Jump to content


WFG Programming Team
  • Content Count

  • Joined

  • Last visited

  • Days Won


Everything posted by elexis

  1. Sure, we can't increase the precision without increasing the performance.
  2. What kind of 'free code' do you mean? Not the already free game code? Funding developers to write new or improve existing code, or do reviews, or artwork, presenting the game software conventions, or whatever else Wildfire Games needs to accomplish the objectives of Software in the Public Interest. There is still a lot to do until this game is finished, and a successor to this game was planned too.
  3. Wildfire Games is a member project of the non-profit organization Software in the Public Interest because we create free software, not only gratis, but also free to modify and redistribute under the same terms. Donations to Wildfire Games are tax-exempt and allow Wildfire Games to continue developing the game without such practices. Wildfire Games can involve donors a bit more than a hidden Donate button on play0ad.com however. About the T-Shirt shop proposal, that actually seems to be possible, given that Blender Foundation also has one. Not sure if it will be possible with the mentioned
  4. Last time I checked, elephants are only garrisonable in Elephant stables. It's certainly inconvenient to not be able to hide one hero in regicide mode. The "Inspired defense" was the garrison aura IIRC, every hero inherits that, and should be removed for units that cannot garrison. Perhaps one can also chose to be practical about the hero and let it garrison.
  5. Sounds like splitting game content (tempates + map + art) from the pyrogenesis engine (code): #5366
  6. Looks good. The entire VisualActor will be the mine that units surround and play the gather animation from? Should work, but what will happen to the actor once the mine is depleted? It sounds like it might look awkward if theres an empty spot in the map that was designed to have this little hill and flora there. So perhaps that could use some 'damage variant' or the decaying 'corpse / rubble' actor mechanism? With damage variants, the transition could be smoothened, but of course costs a lot of additional work. Replacing the active mine actor with a depleted mine actor would be easier. Also if
  7. Depends on what the implementation will look like. Not everyone will be able to implement that, so I suspect few big companies will offer such a service and most might end up using that. Then the change the users see would depend on what that corporation decides to filter for. Also there is a difference between not being affected and not seeing how you're affected. If you can't read the books that were burned, you don't know if you missed something. It's true that they probably can't kill the message itself, but people like investigative journalists, political activists
  8. Depending on the size of the hill necessary to make the mine look more authentic, Most maps would have to be significantly reworked in many cases to integrate large hills. Can we find a visually pleasing model and texture for any existing maptype and biome? Reserving something like 20% more map area for hills will mean either (even) less area for players to build, or require players to play with larger mapsizes (i.e. simulation performance improvements become more important). Gameplay wise, if there are less locations on the map that provide resources, then not capturi
  9. The ratings are in a separate database, so aren't deleted. I think @Hannibal_Barca might help you out with that (should verify that you aren't an impersonator). You may want to backup your user.cfg to preserve the login.
  10. Find one who can configure their router to forward UDP port (by default 20595) for his local IP address, then everybody should be able to join that host (with or without STUN).
  11. That must have been under the assumption of creating maps like we used to (players, one river, random hills, random resources, voila). That were 200-300 lines of code. Now we have 1500 lines or whatnot and the ceiling to map quality is the performance bottleneck. The next barrier to tear down in particular is being able to work with a precision smaller than 1 tile, so that maps can look as good as on those screenshots.
  12. https://wildfiregames.com/forum/index.php?/forum/468-alpha-23/
  13. Vanilla + leftovers that don't change the simstate
  14. @stanislas69 I recall spending a lot of time with ffm trying to reproduce the bug, messing with forged replays, never worked reliably. Looked at the code, looked like it shouldn't be possible to bug. It's a Schrödingers bug. If you can reliably reproduce the bug, or with high likelihood, you should report the steps to reproduce. Should try at least 5 times. Edit: It might not be that bug, if the construction site just appears very late and the screenshot is taken before then.
  15. cmpAttack.template.Capture should be equivalent to testing for GetAttackTypes. I only found "this.template" , no other "*.template" in the source. I suppose it's a convention in order to not have components parse templates unless it's the template of their own component. cmpAttack.PerformAttack should be called if a capture attack is ordered and if it can capture, or if an attack is requested and the entity doesn't have buildingAI. Perhaps that can be phrased as a ternary: (order == capture ? canCapture : !hasBuildingAI). In the model of the current code: cmpAttack.Perf
  16. Those weren't two solutions but one. Players, scripts, AIs may send arbitrary or broken commands, so the component should test for both conditions to be true simultaneously. Only testing for the latter would be wrong, because the idea of the commit mentioned above is that arrows shot are either handled by UnitAI for units that only shoot one arrow at a time, or by BuildingAI if they shoot multiple. (So the purpose of the fix had never been to correct a tooltip, but the broken tooltip was one of the consequences of the oversight addressed in that commit)
  17. Evidently not committing the oneline fix was worse than committing it. In other areas of the file, I see this.order.data.attackType == "Capture". I didn't test, but cmpAttack.GetAttackTypes().indexOf("Capture") != -1 doesn't work for the SiegeTower?
  18. The question is really what people want to achieve. To edit the technical documentation, they can get started right away. The trac wiki is editable by anyone. Posts for play0ad.com can be posted here and we can copy them over if they are good. If we like your contributions and think that you are responsible, you will be invited into the team (whether you wrote an applicaiton or not). One more thing about the technical documentation, I do believe it should be first in the source code (doxygen and JSdoc), and secondly in the wiki. We want everyone who distributes the code to be able to
  19. Sure, but that approach: The recommendations above are to recommend not limiting oneself to that. At least one contributor got himself a review by acting on that post, so it wasn't useless. If you see that people don't have enough manpower to keep up with the review queue, the obvious conclusion is that there must be more team members that do reviews, and thus you and everyone else capable should become one and allow others to upload patches in the principle "here it is, take it if it's any good". That may be true for newcomers, but you have gained much more knowledge than
  20. "I don't understand at all what is the goal of the team now? if it is to kill the project, you are doing a very good job."
  21. I haven't seen compelling evidence yet that the forum attachment and formatting options of the trac wiki are insufficient. The main purpose of the request was to pull in a large crowd (right?), that sounds good, but Wildfire Games should really keep a copy of all revant information on their sites. Even better would be if the information is created there in the first place, rather than having to move it over in retrospect. Especially if we're talking about technical documentation (that should even be inside the code first and on the wiki second), and WFG history. If the wikipage remai
  • Create New...