Blog

Beta 5, Default and Custom Particle Effects

First, X-Plane 11.30 Beta 5 is live. If we find it’s not a screamapiller we’ll get the Steam version going tomorrow.

Update: it is a screamapiller.  The lighting is missing for chunks of the runway, city, etc. When b6 fixes this, we’ll cut a Steam build.

There were some questions in the comments section about whether aircraft would lose the built-in default effects as part of the particle system upgrade in 11.03. The short answer is “no” and the slightly less short answer is “no, unless you do some stuff in Plane-Maker.” Here’s the details:

• In X-Plane 11.25 an earlier, you got default effects, and there was nothing you could do about it. Every aircraft gets every effect.
• In X-Plane 11.30, default effects are enabled by check-boxes in Plane-Maker’s “Invisible Parts” screen. Just like you can hide rendering of the aircraft’s physics model and replace it with OBJ models, you can replace the default physics-based effects and replace them with your own OBJ-based emitters.
• In X-Plane 11.30, most of the built-in effects are built by the new particle system, but a small number (e.g. rotor down-wash) still use the old one, to take advantage of some hard-coded effects code that the new system can’t (yet) simulate.

Note that the question of how we did the default effect (new or old code) is totally unrelated to whether you (the author) want to replace it; if you turn off our effects, that check box will turn off our effects both now and when we re-implement that category of effects with new emitters (e.g. for rotor down-wash).

I still have a pile of effects bugs open, so please don’t ship aircraft based on the beta using the new particle system. (You shouldn’t ship an aircraft based on a beta anyway, but in the particle system, as of beta 5, I definitely have bugs that, when fixed, will require re-tuning of art files, e.g. lighting levels aren’t final).

Here’s a few particle bloopers.

The Mystery of the Collapsing Landing Gear

We’re pushing to get a stable X-Plane 11.30 beta 3 right now, and I think we’re making progress: Chris has fixed the crash when using the radio stack audio (this Windows-only crash was very frequent) and I have found at least one case of collapsing landing gear.

The bug report was the landing gear collapsing when taxiing on orthophoto-based scenery. It turns out that the custom DSF meshes have triangles that are duplicated and then flipped. In other words, a few of the triangles have twins that are upside down.

Here’s a pair of pictures to illustrate.

The image on the left is the “tops” of the orthophoto mesh triangles – this is what you expect to see.  The image on the right is the bottoms!

Now normally the “bottom” view should show nothing – we should see only the light blue background, because the camera is above the mesh, so all of the triangle backs are facing away from us. Those extra triangles you see are copies of parts of the orthophoto mesh that have been duplicated and flipped over, so that their fronts face down and their backs face up.

This DSF violates the DSF specification, which requires all triangles to be wound counter-clockwise when viewed from above. But as of X-Plane 11.26, this wasn’t a huge problem; X-Plane blissfully ignored the flipped triangles. X-Plane 11.30 is intermittently detecting the upside-down triangles and the resulting upside-down normal vectors are making the collision physics go crazy. We should have a fix for this in X-Plane 11.30 where we filter out these flipped triangles.

One last bit of the mystery: the creation agent on both bad DSFs I have looked at is “X-Plane Scenery Creator 0.9a”. Does anyone have a link to this tool or know who the author might be?

Post script: there is a separate bug where the new physics code is reacting to DSF boundary cracks more severely than X-Plane 11.26. I had hoped this would all be one big bug but it’s not, so mesh cracks will be next on my todo list.

Art Controls, Stutters, and Progress

Art controls are internal tuning constants in X-Plane that we put in so that our art team can modify the behavior of X-Plane while they work without having to recompile the sim from source code. We leave them in the final product because they are sometimes useful to third parties, useful for in-field debugging, and because we are generally okay with people hacking the sim if they understand the limitations. X-Plane has always been a relatively open sim to play with and lots of X-Plane authors got their start just poking around. Our first instinct isn’t to encrypt everything.

We don’t go out of our way to break art controls. We don’t go in and change their names in every beta just to mess with third parties. But we also spend absolutely zero time trying to maintain art control compatibility with previous versions of the sim. Backward compatibility takes a lot of planning and effort and there’s just no way we can keep a bunch of internal tweaks the same while modifying the sim.

In X-Plane 11.30, a number of art controls have stopped behaving the way they used to. They aren’t going to go back to the way they were, because we don’t spend time trying to support hacks based on art controls. Here’s what happened and why:

As part of our work to port X-Plane to Vulkan, I rewrote the code that renders a frame in two ways:

1. All off-screen rendering needed to draw the frame is now done before the frame is rendered; up to X-Plane 11.25 some of this work was done as a diversion, mid-frame.
2. The graphic resources (mostly off-screen rendering buffers in VRAM) are allocated once when the configuration of the sim changes, rather than being allocated “just in time” when we get into a frame and realize something has changed.

Change 1 was needed to match how Vulkan and Metal handle off-screen rendering*; change 2 helps avoid stutters by allocating expensive resources early when we are not flying.

The side effect of this is that art control edits take effect only when the sim is reconsidering its graphic setup anyway (e.g. due to a window size change or rendering setting change); if the art controls change mid-frame, any code that does resource allocation ignores them because it is no longer “sniffing” for configuration changes per-frame. Other code that uses the art controls notices the change, and the result is often haywire drawing due to a half-used art control.

* This was an example of the OpenGL driver doing a lot of work for us, and hurting performance, to support an abstraction that doesn’t make sense. The real graphics card has to do real work when a rendering pass begins and ends; the OpenGL driver automatically synthesizes these rendering passes on the fly for the GPU as the app runs. Because Vulkan/Metal requires that we spell out the passes explicitly, we (the app) know exactly when we are doing something expensive (starting a new rendering pass) and we can minimize the cost.