smiley

We should try to make it as unique as possible. China is pretty big, so chances are there are many biomes that visually look similar to existing ones.

With the addition of Han, generic biomes from China are missing. We have to refine and commit this, at least the Han related biomes. We also need some new fauna. Perhaps pandas as they have become a symbol at this point. Livestock and huntables as well.

This was the original plan for 0AD as well. It was called settlements instead of ruins. Map is divided into provinces and each province has a settlement. https://trac.wildfiregames.com/wiki/War_Story

I don't think we can do anything more complicated than some sort of sheet erosion (which does exist currently). But that can't make mountains, only just smoothen the map. Even if we could simulate rock erosion in < 1 minute or so, that still seems very inefficient and that time can be spent on making more natural forests or creating more natural looking decoratives (markov chain based grass patterns or something perhaps). This is the current plan. Make some nice mountains in Atlas and import the heightmap. Or using an actual actor mesh like in AoE2 (but that always looked a bit weird, like it was glued on or something). We did this for the Jebel Barkal map. Not worth to reimplement any fancy image processing algorithms. Players can't wait too long. But if the underlying thing is some sort of gradient noise based, it might be doable.

Cool stuff, I tried using some fractal generators with pretty boring results some time ago. Just ended up being a bit too "rolling hilly" instead of mountainy. I personally would suggest porting the code over to our Javascript map generation library, or at least publishing the scripts. Those mountains look better than handcrafted ones currently existing and we have basically no procedurally generated ones..

I guess for countries with strong ties to the jurisdiction under which the company is based in this is actually relevant as the other jurisdiction would be cooperative, (extradition and such). However, given that we are not a legal entity in the US or anywhere else for that matter, we don't really have that much legal rights anywhere in the world. For any legal matter, someone from this community (probably the person responsible for hosting the service) has to represent as an individual. This is all how I interpreted this stuff, but based on that, its actually better to keep the status quo. At least until an actual lawyer can advise on the legal ramifications. As for the large companies, well they have in-house counsels who are more than happy to shred through any and all red tape. I have absolutely no idea what I am talking about though.

The jurisdiction is probably anywhere we facilitate providing the service. Unless we block IP addresses from the US, we are probably bound to COPPA. Same goes for GDPR. If we can't uphold the requirements, regardless of where the service is provided from, providing the service in EU means we are probably liable. Governing law clauses would probably be entirely superseded or nullified by the jurisdiction under which the trial is taking place. An EU court would probably take issue if we don't adequately handle personal data regardless if we say our terms are based on American law that has no such framework. Notice the use of the word "probably". I am not a lawyer.

Try the official download, https://play0ad.com/download/, which might fix it. We most definitely do not link to an NFT site in our error dialogs. If the modifications were intentional, how are we supposed to know if there aren't any other modifications which could introduce crashes?

It was in response to Stan's question about the reasoning about the current behavior. Commit archaeology suggests its a bug introduced during one of the large rewrites at some point.

If you don't need anything fancy an alias would do fine as well. alias use_clang='CC=clang CXX=clang++' The commands can be prefixed with use_clang which would save a little time, although, you would need to clean your workspace when switching with clean-workspace.sh.

You could also use clang.

https://bugs.gentoo.org/847601 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105756 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105931 GCC bug, updating to gcc 12.2 should fix it.

Clearly, it hasn't. That still doesn't make voluntary effort involuntary. If things get too out of hand, and we cannot resolve the situation, then I guess we just have to admit defeat and leave lobbies to be community hosted.

#4661 -> See comment:1. rP18806 -> Related fix D2483 -> See Chapter 13

Why? If we change our direction every time there is a lack of volunteers, we'll never get anywhere. If you want to help, rather than constantly point out the lack of volunteers, become one, or find someone willing to become one. Everything else is unfortunately not doing anything to remedy this. Stan is very well aware of the situation. And I am not sure babysitting the lobby is his best use of time. Regardless, I won't speak for him.

We could add a support class of units to armies which currently would just be healers, but maybe in the future could also include siege engineers and the like.

Petra has healers with the lowest priority. Combined with the resource requirements etc. the situation where Petra would train healers in normal operations is almost non-existent. It trains healers in regicide because hero management is delegated to the victory manager which assigns guard units and healers to the hero entity.

The executable in the thread is outdated. But you can still open the current exe properties and change hidpi awareness.

Me too because the svn repo only takes 12.5GB with build artifacts. (the git repo takes like 20GB though) The partition is not relevant since you would be building it yourself. I would suggest a user owned directory though. ~/ perhaps.

I suspect rP26956. If units switch back to the idle state after their target dies, this would happen.

You can just follow https://trac.wildfiregames.com/wiki/BuildInstructions with the only difference being the checkout. (Once you get to the https://trac.wildfiregames.com/wiki/BuildInstructions#Gettingthecode1 section, add the revision number to the command) svn co https://svn.wildfiregames.com/public/ps/trunk@25860 0ad Everything else would be identical.

We are literally on the top of every list, so theres that too. https://trac.wildfiregames.com/wiki/0AD_The_Vision. That is as close one can get to unravel the meaning of the name. Changing it now would be changing the status quo of two decades.

We just send the serialized game state. It would be less bandwidth to send the commands instead, but it would require too much record keeping from both host and client to keep track of what to send and request. And I don't think we even store commands in memory. It's just logged out IIRC. Where does the missing commands in between minutes 7 and 10 come from? Or is it a time travel type deal where we just skip the 3 minutes? This requires storing the serialized saved game state which should be loaded in between replays and likely not implementable without starting from scratch. Solution 2 would work and should not consume that much extract bandwidth, (the commands would have been transmitted anyway had they not disconnected). But it seems to be a bit clunky at first glance. I am not thinking of any alternative though.

We currently don't collect enough information to implement a "forgot password" feature. This is for changing passwords after the account has already been authed. Which could be implemented via another tab to the prelobby page or something inside the lobby interface, I am not sure.

An alternative but slightly unattractive solution in terms of maintenance and dependency on ejabberd is writing up a module to handle some type of age query IQ.