JeremyBlum.com

I’m back at home for winter break now, so I of course had to go visit DeepNote!  There have been a bunch of new really tough songs released for GH3, so we figured we’d try playing them.  The video below speaks for itself, but basically, DeepNote destroyed them.  We also had it play Guitar Hero 4: World Tour.  It performed quite well, but will need some programming adjustments before it is as good at it as it is at Guitar Hero 3.

This is the first video I have filmed with my new Canon Vixia HF11 HD Camcorder.  The video below is in standard quality for the sake of your bandwidth, but go check it out in glorious HD on youtube to get the full effect.

We didn’t win!  In case you haven’t figured it out, I’m talking about the Parallax Propeller Design Competition that we entered the DeepNote Guitar Hero Bot into.  That can probably be attributed the fact that 15% of the score was based on the practicality of the project.  We probably got a 0% for that…  Oh well, we still got Special Recognition from Parallax and a place on their website.  You can go check out the other project entries here; there are some extremely well done entries.  Here is our humble little page on the Parallax Site.

In related news, I’ve released all the source code, schematics, parts lists, etc. to the DeepNote “Build Your Own” page.  We don’t have any plans to release a full tutorial on how to build a bot, but with the information provided it shouldn’t be too hard to manage.  In fact, you won’t even have to program the darn thing because we’ve provided the source code!

We have successfully submitted the information for our DeepNote Guitar Hero Bot to the Parallax Propeller (the microcontroller we used) Design Contest.  The contest judges entrants on the following criterion:

• 30% Propeller Design Appropriateness Determined on applicant’s ability to use the Propeller’s unique architecture to achieve a working design. An entry which receives the highest mark in this category will utilize the Propeller to achieve a design not possible with a traditional microcontroller.

• 30% Originality Creativity of the project.

• 25% Professionalism This is based on the use of clear English in the Project Report, a quality design (if needed), clarity of explanations, quality of pictures and source code formatting/commenting.

15% Practicality Based on the project’s usefulness.

Check out all the competition information HERE

Yup, we’ve been posted on engadget yet again!  Go check it out:

http://www.engadget.com/2008/08/07/deepnote-guitar-hero-bot-reveals-its-secrets-on-video/

PS:  We’ve been working on the bot tutorial, and we hope to have it up in the near future.

I’m excited to announce the release of the DeepNote Walkthrough video.  It is a full, detailed explanation of how the bot works.  It will be submitted along with our contest entry to parallax to compete in the upcoming propeller design challenge.  Check it out below:

DeepNote managed to make it onto engadget, a popular tech blog, today:
http://www.engadget.com/2008/07/14/deepnote-guitar-hero-bot-watches-the-game-rocks-you-within-an/

Since Guitar Hero Aerosmith came out, we figured we’d give our bot a crack at it. It managed to play it as well as, if not better than the PC version of Guitar Hero 3, without any modifications at all. That goes to show how easily it can adapt. Here’s some proof.

We are also trying to get a live stream of the bot going at ustream.tv. So, go check it out, and hopefully we should have it running soon. :)

Everybody’s been asking for a video of it, so here it is.  Enjoy:

Oops, I forget to charge my camera battery, so no pictures today we managed to get some the next day (posted below), but here’s what we’ve done recently.

We completely finshed the metal rack that will hold the sensors, and accompanying tripod mount for it.  Using module components, we are able to construct it in such away that sensors can be replaced without unsoldering anything, and using clip-in racks, the positioning of the sensors can be easily adjusted based on screen size.  Our tripod mount is also adjustable, allowing the tripod to stand close to, or far away from the TV using an adjustable arm.  This is beneficial, if for example, the TV is in a cabinet.   There are also felt bumpers on the front of the bar to prevent damage to the television.

Thanks to a suggestion and some help from youtube user al337h4x, we modified the game textures to remove lightning!  That completely fixes the problem we were having with the sensors activating when lightning flashed past them.  This should be good for allowing us to get some high scores, but DeepNote will of course continue to work on standard, unmodified Guitar Hero games.

Enjoy some pictures of the finished sensor rack!

view slideshow

 

As we near the finish line with our Guitar Hero Bot, we are pleased to announce its official name and website!

Drumroll please…..

The bot will be called: The DeepNote™ Guitar Hero Bot

And its official website is: MechanizedRock.com

Go check it out! (it explains how we chose DeepNote).