May 11, 2006

Blog Team Collaboration

Brian, Matthias, and I are having some fun. Taking the RaSPy, we are going to "attempt" the following:

1. Take the prototype bot and improve on its design (Jim)
2. Take the original code and improve it (Brian - already done a couple - see earlier post).
3. Develop an LDRAW file of the final design (Matthias)

The fun part is collaborating on a project where the 3 of us aren't in the same geographical location. Using email and this blog, we hope to demonstrate some good teamwork by distributing the duties. Hopefully, once the NXT is released and more people have access to the parts and software, we'll start seeing more collaborations...

Jim

5 comments:

byronczimmer said...

I'd be interested in knowing the actual randomness of the random function.

A simple project would be to create X storage variables, initizalize them to zero and then run through some very large number of loops, which generate a random number and increment that variable's value.

I'd be further interested to know if the generated curves are identical after resetting the NXT and rerunning the program.

It's a simple program -- indeed no lego building required, just a quick loop and some display at the end (or message the results via bluetooth to your phone).

Any takers?

byronczimmer said...

I had a design recommendation for you too...

Use the ultrasonic sensor to sense someone doing the hand throw motion in front of RaSPy.

Jim Kelly said...

Ultrasonic Sensor will be implemented in version 3.0 ;)

I've got to get some 'real work' done today :)

Jim

Brian Davis said...

Another very simple way to test the Random function is to generate two random numbers and then plot a point at that position on the LCD, and repeat. For a "perfect" random number generator, the screen should fill uniformly. The NXT in my testing doesn't *quite* do this, but it comes very close - certainly close enough for most purposes.

As to the related question, of if the Random function repeats each time the NXT is turned on (or in other words, how is the random number generator "seeded"), I do not have the answer yet. There is no plug on the block to wire in a seed value, so at the very least it's not something the user has a lot of control over.

--
Brian Davis

byronczimmer said...

Does the NXT know absolute time (or at least whatever you've told it is 'now')?

With time as the random seed, it should seem random enough to us mere humans, assuming they use enough of a range.

Related Posts Plugin for WordPress, Blogger...