Tag Archives: unity 5

First Experimental Build

So we’ve moved over to Unity 5 and onto our first experimental build 🙂

For instructions on how to opt into the experimental builds, see here:

The builds are only for 64-bit Windows and OSX operating systems. We’d love to get feedback on whether the move to U5 and some changes has alleviated loading problems (10-15 mins) for Yosemite users.

There’s also now a dedicated experimental sub-forum here:

I mentioned in the last update, that we were doing a lot of custom stuff in Stranded Deep. Well the upgrade broke a lot of it! 😛 We had a few problems that delayed last week’s release. There’s a problem with the way we manually batch our materials through code with Unity 5. The bug has been reported but for now there’s no item highlighting. We also had a collider related problem with the new physics system that borked the yellow raft, so enjoy your free wooden one 😛

One of the things we wanted to address with the upgrade to Unity 5 was the UI. In particular, the UI performance. The system we were using previously was great, but a bit of a performance hog. Unity now have a nice native UI system now. So last week was spent getting up to speed and switching over to the new UI system. It’s much better for performance, as it should be being built into the engine, but the feature set isn’t as comprehensive as our old UI system. So a fair amount of time had to be invested creating controls and widgets for the new UI that we used previously. A lot of time was also spent making sure it was setup in a manner that’s nice and scalable for the future. If anything needs to change or be added, it’s a simple matter of dragging a new control or group in and it will automatically layout everything correctly. So as of 0.04.E1, the main menu is running on the new UI. All of the in-game UI is yet to be converted.

You’ll see there’s a few pre-emptive settings in the audio menu. These will be implemented when we take another pass at the audio system. Unity 5’s got a few new audio tools we need to check out.

We’ve just started to integrate Speedtree for better trees and foliage. One of the things that’s not immediately apparent about the Speedtree integration in Unity is that there’s no official physically based shader set for them. So you have a great scene full of shiny PBR materials and trees that stand out like you know what. All the trees and foliage look horrible right now. We’re on it and trying to get it sorted out asap.

All in all, we’re seeing a nice performance boost with the upgrade. We’ve moved to a fully deferred renderer now. Garbage savings from new UI system and the new multi-core physics system is helping smooth things out too. Freeing up all these precious milliseconds mean we can spend them on more important things like effects, more complex AI, etc.

We’re also finalising plans for the new terrain system and world. We’re really excited with where it’s all heading. It’s also going to be perfect for community mods. We’ll be filling you in in a future post once we’ve got it all figured out.

So that’s it this week, till next time!
Stay Alive!
:: Beam Team


Tagged ,


As mentioned in our last update post, we’re going to be running a new ‘experimental’ build alongside the stable version of Stranded Deep very soon. We’ll keep you posted when the first experimental build goes live (hopefully next week sometime) and we’ll post some instructions on how to opt in. But for now, we thought we’d update you with some details of the immediate plans.

The first thing on the experimental agenda is moving over to Unity 5.

One of the biggest sales pitches of Unity 5 is the new fully deferred renderer and physically based shading. The switch to a proper deferred renderer will be great, but unfortunately not as easy as pressing the “upgrade” button. There’s a lot of custom stuff going on in Stranded Deep which was necessary to get the most out of the engine we’ve been working with. If you’ve been following our blog posts for a while, you’ll know that we are already doing physically based shading along with various other effects like faking skybox global illumination and a lot of different depth techniques for our effects. So in visual terms, the engine upgrade won’t really do that much for now. But hopefully later on we can integrate even more graphical features like dynamic global illumination, screen space reflections and such.

There’s a few new features that should improve performance though. The physics engine has been upgraded both in speed and the quantity of objects it can handle. Stranded Deep uses a lot of physics so this should be a great improvement. There’s also the new UI system that we’ll be transitioning over to. We haven’t had much of a chance to look at it, but hopefully there’s some performance to be gained. The current system we’re using has some annoying workarounds and a fairly significant performance overhead. Every bit counts 🙂

After we’ve ported over, we’ll get back into the guts of the game. The construction system was in the middle of a rework before the last hotfix so that’ll be high on the list. We’ve got a lot of great ideas for new systems and they all centre around the theme of scavenging. We really want to incorporate more scavenging for more unique structures and tools. We’ll fill you in with more details when we get to these, but here’s a few teasers of some concepts —



There’s a heap of other systems that will be getting re-worked as well and we’ll keep you informed as it happens. Starting from this post, we’re going back to fortnightly blog updates as mentioned in the last update. If you’re not keen on trying out the experimental build first hand, this will be a good way to keep up with the progress of the game and see what we’re working on.

That’s it from us for now. We’ll have a hotfix coming out soon for the stable build and then stay posted for further info on the new experimental builds.

Stay Alive!
:: Beam Team

Tagged , , , ,