| *** I_Died_Once <I_Died_Once!~I_Died_On@c-73-184-170-223.hsd1.ga.comcast.net> has quit IRC (Ping timeout: 258 seconds) | 00:40 | |
| *** I_Died_Once <I_Died_Once!~I_Died_On@c-73-184-170-223.hsd1.ga.comcast.net> has joined #bzflag | 01:28 | |
| *** spldart <spldart!~spldart@2601:2c5:c680:87a0:3210:b3ff:fef0:93af> has quit IRC (Ping timeout: 240 seconds) | 04:25 | |
| *** I_Died_Once <I_Died_Once!~I_Died_On@c-73-184-170-223.hsd1.ga.comcast.net> has quit IRC (Ping timeout: 265 seconds) | 05:09 | |
| *** spldart <spldart!~spldart@2601:2c5:c680:87a0:3210:b3ff:fef0:93af> has joined #bzflag | 05:27 | |
| *** Gollum <Gollum!~Gollum@131.93.56.40> has joined #bzflag | 06:28 | |
| Gollum | Is this where the cool kids are? | 06:28 |
|---|---|---|
| Agatha | sadly no | 06:57 |
| Gollum | oh nothing new then haha | 06:57 |
| *** Sgeo <Sgeo!~Sgeo@user/sgeo> has quit IRC (Read error: Connection reset by peer) | 07:05 | |
| *** Gollum <Gollum!~Gollum@131.93.56.40> has quit IRC (Quit: Client closed) | 07:16 | |
| *** SpringTank <SpringTank!~quassel@69-154-138-57.lightspeed.rcsntx.sbcglobal.net> has joined #bzflag | 08:16 | |
| SpringTank | question. what is the server setting that always forces a flag to be put into a random place when dropped? | 08:16 |
| allejo | set _maxFlagGrabs 1 | 08:29 |
| SpringTank | works, thanks | 08:38 |
| *** I_Died_Once <I_Died_Once!~I_Died_On@c-73-184-170-223.hsd1.ga.comcast.net> has joined #bzflag | 09:48 | |
| jfindlay | noo, I hate that config | 10:00 |
| SpringTank | another question. was there every a recent update to the useless mine plugin? I've downloaded the one from 2011 and it works and I've been able to mod it. however, I can't get it to update scores or send the server wide messages. | 10:01 |
| SpringTank | I've narrowed down part of the problem to this conditional statement: if (((bz_PlayerDieEventData_V1*)eventData)->shotID==i+1000) | 10:43 |
| SpringTank | something to do with identifying word weapon shotID's | 10:43 |
| SpringTank | i think... | 10:44 |
| SpringTank | ah, i found the one on github | 10:44 |
| blast007 | A while back, I had someone design a new logo. I cleaned it up a bit. How does this look? https://static.bzexcess.com/tmp/8f.svg | 11:40 |
| blast007 | Is it an improvement over our old logo? (ignore the red background thing) https://github.com/BZFlag-Dev/bzflag/blob/2.4/misc/art/bzicon-red.svg | 11:41 |
| the_map | I think I like the corners and edges of the current, more than the pillowed look of the new | 12:36 |
| the_map | the new has a more modern flat look, which doesn't quite capture the feel of the game | 12:36 |
| blast007 | hmm, I guess now that I have our tank model exported, I could even just render out 3D icons | 12:51 |
| *** I_Died_Once <I_Died_Once!~I_Died_On@c-73-184-170-223.hsd1.ga.comcast.net> has quit IRC (Ping timeout: 268 seconds) | 12:56 | |
| blast007 | https://raw.githubusercontent.com/BZFlag-Dev/bzflag/v1.10.0/win32/bzflag.ico like that again ;) | 12:58 |
| the_map | the real question: is the "32" for Windows 32-bit or for the image being 32px? :) | 12:59 |
| blast007 | Windows 32-bit :P | 13:00 |
| blast007 | even now, we still build for 32-bit on Windows | 13:00 |
| the_map | is it because you like seeing what you can get bzflag to run on? :p | 13:00 |
| SpringTank | The new logo is nice, but IMO the minimilist style that seems to be trending as of lately has got to go. I like the old glassy look, or shoot, maybe even the old "windows XP" plasticy look. but im just being nostalgic. | 13:24 |
| blast007 | the_map: part of it is that some of our dependencies didn't/don't have a 64-bit build in their included Windows build files | 13:55 |
| blast007 | there are issues with 32-bit builds on Linux (floating point issues causes players to get stuck in teleporters that are on/near mesh objects, for instance) but we don't see those on the Windows builds | 13:57 |
| blast007 | so less motivation to make the switch on Windows | 13:57 |
| blast007 | I'll probably try to do it for 2.6.0 though | 13:58 |
| *** Sgeo <Sgeo!~Sgeo@user/sgeo> has joined #bzflag | 14:17 | |
| Agatha | The old logo is better IMO. There are a few things I would change about it, but the new one is a clear regression. | 15:46 |
| Agatha | IIRC I got 64-bit builds to work on Windows when I was working on the Two Tanks project. Given how little I know what I'm doing, it therefore doesn't seem like it should be a big problem. The floating-point issues on Linux are another matter, I suppose. Offhand, perhaps there are some compiler flags (robust FP, etc.) that should be tried; 32- vs. 64-bit changing the behavior is probably only due to chance. | 15:51 |
| Agatha | 32-bit computers . . . I guess still exist, and are maybe over-represented among bz players, but 64-bit by default seems fine. You ask me, it should be an easily-switchable configuration option, so both versions can be released. | 15:51 |
| *** I_Died_Once <I_Died_Once!~I_Died_On@c-73-184-170-223.hsd1.ga.comcast.net> has joined #bzflag | 16:53 | |
| allejo | SpringTank: use the more recent UselessMine plugin (the one I wrote) | 17:45 |
| allejo | https://forums.bzflag.org/viewtopic.php?f=79&t=20011 | 17:46 |
| allejo | there were changes made to the World Weapon API that causes older plugins not to function correctly with point assignment | 17:46 |
| blast007 | Agatha: did that require any modification of build files in our dependencies? | 20:11 |
| blast007 | Using CMake to build them might be a better option than what we're currently doing. Or using Microsoft's vcpkg dependency manager, which I've started to experiment with. | 20:12 |
| Agatha | blast007: I don't remember—which itself is more evidence that the process wasn't significantly more painful than setting up the thing in the first place (which to be fair is very unintuitive; you must rely on the build directions, I see this as a major deficiency by itself). If moving to CMake is on the table, I absolutely endorse that. Everyone who uses C++ knows how CMake works; it's the sine qua non of build systems, laaa | 20:15 |
| Agatha | Another vaguely recollected opinion: remove dependencies or at least bundle them directly into the main source as redistributables. This is especially important for Windows, which lacks sane package management. There's no particular reason not to do this, and it's arguably standard practice to do so. This is, FWIW, fairly easy to do with CMake; you can point the build at some bundled sources, or allow the user to pick an external location—but either way you | 20:27 |
| Agatha | have it ready to go because building without dependencies is more or less useless / impossible in the first place. | 20:27 |
| blast007 | we used to bundle them but that caused problems for linux distributions | 20:28 |
| blast007 | they had to remove them and patch our build files | 20:28 |
| blast007 | I don't see us going back to that. We have at times provided a pre-compiled dependency package that can just be extracted. The master version of the code doesn't even need you to define the BZ_DEPS variable anymore as it just looks in a "dependencies" folder you create. | 20:30 |
| blast007 | vcpkg is a pretty slick dependency management that can even integrate with CMake | 20:31 |
| blast007 | it can automatically download and build the dependencies based on a manifest file | 20:31 |
| blast007 | macsforme prefers premake5 over CMake, but I feel CMake is more mature and well supported | 20:31 |
| JeffM[m] | preeeeeeeeemmmmmmmaaaaaaaaaakkkkkkkeeeeeeeeeeeeeee :) | 20:32 |
| blast007 | premake doesn't have package/library detection from what I can see | 20:32 |
| blast007 | at least few that are already written | 20:32 |
| JeffM[m] | well your premake doesn't :) | 20:33 |
| JeffM[m] | my fancy one does | 20:33 |
| blast007 | :) | 20:33 |
| JeffM[m] | I think there are ways to do it, but I have not done that part. so yes that is a valid concern | 20:34 |
| JeffM[m] | I know it can be done, because at work it pulls from our package repository | 20:34 |
| JeffM[m] | I don't know if that's custom code, or fancyness in our lua | 20:34 |
| JeffM[m] | we do have custom versions of several tools | 20:35 |
| blast007 | until November 25, 2020, Premake5 had a module downloader that pointed to defunct site, opening up a possible remote code execution bug. | 20:39 |
| JeffM[m] | ooof | 20:39 |
| blast007 | and that fix wasn't in a Premake5 release until April 13, 2021. (5.0 alpha 16) | 20:42 |
| SpringTank | Thanks allejo. I already found that one and used it instead. | 23:27 |
| SpringTank | I modyfyed it so that it places a mine on dropping the useless flag as well as the command. I think it's a little more intuitive for new players and adds some interesting mechanics to using the mines. I also limited the shots to 3 for the useless flag so if you run out of shots it drops the flag and places a mine right there. If you die while holding the useless flag it places a mine kinda like a dead man's trigger. | 23:29 |
| blast007 | ah, that's an interesting modification | 23:31 |
| blast007 | I assume that when you die with a flag, the server triggers a flag drop API event, and that's what causes the last part? | 23:32 |
| SpringTank | yes | 23:33 |
| SpringTank | it looks to be possible to check to see if there was a player die event before you check for a flag drop in order to disable that "feature" | 23:33 |
| SpringTank | but IMO I prefer the former. | 23:34 |
| the_map | to play devil's advocate, that removes some of the challenge of using the mine - a player currently has to stop what they're doing to put the mine down, so they have a moment of downtime | 23:34 |
| SpringTank | ^ | 23:34 |
| SpringTank | bzflag needs a dedicated "use" key for this kind of stuff. | 23:34 |
| SpringTank | i know you can bind commands, but that's a little less intuitive for new players | 23:35 |
Generated by irclog2html.py 2.17.3.dev0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
