[pfisch]

No. I don't agree with this at all. We had like 160k with kickstarter for Road Redemption and have made a game on early access that has sold enough to easily continue development.

The problem 100% was that Neal Stephenson was incapable of doing the project himself and was just hiring devs to do it and then ran out of money. It was poorly managed because they had lots of traction and risked basically no capital on development. If you give me half a million dollars I would have no problem getting that project to an early access state. Honestly I could do it with half that.

I can say that because I actually know how to make both the hardware and software side by myself if it came to that.

[Cushman]

This wasn't a game, it was an engine. A novel motion control scheme, more satisfying to wield than any one currently extant, and a realistic and historically accurate simulation which supports different sword fighting styles. I know a thing or two about motion controls, and game development, and sword fighting for that matter, and I am dramatically less convinced than you that this is trivial.

(Even if you are able to donate a year of developer time, valued in the six figures, which in fairness to your point is not something Stephenson brought to the table along with his relatively modest personal assets. Semi-famous authors aren't as rich as people think, either.)

I'm not saying I'm impressed with what they made, but I'm not too surprised either.

[pfisch]

I don't really think it was an engine so much as a peripheral with some mediating libraries. I would do that plus write a C# interop to make it compatible with unity so you could make your die by the sword clone quickly.

What they were trying to make was basically a better wiimote, and a wiimote is essentially an arduino with a single accelerometer hooked up to it and a bluetooth receiver. I think without the budget restrictions of a wiimote you could probably use better/more responsive parts and then basically just mimic a wiimote with the wiimotion plus(which is just a single gyroscope) and call it a day. Or you could add several of each of those components and then average their output or do other clever math with their separate outputs. I think this would take a little more than a year if the person who was doing it knew what they were doing.

Then you need to make the game/demo which you would do in parallel with a team of 3-4 additional people.

"Even if you are able to donate a year of developer time, valued in the six figures"

See this is where misunderstandings with kickstarter begin. We assumed almost no risk in making Road Redemption. In addition to salary we get 100% of revenue less distribution. So not only did we get to draw salary during development, now we get all the sales revenue and we own the IP. It is a crazy good deal, nobody needs to donate anything.

It is basically like you get a bunch of VC money and also the VC's have no equity interest.

The problem comes in when someone who is unqualified tries to middleman the operation and takes a bunch of money but is totally incapable of doing what they said they would do. Now they both don't have enough money to actually get the shit done at cost, and don't want to give away all the equity. The biggest problem though is that they are unqualified to discern who is capable of doing the task because they don't know anything about how to actually do things. So they fritter away the money and then the project falls apart.

[Cushman]

Have you written software for motion controls before? Like not gestures, but actual physical motion? I expect they quickly found out that Wiimote+[0] is not good enough for swordfighting the way they wanted it. Nothing on the consumer market is. The Razer Hydra came closest, which is why the thing they released wound up targeting it, but even then it was hard to find and too unreliable in typical use for a fun experience. (Now that the STEM is finally shipping, this may have changed.)

It's a garbage-in, garbage-out scenario that is really, really hard to solve in software. And come on, why am I explaining this to you? Why are there zero compelling or realistic Wii/U sword games?

It also doesn't sound like you've put any thought into the challenge of simulating realistic swordplay as a generic engine that supports different historically accurate fighting styles, with all the nitty physics, kinematics, and for that matter historical issues to be dealt with.

But I don't know why I'm having to defend the concept that software costs money. I think I'll just quote Neal's apology:

"Members of the team made large personal contributions of time and money to the project before, during, and after the Kickstarter phase. Some members, when all is said and done, absorbed significant financial losses. I am one of them; that has been my way of taking responsibility for this. The team had considerable incentives--emotional and financial--to see CLANG move on to the next round of funding. They showed intense dedication and dogged focus that I think most of our backers would find moving if the whole story were told. I will forever be grateful to them. In the end, however, additional fundraising efforts failed and forced the team to cut their losses and disband in search of steady work." https://www.kickstarter.com/projects/260688528/clang/posts/9...

These were intelligent people, working with passion, on something hard. I'm glad your Kickstarter was a success, but maybe a little humility is called for.

[0] As in accelerometer+gyro, in general. It's fundamentally flawed, and you can't buy your way out of it with better parts.

[pfisch]

Wait this is even worse. They didn't even build any hardware at all. So just to be clear in less then a year they spent half a million and used other peoples hardware and didn't even have a real gameplay demo.

"It also doesn't sound like you've put any thought into the challenge of simulating realistic swordplay as a generic engine that supports different historically accurate fighting styles, with all the nitty physics, kinematics, and for that matter historical issues to be dealt with."

You are controlling the sword so I'm not really sure what you mean. You need to mocap a bunch of moves from someone fighting in a historic style for the AI and then have the AI randomly jump between a half dozen attack sequences. Before our kickstarter we had a great animator and a bunch of playstation eyes and were able to get good motorcycle fighting animations for a price in the thousands.

I'm sorry but this project was obviously mismanaged. The extreme detail that went into designing the arenas seemed to consume all the resources of the project. Look at the videos of what they had. It is like 100% artistic detail and 0% gameplay.

[Cushman]

Sorry, I don't know a lot about these things but it's clear to me now I know more than you do. Its really unkind to speak with presumed expertise about things you're not expert in, and accordingly I'm not interested in talking about this with you any more.

Congratulations on shipping a game. I mean it, that's real. I haven't done that. But stay humble.

[pfisch]

Just to be clear. I do think the accelerometer/hardware part would be very difficult. I also think the software side of that would be super hard which is why I'm guessing it would take over a year once you have the hardware. In fact because it is so hard I think the first way I would approach a solution would be with an ensemble machine learning solution leaning on random forests because they are quite fast and Random Forests is the ML algo I have had the most success with in the past, and then try and post process that. If that worked I would probably have users individually generate their forests during a calibration when the game starts.

I have 2 more ideas if that doesn't work well. Though I would probably end up combining all the ideas in another weighted ensemble on top.

The game part of this however is not really that complex and could be done on a regular schedule.

[jessaustin]

The extreme detail that went into designing the arenas seemed to consume all the resources of the project.

Let me guess, every arena had a bike shed off to one side?

[pfisch]

Look at the arenas, they are beautiful. They have cloth objects blowing in the wind. Everything looks like it was custom made from scratch.

"Let me guess, every arena had a bike shed off to one side?"

I'm guessing that was a shot at Road Redemption, and I do wish the art was better, but having a huge art team to make 50+ miles of track from scratch would cost a lot more than 150k, and leave no money for making the actual game. We used a lot of premade assets from turbosquid and the unity asset store. We used EZRoads to make all the tracks except the rooftop levels. Most of the animations we had to mocap/animate ourselves though because there just are not a lot of solid on motorcycle animations available.

The point is that we actually made a game and it is pretty fun(90% user review score), and we did it with a lot less than 500k. Art does not make a video game. That is why I am saying the clang project was mismanaged. The art they made is beautiful, and looks like it is all custom. They could've easily gotten 90% of that from the asset store and it would be worse, but it would also cost less than $1000. Then they have at least 400k left to make the game. The problem they had to solve was difficult, but with 400k they should've been able to make something awesome.