IRC logs for #openttd on OFTC at 2007-11-10
⏴ go to previous day
00:00:05 <TrueBrain> where M is 3, and K is the start-cells?
00:00:17 <Phazorx> obviously as M<<N and compressing and K is a M coordinates + value
00:00:32 <Phazorx> K is array of start cells
00:00:42 <TrueBrain> yeah, k, guess that one correctly ;)
00:00:44 <Phazorx> each elemnt is a byte value and it's cooredinates
00:01:31 <TrueBrain> any life example of this? :p
00:01:32 <Phazorx> depending on data set and K data you can in theory to compress N^N to N^M+M
00:01:45 <Phazorx> it needs a math engine to be feasible
00:01:46 <TrueBrain> any plans on making it? :p
00:02:08 <TrueBrain> I read more about such things, I try to remember where... but.. complete failure in my head :p
00:02:23 <TrueBrain> it was described differently though..
00:02:27 <Phazorx> it is somewhat borderline with fractal thoery
00:02:56 <Phazorx> and guessing values to comeup with checksums can be based on wavelet theory
00:03:07 <Phazorx> that emphisizes on errors and deviations
00:03:29 <Phazorx> in this case you'd try to come up with data set that would have them limited to 0
00:03:38 <Phazorx> it is somewhat possible
00:03:45 <Phazorx> but how good it will be really based on math behind it
00:03:56 <Phazorx> and i simply lack capacity on that level
00:04:23 <TrueBrain> well, it does sound nice :)
00:04:26 <Phazorx> i'm sure someone else with right approcha and background based on this can come to something much faster than i would
00:04:26 <TrueBrain> and if it works, it really compresses
00:04:37 <Phazorx> it is luck based you see
00:05:02 <Phazorx> but 16M to 772bytes sounds really nice
00:05:03 <TrueBrain> Phazorx: yup.. and I am still wonder about the part: unique ;)
00:05:15 <Phazorx> well why does it have to be unique
00:05:58 <Phazorx> you know binary search
00:06:39 <Phazorx> you need up to N tries to guess a number between 1 and 2^N
00:06:50 <Phazorx> but what do you get if you have less than N tries?
00:07:18 <Phazorx> well you get probability
00:07:26 <TrueBrain> N-1 tries can never be more than +/- 1 wrong
00:07:29 <Phazorx> and also you might get it from 1st attempt
00:07:41 <Eddi|zuHause2> probability is not enough
00:07:46 <Phazorx> Eddi|zuHause2: correct
00:07:49 <TrueBrain> that is what I worry about too yes ;)
00:07:58 <Eddi|zuHause2> you have to be 100% certain that your decompression is correct
00:08:05 <TrueBrain> and that is what I don't see here... the resulting decompression needs to be correct ;)
00:08:09 <TrueBrain> I let Eddi|zuHause2 talk :)
00:08:11 <Phazorx> but what if a number selection within probable subset is predefined
00:08:50 <Eddi|zuHause2> you need additional space to specify the subset
00:08:57 <Phazorx> if you know how "guessing" works at decompression part, you might give it appropriate data to guess it faster
00:09:14 <Phazorx> and perhaps you can guuide it so it would guess each one from 1 attempt rather than N
00:09:26 <Phazorx> Eddi|zuHause2: exactly
00:09:32 <Phazorx> that's what K array does
00:09:44 <TrueBrain> so, basicly, you guess, feed it to the decompresser, and check if the data is useful?
00:09:50 <Phazorx> but after that you still get not unique solution
00:10:03 <Phazorx> TrueBrain: essentialy that is the decompression algorithm
00:10:14 <Phazorx> and same for compression
00:10:23 <TrueBrain> but what if you make XOR values on both rows and columns?
00:10:30 <Phazorx> you feed decompression with K data untill it has enough to guess it from 1st try
00:10:36 <TrueBrain> how much values do you need to make an unique matrix?
00:10:54 <Phazorx> and that is specific for evey case
00:10:55 <TrueBrain> as that would speed up decompression ;)
00:11:10 <TrueBrain> anyway, that guessing from the 1st try is the most trickiest part ;)
00:11:11 <Phazorx> well bumping M and lowering N will too
00:11:19 <Eddi|zuHause2> i really do not think that "speed" is an issue here :)
00:11:52 <Phazorx> Eddi|zuHause2: a math part should be very simple from computer perspective while being infinitely uncomprehendable in human eyes
00:12:18 <Eddi|zuHause2> what makes you think my eyes are human? :p
00:12:19 <Phazorx> and having rigth math parts, especialy when it comes to conclusion "sratch all, lest try another starting cell"
00:12:41 <Phazorx> Eddi|zuHause2: i wasnt refering to thee or myself spefically :)
00:13:10 <Eddi|zuHause2> but i should go to bed
00:13:21 <Phazorx> and i should have dinner
00:13:38 <TrueBrain> Phazorx: tnx for the input :)
00:13:54 <Phazorx> thinking is my kind of fun
00:14:13 *** Wolf01|AWAY is now known as Wolf01
00:15:05 <TrueBrain> Phazorx: I miss it at my study currently ;) (hehe, and I am doing CS...)
00:15:31 <Phazorx> as far esuducation goes i dont have any formal one :)
00:15:33 <Belugas> technical question: is it possible to enabled IDE drives on a SATA system? I have to transfert quite a lot of data
00:15:55 <Phazorx> dont taky it as a guideline tho, but my point is thinking is a process and education is a base
00:16:12 <Phazorx> it would be nice if they come together but it doesnt always
00:16:32 <Phazorx> Belugas: dos your south bridge has IDE controller?
00:16:51 <Phazorx> there are sole el cheapo sata>pata addapters
00:16:57 <Phazorx> but i have not heard of reverse
00:23:53 <stillunknown> Belugas: There are converters which allow PATA drives to connect to a SATA controller.
00:27:25 <Belugas> there is one IDE port-controler and 5 SATA, but somehow, if IDE is detected, it looks like disabling the SATA
00:29:20 <Tefad> i think i finally found the source
00:29:26 <stillunknown> Strange, have you tried using the 4 normal ports and not the 1 "seperate" one.
00:30:13 <glx> ide works well with sata in my machine
00:30:20 <Belugas> not mine, mother's in law
00:30:27 <Belugas> anything special you did?
00:31:05 <glx> I have an HDD in SATA, a CD-RW and a DVD+-RW on IDE
00:33:08 <Gonozal_VIII> same here plus an ide hdd
00:33:42 <Belugas> so, sata and ide HD can coexist
00:33:55 <Belugas> so , why is it not the case here???
00:34:03 <Belugas> anythng special in the BIOS maybe?
00:34:16 <Wolf01> Belugas, try to set ide compatibility on bios for sata, i used it when i needed to clone the ide hard disk on a sata one
00:34:36 <Gonozal_VIII> don't know... i did nothing special, it just worked
00:37:22 <Wolf01> some motherboards allow 3 settings for hdd: sata only, sata & ide, sata as ide (or something like this), the last is more compatible but disables raid if i'm not wrong
00:42:59 <CIA-1> OpenTTD: glx * r11398 /trunk/src/newgrf_industries.cpp: -Fix [FS#1414]: industry variables 67 and 68 return was incorrect
00:45:40 <Belugas> good point, Wolf01. i will try that tomorrow
00:46:09 <Belugas> it sounded so weird to me that ide and sata on same machine were not meant to communicate
00:46:25 <Wolf01> eheh, that's my job :P
00:59:13 <Eddi|zuHause2> i have 2 ide HD, 2 ide DVD and 1 sata HD, works like a charm
00:59:40 <Eddi|zuHause2> i somehow have a 3rd ide port, but i never got that one running...
01:00:00 <glx> I have 2 IDE ports and 4 SATA
01:00:30 <glx> I'm free to updgrade when needed ;)
01:01:21 <Eddi|zuHause2> according to my board documentation, i have 2 dual IDE controllers, and a raid controller with the 3rd ide port (suitible for one drive) and 2 sata ports
01:04:05 <Eddi|zuHause2> but you don't have to put the drives into a raid
01:04:15 <Eddi|zuHause2> and i only have 1 drive connected
01:17:32 <CIA-1> OpenTTD: belugas * r11399 /trunk/ (25 files in 3 dirs):
01:17:32 <CIA-1> OpenTTD: -Feature(ette): transparency settings can now be saved and thus remembered.
01:17:32 <CIA-1> OpenTTD: -Codechange: Wrap all transparency settings in accessors, hiding the implementation
01:17:32 <CIA-1> OpenTTD: -Change: Clicking "transparent building" menu will toggle Houses And Trees only. The other options can be used in the transparency gui.
01:17:33 <CIA-1> OpenTTD: Initial patch by Smatz (FS#1349), with some rework by BigBB and your humble servant.
01:31:35 *** Eddi|zuHause3 has joined #openttd
03:28:25 *** XeryusTC has joined #openttd
03:52:44 *** Dephenom has joined #openttd
03:59:56 *** Dephenom has joined #openttd
05:05:01 *** Diabolic-Angel has joined #openttd
06:21:14 *** DaleStan_ has joined #openttd
06:58:32 *** Frostregen_ has joined #openttd
07:04:12 *** Frostregen_ is now known as Frostregen
07:32:10 *** ThePizzaKing has joined #openttd
08:11:05 *** Ammller has joined #openttd
08:23:13 *** Diabolic-Angel has quit IRC
08:51:38 *** Gonozal_VIII has joined #openttd
09:39:19 *** dihedral has joined #openttd
09:40:58 *** Eddi|zuHause has joined #openttd
09:45:08 *** Eddi|zuHause has joined #openttd
09:47:56 *** stillunknown has joined #openttd
09:53:40 <_42_> dihedral, I found 2 matches to your query: [1]Soup, Soup. [1]Soup (~Soup@66-230-114-105-dsl-rb1.nwc.acsalaska.net) was last seen being kicked from #openttd by TrueBrain ( Go kick) 11 hours 52 minutes ago (09.11. 22:01), after spending 16 seconds there.
09:53:40 *** Progman has joined #openttd
10:07:21 <Wolf01> does the intro screen viewport scrolling (like locomotion or rollercoaster) interests somebody?
10:09:24 <Gonozal_VIII> you could scroll around in an earlier version when you desynced from multiplayer while the right mouse button was pressed
10:10:05 <Gonozal_VIII> not very interesting there
10:13:13 <Eddi|zuHause> i'm pretty sure he means automatically jumping between starting screen locations (useful with a starting screen that has multiple sceneries set up
10:15:06 <Wolf01> i'm trying to add jumping locations with signs
10:16:11 <Gonozal_VIII> the currently used map is not interesting outside the standard view, would have to use a different one
10:18:04 <Wolf01> but you can always put your best game there
10:25:44 <Eddi|zuHause> Gonozal_VIII: you can replace the title screen with any savegame
10:26:02 <Eddi|zuHause> just rename it to opntitle.dat
10:26:21 <Gonozal_VIII> oh... i didn't know that
10:27:19 <Eddi|zuHause> likewise, you can play the title screen by renaming it .sav
11:00:02 *** LeviathNL has joined #openttd
11:26:39 *** Gonozal_VIII has joined #openttd
11:28:54 *** Diabolic-Angel has joined #openttd
11:30:00 <Gonozal_VIII> i've changed my title screen a bit now, interconti airport instead of the normal one, helidepot instead of heliport, drive through bus/mail station and i also added a bridge over some diagonal rail tiles :-)
11:30:18 <Gonozal_VIII> and some presignals...
11:33:20 <LeviathNL> glx, nice work with the ind var 67 68 fix.
11:35:27 <LeviathNL> Is it correct still more than 1 tourist centre of the same type is generated (is this because of bigger maps?)
11:36:26 *** Vikthor has joined #openttd
11:42:58 *** HerzogDeXtEr has joined #openttd
11:46:15 *** tokai|ni has joined #openttd
11:56:00 <mikl> Gonozal_VIII: hvad er det sjove ved det?
11:56:39 <mikl> ah, sorry - wrong channel...
12:38:58 *** Farden123 has joined #openttd
12:45:09 *** Farden123 is now known as Farden
12:49:49 <csaba> how can change the road direction?
12:50:26 <csaba> there are onliy two directions in the build road menu
12:50:42 <Gonozal_VIII> there is no diagonal road
12:50:53 <fjb> Roads can be build only in two directions.
12:57:36 <csaba> on more : can i rotate the live map?
12:58:33 *** Greyscale has joined #openttd
12:58:52 <Gonozal_VIII> but you can make the buildings/trees transparent to look behind them
13:00:53 <Gonozal_VIII> hold the mouse button down on the third menubutton and select transparency options
13:04:07 <fjb> Or just press x to toggle the transparency.
13:07:19 <csaba> sorry i play with standard ttd because im practicing
13:08:53 <Ammller> csaba: really TTD or at least TTDPatch?
13:09:46 <csaba> i dont know exactly but i had to apply something patch for xp
13:11:20 <Ammller> csaba: there is also an official TTDPatch channel, #tyccon @ quakenet
13:12:02 <Ammller> yeah, they can help you better with i.e. how to make something transparent
13:12:21 <Ammller> or you switch to OTTD, should be easier
13:13:31 <csaba> because im a beginner i dont want to wander on the map without aim on the online wersion
13:14:46 *** Hendikins has joined #openttd
13:14:55 *** |Bastiaan| has joined #openttd
13:15:08 <Gonozal_VIII> openttd is not only multiplayer, it can do anything that the original ttd could
13:16:14 *** Brianetta has joined #openttd
13:18:14 <csaba> that is all i know tell :)
13:20:17 <csaba> there is 4 kind of bridge:)
13:21:04 <Gonozal_VIII> there are more than 4
13:22:12 <Gonozal_VIII> oh no! the pressure! it's too much!
13:23:54 <Gonozal_VIII> but i think there are also more then 4 bridges in ttd
13:24:10 <csaba> caused by me, ok wont ask
13:26:49 <Gonozal_VIII> you can ask anything you want
13:30:37 *** Vikthor has joined #openttd
13:30:45 <fjb> But don't expect an answer to every question. ;-)
13:40:24 <Gonozal_VIII> you have to think ahead ;-)
13:41:14 <Gonozal_VIII> what about them?
13:41:31 <fjb> You will learn how the slopes are working. Every beginner has some trouble with slopes.
13:47:20 <csaba> thanks everything see you soon:)
13:51:37 *** XeryusTC has joined #openttd
13:53:43 *** |Bastiaan| has joined #openttd
13:56:33 <Ammller> [14:40] <csaba> the slopes the slopes <-- they are really hard in TTD
13:57:39 <Gonozal_VIII> can't remember if i had problems with them at first, that's so many years ago
13:58:09 <Ammller> you had no foundations, if I remember right?
14:02:16 *** |Bastiaan| has joined #openttd
14:39:05 *** Farden123 has joined #openttd
14:45:47 *** Farden123 is now known as Farden
14:54:32 *** Wolf01 is now known as Wolf01|AWAY
15:06:59 *** Eddi|zuHause has joined #openttd
15:10:23 *** skidd13 has joined #openttd
15:12:50 <skidd13> TrueBrain: I've been thinking over the code. There are functions missplaced in various files. And I come to the conclusion that file cleanup would be usefull. My thinking got into the direction to split the engine from the game (or at least to clarify the border a bit better)
15:13:10 <TrueBrain> skidd13: many tried, many failed :)
15:13:18 <TrueBrain> resolving the include mess for example
15:13:21 <TrueBrain> is always a good thing
15:13:31 <skidd13> I knew. But I'm not many :D
15:13:49 <TrueBrain> a while back we made a purposal for directory structure
15:13:57 <TrueBrain> which should help for things like that
15:14:00 <TrueBrain> but I guess it got lost...
15:14:06 <skidd13> Do you have a backup?
15:14:08 <TrueBrain> (like newgrfs in newgrf/ )
15:14:57 <TrueBrain> if I remember correctly, Rubidium had that file on his http, but I am sure I am making things up
15:15:32 <skidd13> Moving the random functions and the templates of macros.h should be collected in something like mat.h
15:15:54 <TrueBrain> skidd13: wrote down a purposal, and I am sure someone will look into it :)
15:16:07 <Gonozal_VIII> hehe how long will you remember that "making things up" comment?
15:16:30 <TrueBrain> skidd13: wiki sounds nice
15:39:33 *** Farden123 has joined #openttd
15:46:24 *** Farden123 is now known as Farden
16:05:46 <TrueBrain> skidd13: I miss newgrf :p
16:05:55 <TrueBrain> maybe 'command'? :)
16:06:18 <skidd13> Add it where you thing it should be!
16:18:39 *** thgergo has joined #openttd
16:22:30 <Eddi|zuHause> that sounds like yet another patchkilling feature ;)
16:23:48 <MaSch> Is there a option to enable/disable Pre/Combo/Exit Signals?
16:24:11 <Eddi|zuHause> i don't think so
16:24:21 <MaSch> because i cant find them O.o i know that some time ago i used them, but there are no such signals
16:24:39 <fjb> Just right click on a signal.
16:24:41 <Eddi|zuHause> ctrl+click on a signal with the signal tool
16:24:58 <fjb> Ok ctrl-click, I was too fast.
16:25:22 <MaSch> but i remember a window to control these things
16:25:39 <Eddi|zuHause> that existed but was never in trunk
16:26:14 <MaSch> okay.. so its only in the nightly builds?
16:26:35 <Eddi|zuHause> the nightly builds are done directly from trunk
16:27:14 <MaSch> okay.. but how i got this window ?? i only use nightly builds
16:27:36 <Eddi|zuHause> my best guess, you used an "integrated nightly" once
16:27:57 <Eddi|zuHause> which has not much to do with a regular nightly
16:28:45 <MaSch> okay.. whatever.. with ctrl + click it works, too ^^ thanks a lot
16:30:55 <Eddi|zuHause> while we are at signals... what happened to tecky and his PBS attempt?
16:33:19 <fjb> Maybe he is burried under the signals. He was senn some months ago the last time.
16:44:05 <ln-> is there a nice design pattern for adding new data members to a class without adding them to the class definition?
16:47:07 *** skidd13 has joined #openttd
16:47:46 <Eddi|zuHause> i think C++ allows to split the class definition into several parts
16:48:02 <Eddi|zuHause> but what exactly are you trying to do?
16:50:49 <ln-> something like class A { Z z; }, but Z is something that requires too many headers included and isn't available for all platforms, so i wouldn't want the A header file to depend on it at all.
16:52:58 <ln-> omfg, an #ifdef inside a class definition is so ugly
16:53:09 <ln-> and doesn't remove the dependencies
16:53:11 <Eddi|zuHause> ln-: like i said, i think you can have several "class A {}" definitions which then are merged
16:54:00 <ln-> Eddi|zuHause: interesting, i haven't been aware of that. gotta test.
16:54:12 <Eddi|zuHause> so in the header file you put all public members, and in the .cpp file you put the private members and implementation details
17:02:14 *** wolfy is now known as Wolfensteijn
17:10:53 <fjb> How about letting A inherit from Z and provide an empty Z for some platforms?
17:12:21 <ln-> still the dependency problem
17:13:40 <fjb> I don't know another way around the dependency beside #ifdef.
17:15:07 <fjb> Ups, banks get placed far outside the towns with ECS.
17:15:08 <ln-> the header with A is included either directly or transitionally in a lot of places. making A include something extra makes everything #including A depend also on Z.
17:16:05 <fjb> Then put the #ifdef into the header for A.
17:16:21 <Eddi|zuHause> how about adding a portable (abstract) superclass to Z instead?
17:17:37 <ln-> Eddi|zuHause: that's the least bad alternative so far.
17:18:20 <ln-> fjb: #ifdefs are not acceptable and do not solve dependency problems.
17:20:31 <ln-> basically one could also have a table with pointers to A and pointers to Z in the .cpp, to connect instances of A with instances of Z, and so the header file of A wouldn't need to know anything about Z.
17:20:55 *** DaleStan_ is now known as DaleStan
17:21:02 <ln-> but updating and accessing such a table would of course need thread synchronization which is unpleasant.
17:21:37 <Eddi|zuHause> you could also have void pointers and then cast them to Z pointers everywhere...
17:21:58 <Eddi|zuHause> urgh... that sounds even ugly to me :p
17:22:16 <fjb> Sounds like many page faults...
17:26:56 *** Brianetta has joined #openttd
17:32:02 *** Vikthor has joined #openttd
18:05:42 *** LeviathNL has joined #openttd
18:19:18 *** RamboRonny has joined #openttd
18:19:41 *** RamboRonny has joined #openttd
18:28:02 *** skidd13 has joined #openttd
18:30:47 *** fjb is now known as fjbAWAY
18:46:15 <csaba> i can not travel grain or livestock between the farm and factory
18:46:27 <csaba> i have depots next to them
18:47:41 <ln-> csaba: i have no obligation to answer you
18:47:42 <csaba> ln if you cant help pls leave me alone
18:48:42 <csaba> i have never said that fuck up
18:48:50 <skidd13> csaba: depots? You need a station. a depot is only needed to handle trains
18:50:57 <csaba> where is the station icon ?
18:52:19 <skidd13> You might try the newbee guide in the wiki.
18:52:47 <ln-> skidd13: why would you want to kick me, sir?
18:53:22 *** Gonozal_VIII has joined #openttd
18:53:32 <ln-> skidd13: in response to stupid and rude behavior.
18:54:34 <skidd13> You should be friendly. OpenTTD is such a peacefull game. So please calm down. Thanks
18:55:16 <ln-> skidd13: i started by giving friendly instructions on how to ask for help on irc.
18:55:43 <csaba> i have found the lorry station
18:56:30 <skidd13> ln-: Some people are not native english, so their questions might sound not nice... Keep that in mind ;)
18:56:57 <csaba> sorry my bad english i mixed the meaning of lorry
18:58:04 <csaba> yes i mixed the meaning of ferry and lorry
18:59:31 <skidd13> ln-: And there is no need to insult someone
19:00:26 <Phazorx> TrueBrain: i got some reading material for ya
19:00:37 <TrueBrain> either http or mail
19:02:04 <ln-> skidd13: i was not insulting, i merely ordered him to shut his mouth after he ignored my help.
19:03:13 <ln-> skidd13: not significantly worse than "not funny! end of discussion."
19:03:16 *** SERVEPRO has joined #openttd
19:03:51 <TrueBrain> Phazorx: tnx, reading..
19:04:15 <Phazorx> i hope it at least worthy of time spent
19:04:19 <TrueBrain> Phazorx: btw, I did some matrix compression today, I could reduce any 18 bits with 3 bits :p
19:04:23 <TrueBrain> (and any 50 bits with 11 bits)
19:04:23 <csaba> fullload buttton has been activated:)
19:04:39 <Phazorx> TrueBrain: any bits or some ?
19:04:45 <TrueBrain> Phazorx: any combination of bits
19:04:53 <TrueBrain> some cases it was better, but never worse
19:05:16 <Phazorx> 266^3 chunk would be nice tho
19:05:36 <TrueBrain> I did it by hand, the checking of the matrix, as I was too lazy to write an application for it :p
19:05:46 <TrueBrain> so ... 6x6 was biggest I tried, with each cell containing 2 bits
19:06:45 <Phazorx> one thing i can say for sure - checksum validation gains value with number of dimentions
19:07:18 <Phazorx> so each next dimention used probably reduces needed data exponentialy
19:07:29 <TrueBrain> in my method, not exponential
19:07:52 <TrueBrain> but still, I found it suprising that you can reduce any amount of bitstream with several bits (up to a given limit)
19:07:55 <TrueBrain> just it takes a bit of CPU time :)
19:08:01 <Phazorx> well given that number of dimentions is much less than data row/column
19:08:26 <Phazorx> yeah,i can foresee that being a storage method for future
19:08:45 <Phazorx> where it is always a balance between time spend and final sized
19:08:51 <Phazorx> rather thana fixed number
19:09:19 <Phazorx> and i can assure you that it works better on noneuniform data, so you ebtter of compressing it with soemtihng very fast 1st
19:09:20 <TrueBrain> your document finally clears up a few things ;)
19:09:54 <TrueBrain> I didn't use XOR, as the results can be doubtful, but I used SUM
19:09:55 <Phazorx> the idea is very close to encrypton and can be used to encrypt as wellas compress
19:10:18 <Phazorx> TrueBrain: well sum with overflow?
19:10:27 <TrueBrain> Phazorx: no, that is why it isn't exponentional :)
19:10:49 <Phazorx> well my aprach is mroe generic
19:11:01 <Phazorx> but it could be any of known hash methods
19:11:05 <Phazorx> doesnt matter that much
19:11:20 <Phazorx> xor is just faster and has enough invariety to it to make it work
19:12:01 <TrueBrain> downside of XOR is that when you have something like: 10 ?? ?? => 00, the last two ?? are set, but the middle row can be either 10 or 01
19:12:10 <TrueBrain> (where => is the XOR of the first 3 pairs of 2 bits)
19:12:59 <Phazorx> well that's why you have more than just one dimension
19:13:08 <Phazorx> so a cross check would clear up soem things
19:13:16 <TrueBrain> but still doesn't make it unique
19:13:20 <Phazorx> and 3rd dimetion will give you even more detail
19:13:21 <TrueBrain> or worse: the checksum isn't unique
19:13:49 <Phazorx> as i mentined yesterday uniqueness isnt the goal
19:14:02 <TrueBrain> for lossless compression, in the end it is :)
19:14:32 <TrueBrain> euh, in the end you want to be able to decompress
19:14:41 <Phazorx> if you can guess from first try of many guess and have enough guideance to have a 100% probability on 1st guess being correct
19:14:44 <TrueBrain> so, lossless defines that you need to get an unique solution ;)
19:15:00 <Phazorx> it doesnt matter tha there can be 1000000 more incorrect guesses as long as they are not 1st one
19:15:03 <TrueBrain> oh, k, you mean it like that ;)
19:15:13 <TrueBrain> in my book, it means you did define enough uniqueness :)
19:15:28 <TrueBrain> (filling in spots in the matrix, does define uniqueness :))
19:15:28 <Phazorx> you can read it like that
19:15:36 <Phazorx> but a meant from math side of things
19:15:42 <Phazorx> it is not a unique solution
19:15:57 <TrueBrain> anyway, what I ment with the XOR problem, is that you have to give the middle row on every entry. It can not be calculated in any way
19:16:02 <TrueBrain> your method kind of avoids that btw
19:16:22 <TrueBrain> (by pre-defining what it should be, and giving addition information if it is not ;))
19:16:34 <Phazorx> yeah i guess, and i like xor because evey bit is meangfull in every step
19:16:43 <Phazorx> but 2 dimentions is the least to make it mena enoughj
19:16:51 <Phazorx> and more dimentions - more meaning it has
19:17:23 <TrueBrain> btw, the word 'guessing' gets a whole other meaning ;)
19:18:03 <TrueBrain> funny enough, what I wrote today isn't far off from what you wrote down :) Just you present a very clever way to produce enough results to create the matrix
19:18:36 <Phazorx> i think it is a worth idea
19:18:42 <Phazorx> just the math part is lacking
19:21:17 <Phazorx> i'm not sure if you are familiar with simple block variation ciphers but they sort of do same thing but in reverse, where key bits are used to encrypt in same fashion as i position this idea to "guess" the original data
19:22:01 <Phazorx> you can write an app to brute force it tho
19:22:32 <Phazorx> on some smaller chunks like 16x16x16 bytes
19:23:39 <TrueBrain> as a proof-of-concept doesn't care about how much CPU it uses, you can create it without the math
19:23:43 <TrueBrain> by just brute forcing it
19:24:03 <Phazorx> im pretty sure since the concept is trivial someone would already have doe it
19:24:11 <Phazorx> and smatr math part is the key
19:24:24 <Phazorx> to make it somewhat usable
19:24:51 <TrueBrain> the invention of the wheel, don't you think someone was thinking too, someone else would have thought it up?
19:24:58 <Phazorx> could be fun to get some cray time and encoe like a gig into few megabytes :)
19:25:11 <TrueBrain> this Soup guy is funny.. it is joining all channels I am on and starts annoying :p
19:25:43 <Phazorx> you know that i banned all ISPs in his town in #openttdcoop ?
19:25:46 <TrueBrain> Phazorx: I can get CPU time, and a lot
19:26:17 <Phazorx> TrueBrain: when more power to you, i still would like more proper math appraoch than brute force
19:26:28 <Phazorx> as it is now it is possible to be done but not feasible to be used
19:26:40 <TrueBrain> have you tried it? :p
19:27:01 <Phazorx> been a while since i actualy done a C app from scratch
19:27:08 <Phazorx> like 10 years at least :)
19:27:25 <TrueBrain> if you don't mind, I will give it a spin?
19:27:31 <Phazorx> you definately should
19:27:51 <Phazorx> but i;d like to spin it too if you publish the code :)
19:27:58 <Phazorx> and most definately enhance the guesing part
19:28:05 <TrueBrain> just minor things that don't make sense... 71 XORed is 81?
19:28:38 <Phazorx> you mean when cheating was done?
19:28:39 <TrueBrain> let's get an other channel for this ;)
19:29:04 <TrueBrain> anyone who cares, can join too :p
19:52:02 *** Klanticus has joined #openttd
20:08:02 <Brianetta> It's a shame only 255 UK servicemen were lost in the Falklands War. One more and it would have wrapped to zero, what with it being 1982 and all.
20:08:56 <hylje> i *so* saw a bitness joke coming
20:10:26 <Brianetta> Th\at's why I had to get the whole thing out in one line
20:16:28 *** fjbAWAY is now known as fjb
20:37:45 *** TinoDidriksen has joined #openttd
20:51:26 *** xerxes is now known as shodan
21:50:06 <_42_> fjb, please look a bit closer at the memberlist of this channel.
21:51:55 <Rubidium> he's usually not online in the weekends
21:52:30 <Rubidium> and if he's online, it's when it is 'late' for him, i.e. in about 4 to 6 hours
21:53:19 <fjb> Ok, thank you. I think I found a bug in newindustries.
21:54:16 <Rubidium> it's just an undocumented OTTD specific feature ;)
21:55:06 <fjb> But my citicens complain, there is no bank in their town...
21:57:01 <fjb> And the banka are complaining, they have no customers.
22:01:26 <fjb> Yes, looks like others have the same problem.
22:23:19 <SmatZ> wow, rbijker.net, Rubidium's own website ;-)
22:24:15 <SmatZ> hmm yes, a bit boring :-D
22:24:55 *** ThePizzaKing has joined #openttd
22:25:03 <Rubidium> it "just" contains loads of crappy patches
22:36:27 <fjb> One of the biggest providers around here, nothing more.
22:38:35 <Rubidium> I haven't found a compelling reason to go somewhere else
22:55:18 *** lolman is now known as John
22:55:25 *** John is now known as lolman
22:57:08 <CIA-1> OpenTTD: rubidium * r11400 /trunk/src/vehicle.cpp: -Fix [FS#1409]: assign orders during cloning after refitting as it needs the cargo types to be the same.
23:22:44 <CIA-1> OpenTTD: rubidium * r11401 /trunk/src/macros.h: -Fix [FS#1391]: make all min functions do exactly the same instead of branching on either < or <=.
23:44:01 *** Anon1180 has joined #openttd
23:51:01 *** Diabolic-Angel has quit IRC
continue to next day ⏵
