| --- Log opened Thu Dec 20 00:00:00 2012 |
| --- Day changed Thu Dec 20 2012 |
| 00:00 | <@Namegduf> | It might be a good idea to use a byte array. |
| 00:00 | < AbuDhabi_> | Okay, so. With an unsigned int, I'll be relatively safe to use 31? |
| 00:04 | <&McMartin> | log_2(sizeof(int)) |
| 00:04 | <&McMartin> | Well |
| 00:04 | <&McMartin> | log_2(sizeof(int) * 8) |
| 00:04 | <&McMartin> | You can do a check on startup to ensure that sizeof(int) is at least 4. Once that is confirmed you can use 32. |
| 00:05 | <&ToxicFrog> | The actual answer is "use uint32_t" |
| 00:05 | <&McMartin> | ^ |
| 00:05 | <&McMartin> | However, not all extant C compilers will recognize that |
| 00:05 | < AbuDhabi_> | And it will make my fingers go all spindly. |
| 00:05 | <&McMartin> | It's shorter than "unsigned"! |
| 00:06 | < AbuDhabi_> | Unsigned doesn't need a shift, any numbers, or a key very close to Enter. |
| 00:07 | <&ToxicFrog> | typedef uint32_t flagword |
| 00:07 | | himi [fow035@D741F1.243F35.CADC30.81D435] has joined #code |
| 00:07 | | mode/#code [+o himi] by ChanServ |
| 00:07 | <&ToxicFrog> | I'm pretty sure that <stdint.h> is available on everything but windows, and that windows has an equivalent |
| 00:08 | <&McMartin> | On Windows you *are* guaranteed that sizeof(int) >= 4, anyway |
| 00:08 | < AbuDhabi_> | Even Win3.1? :P |
| 00:09 | <&McMartin> | Win3.1 executables do not run on Win32 and vice versa |
| 00:09 | <&McMartin> | It's called "Win32" for a reason~ |
| 00:09 | < gnolaptop> | TL;DR: With your initial condition of "on a 32 bit machine that's at least within forlorn-look distance of the bleeding edge", then the answer is practically but not formally "32". |
| 00:10 | <&McMartin> | (In fact, x64 systems *cannot* run 16-bit executables, AIUI) |
| 00:10 | <@Alek> | thought they could emulate them. |
| 00:10 | <&McMartin> | It requires additional machinery - Win64 does not include it. |
| 00:11 | <@Alek> | boo |
| 00:11 | <&ToxicFrog> | Yes, but they can't do native execution the way win98 could, for example |
| 00:11 | <&McMartin> | One of my mom's apps stopped working when she got a 64-bit laptop for this reason. |
| 00:11 | <&McMartin> | I think it was Printmaster Premier |
| 00:11 | <@Alek> | how about Compatibility Mode? |
| 00:11 | <&ToxicFrog> | They've actually needed emulation since WinNT, it's just that Win7 no longer includes even the emulation software |
| 00:12 | <&McMartin> | My compat mode only goes back to Windows XP. |
| 00:12 | <&ToxicFrog> | Compat mode is for emulating behaviours of old versions of the win32api, not loading 16-bit executables on a 32- or 64-bit machine. |
| 00:13 | <&ToxicFrog> | Even on XP, where compat mode goes back to win95 by default, it won't let you run 16-bit apps without invoking NTVDM, IIRC. |
| 00:14 | | You're now known as TheWatcher[T-2] |
| 00:14 | <&McMartin> | And yeah, IIRC compat mode actually does the Windows equivalent of hacking LD_PRELOAD_PATH to point at entirely different versions of libc and the kernel |
| 00:14 | <&ToxicFrog> | Pretty much |
| 00:16 | | himi [fow035@D741F1.243F35.CADC30.81D435] has quit [Ping timeout: 121 seconds] |
| 00:16 | | You're now known as TheWatcher[zZzZ] |
| 00:29 | | AbuDhabi_ [annodomini@Nightstar-4cc7c7c5.adsl.tpnet.pl] has quit [Ping timeout: 121 seconds] |
| 00:37 | < Reiv_> | McMartin: Why is this the case? |
| 00:37 | <&McMartin> | What's the "this" there? |
| 00:38 | < Reiv_> | that 16-bit *cannot* run on 64bit |
| 00:38 | < Reiv_> | The emphasis caught my eye. |
| 00:39 | <@rms> | It's called deprecation |
| 00:39 | <@rms> | You know, not supporting old broken shot |
| 00:39 | <&McMartin> | The 16bit instruction set, the 32bit instruction set and the 64bit instruction set are all totally different instruction sets, requiring different hardware |
| 00:41 | < Reiv_> | Aaah. |
| 00:41 | < Reiv_> | But the 64bit chips have been rigged to run 32bit stuff cleanly? |
| 00:41 | < Reiv_> | I vaugely recall that was one of the big coups that put the AMD architecture as the standard, or something. |
| 00:42 | <@Tamber> | ...thats *possible* with x86? |
| 00:42 | <&McMartin> | The 64-bit chips include a 32-bit instruction decoder that operates on the same registers. |
| 00:42 | < Reiv_> | Ah, cute. |
| 00:42 | <&McMartin> | IIRC, you can write to EAX (a 32-bit register) and affect a predictable part of a specific 64-bit register |
| 00:43 | < Reiv_> | Tamber: hells yes. For years people were buying 64bit chips for their speed and multiple cores, and promptly installing 32bit WinXP on them because Vista was terrible and WinXP64 was even worse :p |
| 00:43 | < Xon> | <McMartin> The 16bit instruction set, the 32bit instruction set and the 64bit instruction set are all totally different instruction sets, requiring different hardware |
| 00:43 | < Xon> | this is actually wrong |
| 00:43 | <&McMartin> | How would you phrase it? |
| 00:43 | < Xon> | 16bit/32bit/64bit instruction set decoders are still on all 64bit x86 chips |
| 00:44 | <&McMartin> | ... that dosen't contradict the statement you quoted, it confirms it |
| 00:44 | < Xon> | it's a deliberate design choice that you dont' get to play with 16bit instruction sets when the cpu is in 64bit mode |
| 00:44 | < Xon> | but you can still easily reset the cpu back to 32bit mode and regain access to 16bit mode |
| 00:44 | < Xon> | it makes *no sense* todo so |
| 00:45 | <@Tamber> | Put shortly, then, "16-bit *cannot* run on 64-bit"? :p |
| 00:45 | <&McMartin> | The statement you're quoting is me saying "using different bits of silicon" |
| 00:45 | < Xon> | yeah, but not because it's different hardware |
| 00:45 | <&McMartin> | It's not the same decoder logic |
| 00:45 | < Xon> | (it's all just micro-ops/software in the CPU anyway) |
| 00:45 | <&McMartin> | Because the ISAs are different =P |
| 00:46 | < Xon> | most of the instructions are binary identical |
| 00:47 | < Xon> | and only a few instructions where actively removed in the migration to 64bit mode |
| 00:47 | | AbuDhabi [annodomini@Nightstar-4cc7c7c5.adsl.tpnet.pl] has joined #code |
| 00:48 | <&McMartin> | That's only true if you're referring to sub-registers, though, right? |
| 00:48 | <&McMartin> | MOV may be the same, but AX, EAX, and the 64-bit EAX are all different things |
| 00:48 | <&McMartin> | I forget the 64-bit register mnemonics, since I haven't had to write assembler for the chip in a decade |
| 00:48 | < Xon> | this is where variable length x86 op-codes come into play =p |
| 00:51 | | AnnoDomini [AnnoDomini@Nightstar-27936bbc.dynamic.chello.pl] has quit [Ping timeout: 121 seconds] |
| 00:54 | <&ToxicFrog> | McMartin: AMD64 registers use an R prefix in source, IIRC |
| 00:54 | <&ToxicFrog> | AX - EAX - RAX |
| 00:55 | <&McMartin> | Actually, AL - EAX - RAX~ |
| 00:55 | <&McMartin> | Er |
| 00:55 | <&McMartin> | AL - AX - EAX - RAX |
| 00:55 | <&McMartin> | Probably originally just "A" in the 8088~ |
| 00:56 | | * McMartin used the 6502. |
| 00:56 | | McMartin [mcmartin@Nightstar-a682fef2.pltn13.sbcglobal.net] has quit [[NS] Quit: kernel upgrade] |
| 00:56 | | AnnoDomini [AnnoDomini@Nightstar-27936bbc.dynamic.chello.pl] has joined #code |
| 00:56 | < Xon> | the naming hasn't been that imaginative |
| 00:56 | <&ToxicFrog> | It's been a while, but isn't it AL/AH? |
| 00:56 | <&ToxicFrog> | (being the low and high bytes of AX, respectively) |
| 00:56 | < Xon> | yeah |
| 00:58 | | AbuDhabi [annodomini@Nightstar-4cc7c7c5.adsl.tpnet.pl] has quit [Ping timeout: 121 seconds] |
| 00:59 | | himi [fow035@D741F1.243F35.CADC30.81D435] has joined #code |
| 00:59 | | mode/#code [+o himi] by ChanServ |
| 01:01 | < Xon> | but it's a pure design choice of not exposing the ability to switch to 16bit mode because it's a pile of cruff, the underlying ability to execute 16bit code is still present on a 64bit cpu. It just requires a full CPU reset to reach it rather than a fast mode-change |
| 01:02 | < Xon> | so it's wrong to say the machinary doesn't exist for it =p |
| 01:11 | | VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code |
| 01:21 | | McMartin [mcmartin@Nightstar-a682fef2.pltn13.sbcglobal.net] has joined #code |
| 01:21 | | mode/#code [+ao McMartin McMartin] by ChanServ |
| 02:02 | | Vornicus [vorn@Nightstar-221158c7.sd.cox.net] has joined #code |
| 02:02 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 02:11 | < simon`_> | I'm looking for cycles in a graph in Java, and I wonder how to represent the graph acyclically. in functional programming I'd just keep a list of tuples. |
| 02:11 | < simon`_> | as in, represent the graph using an acyclical datastructure. |
| 02:12 | <&McMartin> | An ArrayList of node objects (which may, themselves, be ArrayLists) is *like* tuples |
| 02:12 | < simon`_> | ugh. |
| 02:13 | < simon`_> | why isn't there a java.util.TwoTuple! :) |
| 02:13 | <&McMartin> | Well, you can make your own Pair easily enough |
| 02:13 | < simon`_> | I feel that most of my simple container classes are just heavy tuples. |
| 02:13 | <&McMartin> | nod |
| 02:13 | <&McMartin> | Yeah |
| 02:13 | <&McMartin> | One of the things that are A Bummer about most OO languages. |
| 02:14 | <~Vornicus> | My typical graph representation is the one where it's just a list or dict of lists. |
| 02:15 | <~Vornicus> | I also used a sparse matrix (quad-linked) once but that was for a data structures class. |
| 02:18 | <~Vornicus> | (advantage of using an adjacency matrix is that you can use a matrix-multiplication-like operation to do pathfinding and connectivity) |
| 02:23 | < simon`_> | cool |
| 02:23 | <&ToxicFrog> | simon`_: the traditional approach for graphs is a two-dimensional array where the two dimensions are both indexed by node ID and graph[from][to] is the edge weight from $from$ to $to$ |
| 02:24 | <&ToxicFrog> | For flat datastructures, I mean |
| 02:24 | < simon`_> | ToxicFrog, that sounds reasonable. |
| 02:24 | <~Vornicus> | That's the adjacency matrix method |
| 02:25 | < simon`_> | ToxicFrog, so now that I don't know exactly how many vertices and edges is in my graph because I'm building it from a set of objects that refer to one another cyclically, I'd have to use a sparse representation I think, or at least convert it to a two-dimensional array afterwards. |
| 02:25 | <~Vornicus> | Sparse is your friend. |
| 02:26 | <~Vornicus> | What language? |
| 02:26 | <~Vornicus> | oh, java |
| 02:26 | <&McMartin> | Which really doesn't have good options, but you can throw together some bad ones |
| 02:26 | <&ToxicFrog> | Why do you want an acyclical data structure for this, anyways? |
| 02:42 | <@Alek> | soooo... to run a 16-bit application... you'd dual-boot over to a 32-bit OS. right? |
| 02:43 | <&McMartin> | Alek: Actually, these days, you'd probably load it into DOSBOX |
| 02:43 | <&McMartin> | So that you could also properly simulate the hardware your application will expect |
| 02:44 | <@Alek> | mhm. |
| 02:44 | <@Alek> | not always reliable, though. |
| 02:44 | <&McMartin> | A better bet than trying to run it under Win7-32, though. |
| 02:50 | <@Alek> | mmh |
| 02:55 | <&ToxicFrog> | Also a better bet than trying to find drivers for modern hardware for an old OS. |
| 02:55 | < simon`_> | ToxicFrog, it's for an OO course assignment in making a spreadsheet. there are cell references that can be cyclical, and this has to be detected. |
| 02:57 | < simon`_> | ToxicFrog, incidentally, it matters a great deal if this is checked when creating the cell references (during which one can assume that no existing cycles exist) or when getting the cells (during which any amount of cycles may exist). unfortunately the assignment indirectly suggests the latter as the only correct option, although it is slightly more difficult. |
| 02:57 | | thalass [thalass@Nightstar-724ec5eb.bigpond.net.au] has joined #code |
| 02:58 | < simon`_> | i.e., do I just look for a cycle back to the cell itself, or can cycles occur elsewhere, in which case I want to use Dijkstra's or Tarjan's algorithm for determining strongly connected components. but ah, this is a 1st semester course. *sigh* |
| 03:02 | < simon`_> | the suggested solution (which is incorrect) says that you can just mark the references as you visit them, and if an already-marked one occurs, you've got a cycle. |
| 03:04 | < simon`_> | set 0 0 add ref 1 1 ref 2 2; set 1 1 ref 2 2; set 2 2 const 5; get 0 0; (postfix-notation) |
| 03:04 | < simon`_> | s/post/pre/ |
| 03:09 | <&ToxicFrog> | Why is that incorrect? |
| 03:12 | <~Vornicus> | The correct thing is to generate predecessors from each node and if you end up seeing yourself in the predecessor list you're screwed |
| 03:13 | <&ToxicFrog> | ...there's a distinction between that and the "incorrect" solution that I'm not seeing |
| 03:13 | <~Vornicus> | The suggested solution appears to fail on A1*A1, if I understand it |
| 03:13 | <~Vornicus> | (in A2) |
| 03:13 | <&ToxicFrog> | I can't parse "A1*A1 in A2" |
| 03:14 | <~Vornicus> | Spreadsheet, right |
| 03:14 | <~Vornicus> | in cell A2, write =A1*A1 |
| 03:15 | <&ToxicFrog> | Ok... |
| 03:15 | <~Vornicus> | You're referring to a single cell, twice |
| 03:15 | <&ToxicFrog> | Oh, I see what you mean |
| 03:17 | <~Vornicus> | By marking cells you've visited and calling out Circular! whenever you visit a cell twice, you lose because visiting a particular cell twice happens a lot. |
| 03:17 | <&ToxicFrog> | I'm making an implicit assumption here that either there is at most one edge from A to B foreach [A,B], or that you normalize from "set of edges" to "set of connected cells" when processing each cell |
| 03:17 | <&ToxicFrog> | I also missed the "spreadsheet" part |
| 03:17 | <~Vornicus> | Circular references hit me ridiculously hard when I was trying to build my spreadsheet to calculate resource requirements for certain Minecraft objects. |
| 03:18 | <~Vornicus> | Basically I had to give each item a "level" that was max(p.level for p in predecessors) + 1 |
| 03:34 | | VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [[NS] Quit: Program Shutting down] |
| 03:44 | | harlow [mac@Nightstar-fe8a1f12.il.comcast.net] has quit [Connection closed] |
| 03:59 | | syksleep is now known as Syk |
| 04:20 | | gnolaptop [lenin@Nightstar-bb103c17.cust.bredbandsbolaget.se] has quit [Ping timeout: 121 seconds] |
| 04:36 | | Kindamoody[zZz] is now known as Kindamoody |
| 05:11 | | thalass [thalass@Nightstar-724ec5eb.bigpond.net.au] has quit [[NS] Quit: *trudge*] |
| 06:15 | | himi [fow035@D741F1.243F35.CADC30.81D435] has quit [Client closed the connection] |
| 06:18 | | ErikMesoy|sleep is now known as ErikMesoy |
| 06:23 | | Kindamoody is now known as Kindamoody|out |
| 06:30 | | 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!] |
| 07:42 | | You're now known as TheWatcher |
| 07:57 | | AbuDhabi [annodomini@Nightstar-4cc7c7c5.adsl.tpnet.pl] has joined #code |
| 08:05 | | McMartin_ [mcmartin@Nightstar-312b12c0.pltn13.sbcglobal.net] has joined #code |
| 08:08 | | McMartin [mcmartin@Nightstar-a682fef2.pltn13.sbcglobal.net] has quit [Ping timeout: 121 seconds] |
| 08:09 | | Courage [Moltare@583787.FF2A18.190FE2.4D81A1] has quit [Ping timeout: 121 seconds] |
| 08:12 | | Moltare [Moltare@583787.FF2A18.190FE2.4D81A1] has joined #code |
| 08:54 | | AbuDhabi_ [annodomini@Nightstar-b954fa08.adsl.tpnet.pl] has joined #code |
| 08:57 | | AbuDhabi [annodomini@Nightstar-4cc7c7c5.adsl.tpnet.pl] has quit [Ping timeout: 121 seconds] |
| 09:03 | | McMartin_ is now known as McMartin |
| 09:04 | | mode/#code [+ao McMartin McMartin] by ChanServ |
| 09:24 | | gnolaptop [lenin@Nightstar-bb103c17.cust.bredbandsbolaget.se] has joined #code |
| 09:44 | | You're now known as TheWatcher[afk] |
| 10:30 | | RichyB [richardb@Nightstar-3b2c2db2.bethere.co.uk] has joined #code |
| 11:10 | | ErikMesoy [Erik@Nightstar-be32adc8.80-203-17.nextgentel.com] has quit [Client closed the connection] |
| 11:21 | | ErikMesoy [Erik@Nightstar-be32adc8.80-203-17.nextgentel.com] has joined #code |
| 11:41 | | Vornicus [vorn@Nightstar-221158c7.sd.cox.net] has quit [[NS] Quit: ] |
| 11:58 | | gnolaptop [lenin@Nightstar-bb103c17.cust.bredbandsbolaget.se] has quit [Ping timeout: 121 seconds] |
| 12:02 | | Pandemic [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [Connection reset by peer] |
| 12:02 | | Pandemic [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code |
| 12:02 | | mode/#code [+o Pandemic] by ChanServ |
| 12:18 | | ErikMesoy [Erik@Nightstar-be32adc8.80-203-17.nextgentel.com] has quit [Client closed the connection] |
| 12:19 | | ErikMesoy [Erik@Nightstar-be32adc8.80-203-17.nextgentel.com] has joined #code |
| 14:44 | | Syloq [Syloq@NetworkAdministrator.Nightstar.Net] has quit [Ping timeout: 121 seconds] |
| 14:51 | | iospace [Alexandria@Nightstar-e67f9d08.com] has quit [[NS] Quit: screen is being a dick again] |
| 14:53 | | iospace [Alexandria@Nightstar-e67f9d08.com] has joined #code |
| 14:53 | | mode/#code [+o iospace] by ChanServ |
| 14:54 | | iospace [Alexandria@Nightstar-e67f9d08.com] has quit [[NS] Quit: leaving] |
| 14:54 | | iospace [Alexandria@Nightstar-e67f9d08.com] has joined #code |
| 14:54 | | mode/#code [+o iospace] by ChanServ |
| 15:02 | < ErikMesoy> | hm. IDLE does not play nice with Tkinter |
| 15:08 | <@iospace> | > Tk |
| 15:08 | <@iospace> | > Tk was made for Tcl |
| 15:08 | <@iospace> | > Tcl |
| 15:08 | <@iospace> | > Tcl |
| 15:08 | <@iospace> | there's your problem :P |
| 15:09 | < Syk> | lol |
| 15:20 | < ErikMesoy> | Butbutbut the python wiki asserts that Tkinter is de-facto standard for Python GUI. >_> And what is with Tcl anyway? |
| 15:22 | <@Tarinaky> | Yeah, there's your problem :p |
| 15:28 | < ErikMesoy> | Using Python? |
| 15:51 | < ErikMesoy> | Hello? |
| 15:52 | | * Pandemic spreads to ErikMesoy |
| 16:05 | | McMartin_ [mcmartin@Nightstar-dc5686e1.pltn13.sbcglobal.net] has joined #code |
| 16:08 | | McMartin [mcmartin@Nightstar-312b12c0.pltn13.sbcglobal.net] has quit [Ping timeout: 121 seconds] |
| 16:14 | | Kindamoody|out is now known as Kindamoody |
| 16:29 | <@iospace> | ErikMesoy: Tcl is a horrible, horrible language |
| 16:29 | <@iospace> | it gives brainfuck a run for its money interms of stupid syntax |
| 16:30 | < ErikMesoy> | Ah. Now I am worried when I see the python doc on the subject. |
| 16:31 | < ErikMesoy> | 'When trying to answer questions of the form "how do I do blah", it is often best to find out how to do"blah" in straight Tk, and then convert this back into the corresponding Tkinter call.' |
| 16:33 | | * ErikMesoy is attempting to make a basic button interface with this. Has packed buttons A and B to TOP, and buttons C and D to LEFT. Result: An up-tack shape (?) with buttons C and D forming the horizontal and A and B forming the vertical. Not quite as expected. |
| 16:36 | < ErikMesoy> | May you live in interesting time. May you have an educational experience. |
| 16:36 | < ErikMesoy> | *times |
| 16:42 | < RichyB> | The only UI framework really worth using in Python? The built-in "turtle" module. :D |
| 16:44 | < ErikMesoy> | So I'm in the wrong language for a UI with buttons, basically? |
| 16:44 | < RichyB> | import turtle; [ turtle.forward(10) or turtle.right(10) for x in xrange(0, 36) ] |
| 16:44 | | celticminstrel [celticminst@Nightstar-e83b3651.cable.rogers.com] has joined #code |
| 16:44 | | mode/#code [+o celticminstrel] by ChanServ |
| 16:44 | < RichyB> | Nah, I'm just teasing. I've used GTK from Python; it works great. |
| 16:52 | < ErikMesoy> | But it appears to be obscure, since I can import Tkinter with no fiddling, while import pygtk throws ImportError. Any particular version you suggest I should get? |
| 16:53 | <@gnolam> | Ugh, GTK |
| 16:54 | < RichyB> | pygtk will prolly need to be installed separately, yo. |
| 16:54 | < RichyB> | Tkinter is (almost?) in the standard library. |
| 16:55 | < ErikMesoy> | Should I be looking at gtk.org or pygtk.org for downloads? Is there a relevant difference? |
| 16:55 | < ErikMesoy> | I suspect there are dependencies here. |
| 16:56 | < RichyB> | I don't know what platform you're using. OS X? Linux? |
| 16:56 | < RichyB> | On Linux, it should be in your repos. |
| 16:57 | <@gnolam> | If you want a proper GUI library, go directly to Qt. |
| 16:57 | < RichyB> | On OS X, I don't know, I don't even know if GTK looks nice on OS X or not. |
| 16:57 | < ErikMesoy> | I'm getting into a holy war, aren't I? |
| 16:57 | <@gnolam> | If it's just "quick and dirty and nobody except me will see it", then Tkinter should work. |
| 16:57 | < RichyB> | Well... yeah. |
| 16:57 | < RichyB> | On Linux, I find GTK really nice. |
| 16:58 | < RichyB> | I will concede to gnolam that Qt is prettier on other platforms than GTK is.] |
| 16:58 | < RichyB> | It also does more. (Qt has cross-platform GUI, networking, filesystem stuff...) |
| 16:59 | < ErikMesoy> | I'm mainly on Win7 with Linux access if it should prove necessary, and aiming to write a minigame that should be cross-platform-compatible. |
| 17:00 | < ErikMesoy> | Mostly as practice for later projects, so I want structural features like "having an interface that other people can use without needing to clone my computer setup". |
| 17:00 | < ErikMesoy> | BBL, dinner. So, overall, tips on where I should reboot from? Switch language? :p |
| 17:01 | <@Tamber> | You *want* to start another holy war? :p |
| 17:01 | < RichyB> | A brief look at stackoverflow answers to "what's the best Python GUI framework for Win32?" says that wxPython and PyQT are your best options. |
| 17:02 | < RichyB> | wxPython is based on wxWindows, which tends to look great on Windows, okay on OS X, kind ugly on Linux. |
| 17:02 | < RichyB> | buuuut wxPython is very definitely cross-platform. |
| 17:03 | < RichyB> | PyQT uses Qt which is awesome. Only real downside is that AIUI the license is your choice of (free, GPL) or (costs money, proprietary). |
| 17:04 | < RichyB> | If you're fine with the GPL (personally, I prefer to be able to release stuff as 3BSD) then PyQT is supposed to be lovely. |
| 17:11 | <@Tarinaky> | Crowd Sourcing: What's a good 'fuel economy' for a small mecha? |
| 17:16 | < ErikMesoy> | Tamber: As I understand it, holy wars are often about what's 'good' or 'best' in the abstract, and I'm hoping for more sedate responses when I ask what has specific features like 'good for UI with cross-platform usability'. :-) |
| 17:17 | < Syk> | Tarinaky: define fuel economy |
| 17:17 | < Syk> | as in miles per gallon, or biofuel vs fossil :P |
| 17:18 | <@Tamber> | Tarinaky, 160 babies/minute |
| 17:18 | <@Tamber> | (Or equivalent number of kittens.) |
| 17:18 | <@Tarinaky> | Syk: ml/s |
| 17:20 | < Syk> | hmm |
| 17:20 | <@Tarinaky> | I don't want to do an arsepull. |
| 17:21 | <@Tarinaky> | But I also don't want them to be walking around carrying obscene amounts of fuel. |
| 17:23 | < ErikMesoy> | I think you have to do one of the two. Maybe soften the arsepull by invoking "internal reprocessor plants" that increase efficiency. Or a small nuclear power source that provides the "base" level of power, and marginal other fuel is used to make the mecha do things other than be powered on. |
| 17:24 | <@gnolam> | RichyB: PySide. |
| 17:24 | <@gnolam> | LGPL. |
| 17:25 | < RichyB> | gnolam: sounded (from the stackoverflow posts) like it might not be stable? |
| 17:25 | <@Tarinaky> | ErikMesoy: Cars don't need massive amounts of fuel to get anywhere? |
| 17:25 | < RichyB> | Tarinaky: wolfram alpha + wikipedia + a little farting around say you could get 1e12 J of energy from 5g of mixed H_2 and H_3. :) |
| 17:25 | < RichyB> | I assume that mecha use fusion reactors. |
| 17:26 | <@gnolam> | It is - or at least was - backed by Nokia, so it should be stable enough. |
| 17:26 | < RichyB> | gnolam: Qt is stable, it was PySide I was asking after? |
| 17:26 | <@Tarinaky> | Like I said. These aren't big mecha, they're small mecha. |
| 17:26 | < ErikMesoy> | Tarinaky: Cars generally don't walk or get into combat operations or carry around a small armory+plating. |
| 17:26 | <@Tarinaky> | ErikMesoy: Tanks do two of those. |
| 17:27 | <@Tarinaky> | The third can be handwaved with Unobtanium/Myomers. |
| 17:27 | <@Alek> | how about a fluoride fuel? XD |
| 17:27 | <@Tarinaky> | I/soldiers don't walk around with a small tank of glucose stapped to my back with a straw >.> |
| 17:27 | <@gnolam> | RichyB: exactly. |
| 17:27 | <@Tarinaky> | Miniture Nuclear Fusion is supposed to be in here as Lostech/Holy Grail McGuffin. |
| 17:27 | < RichyB> | gnolam: cool. |
| 17:28 | < ErikMesoy> | Tanks live on supply lines. |
| 17:29 | <@Tarinaky> | Yeah, /that/ has an in-story explanation for. |
| 17:29 | <@Tarinaky> | (Any vehicle lives on supply lines~) |
| 17:30 | <@Tarinaky> | The team HQ is a giant resupply/refueling/transport vehicle which can carry tons of fuel. |
| 17:31 | < ErikMesoy> | Aha. |
| 17:33 | < ErikMesoy> | Well, then I withdraw my objections and I'd suggest grabbing a tank fuel efficiency value and dividing it by some number between 1 and 10. |
| 17:34 | <@Tarinaky> | I'm having trouble finding figures for 'per second'. |
| 17:39 | <@gnolam> | Take fuel efficiency per km and divide by the average speed? |
| 17:39 | <@gnolam> | Err |
| 17:39 | <@gnolam> | Fuel consumption |
| 17:39 | | * gnolam really can't brain anymore today. |
| 17:48 | <@gnolam> | ... as can be seen as what I said is quite off. But you should get what I was /trying/ to say. |
| 17:49 | < ErikMesoy> | Yeah, dimensional analysis. |
| 17:49 | < ErikMesoy> | (That got me through high school physics ridiculously well.) |
| 18:04 | | * Rhamphoryncus attempts to recover from soft-bricking his phone |
| 18:21 | <@Tarinaky> | I need a sanity check. |
| 18:22 | <@Tarinaky> | I'm getting something like a 27kW Alternator to drive my 3ton Mech. |
| 18:30 | <@Tamber> | I wouldn't be surprised by that, tbh. |
| 18:31 | <@Tarinaky> | Sweet. |
| 18:31 | <@Tarinaky> | This also gives me 18.8 mpg. |
| 18:31 | <@Tamber> | ...actually, that appears to be very, very low |
| 18:31 | <@Tamber> | If this quick poke around seems to be accurate, that is. |
| 18:32 | <@Tarinaky> | I'm not a petrol head so I've no real idea how much energy/power I should be getting. |
| 18:32 | <@Tamber> | "power required to accellerate 1400kg vehicle from rest to 25m/s in 5s: 87.5kW." |
| 18:33 | <@Tarinaky> | Ah. I'm only going 6 m/s. |
| 18:33 | <@Tamber> | You're also using a much bigger vehicle. |
| 18:33 | | * Tamber goes to calculate things |
| 18:33 | <@Tarinaky> | Only x2. |
| 18:35 | | * gnolam looks up a conversion into civilized units. |
| 18:35 | <@gnolam> | ... yes, that sounds very low indeed. |
| 18:36 | | * Tarinaky is pleased he took the decision to make a spreadsheet for this. |
| 18:36 | | Vornicus [vorn@Nightstar-221158c7.sd.cox.net] has joined #code |
| 18:36 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 18:36 | <@Tamber> | Let's say it's it's going to 6 m/s in 5s, for sake of argument. |
| 18:36 | <@Tamber> | 161kW, by my calculations. |
| 18:37 | <@Tarinaky> | Ah, wait. I've spotted an issue. |
| 18:37 | <~Vornicus> | acceleration 1.2 m/s^2 ; travel distance 15 m |
| 18:37 | <@Tarinaky> | This is to /remain at rest/ at speed. |
| 18:37 | <@Tarinaky> | Not to accelerate. |
| 18:37 | <@Tamber> | aha |
| 18:37 | <@gnolam> | "to remain at rest at speed"? |
| 18:38 | <@Tamber> | "to maintain current speed", I read that as. |
| 18:38 | <@Tarinaky> | The mechs starts at 6m/s. |
| 18:38 | <~Vornicus> | energy requirements 12.5 * mass J; power requirements 2.5 * mass W |
| 18:39 | <@gnolam> | And... you're doing drag calculations, or what? |
| 18:39 | <@Tarinaky> | I'm handwaving drag and friction in to a linear function of mass. |
| 18:40 | <@Tarinaky> | Because I have no concept of shape. |
| 18:40 | <@Tamber> | Bricks aren't very aerodynamic. |
| 18:40 | <@Tarinaky> | So my speed is just my power / mass. |
| 18:40 | <@gnolam> | So we're talking ex recto values here? In that case, just make up whatever values you like. |
| 18:41 | <@Tarinaky> | ex recto? |
| 18:41 | <~Vornicus> | "from the ass' |
| 18:41 | <@Tarinaky> | I want to avoid the numbers being ex recto as much as possible. |
| 18:41 | <@Tarinaky> | It's just I'm only approximating friction and drag. |
| 18:43 | <@Tarinaky> | If I add a factor of 10 and make it 270kW does that put me any closer to sane? |
| 18:43 | <@Tarinaky> | Puts my fuel economy to 1.88 mpg. |
| 18:47 | <@Tarinaky> | Apparently my mpg should be somewhere between 50 (hybrid car) and 8 (humvee). |
| 18:48 | <@Tarinaky> | So I think, maybe, 27kW isn't that bad? |
| 18:50 | <@Tamber> | Ex recto here, but I'd think a mecha would be more towards the "Hummer H2" end of the scale, if not further. :p |
| 18:51 | <@Tarinaky> | I can't seem to get Wolfram Alpha to understand "how much power to accelerate 3tons from 0 to 6 m/s in 1 second" |
| 18:53 | <&ToxicFrog> | 3 tonnes, 6m/s^2 for 1 second, 6N/kg == 6*3000 == 18kN, distance of 3m, power in J = force x distance, so you need 54kJ. I think. |
| 18:54 | <@gnolam> | Ek = mv^2/2. P = Ek/t. |
| 18:55 | <@Tarinaky> | If I divide by two I get 54kW for the Alternator. |
| 18:55 | < ErikMesoy> | Tarinaky: "3 tons * 6 m/s^2 * 1 second" ? |
| 19:12 | | Syk is now known as syksleep |
| 19:23 | <@Tarinaky> | Oh crap, Chrome crashed :/ |
| 19:25 | <@Rhamphoryncus> | Afaik robots are extraordinarily inefficient. We don't have effective leg designs (although there's some interesting research on it) |
| 19:29 | <@Tarinaky> | Science Fiction. |
| 19:29 | | * Tarinaky handwaves |
| 19:29 | < RichyB> | Rhamphoryncus: "robots" are fine. "Humanoid machines", you mean. :P |
| 19:29 | <@Rhamphoryncus> | Yes, but if you're trying to figure out efficiency you should look at the parts that matter :) |
| 19:30 | <@Tarinaky> | Which don't exist so direct comparison cannot be drawn. |
| 19:30 | <@Tarinaky> | Yes, I get it :p |
| 19:30 | <@Rhamphoryncus> | RichyB: no, there's more in the human leg than we can duplicate in robots as of yet |
| 19:31 | < RichyB> | What I mean is, there are plenty of real-life robots which are perfectly efficient and don't look anything like people. |
| 19:31 | <@Tarinaky> | Ended up with more mass and a lower speed, more mass, and a bigger engine still. But I think I have a vague idea where some of the parts need to be, stats wise. |
| 19:31 | < RichyB> | Bomb-disposal bots, UAVs, manufacturing bots. |
| 19:31 | <@Tarinaky> | I just made a guess for armour though. |
| 19:31 | <@Tarinaky> | My numbers are far too small for armour :/ |
| 19:32 | < RichyB> | That DARPA competition where a bunch of people computerised a bunch of cars and then conducted a cross-desert rally - those cars were all robots. :) |
| 19:32 | <@Rhamphoryncus> | RichyB: robots but not mecha, which is what I meant, sorry |
| 19:32 | < RichyB> | I was just objecting to your use of the word "robot" as if it meant "mecha" or "humanoid machine". :) |
| 19:33 | <@Rhamphoryncus> | alright |
| 19:33 | < RichyB> | Sorry for being an annoying pedant. |
| 19:33 | <@Tarinaky> | You should be sorry. Grr. |
| 19:38 | <~Vornicus> | wouldn't want to make a humanoid machine more than about ten feet tall anyway |
| 19:43 | <@Tarinaky> | I need to figure out how damage works at some point. |
| 19:43 | | Kindamoody is now known as Kindamoody[zZz] |
| 19:46 | <@Rhamphoryncus> | welp, I've yet to find anything specific about energy efficiency. Closest is one old robot that was as efficient as human walking |
| 20:09 | < RichyB> | How well does an ASIMO do? |
| 20:09 | < RichyB> | Alternately, you could just assume the same efficiency as a walking human. |
| 20:10 | < RichyB> | Evolution has done a reasonable job of making us walk fairly efficiently after all, because wasting energy by travelling inefficiently is selected against. |
| 20:11 | < RichyB> | My thought process there is "it would be evolutionarily advantageous to be more efficient at walking; humans are evolved machines; humans' walking is probably within some constant factor of being optimally efficient" |
| 20:12 | < RichyB> | ALSO |
| 20:12 | | Syloq [Syloq@NetworkAdministrator.Nightstar.Net] has joined #code |
| 20:12 | | mode/#code [+o Syloq] by ChanServ |
| 20:12 | < RichyB> | Why the Hell are we discussing sci-fi mecha on #code? |
| 20:12 | < RichyB> | This is wildly off-topic for #code |
| 20:13 | | Vornicus [vorn@Nightstar-221158c7.sd.cox.net] has quit [[NS] Quit: ] |
| 20:13 | < RichyB> | clearly it belongs in #fleet instead :D |
| 20:13 | <@Tarinaky> | It's related to a project I am eyeing up. |
| 20:13 | <@Tarinaky> | So it's related to #code that way. |
| 20:13 | <@Tarinaky> | Although at this point we're a little tangential to it. |
| 20:13 | < RichyB> | Well... it's setting, rather than an issue with the implementation. |
| 20:14 | < RichyB> | Also, #fleet is just a good channel to be in when talking about science fiction. ^_^ |
| 20:14 | <@Tarinaky> | My queries were specifically to do with quantities/constants in the equations/implementation. |
| 20:14 | <@Tarinaky> | Which makes it algorithms and pseudocode~ |
| 20:14 | < RichyB> | Yar. |
| 20:14 | <@Tarinaky> | What is #fleet? |
| 20:15 | < simon`_> | I intend to find out. |
| 20:15 | <@Tarinaky> | Report back within the hour, or we'll declare you MIA. |
| 20:15 | < simon`_> | a channel much like this one, it seems. |
| 20:15 | < simon`_> | presumably with less code. |
| 20:15 | <@Tarinaky> | And more... ships? |
| 20:16 | <@Tarinaky> | *badum-psch* |
| 20:16 | < RichyB> | AIUI it was originally concocted to discuss the crafting of a specific fictional universe that involved a lot of space-ships. |
| 20:17 | < RichyB> | Right now, it's mostly people with lovely personalities and science fiction. ^_^ |
| 20:19 | <@Alek> | quite. |
| 20:44 | | You're now known as TheWatcher |
| 20:45 | | Reiv [Reiver@Nightstar-2aaa792a.vf.net.nz] has quit [Ping timeout: 121 seconds] |
| 21:35 | | RichyB [richardb@Nightstar-3b2c2db2.bethere.co.uk] has quit [[NS] Quit: Leaving] |
| 22:05 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 22:05 | | mode/#code [+o himi] by ChanServ |
| 22:21 | | Netsplit *.net <-> *.split quits: @jerith, simon`_, @celticminstrel, @Rhamphoryncus, Reiv_, @himi, @rms, @froztbyte, syksleep, @PinkFreud |
| 22:21 | | gnolam [lenin@Nightstar-bb103c17.cust.bredbandsbolaget.se] has quit [Connection reset by peer] |
| 22:22 | | Netsplit over, joins: @PinkFreud, &jerith, @himi, @froztbyte, simon`_, Reiv_, @celticminstrel, @Rhamphoryncus, syksleep, @rms |
| 22:23 | | gnolam [lenin@Nightstar-bb103c17.cust.bredbandsbolaget.se] has joined #code |
| 22:23 | | mode/#code [+o gnolam] by ChanServ |
| 22:29 | < simon`_> | I asked this a while ago and stupidly forgot it. what do you call energy that is lost in an electric circuit as a result of the materials? (as opposed to "signal energy", I think) |
| 22:30 | <@Tamber> | "heat"? |
| 22:31 | < simon`_> | I could call it that. :) |
| 22:32 | <@Tamber> | And if it isn't heat, it will be, sooner or later. |
| 22:32 | <@Tamber> | :p |
| 22:32 | < ErikMesoy> | "waste heat"? |
| 22:34 | < simon`_> | when a computer gets hot, my impression is that it is primarily because electrons leave the intended paths within the grid of semiconductors. secondarily, it is because information is intentionally discarded in the computations and the electrons have no paths that prevent them from heating their surroundings. |
| 22:34 | < simon`_> | so to distinguish between heat produced in these two ways, I want a word for the first. |
| 22:35 | < simon`_> | (the second way is only a small, theoretical fraction) |
| 22:36 | <@Namegduf> | Uh... electricity doesn't flow if there's not a complete circuit. |
| 22:36 | < simon`_> | Namegduf, no, so you have those... what are they called... heat-out? |
| 22:37 | <@Namegduf> | There's no like, paths leading nowhere which just have electrons flying off into nowhere. That'd leave the circuitry positively charged. |
| 22:37 | < simon`_> | fan-out |
| 22:37 | <@Namegduf> | Fan out is something else. |
| 22:37 | <@Namegduf> | It's the number of outputs of a logic gate. |
| 22:38 | < simon`_> | oh! |
| 22:38 | <@Namegduf> | Or number of places a single output is attached to, equivalently. |
| 22:39 | <@Namegduf> | I think mostly it's just referred to as waste heat. |
| 22:40 | < simon`_> | ah, it seems like the word I'm looking for is signal energy, and the theoretical lower limit is the other one. (I thought signal energy was another name for the theoretical lower limit of the energy required to preserve one bit of information) |
| 22:41 | < simon`_> | so the signal energy eventually becomes waste heat because of thermodynamics, right? |
| 22:43 | <@Namegduf> | Most energy lost when operating an electrical device becomes waste heat. |
| 22:45 | < simon`_> | when you say most, is that because there is something else it might become? |
| 22:47 | <@Namegduf> | Electromagnets can do stuff. |
| 22:49 | | VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code |
| 23:11 | | Attilla_ [Obsolete@Nightstar-9e7fa2b2.range86-162.btcentralplus.com] has joined #code |
| 23:14 | | Attilla [Obsolete@Nightstar-9e7fa2b2.range86-162.btcentralplus.com] has quit [Ping timeout: 121 seconds] |
| 23:17 | <&ToxicFrog> | It can also become electromagnetic radiation. |
| 23:18 | <@iospace> | Dear Arduino: I understand you want to make using your stuff easy for new people, but when i do #include "file.h", i do expect to get functions from that file without "Undefined refrence" errors -_- |
| 23:19 | | ErikMesoy is now known as ErikMesoy|sleep |
| 23:20 | <&ToxicFrog> | iospace: why? |
| 23:21 | <@iospace> | because i wrote some helper functions in a different file |
| 23:21 | <@iospace> | there |
| 23:21 | <@iospace> | Arduino really uses C++, but i prefer C |
| 23:22 | <@iospace> | #ifdef __cplusplus \ extern "C" { \ #endif |
| 23:33 | | Attilla [Obsolete@Nightstar-9e7fa2b2.range86-162.btcentralplus.com] has joined #code |
| 23:34 | | Attilla_ [Obsolete@Nightstar-9e7fa2b2.range86-162.btcentralplus.com] has quit [Ping timeout: 121 seconds] |
| --- Log closed Fri Dec 21 00:00:03 2012 |