Itms

Welcome! @Doktoreus is still active and looking into some parts of the documentation, but we will never have enough contributors in this area, so feel free to start proposing content and an application. I second elexis in pointing out that we'd like to know which part of the documentation you are interested in (the technical one, or one aimed at players).

Follow the steps in this guide: https://trac.wildfiregames.com/wiki/BuildInstructions to get the development version of the game. No need to build anything. Then, locate the translation files (everything starting with your language code) under binaries/data/mods/public/l10n/. There are a few strings translated using files in binaries/data/mods/mod/l10n/ and binaries/data/l10n/. Put them in your "user" mod (locate it using https://trac.wildfiregames.com/wiki/GameDataPaths), in a folder /path/to/your/mods/user/l10n. Yeah, we could easily package a mod with the incomplete translations on each release.

Ah I see what you mean. But the engine does not weigh 2GB, so it can be updated just by downloading the new version. The public mod, on the other hand, is identical on all platforms, and could be updated incrementally in theory. Splitting the public mod into smaller pieces (the art aside, typically) would help, because we could release small balance patches not touching the art, hence just modifying a small zip file, and launchers like Steam would be able to make users only download the changed file.

What do you mean Stan? Are you thinking about package managers? To answer the top post, it would be doable but difficult. We could include a torrent client in the engine so that the game contents, especially the large assets, are updated efficiently. But self-upgrading the engine would be even more involved. Not even speaking about the security issues we would have to prevent. I think the simpler way to do it would be to use a separate launcher for the game. That already exists (see Lutris for instance, on Linux) and, when the game is released, you will be able to use Steam or your preferred game launcher, since we will try to launch on as many platforms as possible. There is very little point in creating our own dedicated launcher.

In theory, you will need to wait for a release. If you want, you can get the translations from the development version and use them, but they won't work if some strings in the original English were deleted or modified between the last release and the current date.

Hello! It all depends on the historical period you want to focus on. In particular, since you seem to focus on Turks, there is no civilization in the time period of 0 A.D. that can be directly linked to modern Turkish culture. You may want to take a look at the Millenium A.D. mod which could include the Seljuk Empire, the Sultanate of Rum, or even the emerging Ottoman Empire, in its part II. Similarly, for Chinese powers, the Han dynasty is present in Terra Magna, but other dynasties could be added in other mods with other time frames (Tang/Song in 1000 A.D., etc).

Would it be fine if Dennis would report only on Jenkins (the detected issues wouldn't appear on Phabricator, only on Jenkins)? Else I can maybe roll up my sleeves and write a patch for the most annoying issue in the plugin.

fc-28 is for Fedora. Yeah we need the crash message. The best way to do this is to run pyrogenesis -editor in a terminal to launch Atlas directly and get the errors in the standard output.

Hello, if you folks create a mod, put it on mod.io and PM to sign it, it shouldn't be an issue! As said above, checking maps is really easy and quick to do, even if there are a lot of them. As a potential user, may I suggest that the mod comes with a pinch of GUI code, adding a map filter for the new maps? (So that the game setup drop-down shows "Default", "Community Maps Pack", "Naval Maps", "Demo Maps" etc). I like the idea

Hello! I am sorry about the situation here. As far as I understand, the report was followed by action, but the report itself was not properly acknowledged. I just sent a PM to user1, telling him to answer here to clarify the misunderstanding. pesem's messages are definitely unacceptable and unwelcome on the lobby, and they are indeed covered by the Terms of Use. So let's not spiral down, and as far as possible let's not involve politics here. Let's just solve the misunderstanding that happened between Sundiata's report and user1's answers to it.

Hello! What platform are you on, are you using the stable or the development of the game, and what do you mean by "never works" (does it crash with a message, or nothing happens)? Thanks in advance

Fixed by Angen in r22125! Thanks for the report Andrettin

Hi @Basshunter, and welcome! The idea sounds very cool. We'd be happy to have you create the assets for that! Putting them into place in the game shouldn't be the main difficulty. So I'm moving this thread to the Art forums, if you don't mind, where you'll be able to work on this with the help of our artists. I also encourage you to get the latest development version (see instructions here, no need to build the game, just download the SVN contents) so that you can dig into our game assets, tweak them, and test your creations as you go.

Hello everyone! Just a very small announcement for the forum users that the support for Gravatar profile pictures has been removed in the latest version of the forum software. You may have noticed that some well-known forum users have become faceless You can find the official notice in the changelog of the forum software: https://invisioncommunity.com/release-notes/440-r81/ Thus this is not a temporary disruption of the Gravatar service, and you should re-upload your profile picture here on the forums if you want to be easily recognized again. Thanks for your understanding, and have a nice weekend y'all

Well, the strange value is confirmed... I'll do some experiments and probably give you a small minimal program reproducing the inconsistency, but I'm not sure what I will find. In the meantime, if you want to play, you can safely comment out the // ENSURE(tlb.Validate()); on line 54 of cache.cpp. It is only a sanity check (something insane is indeed happening! ) which doesn't prevent you from playing at all. Unless the root cause of this issue causes other issues down the line, in which case you can report here.

Hello! I looked into this, I am a bit puzzled: Based on the line numbers of the stack trace, the validation fails for the output of the function 0x80000006/ebx. L2 TLB/cache information: 4K pages & L2 TLB (0x80000006/ebx): instruction # entries = 0x400 (1024) instruction associativity = 8-way (6) data # entries = 0x600 (1536) data associativity = 0x5 (5) So according to cpuid, your processor answers 0x400 6 600 5 (I added the spaces for clarity). However the stack trace says that the value of the register is 1442866176, which is 0x56006400. The byte fields are the other way around (not per byte but per field, so it's not a little/big endianness confusion). I looked at the source code of the cpuid tool, and it looks like our code is doing the same. So I don't see why the value you are getting in our engine is bogus. In order to dissipate any doubt that I can have on the output of gdb, would you mind applying this patch (and recompiling) that just outputs the values sent by the call to cpuid. And share the output here cpuid-debug.patch Thank you very much!

As far as I am concerned, I would love to see specific names eventually in the original writing (thus eliminating the headache that is transliteration). This is blocked by the same issue that makes us package Asian languages as a mod, which is our poor handling of fonts. I am also very much in favor of having the specific names displayed optionally, so that competitive players, players with accessibility needs, or just players with a different taste than mine, can remove those names (which would be even more visible in the original writing). I wouldn't mind if we remove the specific names for civs where we lack information. I don't have a strong opinion on this: reconstructing some words and explaining our rationale, or limiting ourselves to accurate sources, seem both interesting to me.

That is my hope as well

Wonderful timing for coordinated posts full of positivity @Sundiata

You are totally correct, and I also thank you for the kind words. We are in a dip in the programming activity, due to a low availability of the programmers. On the other hand, the output of our artists and of the modding community is staggering these days! There are up and downs in all the areas of the development, and that's the reality of the life of the project (or any project, for that matter). I am very likely to prioritize programming over leadership in the upcoming months, because the state of the review queue is indeed now one of my biggest concerns (personally, I'm happy that I have no bigger problems than that to fix right now ). But I can't make miracles, I can only do my best. I usually don't answer this kind of thread because they are nerve-wracking. I am genuinely sorry that our best is not good enough for you. It is very difficult to see you guys angry or disappointed when you make such great contributions! Not being able to allocate time to include your work is heartbreaking. The best thing you can do to help is to be supportive (like Sundiata). It really helps us to read nice messages: motivation is our only fuel. I understand the need to complain and push people into doing things, that is something we do a lot in other situations, but that doesn't work for volunteer work. I don't blame you for complaining, but it doesn't help, and sometimes it goes too far. I am really encouraging you to give us ideas about concrete and realistic things that we could try to do, that is always welcome! (especially with the matter at hand, which is, how to make the project attractive for new AI contributors). But just stating the problems we have or might have is not helping, as we are painfully aware of them already.

Hello and thanks for the trace! It looks a lot like this ticket: #2074 and the related forum post: Could you, just like then, run the cpuid tool and give us its output? It seems to be in the AUR. Thanks for the report, if we fix it, that will be good for all the users of your processor model

You can open a ticket, that would be nice to have! I suppose what you'd need is to make an unselectable gaia entity of them (just like hawks, indeed), and tweaking the motionflying component so it doesn't need to take off etc.

Could you provide a stack trace? You can use gdb for that, we have instructions on this page, especially in this section. That would help us know what is the root cause of your issue. Thanks in advance and please let us know if you need some help

The AI is not actually threaded, it is mostly async, to be easily threadable, but the latter needs to be done. Some other parts of the code (such as the pathfinder) are also properly designed to be threaded eventually, but they are not yet. We need someone to implement the threading, we do not need yet to make more parts of the code asynchronous.

You don't need my permission Indeed their license is not compatible with CC, so they would need to release some of the assets under that (apparently that's what they want to do with their "Open source package" (that is not very clear)). Contacting them would be the best way to know what we can do.