If you go to DEFCON next year (and you should), prepare for extreme sleep deprivation. If you’re not sleep deprived you’re doing it wrong. This was the state in which we ran into [LosT] and [J0nnyM@c], the brains behind the DEFCON 22 badge and all of the twisted tricks that torture people trying to solve the badge throughout the weekend. They were popular guys but wait around until late into the night and the throngs of hint-seekers subside just a bit.
Plans, within plans, within plans are included in the “crypto” which [LosT] talks about in the interview above. We were wondering how hard it is to produce a badge that is not only electrically perfect, but follows the planned challenge to a ‘T’. This includes things like holding off soldering mask from some pads, and different ones on a different version of the badge. Turns out that you just do as well as you can and then alter the puzzle to match the hardware.
Speaking of hardware. A late snafu in the production threw the two into a frenzy of redesign. Unable to use the planned chip architecture, [J0nnyM@c] stepped up to transition the badges over to Propeller P8X32a chips, leveraging a relationship with Parallax to ensure they hardware could be manufactured in time for the conference.
If you haven’t put it together yet, this is that same chip that Parallax just made Open Source. The announcement was timed to coincide with DEFCON.
16 thoughts on “DEFCON 22: The Badge Designers”
What would have been really neat (and prohibitively expensive) would have been to use an FPGA with the propeller verilog onboard rather than an actual propeller chip. Good luck cracking the badge when you dont have a clue whats inside it.
If it’s traditional, then it’s just intercepting the bitstream (unless encrypted) and reversing the RTL. That may take more time than they’d have at the conference, however. Otherwise, a COB Microsemi Igloo2 in epoxy blob encapsulation would really drive people nuts.
> That may take more time than they’d have at the conference
I’m guessing you haven’t been to Defcon. The main contest should take weeks to solve, somehow people do it in a few days.
Here is a write up of last years solution: http://elegin.com/dc21/
Can’t you find quieter room to record interviews. All tha background noise is really annoying.
Ha! Not at DEFCON with busy people. The closest quiet places (relatively) are the rooms and those are a decent trip from the convention floor.
It’s not the room, it’s the audio gear. They need to upgrade to better microphones and a recorder where you can set audio gain manually. I have recorded interviews in a crowd and the audio came out great. You need short shotgun mics and the proper gear to get good audio.
There is a lot of background noise but the foreground is very clear… They actually have what sounds like very good equipment, keeping costs and the environment in mind.
I’m just happy to see any actual interviews from a conference. For a first pass this is great to see.
Attend more conferences HAD!
We’re working on it.
What was the planned chip architecture? Or was this fact mentioned off-camera? I didn’t hear anything about another architecture in the video.
I’d like to know this too. All I know is from the 2013 opening ceremony Joe said it was going to be different from the previous year, which was already the Propeller. Rumor was it would be the Propeller 2.
I am not at liberty to tell you which processor Ryan used in the original design, but it was not a Parallax processor — it was from another company. That company’s actions (pulling out and rescinding their donation at the last minute) created a huge headache for Ryan and DEF CON. He called me because I’m a very close friend and would not mind giving up a few nights sleep to help him with a new badge design.
It must be stated that Parallax actually helped save the day. Ryan and I worked with their engineers to get the badge designed and into production on a ridiculous schedule. The P2 is not ready yet, so we stuck with the P1. Yes, it was used before, but with new x-platform tools for Spin/PASM and C, as well as the Verilog release of the P1, it seemed worth using it again. This year’s contest required badge reprogramming, and we decided that making the badge into a shield-compatible development tool might be an idea worth doing. It seems to have worked out. That voice I have in the video is not normal; I nearly lost it talking with so many people. Programming the Propeller is fun, and a lot more people know that now.
Everything is open source. You can get the badge design and code files, as well as my IronMac shield design and code files in the Parallax forums.
Sounds like Parallax really came through. And agreed, the new xplat tools did provide a new sense of accessibility to the uc. I had fun tweaking mine with Propeller IDE.
While the Defcon badge is largely meant to facilitate the badge puzzle, it would be nice if future electronic badges could emphasize a socialization aspect, perhaps with wireless games like the Defcon 18 Ninja badge. The 128×32 display from the Defcon 18 badge could probably pull it off.
Believe me, there was a lot of “Can we do this? Can we do that?” questions during the first conversations. It came down to having ONE SHOT at producing the badge (which actually has about 13 variations), hence we were forced to use proven circuits and code.
Sounds like I know what company. Could be wrong, but there’s one company I can think of that loves to do things like that… ah well, I understand, I’ll just imagine it is that one :)
Apart from the LEDs shearing off on various things I did like the badge design this year, specifically the I/O expansion capability – great work.
Bleh, meant Ryan, not Joe.
Please be kind and respectful to help make the comments section excellent. (Comment Policy)