smiley

Its not entirely useless. Even if you manage to do the NAT traversal, the ports still need to be opened. I am not sure if 0AD STUN use a TURN server too. If there is no such thing, (I assume you are behind a symmetrical NAT since it failed even after you used STUN), its pretty much impossible to have an e2e session. ISPs should just commit to ipv6 and do away with NATs.

Tangentially related since this has nothing do with 64 bit builds, but wdbg code is written in a really old win api and has not been updated since it was written more than a decade ago. Better ways to get a backtrace are now available in the win api. Someone should really refactor those early 2000's code.

I think it's not possible. Ninja doesn't support those. I personally didn't try using it with anything besides vs2019 with win10 toolkits, so I am not too familiar with that, but I recall reading something like that, not entirely sure.

None of the potential causes I can think of sounds particularly hard to fix. Then again, I have no clue as to which of them actually caused the bug, or if it is even one of them. No interest in finding out either. A laughing reaction might fit the most since the message of doom and gloom isn't as serious when you start debugging. However, at this point when this seems to be the norm and smooth releases doesn't seem to exist anymore, it might be worth it to consider the reasons why this is the case. Removing a singular oddly specific map script might sound like an attractive option, but people making that choice is the reason for this mess right now. Release with the bug, or fix it. Just don't bury it.

Non deterministic execution with a seeded psuedo rng. Good luck to the lucky soul who is going to fix that.

I think this already happens naturally. ELO inflation is a thing. When I started to play three alpha versions ago, the second on the leaderboard was 1800 something. I don't know the exact rate.

Yes, its pretty much the same thing to be fair, except that visual studio plays nicely with vcpkg, in that it tries to emulate a /usr/lib/. What other approach is there for Win? For certain libraries like nvtt, one would need to patch the portfile too. Painful, but its the least painful way I have found.

That's not surprising. Download counts are no where near active user counts. For the true user count, you would need telemetry, but who cares, this isn't corporate software.

https://github.com/microsoft/vcpkg I suggest devs to drop third party libraries and just rely on vcpkg aside from SM and FCollada.

There is no such thing as physics. Projectiles themselves are delayed damage handlers working with rather straightforward linear transformations. There are purely visual actors accompanying them, but by all means, as far as game logic is concerned, they don't do anything. Actual ballistics, and projectiles as entities dealing damage on collision is an interesting concept. None of the math will be extremely complex, so maybe an experiment for modders. I don't think several hundred projectiles would be that big of a performance concern even for JS mods since projectile paths would be approximated. Also, while on the topic of ships and realism, there is also this, I don't know where the code is though.

First of all, even if your Radeon card isn't used, and integrated graphics is being used, the OpenGL version provided by " Intel® HD Graphics 4600" is 4.3. The integrated graphics can run the game. At least, in the case of the card that's used, the driver has issues and the fallback is used hence, the OpenGL version being 1, and the driver name not being reported as neither Intel nor Radeon. Check your drivers. The game should run on either of your cards without errors.

Not to mention the storage requirements. I ended up refactoring the code to compile it with vs2019. vs2021 is coming out soon.

Never tried fedora, but I run Arch pretty much on a daily basis. I don't like bundled software within installation images. Half the software from a default manjaro installation, I wouldn't ever use. But if someone ever asks me for a distro, I just say manjaro, because, I want them to have up to date software since the machines are pretty new, and most people who would ask me definitely won't know how to install arch.

Manjaro comes with pamac which supports seamless AUR installations. I am pretty sure there is a GUI for that too. It's just a kernel along with userland software. Like I said, aside from the repos, the only difference is the logo.

Yes it is. You get rolling updates, repos are sourced from arch, installation is easy and KDE images comes with great out of the box config. The manjaro and arch community don't get along, half because arch users are douches, half because manjaro users can't read the manual and somehow always end up in arch discussions. But that's a battle for people with way too much free time. The logo doesn't matter in the least, so if you want Arch repos, rolling updates and don't want to manually install a system, go for it. AUR works the exact same way regardless of distro. Regarding dependency hell, manjaro Devs have messed up before, don't update after pushes before a few days, let someone else check for breakage.