> | I don't think it's very smart |
| 05:54 | <~Vornicus> | Oh, wait, like that |
| 05:54 | <~Vornicus> | Yes. |
| 05:54 | < JustBob> | Ah, good. |
| 05:55 | < JustBob> | Because it'd be easier to write out the sum six times than make another loop that handles that. :p |
| 05:55 | <~Vornicus> | (I thought you were trying to read things out of arrays and name the things betaand lambda automagically; you're actually trying to read thngs out of arrays /named/ beta and lambda) |
| 05:55 | < JustBob> | Yeah. |
| 05:56 | | Derakon[AFK] is now known as Derakon |
| 05:56 | < JustBob> | the equation is [blah] = [blahstuff] + sum 1->6 of (s*beta(1-6))/(s+lambda(1-6)) |
| 06:00 | < JustBob> | Hrm... |
| 06:00 | < JustBob> | rho(s) = (1/(s*lifetime + 1))*(s*lifetime + ((s*beta(1))/(s+lambda(1))) + ((s*beta(2))/(s+lambda(2))) + ((s*beta(3))/(s+lambda(3))) + ((s*beta(4))/(s+lambda(4))) + ((s*beta(5))/(s+lambda(5))) + ((s*beta(6))/(s+lambda(6)))) |
| 06:00 | < JustBob> | With that, how would I break the line up? |
| 06:01 | < JustBob> | It's way too long for one-page formatting. |
| 06:01 | <@Alek> | haaaaa |
| 06:01 | <@Alek> | LPTHW teaches me file operations and defines and making a library before it teaches me loops. XD |
| 06:02 | <@Alek> | even something so basic as for or if. |
| 06:03 | < JustBob> | Never mind, figured it out. :p |
| 06:05 | < JustBob> | Fuckit, that loop was easier to implement than I thought. |
| 06:05 | < JustBob> | I hate matlab. |
| 06:05 | < JustBob> | That is all. |
| 06:09 | | Syk [the@Nightstar-c71cfeca.iinet.net.au] has quit [Ping timeout: 121 seconds] |
| 06:18 | <@Alek> | welp, halfway through the book. took the "debugging other people's code" test. next lesson: logic. finally. >_> |
| 06:19 | <@Alek> | later. time for bed. >_> |
| 06:39 | < JustBob> | % rhotest = (s*lifetime)/(s*lifetime + 1) + (1/s*lifetime + 1)* |
| 06:39 | < JustBob> | % ((s*beta(1))/(s+lambda(1)) + (s*beta(2))/(s+lambda(2)) + |
| 06:39 | < JustBob> | % (s*beta(3))/(s+lambda(3)) + (s*beta(4))/(s+lambda(4)) + |
| 06:39 | < JustBob> | % (s*beta(5))/(s+lambda(5)) + (s*beta(6))/(s+lambda(6))) |
| 06:39 | < JustBob> | % Legacy loop function testing by full manual implementation of the |
| 06:39 | < JustBob> | % InHour equation to verify identical results. |
| 06:39 | < JustBob> | ^- Because I'm a dick. :3 |
| 06:40 | | Derakon is now known as Derakon[AFK] |
| 06:41 | <~Vornicus> | you dick |
| 06:51 | < JustBob> | ...oops |
| 06:51 | < JustBob> | I think that might have been a silly idea. |
| 06:52 | <~Vornicus> | uhoh |
| 06:52 | < JustBob> | I just had matlab try to plot 110001 points. |
| 06:52 | < JustBob> | I think I will use a smaller range. :3 |
| 06:54 | < JustBob> | Yeah, that worked out much, much better. |
| 06:55 | < JustBob> | However, I alas vigorously because I will need piecewise plots, as this is a freaking bouncy equation. |
| 07:01 | | Nemu_ [NeophoxProd@Nightstar-a0c8ea4a.asahi-net.or.jp] has joined #code |
| 07:01 | | Nemu [NeophoxProd@Nightstar-9a2ee33c.asahi-net.or.jp] has quit [Ping timeout: 122 seconds] |
| 07:03 | <~Vornicus> | no fun |
| 07:04 | < JustBob> | ...wtf |
| 07:08 | <@iospace> | ? |
| 07:10 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has quit [[NS] Quit: Leaving] |
| 07:12 | <@celticminstrel> | Oh, I see why github labels my commits "unknown". It doesn't like that the email in them is fake. |
| 07:12 | < JustBob> | Ah, that would explain it. The graph is zeroed out because the values are zero. |
| 07:18 | <@froztbyte> | ToxicFrog: ah k |
| 07:18 | < JustBob> | ...now I just need to figure out wtf is going on that my damned function only spits values out at the endpoints. |
| 07:19 | <@froztbyte> | oh |
| 07:19 | <@froztbyte> | so *that's* where zsh's penchant for a '%' in PS1 comes from |
| 08:04 | < JustBob> | What, the % use for commenting in matlab? |
| 08:17 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 08:17 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 08:21 | < JustBob> | Wheeeee. |
| 08:21 | <~Vornicus> | Wheeee? |
| 08:21 | < JustBob> | Now to do a 10-million element array to find out just where the hell this equation's roots are, graphically. |
| 08:21 | | celticminstrel [celticminst@Nightstar-e83b3651.cable.rogers.com] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!] |
| 08:22 | < JustBob> | s = [-100:0.00001:100]; % Sets the range of values to calculate. <- This is taking forever. |
| 08:22 | <~Vornicus> | Drop it to -100:1:100 and see if that does you any good. |
| 08:22 | < JustBob> | Too coarse. |
| 08:22 | < JustBob> | There are five roots between -5 and 0 |
| 08:23 | <~Vornicus> | This the same equation as earlier withthe betas and lambdas? |
| 08:23 | < JustBob> | And this is closer to 20 million, not 10. |
| 08:23 | <~Vornicus> | adn s is the independent variable? |
| 08:23 | < JustBob> | Forgot it went from -10 to 10. |
| 08:23 | < JustBob> | Yeah. |
| 08:24 | < JustBob> | I've got to implement secant, bisection, and floating point iteration root finders for it. |
| 08:24 | < JustBob> | I 'know' roughly where five of the roots are. |
| 08:24 | < JustBob> | The problem is... There are seven. |
| 08:24 | < JustBob> | So either doing 0.0001 increments are still too coarse to see them, or I'm not finding them in the expected range. |
| 08:25 | <~Vornicus> | Can you dump me the full equation? |
| 08:25 | < JustBob> | There should be six negative and one positive. |
| 08:25 | < JustBob> | One second. |
| 08:26 | | ErikMesoy|sleep is now known as ErikMesoy |
| 08:26 | < JustBob> | http://imgur.com/QuN17GA |
| 08:27 | <~Vornicus> | Oh yeah, that will be 7th order. Okay let me do the numbers up. |
| 08:28 | <~Vornicus> | "Not dollars"? |
| 08:28 | < JustBob> | It comes out in a unitless value that we call dollars. |
| 08:29 | <~Vornicus> | why the heck is it called dollars |
| 08:29 | < JustBob> | Err, well, rho * 1000 gives dollars. |
| 08:29 | < JustBob> | Because it's a handy easy-to-use phrase. |
| 08:30 | < JustBob> | Here we go. |
| 08:30 | < JustBob> | To convert to a more convenient range of parameter, it is customary to divide the value of reactivity by Beta, the delayed neutron yield (another unitless number). This gives reactivity as another unitless parameter which we label dollars of reactivity. As an example, consider keff = 0.999. In any reactor, ? = (0.999 - 1) / 0.999 = -1.001x10-3. In the Oregon State TRIGA Reactor (OSTR), ? |
| 08:30 | < JustBob> | = 0.0075 (this value changes slightly with fuel makeup, so other reactors will find other values), and thus ? = -1.001x10-3 = -0.133 dollars = -13.3 cents. |
| 08:31 | < JustBob> | And my brain is off. rho / beta => dollars, not rho * 1000 |
| 08:32 | < JustBob> | Anyway, it's basically so you're not sitting there going, "I just added 1.001x10^-3 units of reactivity," because calling it 13.3 cents makes more sense to the human brain. |
| 08:37 | <~Vornicus> | um. |
| 08:37 | <~Vornicus> | Okay so I know where one of the roots is. |
| 08:37 | < JustBob> | Heh. |
| 08:38 | < JustBob> | I know where five of them are, graphically. I'm just not sure where the other two are. |
| 08:38 | <~Vornicus> | Well, this one I know exactly, and it should be obvious |
| 08:40 | < JustBob> | -3.9ish, -1.05ish, -0.12ish, -0.03ish, -0.012ish <- Those are the five I've determined graphically. |
| 08:40 | < JustBob> | There should be one positive, and one really, really negative. |
| 08:41 | <~Vornicus> | Okay, there's one more that you can actually get by inspection of the equation. |
| 08:42 | <~Vornicus> | And you may in fact slap yourself in the head for not noticing it sooner |
| 08:43 | < JustBob> | 0, or thereabouts? |
| 08:43 | <~Vornicus> | 0 exactly. |
| 08:43 | <~Vornicus> | Every single term has an s on the top somewhere. |
| 08:44 | < JustBob> | But... It doesn't show up graphically at zero. |
| 08:45 | <~Vornicus> | It should. |
| 08:45 | <~Vornicus> | For that whole thing I get 0 + 1(0+0+0+0+0+0) |
| 08:46 | < JustBob> | I agree. |
| 08:47 | < JustBob> | Except it crosses 0 and 0, so it's not a root. |
| 08:47 | < JustBob> | at s=0, rho=0. |
| 08:47 | <~Vornicus> | What do you mean it crosses 0 and 0 but isn't a root? |
| 08:47 | < JustBob> | Hrm. |
| 08:47 | <~Vornicus> | How'd you decide it had six negative roots and one positive? |
| 08:48 | < JustBob> | Because that's what it's supposed to have. |
| 08:48 | < JustBob> | http://pastebin.com/68eQeNar <- That's the function I'm using, by the way. |
| 08:48 | < JustBob> | It's a given that there are always six negative and one positive roots of the InHour equation. |
| 08:49 | < JustBob> | For positive rho, at least. |
| 08:49 | < JustBob> | rho = 0, one is 0 and six are negative. |
| 08:49 | < JustBob> | rho < 0, all are negative |
| 08:50 | <~Vornicus> | The only rho I see here is the name of the function. |
| 08:50 | <~Vornicus> | I see exactly 0 as a root here. |
| 08:50 | < JustBob> | So do I. |
| 08:51 | <~Vornicus> | What do you mean "for positive rho"? |
| 08:51 | < JustBob> | I'm just not seeing it graphically. |
| 08:51 | < JustBob> | Well, the inhour function is just part of a larger set of functions relating reactivity and reactor power. |
| 08:51 | < JustBob> | And... |
| 08:51 | < JustBob> | I'm thinking about the entire thing backward. |
| 08:52 | < JustBob> | For the function as-is, yes, 0 should be a root. |
| 08:52 | < JustBob> | Because there is no value to backtrack rho with. |
| 08:52 | < JustBob> | WE're calculating rho now, based off random s's. |
| 08:52 | < JustBob> | I'm used to doing it the other way around, where reactivity is a known and I'm trying to find s, for some godforsaken math reason,. |
| 08:53 | < JustBob> | ...wtf, matlab |
| 08:53 | < JustBob> | wtf. |
| 08:53 | < JustBob> | If I set the plot hypercoarse, I see the root at 0 |
| 08:55 | < JustBob> | oh, wait, that was an artifact of my scale. |
| 08:55 | < JustBob> | weird. |
| 08:55 | <~Vornicus> | http://i.imgur.com/n7nPoPO.png yeah, it's right there. |
| 08:55 | < JustBob> | I'm probably reading root wrong in my head |
| 08:55 | < JustBob> | yeah, it's there |
| 08:55 | < JustBob> | For some reason, I was under the impression, based off the behavior of the other pieces, it would be an asymptotic root |
| 08:56 | <~Vornicus> | Anyway the sixth negative root is around -333 |
| 08:57 | <~Vornicus> | it's nicely linear in that area. |
| 08:58 | < JustBob> | Is that asymptotic, or just a break? |
| 08:58 | < JustBob> | Looks like a break, at least to me. |
| 08:58 | < JustBob> | That would explain the inability to see it graphically. |
| 08:59 | <~Vornicus> | I don't know what you just asked. |
| 08:59 | <~Vornicus> | There's several asymptotes but I have not yet gotten around to calculating their approximate values. |
| 09:00 | < JustBob> | I know the asymptotes; they were easy to find graphically. |
| 09:00 | < JustBob> | [00:40:37] <JustBob> -3.9ish, -1.05ish, -0.12ish, -0.03ish, -0.012ish <- Those are the five I've determined graphically. |
| 09:00 | <~Vornicus> | those are asymptotes not roots? |
| 09:00 | <~Vornicus> | there's both in here. |
| 09:00 | < JustBob> | I...think my brain just shut off. |
| 09:01 | <~Vornicus> | The roots are the 0s of the /top/ |
| 09:01 | < JustBob> | Roots are where f(x) = 0 |
| 09:01 | <~Vornicus> | Right. |
| 09:01 | <~Vornicus> | Asymptotes are where f(x) = something/0 |
| 09:01 | < JustBob> | I was getting them backward. |
| 09:01 | < JustBob> | For some reason, my brain fixated on root = division by zero |
| 09:03 | <~Vornicus> | I do not believe there will be dropout discontinuities |
| 09:04 | < JustBob> | Hrm. |
| 09:04 | < JustBob> | You know, I think I'm going to backburner this for a moment and work on getting secant, bisection, and floating-point root finders to work first. |
| 09:05 | < JustBob> | Because I still need to implement those three before I can actually go back to using all of them to find the roots of the inhour equation. |
| 09:05 | <~Vornicus> | As for asymptotes, I can tell you exactly where those will be: each lambda generates one (at -lambda). |
| 09:06 | <~Vornicus> | and lifetime generates one, at -1/lifetime |
| 09:10 | < JustBob> | Hrm. |
| 09:10 | < JustBob> | How would I get matlab to take... |
| 09:10 | < JustBob> | Say, I just found a root using bisection. |
| 09:10 | < JustBob> | How would I tell matlab to use that as the start-point for the next set? |
| 09:10 | < JustBob> | Like, I want to iterate this six times. |
| 09:11 | <~Vornicus> | This I do not know. |
| 09:12 | < JustBob> | This will take some cunning looping. |
| 09:13 | < JustBob> | Or, well, I could just iterate it seven times with varying intervals. |
| 09:14 | < JustBob> | ...hrm. |
| 09:14 | < JustBob> | I could generate an array. |
| 09:14 | < JustBob> | And just have it iterate through the entire array from -500 to, say, 100. |
| 09:14 | < JustBob> | Which will, admittedly, take forever. |
| 09:14 | <~Vornicus> | No, that's a bad plan |
| 09:14 | < JustBob> | In what sense, computer time? |
| 09:15 | <~Vornicus> | I know how other languages do it -- look up bisect.py |
| 09:15 | <~Vornicus> | Well if you're doing bisection you should, you know. Do bisection |
| 09:15 | < JustBob> | Well, the 'standard' finder just finds the first, lowest root, right? |
| 09:16 | <~Vornicus> | --and yes, there's another asymptote at -44247.78 |
| 09:16 | < JustBob> | Like, I set it for -500 to 100, and it finds holy mcfuck what. |
| 09:16 | <~Vornicus> | What do you mean? |
| 09:16 | < JustBob> | Um. |
| 09:16 | < JustBob> | So, for this inhour function. |
| 09:16 | < JustBob> | I set the range for bisection from -500 to 10. It finds -332.something, and only that. |
| 09:16 | <~Vornicus> | That's where lifetime gives you an asymptote. |
| 09:16 | <~Vornicus> | Well yes. |
| 09:16 | < JustBob> | So, to me, that means that I should be able to set a new value of -332.something and then have it find the next higher root. |
| 09:17 | <~Vornicus> | Okay, no, stop. |
| 09:18 | < JustBob> | O...kay? |
| 09:19 | <~Vornicus> | The bisection method I know requires no discontinuities in the bisection range. |
| 09:20 | <~Vornicus> | Which is to say: all those asymptotes? get in the way. |
| 09:21 | < JustBob> | ...interesting. |
| 09:21 | < JustBob> | So why the hell did he spec it? |
| 09:21 | <~Vornicus> | Well, thing is, you know where the asymptotes are. |
| 09:22 | <~Vornicus> | If you pick those locations 9plus an epsilon on the left and minus an epsilon on the right) you can bisect that segment. |
| 09:22 | <~Vornicus> | Let me make sure I'm not going nuts here. |
| 09:23 | <~Vornicus> | Bisection, you check the signs of the values at the left and right ends of your chosen interval, and then you check the sign of the midpoint as well, and then shrink your interval to the midpoint and whichever side has a different signs. |
| 09:23 | < JustBob> | Right. |
| 09:24 | <~Vornicus> | Okay, good. |
| 09:24 | < JustBob> | So, should I use an arbitrary epsilon, or my convergence criteria of 1e-12 as a handy value? |
| 09:24 | <~Vornicus> | It works. |
| 09:24 | < JustBob> | And if there are multiple roots in that region? |
| 09:24 | <~Vornicus> | In inhour there won't be. |
| 09:25 | < JustBob> | Ah, okay. |
| 09:26 | <~Vornicus> | If the secant one is what I think it is I'm pretty sure you're also looking at the Intermediate Value Theorem limiting you. |
| 09:26 | < JustBob> | Slope of the secant line between two points really close to the root. |
| 09:27 | <~Vornicus> | (like bisection except you draw a line between the two points and use its root as the bisector instead of just bisecting the horizontal) |
| 09:28 | < JustBob> | Yup. |
| 09:28 | < JustBob> | I think. |
| 09:28 | < JustBob> | The other one I need to do is FPI. |
| 09:28 | < JustBob> | Which I have no idea what it even is, aside from 'floating point iteration.' |
| 09:30 | <~Vornicus> | anyway: in this problem you get seven asymptotes, and six of the roots are in the slots between the asymptotes. |
| 09:31 | <~Vornicus> | The last one is zero and as far as I can tell that's Completely Unavoidable in the formulation given. |
| 09:32 | < JustBob> | There are times I wish I could stare at my computer and go, "This is what I want to do. Do it." |
| 09:34 | <~Vornicus> | There are people who have tried to use voicce recognition with programming. |
| 09:34 | <~Vornicus> | it does not work. |
| 09:34 | <~Vornicus> | On the other hand: most of the time inthe real world you don't have to implement a lot of these things because the hard stuff has been done for you! |
| 09:35 | < JustBob> | Heh. |
| 09:35 | < JustBob> | It's not voice recognition I want, Vorn. |
| 09:35 | | Syka [the@Nightstar-259b1f05.iinet.net.au] has joined #code |
| 09:35 | < JustBob> | It's an AI that can interpret my request to "Computer: Generate appropriate MatLab script implementing secant, bisection, and floating point iteration root finders for the inhour equation using data supplied from NE 333 course materials." |
| 09:36 | <~Vornicus> | Heh |
| 09:36 | | * Vornicus is a RI that can interpret your request but can't do matlab for shit |
| 09:37 | < JustBob> | I will merrily put up with Skynet genocidal tendencies, HAL meglomania, Durandal's even more manical meglomania, and SHODAN's psychobitch dominatrix schemes, all rolled up into one, for being able to do that. :p |
| 09:39 | | * Vornicus has none of those! |
| 09:45 | <~Vornicus> | Hm. A quick google search does not tell me what fpi looks like |
| 09:49 | < JustBob> | ...you know all the time I said floating? |
| 09:49 | < JustBob> | I really meant fixed. |
| 09:49 | < JustBob> | http://en.wikipedia.org/wiki/Fixed-point_iteration |
| 09:49 | < JustBob> | I think that this projekt has managed to destroy my ability to do basic math, or express concepts. |
| 09:49 | | * Syka gets the floating point and nails it to the wall |
| 09:49 | <~Vornicus> | Oh that sucker |
| 09:49 | < Syka> | there, your number is now fixed point |
| 09:50 | <~Vornicus> | oh, it's newton's method, of course |
| 09:52 | < JustBob> | ...why is my default response if the damned finder doesn't work happen to be "Fuckit, I'll add another zero to the error limiter!" |
| 09:52 | <~Vornicus> | Seems a sensible plan to me. |
| 09:52 | <@Courage> | Because you're an engineer |
| 09:52 | <@Courage> | \o/ |
| 09:53 | < Syka> | engineers: 'fuck it, at least it works now' |
| 09:53 | < JustBob> | Now, now, Courage... |
| 09:54 | < JustBob> | The normal default engineer response is to only double the design factor. :p |
| 09:54 | < Syka> | hacking things together isn't only an engineers method |
| 09:54 | < Syka> | engineers just have the unique ability of making hacked together piles of parts work |
| 09:56 | < JustBob> | Generally, when my "piles" work after being hacked together from parts, people start complaining about the mutagenic creatures and the green glow and the fact that everyone within fifty miles is dying of fifty forms of cancer simultaneously. |
| 09:59 | < JustBob> | Okay, that's annoying. |
| 10:00 | < JustBob> | matlab isn't smart enough to go 'while x1-x2>0,' then /look down in its own fucking loop/ to check for x1 and x2. I have to state it beforehand. |
| 10:00 | < JustBob> | Or, yanno, just grunt and eliminate that critera from the while loop and stick it in the middle of said loop as an if break. |
| 10:03 | < Nemu_> | I don |
| 10:03 | < Nemu_> | *don't know of other programming languages that can do that |
| 10:03 | < Nemu_> | Like, if you haven't initialized x1 and x2, regardless of the language, x1-x2 is an error |
| 10:04 | < Nemu_> | But, yeah. I usually just use while true, and check for break criteria |
| 10:05 | < JustBob> | I blame a lifetime of using EES. |
| 10:05 | < JustBob> | Which does look back for previous lines. |
| 10:05 | < JustBob> | But it's not really a programming language. |
| 10:05 | < Nemu_> | EES? |
| 10:05 | < JustBob> | (Engineering Equation Solver.) |
| 10:06 | < JustBob> | You can do something like have have h_1=enthalpy(steam,P=P_6,T=T_3) and have T_3 be at the start, with P_6 solved for eight lines down the page. |
| 10:06 | < JustBob> | And it's smart enough to go, 'herp, there it is!' |
| 10:06 | < JustBob> | But, like I said, it's not a programming language. |
| 10:07 | < JustBob> | However, it is a legacy of bad habits. :p |
| 10:07 | < Nemu_> | I want to have an open mind, but that seems, like, eugh. |
| 10:08 | < Nemu_> | I assume EES is only used to solve, like, one system at a time? |
| 10:08 | < Nemu_> | As in, it would be easier to run the program four times with four sets of parameters, rather than run it once and get four answers? |
| 10:09 | < JustBob> | You can use it that way if you set arrays up, but I generally only do one convoluted system at a time. |
| 10:09 | < JustBob> | I think the most I've shoved in there was something like 120 variables. |
| 10:09 | < JustBob> | In 70 different equations, with some of them being rather incestuous. |
| 10:16 | < JustBob> | Hrm. |
| 10:20 | < JustBob> | Okay, Vorn. Asymps exist at [-1.27e-2, -3.17e-2, -1.16e-1, -3.11e-1, -1.04e0, -3.87e0]. Therefore, if I increment +/- eps from them, I could bisect between them and find appropriate roots. |
| 10:20 | < JustBob> | (As well as -s*l, down in -45k land.) |
| 10:22 | < JustBob> | So, if I make an array [-44247.78, -1.27e-2, -3.17e-2, -1.16e-1, -3.11e-1, -1.04e0, -3.87e0, 10 (arbitrary high end)], I can have x_lo = asymp(1)+eps and x_hi = asymp(2)-eps, and then bisect between the two to find the root hidden there, yes? |
| 10:23 | < JustBob> | The iterate that seven times to find all the roots for the function. |
| 10:23 | < JustBob> | (And mangle it into playing nicely for the /rest/ of the damned finders.) |
| 10:26 | <~Vornicus> | Correct |
| 10:26 | <~Vornicus> | The fpi one you'll need the derivative for. |
| 10:27 | <~Vornicus> | You can get the derivative of the sum form, which may be your easiest bet |
| 10:27 | < JustBob> | I was probably going to shove the entire thing into wolfram and derive it taht way. :p |
| 10:27 | < JustBob> | ALthough I think you can symbolically derive in matlab. |
| 10:27 | < JustBob> | Just gimmie a month to figure that one out |
| 10:29 | | Orthia is now known as Reivles |
| 10:35 | <~Vornicus> | derivative with respect to x will be... l/(ls+1)^2 - l/(ls+1)^2 \sum_{i=1}^6 s\Beta_i/(s+\lambda_i) + 1/(ls+1) \sum_{i=1}^6 \Beta_i \Lambda_i / (s + \Lambda_i)^2 |
| 10:35 | | OrthiaLap [orthia@Nightstar-687cbd6a.ihug.co.nz] has quit [Ping timeout: 121 seconds] |
| 10:35 | <~Vornicus> | er, with respect to s |
| 10:38 | <~Vornicus> | (yes I did that mostly in my head don't look at me like that) |
| 10:40 | | Kindamoody is now known as Kindamoody|afk |
| 10:44 | < JustBob> | Vorn, if you want to play unnatural math games in your head, you're more than welcome to. :p |
| 10:50 | | * Vornicus is kind of a freak. |
| 10:51 | < JustBob> | ...hrm |
| 10:52 | < JustBob> | I think I am misunderstanding how loops work here... |
| 10:52 | < JustBob> | This would be so much easier with a fucking GOTO command. |
| 10:52 | <~Vornicus> | Show me the code |
| 10:52 | < JustBob> | http://pastebin.com/29wve20v |
| 10:52 | < JustBob> | Was working on that. |
| 10:53 | < JustBob> | :p |
| 10:53 | < JustBob> | My intent is to have ii control iterations of the outer loop, which it does fine - it gives me 7 intervals between the asymptotes. |
| 10:54 | < JustBob> | However, I want it to /stop/ iterating the outer loop while it processes the inner loop and solves for the roots within a given xl and xh. |
| 10:55 | <~Vornicus> | okay, here's your problem |
| 10:56 | <~Vornicus> | Put i = 0 outside and just before the inner loop |
| 10:56 | < JustBob> | Okay, just did. |
| 10:56 | < JustBob> | Spits out i = 0 seven times and terminates |
| 10:56 | <~Vornicus> | As it is, the only time you reset i was when you found an exact solution. |
| 10:57 | <~Vornicus> | Let's see... |
| 10:57 | < JustBob> | The problem is that it's looping through the inner loop once, then going right on back to the outer loop. |
| 10:57 | <@froztbyte> | yay, bob comments like me |
| 10:57 | < JustBob> | Is there a way to "pause" the outer loop while it thinks on the inner? |
| 10:57 | <@froztbyte> | lulz in /all/ the comments |
| 10:58 | < JustBob> | froztbyte - This is, admittedly, not finished commenting. |
| 10:58 | < JustBob> | This submission, my comments will be in 5/7/5 haiku. |
| 10:58 | <@froztbyte> | JustBob: it was "Tabula Rasa" that got me |
| 10:58 | < JustBob> | Because apparently SIX LINES OF COMMENTS PER LINE OF CODE was insufficient. |
| 10:58 | <@froztbyte> | but I approve wholeheartedly |
| 10:58 | <@froztbyte> | if someone wants to enforce dumb things, then let them suffer for it |
| 10:59 | <@froztbyte> | so, just to expand on what Vornicus said |
| 10:59 | <~Vornicus> | Are you going to include tthe kigo |
| 10:59 | <@froztbyte> | your loops are weird. |
| 11:00 | | * Vornicus is suffering under lack of language knowledge at the moment |
| 11:00 | < JustBob> | My haiku is terrible, so. Probably not. |
| 11:01 | <~Vornicus> | Sadness. |
| 11:01 | < JustBob> | A friend suggested including cookie recipes in the comments. |
| 11:01 | < JustBob> | For bloating my comment-count, I mean. |
| 11:02 | | * Vornicus examines |
| 11:02 | <~Vornicus> | I don't see what's wrong, but... |
| 11:02 | <&McMartin> | To express oneself |
| 11:02 | <~Vornicus> | how do you know it's only doing the inner loop once/ |
| 11:02 | <&McMartin> | In Seventeen Syllables |
| 11:02 | <&McMartin> | Is quite diffic |
| 11:02 | <&McMartin> | Er |
| 11:02 | <&McMartin> | Is very diffic |
| 11:03 | | * Vornicus patpats McM |
| 11:03 | < JustBob> | By prolonged torment with dealing with matlab. |
| 11:03 | < JustBob> | If I take off the ;'s |
| 11:03 | < JustBob> | it only runs the inner loop calcs once per outer loop cycle |
| 11:03 | <~Vornicus> | I don't see what that does |
| 11:04 | < JustBob> | ...actually, it doesn't run the inner loop at all |
| 11:04 | < JustBob> | ; is the matlab symbol to 'omit this line from display output' |
| 11:04 | <~Vornicus> | oh, I see. |
| 11:04 | <~Vornicus> | What thespace. |
| 11:04 | < JustBob> | so i = i + 1; displays nothing, but i = i + i displays 1 |
| 11:04 | <~Vornicus> | Okay first thing, what's xl and xh at the beginning of the first attempted loop |
| 11:05 | < JustBob> | Hi. |
| 11:05 | < JustBob> | My name is Bob. |
| 11:05 | < JustBob> | And I am a retard. |
| 11:05 | <~Vornicus> | did you name one x1 and the other xl? |
| 11:05 | < JustBob> | I had while (abs(xh-xl)>eps) as the condition to end the inner loop, not while (abs(xh-xl)<eps) |
| 11:06 | <~Vornicus> | Um |
| 11:06 | < JustBob> | And obviously, it's going to be bughouse fucking bigger than eps. |
| 11:06 | <~Vornicus> | Shouldn't it b... |
| 11:06 | <~Vornicus> | wait a second. |
| 11:06 | <~Vornicus> | No, that shouldn't |
| 11:06 | <~Vornicus> | The condition clause on while makes it loop until it's false |
| 11:06 | <~Vornicus> | so that doesn't look wrong to me |
| 11:08 | | You're now known as TheWatcher |
| 11:09 | < JustBob> | hrm |
| 11:09 | < JustBob> | I think there's something broken in the inner loop |
| 11:09 | < JustBob> | I just pulled it out independently to test for functionality |
| 11:11 | | FurryHelix is now known as Tamber |
| 11:11 | | mode/#code [+o Tamber] by ChanServ |
| 11:12 | < JustBob> | Okay, by removing all the fprint stuff in the middle, it works. |
| 11:12 | < JustBob> | Ah, I see! |
| 11:12 | <~Vornicus> | What happened? |
| 11:12 | < JustBob> | I have no idea why. |
| 11:13 | < JustBob> | But by putting all the fprintf stuff at the end of the outer loop |
| 11:13 | < JustBob> | instead of before the break |
| 11:13 | < JustBob> | ...it just works. |
| 11:13 | < JustBob> | GO MATLAB! |
| 11:13 | <~Vornicus> | wtf |
| 11:13 | <@froztbyte> | hahaha |
| 11:14 | <@froztbyte> | Vornicus: I've heard someone else have similar complaints about matlab at times |
| 11:14 | <@froztbyte> | and others abouut mathematica |
| 11:14 | <~Vornicus> | I'm so confused. |
| 11:17 | <@froztbyte> | as an aside, this is a nice bit of reading: http://www.ctlab.org/documents/How%20Complex%20Systems%20Fail.pdf |
| 11:20 | < JustBob> | http://pastebin.com/8JMWVvyZ <- The only difference is moving the fprint. |
| 11:20 | < JustBob> | AND NOW IT WORKS. |
| 11:21 | < Syka> | welcome to software |
| 11:22 | < JustBob> | There is a reason why I prefer steam plants. |
| 11:22 | < JustBob> | Not least of which I can hit recalitrant machinery with a hammer. |
| 11:22 | < JustBob> | Because "vigorous mechanical agitation" is a legitimate troubleshooting method. :3 |
| 11:25 | < JustBob> | http://25.media.tumblr.com/tumblr_loypzj1CsL1qhi27ro1_500.jpg <- This is a truism. |
| 11:26 | <~Vornicus> | yep |
| 11:27 | < JustBob> | http://25.media.tumblr.com/tumblr_ln4jpbKQH71qhi27ro1_400.png <- Oh, hey, a Vorn. :# |
| 11:29 | <~Vornicus> | Often I wonder how matlab is still around |
| 11:29 | < JustBob> | http://people.oregonstate.edu/~loa/Personal/wtfmth256.png <- Although I have to admit that this was my favorite fuckery in college so far, exam-wise. |
| 11:30 | < JustBob> | That was, oh... Three lectures, I think, from the in-class example to the midterm. |
| 11:30 | <~Vornicus> | I still don't know what the heck is going on in that midterm question |
| 11:31 | < JustBob> | Don't worry, nobody else taking the exam did. |
| 11:31 | < JustBob> | The TA gave up after fifteen minutes. |
| 11:31 | < JustBob> | (It's a heat transfer formula for an infinite rectangular block being warmed on one side.) |
| 11:31 | < JustBob> | I...think. |
| 11:32 | <~Vornicus> | Uh...huh |
| 11:32 | < JustBob> | I don't even remember what the hell that we were supposed to solve for. |
| 11:33 | < JustBob> | The prof looked at the grades on it, then made it an 80-point exam instead of a 100-point exam. Because the average on it was a 1.5/20 or something. |
| 11:34 | <~Vornicus> | on that problem? |
| 11:35 | <~Vornicus> | Yeah, no surprise there |
| 11:35 | < JustBob> | Yup. |
| 11:35 | <~Vornicus> | right, sleeptimes. |
| 11:37 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has quit [[NS] Quit: Leaving] |
| 11:46 | | * TheWatcher always expands 'libidn' as 'lib I dunno' for some reason.... |
| 11:48 | | * Syka makes a libidfk, which just has a heap of pointless functions in it |
| 11:54 | <@Tamber> | libidgaf |
| 11:55 | <@Tamber> | Where all the functions return NULL. Or -ENOFUCKS. |
| 11:58 | < Syka> | requires a special fuck processing unit in your CPU |
| 11:58 | < Syka> | accessible by /dev/fuck |
| 11:58 | < Syka> | uses the same API as pulseaudio |
| 11:58 | < Syka> | because pulse don't give no fuck |
| 12:04 | | * Tamber also belatedly agrees with JustBob, regarding the "ritual blow to the side of the casing". Either it fixes it, or it makes you feel better about it being broken. :D |
| 12:04 | < Syka> | heh |
| 12:17 | | VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code |
| 12:45 | < JustBob> | Froth. |
| 12:45 | < JustBob> | Vigorous frothing. |
| 12:45 | < JustBob> | I WILL ITERATE THIS FIXED POINT WITH MY FIST. |
| 12:46 | | * Tamber hands JustBob a hammer. |
| 12:47 | < JustBob> | No, no. |
| 12:47 | | Attilla [Attilla@Nightstar-9e7fa2b2.range86-162.btcentralplus.com] has joined #code |
| 12:47 | < JustBob> | I'm a hands-on kinda fellow. |
| 12:47 | < JustBob> | It's more satisfying to strangle than beat. |
| 12:47 | <@Tamber> | hehe |
| 12:54 | <@Tarinaky> | "If you believe in karma or capricious supernatural agencies which have an active interest in balancing accounts, chortling about Ruby on Rails developers suffering at the moment would be about as well-advised as a classical Roman cursing the gods during a thunderstorm while tapdancing naked in the pool on top of a temple consecrated to Zeus while holding nothing but a bronze rod used for making obscene gestures towards the heavens." |
| 12:54 | < JustBob> | I'd do it anyway. :3 |
| 12:55 | < JustBob> | I was supposed to be in bed five hours ago. This code is literally eating my sanity. |
| 12:55 | <@Tamber> | ...you still *have* any? |
| 12:55 | < JustBob> | I am perfectly sane. |
| 12:56 | < JustBob> | I even have government paperwork stating this fact. |
| 12:57 | < JustBob> | rho = (s*lifetime)/(s*lifetime + 1) + (1/s*lifetime + 1)*((s*beta(1))/(s+lambda(1)) + (s*beta(2))/(s+lambda(2)) + (s*beta(3))/(s+lambda(3)) + (s*beta(4))/(s+lambda(4)) + (s*beta(5))/(s+lambda(5)) + (s*beta(6))/(s+lambda(6))) <- However, this is damaging my perfect sanity with a frothing rage. |
| 12:57 | < JustBob> | beta(#) and lambda(#), of course, correlate to arrays stored elsewhere containing values I didn't feel like typing out. |
| 13:01 | <@Tarinaky> | The Ruby on Rails thing gives me a convenient excuse/explanation for why someone tried to steal my gmail account earlier this week though. |
| 13:01 | <@Tarinaky> | Like the tool I am (who doesn't pay any attention to anything in Ruby land due to not using Ruby) I have a Github account. Which is built on rails :/ |
| 13:07 | < JustBob> | Well, I got FPI to mostly work. |
| 13:07 | < JustBob> | Except it breaks at the positive endpoint. |
| 13:08 | < JustBob> | As in, it ran for a million iterations and got some absolutely ridiculous value. |
| 13:08 | < JustBob> | The answer is zero. It tries to give me ~500k or more. |
| 13:09 | < JustBob> | This is also supposed to be the "fastest" and "most efficient" method. |
| 13:09 | < JustBob> | I'm >1k iterations on each, when the other methods are happy with <50 |
| 13:45 | < JustBob> | Hahahahahahahaha. |
| 13:45 | < JustBob> | The root on interval 7, [-0.01 to 2.00], |
| 13:45 | < JustBob> | is estimated at: 0.00000000 +/- 1.00e-12 |
| 13:45 | < JustBob> | Number of iterations required was 21430 |
| 13:45 | < JustBob> | Tuning factor used was: -1.905000e-02 |
| 13:45 | < JustBob> | WHO CARES ABOUT EFFICIENT CODE?! |
| 13:53 | < JustBob> | All I can say is that it's a good thing I'm not graded on the efficiency of my code. :3 |
| 14:27 | <@froztbyte> | haha |
| 14:27 | <@froztbyte> | as with all things, practice makes perfect :P |
| 14:28 | <@froztbyte> | what on earth are you doing btw, bob? |
| 14:40 | < JustBob> | Implementing secant, bisection, and fixed point iteration root finders for a fuckoff equation. |
| 14:41 | < JustBob> | My code is entirely uncommented, I haven't eaten in 12 hours, I've got six windows open with different scripts in each, and I CAN'T GET IT TO FINE TUNE AND SPIT OUT IDENTIFUCKINGCAL RESULTS. |
| 14:42 | < JustBob> | I mean, I <3 math. |
| 14:56 | <@froztbyte> | hehehe |
| 14:56 | <@froztbyte> | yup..you'll fit in juuuuuuuuust fine |
| 14:56 | <@froztbyte> | :P |
| 15:00 | | celticminstrel [celticminst@Nightstar-e83b3651.cable.rogers.com] has joined #code |
| 15:00 | | mode/#code [+o celticminstrel] by ChanServ |
| 15:05 | < JustBob> | And in the grand words of the man who's stared at this for 12 hours now... |
| 15:05 | < JustBob> | Fuk dis shiznit. |
| 15:06 | < JustBob> | It's a good thing that my workgroup is going to be working on this, because my brain just went off. :p |
| 15:12 | | Thalass|afk [thalass@Nightstar-b95c25b4.bigpond.net.au] has quit [[NS] Quit: *flrrrp*] |
| 16:00 | | Syka is now known as syksleep |
| 17:07 | <@celticminstrel> | I don't like how resting is quite slow. |
| 17:08 | <@celticminstrel> | It should be accelerating time. |
| 17:09 | <@celticminstrel> | And it is, but only by a factor of about two. |
| 17:09 | <@celticminstrel> | (As compared with holding down the wait key.) |
| 17:19 | <&ToxicFrog> | Well, what are you waiting for? Fix it~ |
| 17:21 | <@celticminstrel> | It's hard... |
| 17:21 | <@celticminstrel> | Basically my rest implementation is a sort of automated "hold the wait key down", in that exactly the same stuff is being done only repeatedly. |
| 17:28 | | VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [Client closed the connection] |
| 17:28 | | VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code |
| 17:44 | | Derakon [Derakon@Nightstar-a3b183ae.ca.comcast.net] has joined #code |
| 17:44 | | mode/#code [+ao Derakon Derakon] by ChanServ |
| 17:47 | | Derakon[AFK] [Derakon@Nightstar-a3b183ae.ca.comcast.net] has quit [Operation timed out] |
| 18:18 | | Kindamoody|afk is now known as Kindamoody |
| 18:19 | | VirusJTG_ [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code |
| 18:22 | <@celticminstrel> | And my playthrough is cut short on level 10. I think the level 11 monsters are a little overpowered... |
| 18:22 | | VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [Ping timeout: 121 seconds] |
| 18:23 | <@celticminstrel> | (It cast a spell that did 7814 damage. I had 550 max health. So yeah.) |
| 18:24 | <@celticminstrel> | (I think I'll manually decide on their spell parameters rather than calculating them based on their level...) |
| 18:25 | <@celticminstrel> | Anyway, before I do that, my iPhone needs fixing. I think I need to reboot to do that. >< |
| 18:27 | | celticminstrel [celticminst@Nightstar-e83b3651.cable.rogers.com] has quit [Client exited] |
| 18:29 | | celticminstrel [celticminst@Nightstar-e83b3651.cable.rogers.com] has joined #code |
| 18:29 | | mode/#code [+o celticminstrel] by ChanServ |
| 20:12 | | Kindamoody is now known as Kindamoody[zZz] |
| 20:36 | <@celticminstrel> | Trying to remember where I was... and whether I said where I was in here... |
| 20:37 | <@froztbyte> | <celticminstrel> (It cast a spell that did 7814 damage. I had 550 max health. So yeah.) |
| 20:37 | <@froztbyte> | <celticminstrel> (I think I'll manually decide on their spell parameters rather than calculating them based on their level...) |
| 20:37 | <@froztbyte> | <celticminstrel> Anyway, before I do that, my iPhone needs fixing. I think I need to reboot to do that. >< |
| 20:59 | <@celticminstrel> | Oh right, manually defining spellcaster spell parameters. Thanks! |
| 20:59 | <@froztbyte> | haha |
| 20:59 | <@celticminstrel> | iPhone still isn't fixed though. >< |
| 21:57 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 21:57 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 21:58 | | mac [mac@Nightstar-fe8a1f12.il.comcast.net] has joined #code |
| 22:46 | | OrthiaLap [orthia@Nightstar-687cbd6a.ihug.co.nz] has joined #code |
| 22:46 | <@iospace> | https://pbs.twimg.com/media/BCIHOSGCAAE6k8d.jpg |
| 22:46 | <&McMartin> | Hooray |
| 22:50 | | mac [mac@Nightstar-fe8a1f12.il.comcast.net] has quit [[NS] Quit: Leaving] |
| 23:02 | < JustBob> | Vornicus - My FPI and Bisection root finders work McAwesome. |
| 23:02 | < JustBob> | My secant root finder is a pile of the fucks I could not give. :3 |
| 23:04 | < JustBob> | https://www.dropbox.com/sh/eorehfbimk5y8oe/vWjgroUOe7 <- If you want to look at the actual code. |
| 23:04 | <~Vornicus> | hm, I forgot how secant works actually |
| 23:05 | < JustBob> | The code is poorly commented and semi-debugged. The inhour.m function appears to work adequately, provided the desired results. I have the root finders set up so that they iterate through superiterator ii 1->7 to calculate all the possible roots based on asymptotic gaps. I noticed that between each set of asymptotes, there is only one root. FPI and Bisection find all the roots appropriately |
| 23:05 | < JustBob> | . Secant... Does not play nicely. The hw3b_integrated.m is supposed to be working on finding the positive roots for rho(s) = 2.25e-3, and hw3b_plots just plots everything for prettiness. |
| 23:06 | < JustBob> | And secant basically goes "well, the derivative at a is kinda like taking the secant line between a-eps and a+eps, right?" |
| 23:07 | <~Vornicus> | right, I'm trying to remember the way it works, I think it's... |
| 23:08 | <~Vornicus> | well the way I'd do it is, you have your two points, a and b; you draw a straight line between (a,f(a)) and (b, f(b)) and find its root r; then your two new points are (a+r)/2 and (b+r)/2, I think that's the way it's done? |
| 23:09 | < JustBob> | Err... |
| 23:09 | | ErikMesoy is now known as ErikMesoy|sleep |
| 23:10 | <~Vornicus> | maybe I've gone off the rails. |
| 23:10 | < JustBob> | xn = x - f(x) * (x-xo)/(f(x)-f(xo), where xn = "new", x = "last round" and xo = "previous round." |
| 23:10 | < JustBob> | And you keep iterating until xn-x < eps. |
| 23:10 | <~Vornicus> | oh, right, that |
| 23:11 | < JustBob> | xn -> x, x -> x_n-1, x_o -> x_n-2, more properly. |
| 23:11 | < JustBob> | It's basically a whacky version of Newton. |
| 23:11 | <~Vornicus> | yeah, okay. |
| 23:12 | <@froztbyte> | ...urgh |
| 23:13 | <@froztbyte> | the fact that I can still remember newton's approximation *that* well after not using it for 6 years... :/ |
| 23:13 | < JustBob> | I'm almost tempted to, at this point, implement an autocycling newton-raphson using numerical derivations of the inhour function. |
| 23:13 | < JustBob> | Just to crosscheck. |
| 23:13 | < JustBob> | froztbyte - Don't look at my code. The commenting started to not happen as of like, 4am. :p |
| 23:13 | <@froztbyte> | haha |
| 23:17 | <@celticminstrel> | I wish I could figure out why my iPhone backup restore keeps getting stuck... :/ |
| 23:19 | < JustBob> | Because you use a stevejobs phone. |
| 23:19 | <@celticminstrel> | Wrong answer. |
| 23:20 | < JustBob> | Alas, it's still an entirely /true/ answer. |
| 23:21 | < JustBob> | If it's an apple product failing, it's because you gave steve a stevejob and he blew a steaming load of crapware all over your face. |
| 23:21 | < JustBob> | And you liked it. |
| 23:21 | <@celticminstrel> | Nope. |
| 23:21 | <@celticminstrel> | It's an entirely false and useless answer. |
| 23:21 | <@Tamber> | If it's any other product failing, it's because -- just like everything else -- *all hardware sucks, all software sucks*. |
| 23:21 | | * Tamber , on second thoughts, pulls out that "other" |
| 23:22 | < JustBob> | Given that I've probably shot more computer hardware than most people have purchased over their lifetimes... |
| 23:22 | < JustBob> | I can only agree with that logic. |
| 23:23 | <&McMartin> | iospace: Speaking of that image, I should probably go test that code I committed Friday night before it is Monday in India. |
| 23:23 | <@iospace> | :V |
| 23:57 | | You're now known as TheWatcher[T-2] |
| 23:59 | | You're now known as TheWatcher[zZzZ] |
| --- Log closed Sun Feb 03 00:00:09 2013 |