Share this project

Done

Share this project

Done
We want to put 3D game development on Linux, so you can build and play games without leaving the Linux operating system.
We want to put 3D game development on Linux, so you can build and play games without leaving the Linux operating system.
787 backers pledged $42,358 to help bring this project to life.

You did it! Native game development is coming to Linux

In just three weeks, the Linux community has successfully funded the development of Leadwerks for Linux.  This means we're going to bring Leadwerks 3.1 to Linux, with native support for developing Linux games...so Linux games can now be completely free from Windows.

Greenlight

It's been an amazing few weeks.  During this time, we also successfully completed our Greenlight campaign to make Leadwerks available on Steam and take advantage of features like the Steam Workshop.  You can see from the graph below that our campaign did better than any other software in Steam we had data for.  The votes look like they would have gone a lot higher, but Valve approved us early!

Megatextures

I also had a chance to prototype the major feature of Leadwerks 3.1 I was most concerned about.  I wanted to implement a new terrain system that would remove the limitations of our old one, and thought that a technique I call "dynamic megatextures" would be a good solution.  Basically, this works like id Software's megatexture technology, only the virtual textures are generated on-the-fly rather than being paged from the hard drive.  This means the entire terrain can be uniquely textured, but it doesn't require the hard drive space megatextures usually need:

Getting that knocked out of the way makes me confident we can deliver Leadwerks 3.1 for Linux according to our original vision.

Stretch Goals

Before we get too much further in the campaign, I'd like to talk about stretch goals.  These are an unofficial custom Kickstarter campaigns have started using to put extra funding to good use.  They work by saying "If we reach this level, we will deliver these extra things, if we reach the next level, we will deliver more".  So basically it's a linear list of items, proceeding from one to the next.

I'm not sure if that's a great idea for this campaign, because it means I have to choose a linear list of items.  If I pick A, B, and C, and people don't really care about B, we never get to do C.  I took a poll giving backers the chance to rate different stretch goals, but the responses are spread across a lot of different things.  Kickstarter is a fantastic tool for testing demand and seeing what people really want, so why not put that to use?  I want to propose an alternative approach for stretch goals:

What if I add a $100 reward for each of the top stretch goals people want, and pledges for that reward would be used to fund that item?  I can limit the number of pledges, so that if the maximum is reached, then we will deliver that item.  For example, one reward might look like this:

OCULUS RIFT SUPPORT (60 available): If all rewards are claimed, we will deliver support for the Oculus Rift virtual reality headset, running natively in Linux.

In other words, this would allow us to "multithread" the stretch goals, instead of pursuing them one at a time, and me having to guess the order to place them in.  I think this would also give us a clearer picture of the needs of the Linux community, because the stretch goals aren't dependent on one another.  What do you think of this idea?

And congratulations, Linux community!  I'm happy to make Linux a core part of our company's focus moving forward.

Comments

Only backers can post comments. Log In
    1. Liam Dawe on July 12, 2013

      Already nearly 3k over the goal, if you set the goal to 26000 for the OUYA support it would most likely hit it (60x100 basically what you said above only done "normally").

    2. DerRidda on July 10, 2013

      That indeed was a bizarre idea. I suggest that instead you put a new poll with all N possible stretch goals and than you have to rate each and everyone by importance between on 1 and N, least to most important and each number can only have one item assigned to it.
      This way we will see what people care most about.

    3. Josh Klint 2-time creator on July 10, 2013

      @Nicolas, yeah you're right, that wasn't a good idea.

    4. Nicolas Guarin on July 10, 2013

      I didn't really understand the idea with the stretch goals.

    5. Henrik Nordgren on July 10, 2013

      Great that Linux-support/development is coming. I've switched and it would be nice to get some games.

    6. Josh Klint 2-time creator on July 10, 2013

      Yeah, scratch that idea. We'll just use conventional stretch goals. :)

    7. ET3D on July 10, 2013

      If all you want is to choose priorities, just create a simpler poll. Have people pick one stretch goal they'd like to see implemented, and decide based on that. Best to have it after the Kickstarter ends, and then you can get all backers to vote and you can decide how much value you want to assign to backers at different reward levels.

      I'd suggest that as a first step you will simply list all the possible goals and estimates of how much is needed to add them. That would let people judge better what a goal is worth. Perhaps they'll decide to vote for a simpler goal which requires less money.

    8. meganothing dread bard of torment BOSB on July 10, 2013

      Two disadvantages to your idea with the parallel stretch goals:
      1) Lets say all rewards need 60 supporters. Now what do you do if each stretch goal gets 50?
      2) the $100 pledges are very inflexible. You lose backers who want to spend only $50 or $10 and you won't get any $500 pledges this way

      I would suggest doing a real poll now with all features collected. If just your Pro Backers vote (as a Pro should definitely do ;-) it should be enough to get a visible trend

    9. ET3D on July 10, 2013

      Reading again, I might have misunderstood. It's possible that you meant that $100 backers will simply be able to choose one of the other $100 rewards. That is, they won't be adding more money, but what they choose will determine what's developed. Though this doesn't chime with using that reward's money to fund its development.

    10. Josh Klint 2-time creator on July 10, 2013

      If for example, ten people chose that reward, that would tell me it wasn't worth pursuing. They could change their reward before the end of the campaign, but maybe my idea is too confusing.

    11. Missing avatar

      Felix Bytow on July 10, 2013

      As for the stretch goals I'd say a hybrid approach might be best.
      Things like the Occulus Rift could be rewards, where other - more integral - features are normal stretch goals.

      Also the pure "stretch goals as rewards" has a problem for me and maybe for others as well:
      I could want a specific reward but just can't afford to pledge more :/

      Although I might have misunderstood you.
      "OCULUS RIFT SUPPORT (60 available): If all rewards are claimed, we will deliver support for the Oculus Rift virtual reality headset, running natively in Linux."
      This could also read as: if 60 people choose that reward, everyone will get it.
      Well that would also be great :-)

    12. ET3D on July 10, 2013

      (I meant to say you won't even get two features covered assuming 60 backers are needed for each. Plus drop the "something" in the last sentence.)

    13. ET3D on July 10, 2013

      Looks to me like $100 would be too much. You only have 116 backers who currently pledge $100 or more. For those who pledge $100 adding another $100 would double the expense, so I assume that many of them won't do that. Even if all of them did, with the current number you won't even get two features covered.

      I do think that the idea of making people pledge more for what they want is fine, and it's a way to get more money, but something like $10 would be a lot more reasonable, and even $50 backers might add that. It won't directly get you the amount of money needed to develop these features, but it will let you prioritise them.

      Couple that with estimates of how much each feature will cost, and you'll have something stretch goals but not linear.

    14. Oscar Norlander on July 10, 2013

      Congrats! This is awesome!