| 00:03.01 | therealtbe | gnomsensical |
| 00:03.03 | therealtbe | :-D |
| 00:21.21 | gnu-sense | , reviewing the last hour or so of chat here, thinks that people need to read a ceratin article about Norbert Wiener: http://www.jstor.org/discover/10.2307/1343893?uid=3739664&uid=2129&uid=2&uid=70&uid=4&uid=3739256&sid=21104587156151 |
| 00:34.39 | *** join/#bzflag rob1n (~rob1n@unaffiliated/rob1n) |
| 00:38.44 | Constitution | Patlabor221: how exactly would you compensate for lag? I know we tried it in the 2.1 (or whatever it was) code? and I can't think of a method of compensation without undesirable side effects |
| 00:39.08 | Patlabor221 | historical state tracking like halflife does |
| 00:41.23 | zlsa | valve has a great document out on lag-compensation |
| 00:41.48 | Patlabor221 | yes |
| 00:41.53 | Patlabor221 | that would be the basis of it |
| 00:42.00 | Patlabor221 | 2.99 attempted to do prediction |
| 00:42.01 | zlsa | https://developer.valvesoftware.com/wiki/Lag_compensation |
| 00:42.04 | Patlabor221 | and that did not work |
| 00:42.33 | zlsa | what was wrong with it? |
| 00:42.43 | Patlabor221 | the idea is keep a history of states, and when you get a shot you do hit detction on what you know the shooter saw |
| 00:43.02 | Patlabor221 | you also make the shot appear in a similar relative postion to the victim so he see's it coming |
| 00:43.18 | Patlabor221 | again it's a very solved problem with lots of samples |
| 00:43.37 | Patlabor221 | but they all involve having a server state to use and manipulate |
| 00:44.23 | Patlabor221 | I know you love reinventing wheels Constitution but this is a place to just use what is proven ;) |
| 00:44.47 | Constitution | in other words, you have to dodge shots from the shooter's perspective? that will completely change your ability to make precise dodges |
| 00:44.58 | Constitution | and haha :P |
| 00:45.05 | Patlabor221 | no no no, the server puts the shots in similar relative postions to you |
| 00:45.10 | Patlabor221 | you dodge the shots you see |
| 00:45.25 | Constitution | ah, so you won't even see them until they've killed you :P |
| 00:45.28 | Patlabor221 | we have no instant hit weapons so there are updates |
| 00:45.36 | Patlabor221 | only in very high lag situations |
| 00:45.52 | Patlabor221 | there is no way around that and to have authorative kills |
| 00:45.59 | Patlabor221 | unless you want to go Turn Based |
| 00:46.06 | Patlabor221 | you do interpolation so it doesn't look bad |
| 00:46.12 | Patlabor221 | it's how EVERY other internet game works |
| 00:46.33 | Patlabor221 | again we have no instant hit weapons other then lazer, and we can't dodge that now :) |
| 00:46.37 | Patlabor221 | you just explode |
| 00:47.14 | Patlabor221 | if someone fires point plank you may not see the shot before you die, but IT WAS POINT BLANK, what would you do? |
| 00:47.38 | Patlabor221 | if there isn't enough time to see the shot there isn't enough time to dodge since you can't dodge what you can't see |
| 00:47.41 | Patlabor221 | that happens today |
| 00:47.45 | Constitution | so in high lag situations, compensation doesn't favor the lower lag players, correct? so one high lag player can throw mayhem into a game |
| 00:47.59 | Patlabor221 | the high lag player just dies more |
| 00:48.03 | Patlabor221 | since they can't doge |
| 00:48.05 | Patlabor221 | doge |
| 00:48.07 | zlsa | lol |
| 00:48.11 | Patlabor221 | they don't get an advantage at shooting |
| 00:48.22 | Patlabor221 | bassicly punsh the laggers not the game |
| 00:48.40 | Patlabor221 | go play online team fortress with just rockets, it would work the same way |
| 00:48.53 | Patlabor221 | that's efectivly what our shots are |
| 00:48.57 | Constitution | but their shots get compensated too, right? so that's time taken off every single shot lifetime |
| 00:49.10 | Constitution | I will, when I can clear 16GB of space on my disk :P |
| 00:49.12 | Patlabor221 | it's not just time taken off |
| 00:49.14 | Patlabor221 | it's more then that |
| 00:49.27 | Patlabor221 | there are postional corrections as well |
| 00:50.02 | Patlabor221 | again I'd find some existing code and reuse it, so the behavor is known |
| 00:50.11 | Constitution | and I don't just love reinventing wheels? I just like to understand what the algorithms are doing |
| 00:50.17 | Patlabor221 | then read up ;) |
| 00:50.27 | Patlabor221 | there are many papers on them |
| 00:50.38 | Patlabor221 | IEEE computer society has many |
| 00:50.58 | Patlabor221 | that valve article is an overview of a more acedemic paper they wrote |
| 00:51.44 | Patlabor221 | and HOW DARE YOU DOUBT THE CARMACK! |
| 00:53.55 | Patlabor221 | tho really what we have now has been extended from his stuff quite a bit |
| 00:54.38 | Patlabor221 | the basic idea is the same, keep states around for each player, so you have a little virtual screen for each player and at that point you compare and see who hit what and who dodged what |
| 00:55.26 | Patlabor221 | you just do it all on the server instead of doing it at the untrusted endpoints |
| 00:55.53 | Patlabor221 | I'd go rip the code out of an existing network game, like a FPS or something |
| 00:56.05 | Patlabor221 | or use an engine that does it already |
| 00:57.03 | Constitution | ok? all based on the latency of each player? |
| 00:57.18 | Patlabor221 | and the input history |
| 00:57.29 | Patlabor221 | it's WAY More then just latency math |
| 00:57.39 | Patlabor221 | you roll back time |
| 00:58.09 | Patlabor221 | read the papers |
| 01:20.44 | Patlabor221 | and really Constitution you don't have 16 gigs free? you system must run quite badly |
| 01:23.15 | Patlabor221 | how does it background defrag with out enough cache? |
| 01:23.31 | Patlabor221 | unless it's an SSD, then nevermind :) |
| 02:03.32 | Constitution | it's a 240GB SSD |
| 02:03.44 | Patlabor221 | still pretty full |
| 02:03.56 | Constitution | I have a 512GB SSD sitting on my desk for when I have time to switch over |
| 02:04.32 | Constitution | and I don't think macs have defrag |
| 02:04.49 | Patlabor221 | if they don't then they would run slow |
| 02:05.03 | Patlabor221 | file fragmentation on physical disks is not OS specific |
| 02:05.35 | Patlabor221 | most modern OSs defag physical disks in the background now |
| 02:06.24 | Patlabor221 | IIRC HFS+ handles it dynamicly |
| 02:06.34 | blast007 | and you don't bother on an SSD |
| 02:06.45 | Patlabor221 | yeah hence my comment about nevermind on SSD :) |
| 02:06.46 | zlsa | is it a problem on an SSD? |
| 02:06.55 | Patlabor221 | no there are no heads or sectors |
| 02:06.57 | zlsa | or does it hurt more to try and fix it? |
| 02:06.57 | Patlabor221 | so it doesn't matter |
| 02:07.00 | Patlabor221 | it's a big flash drive |
| 02:07.12 | Patlabor221 | you have random access to all bits at a consistent seek time |
| 02:07.14 | zlsa | oh, for some reason I'm thinking of the programs that remove unused files |
| 02:07.39 | Patlabor221 | oh no, fragmenting is when a file is spread over multiple seperated physical sectors |
| 02:07.45 | blast007 | though it sounds like it's still good to not fill up an SSD |
| 02:07.51 | Patlabor221 | so a spining disk has to seek to multiple places to get the same file |
| 02:08.11 | blast007 | also, an old KB, but possibly still relevant: https://web.archive.org/web/20110602212017/http://support.apple.com/kb/HT1375?viewlocale=en_US |
| 02:08.45 | therealtbe | we should make an SSD with a tiny built in speaker that makes spinning noise |
| 02:08.54 | Patlabor221 | yeah they do a lot of background stuff to prevent fragmentation |
| 02:09.15 | Patlabor221 | it still needs space to work in tho |
| 02:09.19 | blast007 | (seems that link still works on their site, actually... had found an older one that wasn't being redirected, and archive.org had saved the redirect from an older site) |
| 02:10.20 | blast007 | Patlabor221: I'm meaning more for wear leveling |
| 02:10.46 | Patlabor221 | on an SSD? |
| 02:10.53 | blast007 | yeah |
| 02:11.03 | Patlabor221 | yeah you don't want half your bits to go bad at the same time |
| 02:11.50 | blast007 | there is some extra capacity reserved for that purpose too, but it sounds like it can also use unallocated sectors to shuffle data around (as long as TRIM is working, maybe) |
| 02:11.51 | therealtbe | wear leveling is irrelevant because most systems don't have --noatime specified |
| 02:12.13 | Constitution | what? |
| 02:12.28 | Patlabor221 | Constitution, you know your bits have a finite nubmer of toggles right? |
| 02:12.52 | Constitution | yes? I don't see how wear leveling is irrelevant |
| 02:13.01 | blast007 | therealtbe: you realize that the access time doesn't get stored *inside* the files, right? |
| 02:13.02 | therealtbe | must specify noatime, else wear leveling is irrelevant, because each access will also cause a write |
| 02:13.25 | therealtbe | and where does it happen mr blast007 |
| 02:13.55 | Patlabor221 | goes home to remove Geno |
| 02:14.06 | zlsa | about geno: why remove it? |
| 02:14.09 | Constitution | wear leveling is generally (always?) performed by the firmware |
| 02:14.13 | zlsa | just don't add the flag |
| 02:14.37 | Constitution | zlsa: he was kidding (I think) |
| 02:15.16 | therealtbe | anyways, most things, like ubuntu for example, need specified "noatime" otherwise they will perform a write to the file table on every access |
| 02:15.33 | therealtbe | this will burn out those sectors longggggggggggggg before any wear leveling begins to be relevant |
| 02:16.35 | blast007 | therealtbe: well, in NTFS, it's stored in the MFT (Master File Table) and the directory entry for that file (which is stored in the folder that contains the file) http://technet.microsoft.com/en-us/library/cc781134%28v=WS.10%29.aspx |
| 02:16.50 | Constitution | "those sectors" will be different physical memory locations every time, if the firmware is doing its job correctly? and the way it allocates physical memory locations is very relevant to wear leveling |
| 02:17.29 | therealtbe | blast007, correct and that MFT will wear out updating access times before any wear leveling can be significant |
| 02:18.39 | Constitution | it's not the same physical location every time... |
| 02:19.19 | Constitution | when you change something on disk, it writes it to previously unallocated memory, and then later the old memory is processed by garbage collection |
| 02:19.43 | blast007 | when you read data from an SSD, it also changes data gradually :) |
| 02:21.06 | blast007 | or at least, it can "A read error happens when the data stored in a cell changes as a neighboring cell on the same string is read over and over" http://www.pdl.cmu.edu/PDL-FTP/NVM/flash-error-analysis-and-management_itj13.pdf |
| 02:21.49 | blast007 | so my understanding is that even if you don't write data to an SSD, the controller will still periodically do some writes to ensure that data is not corrupted |
| 02:24.05 | therealtbe | when my brakes sense they are wet, they pulse a the ABS a tiny bit to keep them dry, just in case |
| 02:24.36 | therealtbe | sounds cool... until that sensor gets frozen and intermittently fails |
| 02:26.27 | blast007 | ah, we switching to poor car analogies now? |
| 02:27.46 | zlsa | well bzflag is like driving a car, except there aren't wheels and it's not real |
| 02:32.56 | blast007 | "How's my lagging? Call 1-800-...." |
| 02:37.49 | *** join/#bzflag Patlabor221 (uid48618@gateway/web/irccloud.com/x-dwjuczpzangkvzdp) |
| 02:39.27 | Patlabor221 | So name the new plugin UnfairMurderMachine.dll? |
| 02:40.04 | therealtbe | KernelSlaughter.dll |
| 03:35.29 | *** join/#bzflag Patlabor (~JeffM@2605:e000:7d02:c700:ac3b:1581:55a1:ca86) |
| 04:03.18 | Patlabor | so geno kills everyone on the team of the shooter right? |
| 04:05.24 | blast007 | geno kills everyone on the team of the victim, even if the victim is the shooter or a teammate - or was that not a serious question? |
| 04:05.43 | zlsa | oh, so you can rogue suicide with geno and kill all rogues? |
| 04:06.02 | blast007 | and rogue doesn't do geno by default |
| 04:06.03 | Patlabor | I like my version better |
| 04:06.07 | blast007 | :P |
| 04:06.20 | Patlabor | I'll give it a rogue bzdb flag |
| 04:06.53 | Patlabor | then we can retire rogue geno as a speerate plugin |
| 04:10.38 | blast007 | this is a 2.5 change, I imagine? |
| 04:10.47 | Patlabor | doesn't have to be |
| 04:11.14 | blast007 | well, just looks to be some client-side detection of genocide |
| 04:11.32 | Patlabor | it will just never be true :) |
| 04:12.22 | blast007 | wouldn't a custom flag that sends the 'G' type get handled as the old school geno in a 2.4 client? |
| 04:12.24 | Patlabor | we moved the death to the server IIRC so it should function the same |
| 04:12.32 | Patlabor | maybe |
| 04:12.46 | Patlabor | what does the client do specificly for geno? |
| 04:12.51 | Patlabor | I only have server open now |
| 04:12.52 | blast007 | playing.cxx has a test for geno that triggers a gotBlowedUp |
| 04:13.08 | Patlabor | oh won't matter, I'll have the server kill em |
| 04:13.14 | Patlabor | so it'll still work |
| 04:13.21 | Patlabor | I can blow ANYONE up |
| 04:13.59 | blast007 | except for Chuck Norris |
| 04:13.59 | Patlabor | I was just going to give it the flag code GO, and blow them up at the server level |
| 04:14.41 | Patlabor | I can call Kyle Katarn and he can take care of it |
| 04:15.44 | blast007 | pew pew pew |
| 04:16.56 | Patlabor | all I have to do is enture that the old flag code doesn't stop or crash the server, so I'll have it warn and continue with out the flag |
| 04:22.30 | Patlabor | hmm, where do descriptions for non custom flags come from? |
| 05:04.20 | *** join/#bzflag zuii (~akos@BC06BD4B.catv.pool.telekom.hu) |
| 06:48.17 | *** join/#bzflag Snake12534 (uid45641@gateway/web/irccloud.com/x-wzptzrenxvuenhoc) |
| 06:52.21 | Patlabor | holly crap, why didn't anyone ever ask for bz_addPlayerWins?!?! |
| 06:52.52 | Snake12534 | I don't know |
| 06:53.29 | allejo | Patlabor, because hardly anyone uses the API and bz_setPlayerWins() suffices :p |
| 06:54.33 | Patlabor | makes the custom flag sample have a memory leak :) |
| 06:54.54 | Patlabor | it'll loose a playe record evertime someone shoots with it :) |
| 06:55.11 | allejo | because of the lack of bz_addPlayerWins()? |
| 06:56.43 | Patlabor | well the way they did it |
| 06:56.56 | Patlabor | they get the player recod for the shoter to get the current wins to use as a base for setWins |
| 06:57.08 | Patlabor | but it never frees that pointer |
| 06:57.24 | allejo | oh well that's handy |
| 06:57.26 | allejo | :p |
| 06:57.27 | Patlabor | I changed it to call getWins instead |
| 06:57.33 | Patlabor | then it won't leak |
| 06:57.48 | Patlabor | not that anyone uses that part of the sample |
| 06:59.28 | allejo | a bz_addPlayerWins would be convenient though |
| 06:59.57 | Patlabor | yeah |
| 07:00.01 | Patlabor | same for losses and TKs |
| 07:00.12 | Patlabor | I'll add it with this checkin when I test it |
| 07:01.26 | allejo | cool |
| 07:01.37 | Patlabor | get rid of RogueGeno too |
| 07:01.48 | allejo | ah replacing it with the bzdb var? |
| 07:02.04 | Patlabor | bahhh who put a NOT in this thing? |
| 07:02.40 | allejo | ties a knot |
| 07:02.43 | allejo | oops |
| 07:02.51 | Patlabor | congrats |
| 07:03.12 | allejo | bows |
| 07:05.43 | Patlabor | Jim Melton did it |
| 07:05.55 | Snake12534 | Yessir? |
| 07:06.12 | Patlabor | busted the build several ways |
| 07:06.49 | allejo | that's Flash, no? |
| 07:06.58 | allejo | what's busted? |
| 07:07.17 | Patlabor | can't use not, VC 10 doesn't support it, it's also stupid |
| 07:07.33 | Patlabor | also windows doesn't use BSD sockets and he put a bunc of unix stocket stuf in |
| 07:09.04 | Patlabor | I see he had to back out lot of code before for similar reasons |
| 07:09.24 | Patlabor | probaby should be working on a private fork and doing pull requests after testing |
| 07:10.05 | allejo | on the bright side, i'm not the only one who breaks the build! |
| 07:10.23 | Patlabor | oh everyone breaks the build |
| 07:10.34 | Patlabor | people forget about the primary client OS :) |
| 07:12.06 | Patlabor | I've gotten in the habbit of trying builds on my linux VM, just at least to make sure they compile |
| 07:13.49 | Patlabor | there we go, now it builds and won't get into infinate murder loops |
| 07:14.01 | allejo | cheers |
| 07:21.36 | khonkhortisan | This is what's piped through the hard drive speaker: http://upload.wikimedia.org/wikipedia/commons/a/a9/WD_bad_heads_click_of_death.ogg |
| 07:24.31 | Patlabor | hmm geno was suprizingly easy to remove |
| 07:25.31 | allejo | you actually removed the flag? bah so does geno have to be a plug-in now? |
| 07:26.04 | Patlabor | I wrote the plugin too |
| 07:26.20 | Patlabor | UnfairMurderMachine.dll |
| 07:26.29 | allejo | seems apt |
| 07:26.48 | *** join/#bzflag Constitution (~josh@bzflag/developer/Constitution) |
| 07:29.37 | Patlabor | it's why I can get rid of RogueGeno to, I just made the new plugin look for a BZDB var to know who to kill |
| 07:30.34 | allejo | custom BZDB var? |
| 07:30.46 | Patlabor | all bzdb vars are "custom" |
| 07:30.48 | Patlabor | so sure |
| 07:31.35 | allejo | err.. i mean does the plug-in create the bzdb variable? i.e. the bzdb variable won't exist unless the plug-in is loaded |
| 07:32.08 | Patlabor | it dosn't have to create it |
| 07:32.18 | Patlabor | it's set to false if nobody sets it |
| 07:33.31 | Patlabor | so I can just check it as a bool no mater what |
| 07:33.47 | allejo | that works |
| 07:34.04 | Patlabor | that's how bzdb was designed, you don't have to "Register" them |
| 07:34.12 | Patlabor | they all have a default value if unset |
| 07:34.15 | Patlabor | zero |
| 07:37.29 | allejo | hmm but bzfs still keeps a record of some of them for -set. so if a plug-in sets a value for another bzdb variable, it'd have to be -setforce or -setforced (can't remember which) |
| 07:37.56 | Patlabor | bzfs sets a number of values on startup |
| 07:38.33 | Patlabor | and yeah those defaults determine the security and scope |
| 07:38.42 | Patlabor | for this I don't care about scope |
| 07:38.48 | Patlabor | it's server side only |
| 07:42.06 | Patlabor | I could make it read the rogue state from the load command line if that's an issue |
| 07:42.57 | Patlabor | LoadPlugin UnfairMurderMachine,murder_everyone |
| 07:44.43 | allejo | nah not all, just my curiosity |
| 07:44.54 | allejo | i'd prefer a bzdb variable, allows more freedom to change it mid game |
| 07:49.10 | Flash | you will have to be more specific, if you want to slap my hand |
| 07:49.30 | Patlabor | don't use NOT |
| 07:49.43 | Patlabor | like it says in the devIfnfo |
| 07:49.52 | Patlabor | the other thing I added a define check |
| 07:50.35 | Patlabor | you backed all the other things out yourself |
| 07:53.35 | Flash | devinfo is wrong; not/or/and have been part of the standard for a long time |
| 07:54.01 | Flash | IIRC at least since the original ISO version |
| 07:54.24 | Patlabor | it's not wrong in the part where it says "don't use it" :) |
| 07:55.55 | Flash | so the perennial question, how long do we support obsolete compilers? VC 11 has been out for 2 years already ... |
| 07:55.59 | Patlabor | they were not in the inital C++ stuff, they have been in C and were put into C++ at some later date |
| 07:56.18 | Patlabor | and 2.6 wil move to VC12 |
| 07:56.37 | Patlabor | 2.4 was determined to be compatable with older compilers |
| 07:56.37 | Flash | they weren't in the ARM, but I'm pretty sure they made it into the initial international standard. |
| 07:56.52 | Patlabor | no mater whta, we should be consitent |
| 07:57.02 | Flash | sighs |
| 07:57.03 | Patlabor | that is the point made in the devinfo |
| 07:57.03 | Flash | I know |
| 07:57.23 | Flash | but consistent shouldn't mean frozen in time |
| 07:57.27 | Patlabor | go use VB if you want spelled out operators :) |
| 07:57.42 | Flash | Don't have to ... I have a REAL compiler |
| 07:57.44 | Patlabor | personaly I don't see NOT AND and OR as better |
| 07:58.00 | Patlabor | so do I, it' just not used by evryone who builds 2.4 |
| 07:58.06 | Patlabor | deal with it |
| 07:58.08 | Flash | the main reason I see NOT as better is that ! is too often missed |
| 07:58.11 | Patlabor | I can't use threads ether |
| 07:58.19 | Flash | I deal by grumbling |
| 07:58.21 | Patlabor | get a real editor then ;) |
| 07:58.29 | Patlabor | bold that sucker |
| 07:58.40 | Patlabor | lets all just change to java so we have the same pain |
| 07:59.03 | Flash | not an editor issue. A bold ! is still a very small thing, especially when right next to a paren or other letter |
| 07:59.19 | Patlabor | then use spaces ;) |
| 07:59.29 | Flash | consistently :) |
| 07:59.38 | Patlabor | then use them consitently |
| 07:59.57 | Patlabor | bzflag's space/tab/newline use is jacked up anyway |
| 08:00.17 | Flash | but it's consistent |
| 08:00.26 | Patlabor | except when it isn't |
| 08:01.07 | allejo | bz's spacing is far from consistent lol |
| 08:02.02 | Patlabor | hmm, how do I push my unsynced git changes to my linux system with out pushing them to github? |
| 08:02.02 | Flash | and THAT is why I had to back out a bunch of stuff |
| 08:02.02 | allejo | export your changes as a patch and then apply the patch |
| 08:02.02 | Flash | git is not my friend |
| 08:02.02 | Patlabor | and that is why if you are doing changes that may break you probalby want to do them in a fork and use pull requsts |
| 08:02.05 | Patlabor | then you can ask for them to be reviewed |
| 08:02.10 | Patlabor | before you sync them |
| 08:02.38 | Patlabor | that kinda stuff is supposed to be the power of DVCS |
| 08:02.53 | Patlabor | branches with out branching |
| 08:05.11 | Flash | what got me in trouble was trying to pull synced changes while I had unsynced changes. XCode made me check in first and pushed before I knew what was happening |
| 08:05.35 | Patlabor | again another reason to be on a fork, then that would only happen in your fork |
| 08:05.48 | allejo | fwiw, git stash is your friend in cases like those if you don't want to use a fork |
| 08:05.54 | Flash | All learning process |
| 08:06.20 | Patlabor | yeah I had similar issues until I just made a working fork |
| 08:06.29 | Flash | never had these problems with SVM |
| 08:06.30 | Flash | SVN |
| 08:06.42 | Patlabor | now my mistakes are pushing to my fork instead of the main one |
| 08:06.51 | Patlabor | you can build a svn repo from git IIRC |
| 08:06.58 | Patlabor | it's like a fork to a different sytem |
| 08:07.19 | allejo | and jeff, git diff > changes.patch; git apply changes.patch (in regards to your question about pushing changes to your vm) |
| 08:07.31 | Patlabor | ahh patches |
| 08:07.32 | Patlabor | lame |
| 08:07.54 | Patlabor | I should be able to push to any repo not just the one at github |
| 08:08.13 | allejo | you could add a separate remote if that's what you mean |
| 08:08.22 | Patlabor | I don't know what I mean :) |
| 08:08.30 | Flash | sleeps |
| 08:08.44 | allejo | g'night flash |
| 08:09.10 | allejo | seperate remotes in git could work for forks |
| 08:09.26 | allejo | so instead of git push orgin master, you could do git push my-fork master |
| 08:09.34 | Patlabor | diff seems to only show me uncommited changes not unsynced |
| 08:10.58 | allejo | "git diff <hash> <hash>" would give you the diff for changes you've commited locally |
| 08:11.36 | allejo | or "git diff HEAD HEAD~2" for two commits back |
| 08:11.45 | Patlabor | ahh found it, can get it from the log |
| 08:11.50 | Patlabor | I don't have a silly command line |
| 08:12.00 | allejo | oh eww... gross. |
| 08:12.09 | Patlabor | works great |
| 08:12.19 | allejo | command line <3 |
| 08:12.23 | Patlabor | don't have to remember a ton of commands |
| 08:12.29 | Patlabor | when you get old you forget things |
| 08:12.38 | allejo | hehe |
| 08:13.34 | Patlabor | well that made 6 patches, better then nothing |
| 08:13.47 | allejo | oh and speaking of pull requests, you should look at the two that have been left for dead. either merge them or close them |
| 08:14.02 | Patlabor | what are they? |
| 08:14.42 | allejo | update to swedish's translation. it uses proper utf-8 characters now. |
| 08:15.00 | allejo | the other is a callsign autocomplete but there was some concern about it |
| 08:15.09 | Patlabor | then just apply them |
| 08:15.30 | Not-311d | [02bzflag-import-3] 07allejo pushed 032 commits to 03v2_4_x [+0/-0/±2] 13http://git.io/U4eicw |
| 08:15.31 | Not-311d | [02bzflag-import-3] 07khonkhortisan 03d5e0974 - Use valid UTF-8 characters for the Swedish translation. Hex conversion table: 3f-Ã
-c385 80-Ã-c384 85-Ã-c396 8c-Ã¥-c3a5 8a-ä-c3a4 9a-ö-c3b6 |
| 08:15.32 | Not-311d | [02bzflag-import-3] 07allejo 03988192c - Merge pull request #4 from khonkhortisan/swedish-utf8 Use valid UTF-8 characters for the Swedish translation. |
| 08:15.51 | Patlabor | whats the concern about the autocomplete one? |
| 08:16.07 | allejo | https://github.com/BZFlag-Dev/bzflag-import-3/pull/1 |
| 08:16.35 | allejo | originally the tab key was used but that conflicted with jump so it was changed to shift + tab |
| 08:16.59 | allejo | mainly it's blast's comment in the thread |
| 08:17.17 | Patlabor | not a fan of shift+tab |
| 08:18.13 | Patlabor | Reject it and have them just enhance the existing autocomplete system, we don't need 2 |
| 08:20.20 | Patlabor | translation stuff is pretty much auto accept if it doesn't break anything or blow out known good stuff |
| 08:23.22 | allejo | I actually wonder if this patch even builds on windows. i thought regcomp was a linux thing |
| 08:23.34 | Patlabor | oh yeah it is |
| 08:23.40 | Patlabor | so super reject it |
| 08:24.49 | *** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535) |
| 08:24.53 | BulletCatcher | regcomp should be provided by src/other/regex |
| 08:24.55 | *** join/#bzflag death__ (~I_Died_On@69.136.32.149) |
| 08:25.45 | Patlabor | still we don't need two seperate auto complete systems |
| 08:27.46 | allejo | closed it saying to improve the current autocomplete |
| 08:28.17 | Patlabor | we ended up with 4 different robot systems that way :) |
| 08:28.36 | allejo | the more the better! |
| 08:28.41 | allejo | diversity |
| 08:28.43 | Patlabor | not when you have to maintain them all |
| 08:28.48 | allejo | :p |
| 08:29.11 | Patlabor | or answer the question "how do I run bots" and have to answer "Weeeeeeeeellllllllllll" |
| 08:29.41 | Patlabor | "2 of the systems don't work, one of the others is very dumb and will get you baned, and the last one that is decent won't do what you want it to do " |
| 08:29.49 | allejo | and yea, regcomp is provided by src/other/regex. didn't know that was there |
| 08:30.20 | allejo | wait there are actually 4? |
| 08:30.29 | allejo | thought it was just solo bots and roger |
| 08:30.39 | Patlabor | bzrobots, server side players, solo bots, and roger |
| 08:30.53 | Patlabor | 2 of them don't work like you expect |
| 08:31.03 | Patlabor | but they are techicnaly robot systems |
| 08:31.20 | allejo | oh. thought server side players were just an api thing. didn't think they were actual bots |
| 08:31.23 | Patlabor | guuuuhh GCC so slow..... |
| 08:31.31 | Patlabor | it's an API to implement bos |
| 08:31.32 | Patlabor | bots |
| 08:31.39 | Patlabor | thats why it exists |
| 08:31.55 | Patlabor | to make the server have tanks do stuff automaticly, aka "a bot" |
| 08:32.54 | allejo | when you put it that way... |
| 08:32.57 | allejo | it does make sense |
| 08:33.01 | Patlabor | heh |
| 08:33.20 | Patlabor | yeah it just doesn't have shot or death logic so that'd all have to be implemented in a plugin |
| 08:33.29 | Patlabor | hence why it's not very useful |
| 08:33.48 | Patlabor | at least not for playing bots |
| 08:33.55 | Patlabor | for chat bots it's fine |
| 08:34.00 | allejo | so you can program your own bots to drive in a circle and flail randomly |
| 08:34.17 | Patlabor | you can have the bot API send out movement updates, sure |
| 08:34.35 | Patlabor | it'll drive right thru objects |
| 08:34.40 | Patlabor | SSPs cheat a lot |
| 08:35.56 | allejo | well that's... convenient? |
| 08:36.16 | Patlabor | they arn't done, only rason I checked them in was because they were useful for chat |
| 08:36.31 | Patlabor | if you never tell them to spawn they are good observer clients |
| 08:36.50 | Patlabor | web admin, web chat, and the BZFS IRC bridge used them |
| 08:37.40 | Patlabor | they give you a protocol indepenent way to interact with the chat system and not have the chat come from "the server" |
| 08:37.46 | Patlabor | and they support private messages |
| 08:38.04 | Patlabor | the only way to private message the "server" is a / command |
| 08:38.58 | allejo | perfect use to create a !lart bot |
| 08:39.00 | allejo | ~lart brad |
| 08:39.00 | infobot | drops a baby grand on brad |
| 08:45.04 | allejo | I'm starting to smell burgers... time for me to go to bed before i do a midnight food run |
| 08:49.40 | *** join/#bzflag Patlabor (~JeffM@2605:e000:7d02:c700:ac3b:1581:55a1:ca86) |
| 09:08.12 | *** join/#bzflag I_Died_Once (~I_Died_On@69.136.32.149) |
| 09:08.13 | *** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535) |
| 13:02.00 | *** join/#bzflag infobot (ibot@rikers.org) |
| 13:02.00 | *** topic/#bzflag is http://BZFlag.org || https://www.ohloh.net/p/bzflag || http://wiki.BZFlag.org/Getting_Help || Channel Logs: http://ibot.rikers.org/%23bzflag/ || 2.4.2 Current version http://goo.gl/PI9KFD |
| 13:02.00 | *** mode/#bzflag [+o infobot] by ChanServ |
| 13:54.28 | Notify | 03BZFlag:Alezakos * 8838 /DevelopmentPlans/2.4.2: mark fixed issues |
| 13:57.02 | Notify | 03BZFlag:Alezakos * 8839 /DevelopmentPlans/2.4.2: |
| 14:18.22 | *** join/#bzflag zlsa (~zlsa@static-205-157-147-198.ded.unwiredbb.net) |
| 16:13.57 | *** join/#bzflag Patlabor (~JeffM@2605:e000:7d02:c700:d872:9f93:83f0:c960) |
| 16:34.53 | *** join/#bzflag TimRiker (~TimRiker@bzflag/projectlead/TimRiker) |
| 16:34.53 | *** mode/#bzflag [+o TimRiker] by ChanServ |
| 17:18.29 | *** join/#bzflag Patlabor221 (~JeffM@107-209-61-107.lightspeed.irvnca.sbcglobal.net) |
| 17:24.26 | *** join/#bzflag zlsa (~zlsa@unaffiliated/zlsa) |
| 17:38.01 | *** join/#bzflag zuii (~akos@BC06BD4B.catv.pool.telekom.hu) |
| 17:52.29 | *** join/#bzflag zlsa (~zlsa@unaffiliated/zlsa) |
| 17:57.31 | *** join/#bzflag zlsa (~zlsa@unaffiliated/zlsa) |
| 18:23.29 | Patlabor221 | oooo I should name the new Geno plugin Systemd |
| 18:23.34 | Patlabor221 | then nobody will use it |
| 18:23.39 | blast007 | lol |
| 18:23.53 | Patlabor221 | they will be too busy fighting over if it's cool or not |
| 18:24.53 | blast007 | then someone will make an OpenRC plugin |
| 18:38.39 | Not-311d | [02bzflag-import-3] 07JeffM2501 pushed 031 commit to 03v2_4_x [+0/-0/±1] 13http://git.io/2ZhO-g |
| 18:38.40 | Not-311d | [02bzflag-import-3] 07JeffM2501 03523937e - silly API missing a variable and having teams be used as speeds. |
| 18:39.45 | Not-311d | [02bzflag-import-3] 07JeffM2501 pushed 032 commits to 03v2_4_x [+0/-0/±1] 13http://git.io/-7e5qw |
| 18:39.46 | Not-311d | [02bzflag-import-3] 07JeffM2501 03523937e - silly API missing a variable and having teams be used as speeds. |
| 18:39.47 | Not-311d | [02bzflag-import-3] 07JeffM2501 039dd2efa - Merge pull request #8 from JeffM2501/v2_4_x silly API missing a variable and having teams be used as speeds. |
| 18:43.44 | Not-311d | [02bzflag-import-3] 07JeffM2501 pushed 032 commits to 03v2_4_x [+0/-6/±6] 13http://git.io/YUE9eA |
| 18:43.45 | Not-311d | [02bzflag-import-3] 07JeffM2501 03c0db28a - remove Genocide as a supported flag |
| 18:43.46 | Not-311d | [02bzflag-import-3] 07JeffM2501 037f176b3 - remove RogueGenocide since genocide has been removed |
| 18:45.00 | zlsa | so geno is moving to a plugin? |
| 18:45.13 | Patlabor221 | yeah, I just have to add it to my build |
| 18:45.18 | zlsa | it's how it should have been from the beginning |
| 18:45.30 | blast007 | in the beginning there wasn't plugins :P |
| 18:45.31 | Patlabor221 | we didn't have plug-ins in the begining :) |
| 18:45.33 | zlsa | geno is the sort of plugin that planetmofo should enable, but nobody else |
| 18:45.35 | zlsa | true |
| 18:48.27 | blast007 | is there any other flags that don't have a counter-flag (or at least a workable counter-tactic)? |
| 18:48.55 | Not-311d | [02bzflag-import-3] 07JeffM2501 pushed 032 commits to 03v2_4_x [+6/-0/±3] 13http://git.io/XohrjA |
| 18:48.56 | Not-311d | [02bzflag-import-3] 07JeffM2501 0335f62fc - add geno replacement plugin |
| 18:48.57 | Not-311d | [02bzflag-import-3] 07JeffM2501 032b03d0e - add UnfairMurderMachine to fullbuild |
| 18:49.42 | Patlabor221 | so I can push that to the mainline, I have tested the removal of legacy Geno and it's fine, I have done less testing on the new plug-in |
| 18:49.50 | Patlabor221 | or someone can build from my fork and test for me |
| 19:07.56 | blast007 | fails to run autogen.sh on linux, dying on the UnfairMurderMachine plugin |
| 19:08.15 | Patlabor221 | did I spell it wrong? |
| 19:08.27 | Patlabor221 | I added it to the config.ac and stuff |
| 19:08.34 | Patlabor221 | and it built on my leenox last night |
| 19:09.18 | blast007 | ah, may be due to having SAMPLE_PLUGIN stuff still in that Makefile.am |
| 19:09.24 | Patlabor221 | ahhhh |
| 19:09.32 | Patlabor221 | yeah, I'll fix that |
| 19:09.59 | Patlabor221 | hmmm, ahh I see I fixed that on the linux machine but didn't move it back to windows |
| 19:10.00 | Patlabor221 | thanks |
| 19:10.05 | blast007 | :D |
| 19:10.47 | Not-311d | [02bzflag-import-3] 07JeffM2501 pushed 031 commit to 03v2_4_x [+0/-0/±1] 13http://git.io/_YfoMg |
| 19:10.48 | Not-311d | [02bzflag-import-3] 07JeffM2501 0310b92a3 - make it build on linux |
| 19:10.51 | Patlabor221 | this is why forks are a good thing |
| 20:10.19 | *** join/#bzflag khonkhortisan (~Khonkhort@c-24-18-42-215.hsd1.wa.comcast.net) |
| 20:33.06 | *** join/#bzflag khonkhortisan (~Khonkhort@c-24-18-42-215.hsd1.wa.comcast.net) |
| 21:07.42 | Patlabor221 | well let me know when someone has tried it, I added the pull request but have not merged it. |
| 21:09.40 | gnu-sense | I'd test it but I can't even get 2.4.2 to compile on my linux. It complains about something missing I haven't been able to track down. |
| 21:12.42 | alezakos | Patlabor221: will the windows build work now? |
| 21:13.44 | Patlabor221 | it built for me |
| 21:14.03 | Patlabor221 | gnu-sense, 2.4.2 had some issues on some distros |
| 21:14.13 | Patlabor221 | I'd stick with the github code |
| 21:14.37 | blast007 | "ERROR: Genocide has been removed from the game and will be ignored" ;D |
| 21:14.51 | Patlabor221 | yep |
| 21:14.57 | Patlabor221 | it doesn't exit |
| 21:15.01 | gnu-sense | kk. Blast: ha ha. Nice. |
| 21:15.02 | Patlabor221 | maybe make it say "warning" |
| 21:15.49 | blast007 | or make it a fatal error? other invalid flags make bzfs spit out the 'usage' info and terminate |
| 21:16.03 | Patlabor221 | I didn't want to outright break existing config scripts |
| 21:16.10 | Patlabor221 | it was my compromise |
| 21:16.16 | blast007 | I guess.. |
| 21:16.25 | Patlabor221 | yeah I was on the fence |
| 21:16.36 | Patlabor221 | take out 2 lines and it becomes a fatal |
| 21:16.36 | blast007 | make it a fatal error once we merge this stuff with 2.5 |
| 21:16.46 | Patlabor221 | thats what I was pondering |
| 21:17.43 | blast007 | I'll see about setting up a test server when I get home. I'll need to edit the map file and get some other plugins added to the build to test properly. |
| 21:18.40 | Patlabor221 | CmdLineOptions.cxx:1451 is where I special case the geno error |
| 21:18.54 | Patlabor221 | getDescFromAbbreviation is set to always return nullflag for it |
| 21:19.23 | Patlabor221 | I still keep the flag class around since that code is used by the client too, but I moved out out of goodflags and into "disabled flag" |
| 21:19.41 | Patlabor221 | so new client will function with old server |
| 21:19.52 | Patlabor221 | and new server will just never send it to a client |
| 21:25.32 | alezakos | oh the name of the new plugin is awesome |
| 21:25.49 | Patlabor221 | I wanted it to be accurate |
| 21:25.59 | Patlabor221 | so people knew what they were doing to players |
| 21:38.05 | alezakos | hmm, I'm getting src\game\nethandler.cxx(327): error C2065: 'SHUT_RDWR' : undeclared identifier on MSVC |
| 21:41.16 | Patlabor221 | ahh I didn't move that over |
| 21:41.19 | Patlabor221 | put this at the top |
| 21:41.34 | Patlabor221 | #ifndef SHUT_RDWR |
| 21:41.40 | Patlabor221 | #define SHUT_RDWR -2 |
| 21:41.42 | Patlabor221 | #endif |
| 21:41.49 | Patlabor221 | put that after the include for the header |
| 21:41.55 | Patlabor221 | and check it in for me |
| 21:42.29 | Patlabor221 | throw a tab in front of the #define too :) |
| 21:42.44 | alezakos | ah, thanks |
| 21:42.47 | Patlabor221 | is stuck in the middle of WFOs |
| 21:43.01 | alezakos | after the <errno.h> header of "bzfsAPI.h"? |
| 21:43.50 | Patlabor221 | whatever the last header is |
| 21:43.56 | Patlabor221 | it doesn't really matter where |
| 21:44.02 | Patlabor221 | I just do defines after headers |
| 21:44.09 | alezakos | k |
| 21:44.33 | Patlabor221 | do you get the not issue if you build bzfs? |
| 21:45.12 | alezakos | I haven't tried, wait a sec |
| 21:45.20 | Patlabor221 | thanks |
| 21:46.26 | alezakos | the bzfs build throws that error too |
| 21:46.55 | Patlabor221 | it should share the nework code |
| 21:47.24 | alezakos | oh, the 'not' undeclared identifier issue? |
| 21:47.30 | Patlabor221 | yeah |
| 21:47.32 | alezakos | yeah, I'm getting that |
| 21:47.33 | Patlabor221 | just change that to a ! |
| 21:47.38 | Patlabor221 | and check that in |
| 21:47.54 | Patlabor221 | devinfo is pretty clear about not using not |
| 21:48.57 | alezakos | is there any reason to use not instead of ! ? |
| 21:49.13 | Patlabor221 | some people like that it spells out a word |
| 21:49.21 | Patlabor221 | and it is more visual then ! |
| 21:49.47 | Patlabor221 | I can understand that point, but you have to be able to deal with ! in any legacy codebase |
| 21:49.58 | Patlabor221 | and I don't think we should mix ! and not |
| 21:50.08 | Patlabor221 | and if we go with not, then we should do and and or |
| 21:50.13 | Patlabor221 | and that's all too much typing |
| 21:50.42 | alezakos | good point |
| 21:51.41 | Patlabor221 | so for us it will simply be a style choice |
| 21:51.59 | Patlabor221 | hence why I tried to make it clear in devinfo, even if I did get some of the history wrong |
| 21:52.06 | Patlabor221 | it still says "don't use not" |
| 21:57.13 | Not-311d | [02bzflag-import-3] 07kongr45gpen pushed 031 commit to 03v2_4_x [+0/-0/±2] 13http://git.io/N0RDYA |
| 21:57.14 | Not-311d | [02bzflag-import-3] 07kongr45gpen 032110283 - Fix the build on MSVC, as indicated by @JeffM2501 |
| 21:59.17 | alezakos | I almost have a patch ready that adds an option to automatically hunt your enemy if he carries your team's flag, would that be a desirable feature? |
| 21:59.50 | Patlabor221 | not in 2.4.x, save it for 2.5 so everyone can have it at the same time |
| 22:00.38 | alezakos | ok |
| 22:00.46 | Patlabor221 | and I don't think you need to HUNT them, just draw use a HUD marker |
| 22:00.55 | Patlabor221 | you want some visual clue right? |
| 22:01.07 | alezakos | yeah |
| 22:01.23 | Patlabor221 | then yeah a HUD marker is perfect for that |
| 22:01.30 | Patlabor221 | then it dosn't blow out the hunt target |
| 22:01.55 | Patlabor221 | but yeah I'd wait for the breaking release so it's not a case where some clients have it and some don't since it could be seen as an advantage |
| 22:02.02 | alezakos | there's already a HUD marker that points to the team flag on the HUD, it's just hard to identify the flag carrier on the radar |
| 22:02.20 | Patlabor221 | adding an additional marker to the carier is not a bad thing to do |
| 22:02.32 | Patlabor221 | I'd say your feature is an extension of that, not an extension of hunt |
| 22:03.06 | Patlabor221 | hud marker would show you an out the view direction for where they are too |
| 22:03.18 | Patlabor221 | so those 2 would help both radar and 3d view players equaly |
| 22:03.38 | alezakos | sounds fair |
| 22:04.03 | Patlabor221 | hooray for colaboration :) |
| 22:04.23 | Patlabor221 | now we just need to get MS to buy us :) |
| 22:12.40 | *** join/#bzflag Constitution (~josh@bzflag/developer/Constitution) |
| 23:21.22 | Not-311d | [02bzflag-import-3] 07kongr45gpen pushed 031 commit to 03v2_4_x [+0/-0/±1] 13http://git.io/xhhkOg |
| 23:21.23 | Not-311d | [02bzflag-import-3] 07kongr45gpen 037342962 - Joystick hat fixes for Windows |
| 23:23.25 | alezakos | does anyone have a hat-wearing joystick? |
| 23:28.21 | Patlabor221 | yes, I have an oculus rift |
| 23:29.53 | alezakos | well, would bz recognise it as a hat? |
| 23:30.28 | Patlabor221 | it goes on your head, so yeah |
| 23:31.47 | alezakos | no, not that kind of hat :P |
| 23:43.20 | Patlabor221 | there are 2 kinds of joyhats, polar and fixed postions |
| 23:43.43 | Patlabor221 | basscily any angle, or one of 9 postions |
| 23:53.35 | gnu-sense | if you use an occulus rift for very long, does it make you wall-eyed? Because that's what it sounds like... |