myconid

The water shader code has gotten a little bit too artsy to debug easily, so I can't readily point to where the problem might be. Maybe wraitii has an idea... alpha123 and serveurix, does it make any difference when you turn on/off reflections, refractions and so on?

Already working on it.

2.1 or above. You have 4.3, so you're fine.

preferglsl=false + gentangents=true is a bad combination...

It looks like Intel's drivers for the GMA 3600 are so buggy that nothing works on them. Do any other recent 3d games work on your netbook? Besides, on underpowered hardware like the GMAs, even if the game did run it would probably be far too slow to play (the game should run fine on netbooks with Intel HD graphics cards and above).

wxNavigationEnabled was added in version 2.9, according to the docs. A quick search shows that these errors can be caused by an incompatibility between wxWidgets and Xcode's Clang toolchain. Their suggested workaround is to use a different toolchain, such as llvm-gcc. Not a mac user, so that's all I got.

Intended behaviour, unfortunately. The engine stretches the visible objects so they fill as much of the shadowmap as possible (to get better quality shadows), but doesn't account for terrain while doing so. The result is that shadows from terrain aren't rendered unless that terrain falls within the bounding box of objects on screen. By moving that building around, you're forcing the bounding box to be expanded, so part of a cliff falls inside that bounding box. It's easy to reproduce in Atlas. Make an empty map, set the sun elevation to a low value and create a tall mountain. No terrain shadows will appear. Then pick an object and place it on the north-east side of the mountain. Create a second object and place it on the south-west side; move it around. You'll notice that only the parts of the terrain between the two objects will cast shadows. This is part of what causes the problem that Pureon mentioned (The other part is that the engine does the visibility culling in the simulation (not at all thread-friendly design, mind you), and it does it by eliminating everything that isn't visible from the main camera, thus the renderer doesn't know about things that may cast shadows/reflections once they go off screen).

You so clever. Look up Manhattan distance.

Since we have the same hardware, I guess it must have worked with fglrx. I think fabio had a similar problem with the same driver before... I'll test out a couple of ideas to fix it (hopefully before release) and maybe I could get some feedback from you, if it's okay.

If the sqrt is being used to compute distance, it may be an acceptable approximation to get rid of it completely. There's a built-in profiler, accessible with F11 in-game. I never thought to check.. but please don't tell me it's running always, even in the Release builds!?

Hmm. Have you made sure the versions of your kernel and Xserver are as in the link above? kernel <= 3.2 and Xserver <= 1.11

There's a "legacy" version of 12.6, is this the one you installed?

I haven't. If you want to keep fglrx, I suggest checking if your distro has an LTS release. I have Kubuntu 12.04, which is based off of kernel v3.2 and is supported until 2017 (this laptop will be gone long before that!).

I have the same GPU on my laptop. I'm afraid AMD dropped support for the HD 4670 in the latest Linux fglrx drivers, so we need to use versions <=12.6.

And video games are classified as art in Germany, so I'd say it's almost certainly legal to use Swastikas in games, even Nazi Swastikas. I say keep it. If anyone is silly enough to feel offended by an ancient symbol being used in a historically-correct way, this is an open source game and they should feel free to mod/fork/rewrite history.

http://www.wildfiregames.com/forum/index.php?showtopic=16817 http://www.wildfiregames.com/forum/index.php?showtopic=16791 Similar-sounding reports, all made in the same week... Memory leak??

You've probably tried this already, but have you checked the *nix user permissions of the directory? e.g: chown -R username ~/Games/0AD/ (where username is the local username you use to log in) Nevermind, just noticed the actual bug report was about the singletons, not the permissions. Derp.

Technically inaccurate, but correct in terms of everyday gaming lingo - we use p2p as opposed to a dedicated third-party server. Maybe change the link to point to the definition in the "game server" article?

I can't get over how good that looks! Is it shopped in any way? If it is, we should add those effects to the game ASAP!

All the fancywater is rendered using GLSL. That includes both the water-on-models and the global water planes. I think there are two bugs at play here. One is with ShaderTechnique, as it bugs out when the only option is a GLSL shader but preferGLSL isn't set. The other is somewhere in the GUI, which seems to make assumptions about the caller's state that happen to be invalidated by the water effects.

Ludo, check out this thread for some cool pre-built maps. In Atlas, maybe you can set a couple of AI bots to play against each other and then press the "fast forward" button (on bottom panel of first tab) to populate your maps.

Tried running your code. For me, the JS_EncodeString call in function2 causes an "out of memory" error. I queried the length of the string on the line before that call and it returns "8769178900488". The original "textfromscript" version of your code works fine, because the string is const. The "blablubb"/recursive versions cause the error, for the reason I explained earlier. Don't know why you can't reproduce this, but at least the theory checks out.

Does the dump show the raw memory contents, including deleted objects, or is it a liveness graph? If it's the latter, then it doesn't help prove or disprove what I wrote earlier... If the object is copied, then it persists in the function2 call because of a memory leak and you'd need to free it manually. You could create a bunch of short strings and see if it affects the output. eg: function jsfunction(s1) { var textfromscript = "textfromscript"; function1(textfromscript + s1); } var countInJs = 7; jsfunction("blablubb"); function rec(i) { var s = Math.random().toString(36).substring(7); if (i > 0) rec(i - 1); } rec(10000); // allocate a bunch of strings function2(countInJs);

Right, speaking of "shadow objects", there's also the possibility that since he's passing out a constant, it gets initialised statically so it never gets deleted (though it depends on what those conversion functions do..).

I'm not familiar with Spidermonkey's exact implementation, but I've implemented similar systems in the past. As a result, the following contains a fair deal of assumptions. A useful thing to know is that many GC designs don't depend on system malloc(), but keep around one or more large chunks of memory and do their own allocations inside those chunks (it's faster). This means that the memory is reused, and even if a certain object is "deleted" by the GC, it's still kept around in the process's memory until it's overwritten by another allocation. Spidermonkey uses a simple type of "mark-and-sweep" GC (afaik) that doesn't move allocated objects around, so anything you allocate has a fixed memory address. Thus, you can access objects directly from C++ pointers, such as JSString*, without worrying that they might move... but when the memory is reclaimed by the GC and reused, the C++ is stuck with invalid pointers. So, my guess is that you are dereferencing a dangling C++ pointer that points to an object in GC memory that has been "deleted" but is still in memory and hasn't been overwritten. In theory, if the JS keeps allocating new objects of similar size as that string, you'll eventually overwrite the memory and get nonsense output.