0 FPS

In my spare time I’ve been messing around a bit with node.js, and have been trying out building a browser based multiplayer online game.  Right now I am using sprites/minecraft style levels since they are easy to make and low bandwidth; but ultimately I would like to try out doing some more stuff with meshes.  You can give it a try if you have a browser which supports WebGL.   Here is a screen shot:

The URL for the game is http://mmotest.nodester.com .  Right now I am using the freebie hosting coupon for nodester, but I have no idea how much longer it will last.  The database hosting is from mongolabs, and so again is fairly limited.  However, this does show that you can build something like a minecraft game in the browser pretty easily.  Of course the latency with a shared hosting provider like nodester starts to become a bigger problem as you get more players in the game.  I will try to post the source code once I get a chance to clean some things up.  I am having some issues configuring the nodester deployment so that I don’t have to hardcode in various strings like the database password.  Unfortunately, for the life of me I can’t figure out how to pass those sort of config things in on the command line via nodesters REST API, and so for now they are just stored in version control.  Once I can figure out how to remove this information, I will make the github repository public so that others can try it out.  On the other hand, if I can’t solve the problem I may try just do something irresponsible like release a repository with all the passwords in it and let everyone go crazy hacking my demo accounts.

Well the results for Ludum Dare 21 are in!  Here is the final rank for my entry:

Ratings

I have to say that I am a bit disappointed about only getting 4th place for innovation .  (At least I got in the top 10 in that category, which is somewhat respectable given the huge number of entries!)  I think that the controls definitely could have been done a bit better.  In retrospect it was a mistake to use a separate parallel transport frame for moving the camera and for applying forces.  This caused the camera to drift slightly which made the controls more confusing than they needed to be (plus it would have been a really easy fix had I thought of it during the competition).  I’m not sure if it is worth developing this concept into a full game.  Judging by the rating, it didn’t seem like it was that popular, but perhaps I am reading too much into the scores.  Also I think that the screen shot I picked for the entry was not nearly exciting enough (the wood level would have been much better), and the some of the levels definitely needed more polish.

At least there is a mini-LD coming up next weekend, so hopefully that will go a bit better!

Haven’t updated this blog in a long time, but I figure that since it is summer now maybe I will finally be able to keep this thing regularly updated (see how long that lasts…)  Anyway, some good news is that my latest paper has been accepted to the SIAM/SIGGRAPH conference on solid and physical modeling.  Here is a link to the download:

http://sal-cnc.me.wisc.edu/index.php?option=com_remository&Itemid=143&func=fileinfo&id=183

There’s a bunch of neat ideas in here, but the big idea here is the Minkowski product.  The motivation for this comes from the basic Minkowski sum.  If we recall, for two sets , their Minkowski sum is defined as:

Or alternatively,

The idea behind the Minkowski product is to replace with an arbitrary group, .  If we do this, then we can define the Minkowski product over in the following way:

Much like how the Minkowski sum is useful for collision detection of translating objects, the generalized Minkowski product can be used for collision detection between translating and rotating bodies.  Similarly, we can also define a Minkowski quotient which is analogous to the Minkowski difference:

And even better yet, we show how to compute these quantities using convolution algebras!  These operators turn out to be very useful in understanding things like partial symmetries of solids, mechanism workspaces and robotics.  All of the gory details and more are in the paper!

Hi.  My name is Mikola Lysenko and this is my first post on 0FPS. I am currently a first year graduate student at the University of Wisconsin-Madison, working in the spatial automation lab.  I used to run the website assertfalse.com, but I ended up taking it down to save money on bandwidth expenses.  Anyway, I figure that lately I’ve been too lazy in my writing process and that a blog is a pretty good way to bring discipline back into my life.  I intend to focus on research topics in solid modeling and computer graphics, as well as other ideas I find interesting.