[00:00.000 --> 00:01.540]  What time do we have?
[00:01.600 --> 00:02.720]  Not even 20.
[00:03.240 --> 00:04.300]  This one?
[00:06.280 --> 00:08.900]  So that people know why I got slipped.
[00:14.600 --> 00:15.480]  Yes.
[00:19.920 --> 00:20.800]  Alright.
[00:21.440 --> 00:23.340]  Are you going up there to that microphone?
[00:23.340 --> 00:25.820]  Yeah, it's this one right here.
[00:25.820 --> 00:27.160]  Can you bend me the lighting?
[00:30.220 --> 00:32.400]  I will stand behind the podium.
[00:39.100 --> 00:42.280]  I don't want to wear this for an entire hour though.
[00:45.260 --> 00:47.340]  As the hour goes, it just sheds.
[00:52.840 --> 00:56.000]  I think it's hilarious.
[01:00.000 --> 01:01.380]  I'm recording right now.
[01:04.080 --> 01:20.160]  Alright, we are going to start here.
[01:20.560 --> 01:21.220]  Excellent.
[01:21.220 --> 01:23.580]  Welcome to CypherCon's badge panel.
[01:23.580 --> 01:25.740]  For some reason I got selected to moderate it.
[01:25.740 --> 01:26.940]  Can't imagine why.
[01:26.940 --> 01:29.640]  Thank you very much for that by the way.
[01:30.080 --> 01:33.880]  We have six wonderful badge making people up here.
[01:35.180 --> 01:39.800]  We are going to kick off a conversation about making badges.
[01:39.800 --> 01:41.280]  I got some questions off Twitter.
[01:41.280 --> 01:43.060]  Let me unlock this bad boy.
[01:45.320 --> 01:47.880]  We will start with some of the questions from Twitter.
[01:49.420 --> 01:50.840]  First question.
[01:51.080 --> 01:54.300]  Other than to save money, why don't badges have power switches?
[01:54.300 --> 01:55.700]  I like this question.
[01:55.700 --> 01:56.920]  Why don't they?
[01:57.100 --> 01:58.400]  Who wants to start?
[01:59.020 --> 02:00.000]  I will start.
[02:01.220 --> 02:02.040]  That's not fair.
[02:02.040 --> 02:02.760]  Yours does.
[02:02.760 --> 02:04.860]  It does have a switch.
[02:04.960 --> 02:07.080]  That was one of our biggest critiques last year.
[02:07.080 --> 02:08.660]  He had to remove the battery.
[02:08.840 --> 02:10.740]  Next year's will have a switch.
[02:10.780 --> 02:12.280]  That's a very good question.
[02:14.640 --> 02:17.340]  Speaking from experience on the first few that we did,
[02:17.340 --> 02:18.500]  we know better now,
[02:19.060 --> 02:21.620]  the power switch is kind of an afterthought.
[02:21.620 --> 02:23.700]  Why would anyone want to turn off your badge?
[02:23.700 --> 02:26.260]  It doesn't make sense.
[02:28.080 --> 02:29.720]  That's a lesson we learned also
[02:29.720 --> 02:31.180]  in our badges.
[02:32.160 --> 02:33.040]  Logistically,
[02:33.560 --> 02:36.300]  a lot of the badges have SMT switches.
[02:36.540 --> 02:38.020]  And those actually can get knocked off
[02:38.020 --> 02:38.820]  really easily.
[02:38.820 --> 02:41.400]  Kind of like some of the IRDA modules
[02:41.400 --> 02:43.980]  that you guys have had on our badge.
[02:44.720 --> 02:45.320]  They just,
[02:45.320 --> 02:46.660]  if somebody so much as
[02:47.180 --> 02:48.840]  knocks it the wrong way,
[02:48.840 --> 02:50.880]  then it lifts the pads
[02:50.880 --> 02:52.660]  and then you no longer have a working badge.
[02:52.940 --> 02:54.760]  I know some people don't include power switches
[02:54.760 --> 02:56.040]  because of that.
[02:56.920 --> 02:57.980]  Yeah, that's a pain.
[02:57.980 --> 03:01.120]  I will say you can sometimes use capacitive touch
[03:01.120 --> 03:02.640]  for switching, but we find that
[03:02.640 --> 03:04.680]  that will still drain the battery over time.
[03:05.960 --> 03:07.860]  I got one more bit for that one.
[03:08.000 --> 03:09.360]  Go for it.
[03:10.780 --> 03:12.640]  Actually on our badge,
[03:13.680 --> 03:15.060]  those that have had the badge
[03:15.060 --> 03:16.380]  or seen the badge, there's two buttons.
[03:16.380 --> 03:18.380]  On and then off and throw.
[03:18.380 --> 03:20.320]  On is actually a hardware on switch
[03:20.320 --> 03:23.320]  and off and throw is a software switch.
[03:23.440 --> 03:24.520]  So to get our
[03:24.520 --> 03:26.520]  standby current down to 6 microamps
[03:26.520 --> 03:28.320]  so that the battery didn't go dead
[03:28.320 --> 03:30.420]  in transportation to the con,
[03:30.420 --> 03:32.860]  the regulator gets bootstrapped
[03:32.860 --> 03:33.920]  when you push the on button
[03:33.920 --> 03:36.420]  to actually power it up and bring up the 3.3,
[03:36.420 --> 03:37.940]  at which point the micro takes over
[03:37.940 --> 03:40.220]  and keeps the badge on until you push off.
[03:42.280 --> 03:43.640]  That's cool as hell.
[03:43.640 --> 03:44.880]  Alright.
[03:46.700 --> 03:47.920]  There's another fun one
[03:47.920 --> 03:49.720]  also from Tim V.
[03:51.740 --> 03:53.620]  What do you think of the idea
[03:53.620 --> 03:55.580]  of including sound as a post-con option
[03:55.580 --> 03:57.300]  on a badge since having it live at the event
[03:57.300 --> 03:58.880]  would probably be disruptive?
[03:58.960 --> 04:01.320]  And by probably we mean definitely.
[04:02.300 --> 04:03.800]  Isn't that the point?
[04:04.140 --> 04:05.880]  Well that's why these are this bright.
[04:07.580 --> 04:10.680]  So ours do have a sound option.
[04:11.300 --> 04:12.400]  There are two
[04:13.480 --> 04:14.780]  solder points that you can
[04:14.780 --> 04:16.560]  solder a piezo or a speaker to
[04:16.560 --> 04:19.060]  and if you get your badge into the Morse code
[04:19.060 --> 04:20.800]  mode or to the DTMF
[04:21.880 --> 04:22.400]  payphone
[04:22.400 --> 04:24.660]  then you can actually make DTMF
[04:24.660 --> 04:26.260]  tones that you can use to phone freak
[04:26.260 --> 04:28.900]  or have the Morse code
[04:29.420 --> 04:30.680]  make sounds.
[04:30.680 --> 04:32.540]  So we're going to be releasing those
[04:32.540 --> 04:34.640]  instructions once we get some sleep.
[04:36.990 --> 04:38.450]  I know that feel.
[04:40.750 --> 04:41.670]  All right.
[04:41.670 --> 04:42.850]  This one should take a little bit longer
[04:42.850 --> 04:44.770]  for you guys. Describe your approach
[04:44.770 --> 04:46.610]  to choosing components.
[04:47.410 --> 04:48.770]  Randomly poking at digikey
[04:48.770 --> 04:50.170]  not being a good one.
[04:51.490 --> 04:53.010]  So if you were in my talk
[04:53.010 --> 04:54.450]  in the last hour
[04:55.850 --> 04:56.770]  one thing is
[04:56.770 --> 04:58.650]  mostly cost, right?
[04:58.650 --> 05:00.690]  And then supply. So how easily
[05:00.690 --> 05:02.630]  can we get our hands on it? How easily can we
[05:02.630 --> 05:04.470]  prototype with it? And then
[05:05.890 --> 05:06.750]  who are the
[05:06.750 --> 05:09.830]  vendors that actually sell it?
[05:11.910 --> 05:12.810]  For the ones
[05:12.810 --> 05:14.770]  that we've done, one thing
[05:14.770 --> 05:16.790]  is the more you do this
[05:16.790 --> 05:18.770]  the more you build up a library of stuff
[05:18.770 --> 05:20.550]  you've done before, stuff you know how to do
[05:20.550 --> 05:22.630]  and you always keep track of
[05:22.630 --> 05:24.370]  what was the best part for that.
[05:25.050 --> 05:26.730]  So there's
[05:26.730 --> 05:28.570]  also a lot of reuse.
[05:29.230 --> 05:30.710]  If you can't find
[05:30.710 --> 05:32.170]  ones significantly cheaper
[05:32.170 --> 05:34.610]  or some other good
[05:34.610 --> 05:35.450]  reason to change
[05:35.450 --> 05:38.610]  use a lot of the parts you've
[05:38.610 --> 05:39.730]  used before.
[05:40.630 --> 05:42.290]  And I would second that.
[05:42.350 --> 05:44.410]  For us a lot of it is
[05:44.410 --> 05:45.370]  need
[05:46.450 --> 05:47.490]  then cost
[05:48.490 --> 05:50.390]  and it really depends on your budget
[05:50.390 --> 05:52.490]  for the badge. If you have a nice
[05:52.490 --> 05:54.530]  budget then you can afford to
[05:54.530 --> 05:56.470]  have more protection on your circuit.
[05:56.710 --> 05:58.230]  If you have him designing it
[05:58.230 --> 05:59.910]  then a lot of it is over.
[06:01.190 --> 06:02.470]  Just kidding.
[06:03.450 --> 06:04.530]  No such thing.
[06:04.530 --> 06:07.030]  I don't believe you. I'll second that.
[06:11.090 --> 06:12.390]  No more on choosing
[06:12.390 --> 06:13.350]  components?
[06:13.990 --> 06:16.370]  How about what happens when you've picked it
[06:16.370 --> 06:18.670]  you've prototyped it, you go to do manufacturing
[06:18.670 --> 06:20.490]  and it's not available anymore.
[06:22.250 --> 06:24.090]  We didn't have that problem.
[06:24.090 --> 06:25.410]  We started out with
[06:26.330 --> 06:28.290]  a particular brand of IRDA module
[06:28.290 --> 06:30.530]  got four months
[06:30.530 --> 06:32.930]  into it and then realized
[06:32.930 --> 06:34.650]  that the initial
[06:34.650 --> 06:36.710]  modules we had worked great.
[06:36.710 --> 06:37.990]  We got another batch of them
[06:38.370 --> 06:40.410]  and they were sensitive to electrical noise.
[06:40.410 --> 06:42.910]  We were getting electrical interference
[06:42.910 --> 06:44.890]  on our IRDA modules
[06:44.890 --> 06:46.650]  which caused the whole networking stack
[06:46.650 --> 06:48.770]  just to completely stop.
[06:48.890 --> 06:50.550]  So thankfully we
[06:50.550 --> 06:53.110]  found it before we ordered all the parts for $400.
[06:53.250 --> 06:54.110]  But it still
[06:54.930 --> 06:56.370]  caused a rock hard spin
[06:56.370 --> 06:58.350]  because the pinout was different on the new module.
[06:58.350 --> 07:00.110]  The new module was slightly more expensive
[07:00.110 --> 07:02.130]  but there was no choice.
[07:02.130 --> 07:04.070]  It was either abandon that function
[07:04.070 --> 07:06.130]  or pick a different part.
[07:09.250 --> 07:10.510]  Actually we have an interesting story
[07:10.510 --> 07:12.090]  on this year's badge related to that.
[07:12.090 --> 07:14.390]  If you watch the Kickstarter video
[07:14.390 --> 07:16.850]  we mentioned the APA-102 LEDs.
[07:17.150 --> 07:18.310]  It's actually what's on this
[07:18.310 --> 07:20.370]  vendor badge here, but they look a lot
[07:20.370 --> 07:22.490]  nicer, they're a lot easier to program.
[07:22.630 --> 07:23.630]  What we learned was
[07:23.630 --> 07:26.450]  they don't support a very high temperature
[07:26.450 --> 07:28.230]  and they crack when they're being
[07:28.230 --> 07:30.390]  when they're being put to a reflow
[07:30.390 --> 07:32.170]  oven. And so the manufacturer
[07:32.170 --> 07:34.350]  we had putting this together, they were using
[07:34.350 --> 07:35.530]  lead-free solder
[07:36.110 --> 07:38.810]  and that was cracking all the LEDs.
[07:38.850 --> 07:40.570]  And so we're having to go back and use hot air
[07:40.570 --> 07:41.970]  and remove almost every LED
[07:41.970 --> 07:43.390]  and then hand solder them.
[07:43.430 --> 07:46.190]  So we've switched to the same LED we used last year.
[07:46.190 --> 07:47.590]  It's a little bit more resilient.
[07:48.010 --> 07:50.390]  And we're going to solder 6,000 of these
[07:50.390 --> 07:51.290]  by hand this year.
[07:51.530 --> 07:53.170]  So they don't go through the oven.
[07:53.930 --> 07:56.410]  I know those feels. When I was making those USB blinkies
[07:56.410 --> 07:58.510]  earlier today, I overheated one of them while soldering it on
[07:58.510 --> 08:00.270]  and it did not work anymore.
[08:00.690 --> 08:02.370]  And I guess another thing to add onto that
[08:02.370 --> 08:04.090]  is just size constraints.
[08:04.350 --> 08:06.430]  So for the badges for
[08:06.430 --> 08:08.350]  this conference, they were all hand
[08:08.350 --> 08:10.490]  placed by
[08:10.990 --> 08:12.990]  me, among others.
[08:13.230 --> 08:14.270]  But me.
[08:14.470 --> 08:16.370]  And so I had to decide
[08:16.370 --> 08:18.410]  are my eyes
[08:18.410 --> 08:20.390]  good enough to see what I'm putting?
[08:20.450 --> 08:22.470]  So for resistors
[08:22.470 --> 08:24.470]  and capacitors, we went no smaller than
[08:24.470 --> 08:25.730]  0603s.
[08:26.410 --> 08:28.190]  And we tried not to do any
[08:28.190 --> 08:30.890]  QFNs and stuck with QFPs,
[08:30.890 --> 08:32.530]  SOICs, that type
[08:32.530 --> 08:34.390]  of thing. Pins that I could see.
[08:36.810 --> 08:38.470]  I agree with that.
[08:38.470 --> 08:40.970]  I give Charles a dirty look if he puts any 0402s
[08:40.970 --> 08:41.810]  on there.
[08:42.650 --> 08:43.870]  Fly poop.
[08:44.010 --> 08:46.030]  I haven't done that in a long time.
[08:49.850 --> 08:52.150]  Come on. Live dangerously.
[08:52.170 --> 08:54.150]  010.05s.
[08:55.690 --> 08:57.130]  That's a thing?
[08:57.130 --> 08:58.570]  That is a thing.
[08:58.570 --> 09:00.030]  You don't have to deal with.
[09:01.790 --> 09:03.950]  I wouldn't even want to breathe near that.
[09:03.950 --> 09:05.850]  No, you sneeze. They're gone.
[09:08.050 --> 09:10.570]  All right. Let's start a flame war.
[09:10.570 --> 09:12.050]  Which protocol of LEDs
[09:12.050 --> 09:13.310]  do you prefer?
[09:19.190 --> 09:20.190]  So I
[09:20.190 --> 09:22.350]  definitely prefer the AP102s.
[09:22.350 --> 09:24.270]  And that's because you can use
[09:24.270 --> 09:26.190]  SPI hardware to talk to them, which means
[09:26.190 --> 09:28.110]  the core of your microcontroller
[09:28.110 --> 09:30.270]  doesn't have to babysit them. They can go off and do other
[09:30.270 --> 09:32.270]  things. In fact, another chip that
[09:32.270 --> 09:34.250]  I like and have been using a lot is
[09:34.250 --> 09:36.110]  the X-Mega series from
[09:36.110 --> 09:38.270]  Atmel, which is really cool
[09:38.270 --> 09:40.050]  because they have both hardware SPI
[09:40.050 --> 09:41.870]  and DMA. So
[09:41.870 --> 09:43.850]  you can set up your memory buffer
[09:43.850 --> 09:45.930]  and just point the
[09:45.930 --> 09:48.350]  SPI hardware at the memory buffer, say,
[09:48.350 --> 09:50.270]  dump all this out to the LEDs, tell me when you're
[09:50.270 --> 09:51.170]  done.
[09:52.570 --> 09:54.230]  Very convenient.
[09:54.730 --> 09:56.330]  We use just straight-up
[09:56.330 --> 09:57.210]  LEDs, so
[09:58.590 --> 09:59.870]  fortunately, the micro
[09:59.870 --> 10:02.350]  that we're using has the eight different
[10:02.350 --> 10:04.490]  cores. So I put together
[10:04.490 --> 10:06.170]  bitbangs, you know,
[10:06.170 --> 10:07.810]  live-driving all the
[10:07.810 --> 10:10.350]  charlieplexed matrix stuff.
[10:11.650 --> 10:12.410]  I think
[10:12.410 --> 10:14.230]  it would be really nice to be able to just map things
[10:14.230 --> 10:16.410]  like that. I like that.
[10:16.870 --> 10:18.110]  To continue, the
[10:18.110 --> 10:20.110]  the top
[10:22.590 --> 10:23.190]  end
[10:23.190 --> 10:24.130]  of that X-Mega series
[10:24.130 --> 10:26.110]  has eight
[10:26.110 --> 10:27.830]  hardware SPI
[10:29.030 --> 10:30.210]  and I think
[10:30.210 --> 10:32.250]  eight DMA channels. I'm not sure
[10:32.250 --> 10:34.350]  on that. I'd have to check the datasheet.
[10:34.450 --> 10:36.030]  So you could control eight
[10:36.030 --> 10:37.550]  different strips
[10:38.110 --> 10:39.610]  completely with hardware.
[10:40.890 --> 10:41.270]  So
[10:42.010 --> 10:43.810]  it all depends whether you
[10:43.810 --> 10:46.090]  you're doing a job that you want done
[10:46.090 --> 10:47.650]  and you want it done easy or
[10:47.650 --> 10:49.270]  if you're going for credit.
[10:49.470 --> 10:51.330]  If you're going for credit, it's the
[10:52.130 --> 10:52.730]  WS
[10:53.650 --> 10:54.850]  2012.
[10:55.150 --> 10:57.510]  So I have
[10:58.150 --> 10:59.990]  a project that's been cooking for
[10:59.990 --> 11:01.710]  oh man, it's probably three years now
[11:01.710 --> 11:03.910]  where someone gave me a whole
[11:03.910 --> 11:05.170]  bunch of those LEDs
[11:05.770 --> 11:08.130]  stuck them to a board to make a
[11:08.790 --> 11:10.090]  scrolling text display
[11:10.870 --> 11:13.310]  and I have a
[11:13.310 --> 11:14.030]  20 megahertz
[11:14.030 --> 11:15.750]  8-bit microcontroller
[11:15.750 --> 11:17.950]  that reads out of an SD card
[11:17.950 --> 11:20.050]  and spits it out to those strings.
[11:20.270 --> 11:22.030]  Which, if anyone's worked with that
[11:22.030 --> 11:23.870]  knows how tight that timing is
[11:23.870 --> 11:27.750]  and getting that all fitted into a
[11:27.750 --> 11:28.030]  20
[11:28.750 --> 11:30.470]  megahertz 8-bit part
[11:30.470 --> 11:34.070]  was a real challenge.
[11:34.070 --> 11:34.770]  I'll second
[11:34.770 --> 11:36.430]  the APA 102. Having that
[11:36.430 --> 11:39.430]  clock line is really nice. We went back
[11:39.430 --> 11:42.050]  to the WS2812s this year
[11:40.030 --> 11:42.030]  and we are actually using SPI
[11:42.050 --> 11:43.110]  to do it.
[11:43.910 --> 11:46.130]  We are using Nordic on our
[11:46.130 --> 11:48.150]  MCU and one of the challenges
[11:48.150 --> 11:50.290]  with Nordic is you can only send 255
[11:50.290 --> 11:52.110]  bytes at a time over DMA
[11:52.110 --> 11:53.910]  so the driver
[11:53.910 --> 11:56.750]  it works, but it will only work up to 20 LEDs.
[11:57.390 --> 11:57.490]  So
[11:58.330 --> 12:00.190]  somehow able to make it work through bit banging
[12:00.190 --> 12:02.050]  SPI with DMA.
[12:02.690 --> 12:03.630]  So with
[12:03.630 --> 12:05.070]  that 8-bit PIC
[12:05.070 --> 12:08.790]  I was able to do past that, but same thing.
[12:08.890 --> 12:10.070]  You get to a point
[12:10.070 --> 12:12.370]  in the flash or into the SD card
[12:12.370 --> 12:14.110]  and you get a page boundary.
[12:14.110 --> 12:16.410]  When you hit that, you have to readdress the card.
[12:16.410 --> 12:18.090]  So you have this big dead time before you
[12:18.090 --> 12:19.590]  get your next block of data.
[12:19.770 --> 12:21.930]  So to get around that, I ended up having to implement
[12:22.230 --> 12:23.910]  a circular buffer. So I would read
[12:23.910 --> 12:25.630]  ahead in the SD card.
[12:25.810 --> 12:27.930]  So every time the interrupt came around to go service
[12:27.930 --> 12:29.930]  the LEDs to keep their timing happy,
[12:29.930 --> 12:32.050]  there was always data there for it to do it.
[12:32.050 --> 12:33.810]  Otherwise, you get a delay
[12:33.810 --> 12:36.130]  all the LEDs would latch whatever data they had
[12:36.130 --> 12:37.930]  right now, even if it wasn't
[12:37.930 --> 12:38.990]  the correct data.
[12:40.490 --> 12:41.930]  Another reason I like the
[12:41.930 --> 12:43.710]  APA-102s is
[12:44.470 --> 12:45.950]  they can work at, you can
[12:45.950 --> 12:47.550]  send data up to, I think,
[12:47.550 --> 12:50.010]  32 MHz. I'm not really sure because it's not
[12:50.010 --> 12:51.650]  really been relevant to me yet.
[12:52.830 --> 12:53.910]  But you can send
[12:53.910 --> 12:56.050]  data to them a lot faster. Another project
[12:56.050 --> 12:57.530]  I've been working on recently
[12:57.530 --> 12:59.730]  is we're putting LED strips
[12:59.730 --> 13:01.410]  on a car.
[13:01.650 --> 13:03.990]  So it lights up and does all sorts of cool things
[13:03.990 --> 13:05.550]  and there are strips that are
[13:05.550 --> 13:07.150]  200 LEDs long
[13:07.150 --> 13:09.550]  and of course I wanted a reasonable refresh rate
[13:10.590 --> 13:11.550]  so that
[13:11.550 --> 13:13.810]  translates to having to
[13:14.730 --> 13:15.510]  send the data
[13:15.510 --> 13:16.990]  out very quickly.
[13:20.240 --> 13:21.660]  Side question.
[13:21.860 --> 13:23.480]  How masochistic do you have to be
[13:23.480 --> 13:25.080]  to get into badge design?
[13:28.750 --> 13:30.090]  You just need a
[13:30.090 --> 13:32.170]  pair of people that are like, hey,
[13:32.170 --> 13:34.170]  can you give us a hand with this? It'll be
[13:34.170 --> 13:36.130]  fun. That was a year
[13:36.130 --> 13:36.610]  ago.
[13:37.410 --> 13:38.610]  I remember getting
[13:38.610 --> 13:39.710]  that.
[13:40.510 --> 13:41.130]  So
[13:42.870 --> 13:44.590]  for this badge,
[13:44.590 --> 13:46.230]  for this conference,
[13:46.230 --> 13:48.630]  I started soldering in the
[13:48.630 --> 13:50.550]  beginning of February and it was
[13:50.550 --> 13:51.910]  about 6 hours every day
[13:52.550 --> 13:54.830]  and 16 hours on the weekend.
[13:54.890 --> 13:56.110]  No breaks.
[13:56.450 --> 13:57.130]  So
[13:58.270 --> 14:00.610]  if you're cool with that and you don't have a life
[14:00.610 --> 14:02.450]  otherwise, you're
[14:02.450 --> 14:05.070]  welcome to do that.
[14:07.110 --> 14:07.810]  Yeah,
[14:07.810 --> 14:09.110]  and these two have been programming
[14:09.910 --> 14:11.290]  this badge for
[14:11.290 --> 14:13.030]  about a year?
[14:13.310 --> 14:15.290]  Maybe? 8 months?
[14:15.290 --> 14:17.150]  6 months? Yeah, probably 6 months.
[14:17.150 --> 14:18.650]  6 months in earnest.
[14:18.650 --> 14:21.330]  We had all the circuit design before that.
[14:22.030 --> 14:22.810]  Yeah.
[14:22.810 --> 14:23.070]  So
[14:24.710 --> 14:27.130]  it really depends on your design.
[14:27.130 --> 14:29.010]  So last year's CypherCon badge was
[14:29.010 --> 14:30.990]  mostly solder work because there was
[14:30.990 --> 14:33.410]  no programming. So that was me.
[14:33.410 --> 14:35.370]  And I didn't have a stencil
[14:35.370 --> 14:37.070]  then, so I hand-pasted
[14:37.850 --> 14:39.270]  all of the
[14:40.090 --> 14:41.130]  pins.
[14:42.330 --> 14:43.470]  That took a long
[14:43.470 --> 14:45.490]  time, but I personally find it kind of
[14:45.490 --> 14:46.290]  zen.
[14:48.830 --> 14:49.350]  But
[14:49.350 --> 14:50.270]  yeah.
[14:51.350 --> 14:52.390]  Yeah.
[14:53.710 --> 14:55.690]  So very masochistic.
[14:55.950 --> 14:57.410]  So this year when we started
[14:57.410 --> 14:59.370]  she's like, I'll just hand-paste them. I'm like,
[14:59.370 --> 15:01.470]  no, get a stencil. I'll just hand-paste
[15:01.470 --> 15:03.530]  and get the stencil. She got the
[15:03.530 --> 15:04.850]  stencil, we set her up, and she's like,
[15:05.390 --> 15:07.450]  this is amazing!
[15:08.190 --> 15:09.350]  Well, and the brain board
[15:09.350 --> 15:10.950]  of this badge has like
[15:11.670 --> 15:13.910]  80 parts or something like that?
[15:13.910 --> 15:15.630]  And so I was like, yeah.
[15:15.630 --> 15:17.090]  I don't think I can
[15:17.090 --> 15:19.310]  hand-paste that. And thankfully, my hand
[15:19.310 --> 15:21.790]  started cramping actually with the tweezer action.
[15:22.330 --> 15:23.510]  So I'd have to kind of
[15:23.510 --> 15:25.090]  shake it out every now and then
[15:25.090 --> 15:27.870]  and make sure I wasn't having a heart attack or something.
[15:29.190 --> 15:30.190]  I feel you.
[15:30.190 --> 15:32.170]  Anytime anyone asks
[15:32.170 --> 15:34.270]  us if we have a pick and place, Charles
[15:34.270 --> 15:36.330]  just smiles and looks at me.
[15:36.910 --> 15:37.950]  Yep.
[15:38.510 --> 15:40.590]  I think I'll add something to that.
[15:40.770 --> 15:42.030]  If you're 14,
[15:42.030 --> 15:44.190]  go to school and have a job. It's very
[15:44.190 --> 15:46.190]  difficult. I'm just going to
[15:46.190 --> 15:47.510]  say that right now.
[15:51.560 --> 15:53.180]  It's a labor of love,
[15:53.180 --> 15:55.080]  but I think, like, we've had
[15:55.080 --> 15:56.900]  such great feedback from everybody
[15:56.900 --> 15:58.660]  and part of it is
[15:58.660 --> 16:01.860]  loving to see how people work with your badge.
[16:01.860 --> 16:03.520]  Even if someone just
[16:03.520 --> 16:04.700]  sits there and
[16:05.680 --> 16:08.520]  tries to concentrate on it, even for a little bit,
[16:08.520 --> 16:10.480]  for me, it's worth it.
[16:11.240 --> 16:11.880]  So there
[16:11.880 --> 16:13.820]  is that trade-off.
[16:18.900 --> 16:20.440]  So for me, last year,
[16:21.800 --> 16:22.680]  basically,
[16:22.680 --> 16:24.380]  kids go to bed, I get on my
[16:24.380 --> 16:26.340]  computer. Kids go to bed, I get
[16:26.340 --> 16:28.220]  on my computer. Weekends, they're
[16:28.220 --> 16:30.860]  napping, whatever, I'm on my computer coding,
[16:30.860 --> 16:32.500]  designing, soldering, whatever.
[16:33.360 --> 16:35.220]  Some of us within the community,
[16:35.220 --> 16:36.280]  there's a hashtag we use called
[16:36.280 --> 16:38.900]  Badgelife, because it is a lifestyle.
[16:38.940 --> 16:40.340]  It's something you choose.
[16:41.580 --> 16:42.520]  I didn't have
[16:42.520 --> 16:44.660]  anybody saying, hey, this sounds like a good idea.
[16:44.660 --> 16:46.600]  It was mostly me telling myself that, so
[16:46.600 --> 16:48.180]  I can blame myself.
[16:49.180 --> 16:50.580]  But that's, yeah, that's my
[16:50.580 --> 16:52.740]  experience. Masochism.
[16:52.760 --> 16:54.440]  Yes, and I love it.
[16:57.880 --> 16:58.460]  Alright, so
[16:58.460 --> 17:00.380]  one thing I've noticed, and this is based on
[17:00.380 --> 17:02.800]  another question from Twitter,
[17:02.800 --> 17:03.280]  is that
[17:04.380 --> 17:06.220]  your first badge tends to be
[17:06.220 --> 17:08.180]  somewhat static or animated, and your second
[17:08.180 --> 17:10.600]  or third badge starts to become interactive.
[17:10.600 --> 17:12.640]  So you guys want to tell stories about
[17:12.640 --> 17:14.500]  transitioning from static
[17:14.500 --> 17:16.200]  badges to interactive badges?
[17:17.520 --> 17:19.120]  Oh, yeah.
[17:19.600 --> 17:20.240]  So
[17:20.240 --> 17:22.420]  last year's badge here,
[17:23.040 --> 17:24.180]  it being the first
[17:24.180 --> 17:26.520]  year of the conference, it needed to be
[17:26.520 --> 17:27.580]  pretty light.
[17:29.040 --> 17:29.480]  And
[17:30.200 --> 17:31.680]  Michael had asked me if I
[17:31.680 --> 17:33.600]  had any ideas that would work out
[17:33.600 --> 17:35.520]  for making it cost
[17:35.520 --> 17:37.080]  effective for a first year.
[17:37.260 --> 17:39.600]  And I sort of explained an idea of using
[17:39.600 --> 17:41.400]  just some 7400 logic chips
[17:41.400 --> 17:44.040]  to implement an XOR stream cipher.
[17:44.440 --> 17:45.660]  And he said
[17:45.660 --> 17:47.960]  you're hired. And I said, that's not what I meant.
[17:50.520 --> 17:51.600]  And we were closer
[17:51.600 --> 17:53.460]  to when the con was supposed to happen,
[17:53.460 --> 17:55.160]  when that conversation occurred.
[17:55.160 --> 17:57.560]  So it did have to be really simple.
[17:57.620 --> 17:59.280]  But we didn't want people to think
[17:59.280 --> 18:00.780]  that's all we can do. So
[18:00.780 --> 18:03.160]  this year we just sort of dropped
[18:03.160 --> 18:05.200]  the bomb on the conference a little
[18:05.200 --> 18:06.500]  bit with this badge.
[18:06.840 --> 18:09.000]  Because we all come from software
[18:09.000 --> 18:10.880]  design backgrounds, so
[18:10.880 --> 18:13.060]  we can do all that high-level stuff.
[18:13.060 --> 18:15.240]  But I think next year we might go back to something a little
[18:15.240 --> 18:17.520]  more analog and whatnot.
[18:17.520 --> 18:19.080]  I think it's good to be able to just shift
[18:19.080 --> 18:21.220]  around and change, do different things every
[18:21.220 --> 18:21.880]  year.
[18:23.620 --> 18:24.820]  There is an urge
[18:24.820 --> 18:26.900]  to show off, isn't there? I like the
[18:26.900 --> 18:28.760]  three-dimensional part.
[18:28.760 --> 18:29.760]  Thank you.
[18:29.760 --> 18:32.900]  It came together at a sushi restaurant
[18:32.900 --> 18:34.980]  on a paper napkin.
[18:35.580 --> 18:36.960]  We were like, we need
[18:36.960 --> 18:39.720]  to do something having to do with fringe science.
[18:39.840 --> 18:40.920]  And cube the movie
[18:40.920 --> 18:42.700]  while very sadistic
[18:42.700 --> 18:45.560]  is pretty fringy.
[18:45.660 --> 18:46.980]  But wait, can we
[18:46.980 --> 18:48.160]  do a cube?
[18:48.720 --> 18:50.400]  Wait, can we do a cube?
[18:50.400 --> 18:52.400]  Do we want to multiply everything
[18:52.400 --> 18:54.180]  by six times?
[18:54.700 --> 18:56.760]  Next time I want to see a dodecahedron.
[18:57.340 --> 18:58.600]  I think we were talking
[18:58.600 --> 19:00.760]  about a buckyball or something.
[19:00.940 --> 19:02.100]  That's basically it.
[19:02.100 --> 19:02.860]  Yeah.
[19:03.980 --> 19:05.100]  Nope.
[19:08.200 --> 19:09.640]  So in our experience
[19:10.840 --> 19:12.120]  in going from
[19:12.120 --> 19:13.440]  static to
[19:14.260 --> 19:15.680]  interactive, I just want to say
[19:15.680 --> 19:18.160]  I am still convinced that capacitive touch
[19:18.160 --> 19:20.160]  sensors are black magic.
[19:20.520 --> 19:22.600]  They're awesome because they're cheap.
[19:23.180 --> 19:24.240]  You basically just need
[19:24.240 --> 19:26.160]  like a resistor and a
[19:26.820 --> 19:29.400]  copper core.
[19:29.400 --> 19:30.440]  I did
[19:30.440 --> 19:32.340]  some on the badge that Ben is wearing.
[19:32.340 --> 19:37.200]  The recharge badge.
[19:37.200 --> 19:38.360]  And there's
[19:38.360 --> 19:40.260]  two down in front that work
[19:40.260 --> 19:42.220]  beautifully. And then I
[19:42.220 --> 19:44.160]  tried to do like a d-pad thing
[19:44.980 --> 19:46.180]  somewhere else on the badge.
[19:46.180 --> 19:48.040]  And I can see
[19:48.040 --> 19:50.120]  when stuff touches it, but
[19:50.120 --> 19:52.080]  the capacitance like crosses
[19:52.080 --> 19:54.080]  over all the buttons. There's no way to get
[19:54.080 --> 19:56.140]  any reliable direction out
[19:56.140 --> 19:58.240]  of it. And then I did
[19:58.980 --> 20:00.300]  try to do the same thing
[20:00.300 --> 20:02.160]  on this. I just kind of threw together at the last
[20:02.160 --> 20:03.360]  minute and
[20:04.200 --> 20:04.880]  when I
[20:05.800 --> 20:08.220]  built it, put the code on it,
[20:08.220 --> 20:10.180]  I started trying to use the touch pad
[20:10.180 --> 20:12.600]  and just nothing happened.
[20:12.920 --> 20:14.740]  In the car on the way up here.
[20:14.740 --> 20:16.580]  So I'm convinced
[20:16.580 --> 20:18.620]  they're black magic. If you get it to work
[20:19.560 --> 20:20.460]  amazing,
[20:20.460 --> 20:22.140]  please tell us how.
[20:23.400 --> 20:24.720]  I think another thing
[20:24.720 --> 20:26.280]  with the badge interactivity
[20:26.280 --> 20:28.880]  thought is, so our first
[20:28.880 --> 20:31.200]  one, there's different levels of interactivity.
[20:31.200 --> 20:32.360]  There's one where
[20:32.360 --> 20:34.140]  it's between
[20:36.740 --> 20:38.480]  you and your badge
[20:38.480 --> 20:40.620]  with another person and their badge
[20:40.620 --> 20:42.220]  and helping each other figure out
[20:42.220 --> 20:43.560]  like the
[20:44.520 --> 20:46.260]  encryption method for last
[20:46.260 --> 20:48.240]  year's badge. So there's
[20:48.240 --> 20:50.280]  the collaboration portion of
[20:50.280 --> 20:52.020]  people working together to try to figure out
[20:52.020 --> 20:54.160]  puzzles that are dealt with
[20:54.160 --> 20:55.280]  with the badge.
[20:55.960 --> 20:58.240]  With our second badge
[20:58.240 --> 21:00.340]  that we did for another conference,
[21:00.340 --> 21:02.220]  we thought about not
[21:02.220 --> 21:04.280]  only how people could collaborate, but also
[21:04.280 --> 21:06.160]  just the
[21:06.160 --> 21:08.140]  physicalness of actually getting
[21:08.140 --> 21:10.520]  two badges together to interact
[21:10.520 --> 21:12.420]  with each other. And if you've
[21:12.420 --> 21:14.080]  come to our table, you'll see one of the
[21:14.080 --> 21:16.340]  DNA strand badges.
[21:17.620 --> 21:18.440]  And so we
[21:18.440 --> 21:20.620]  wanted something that was simple enough to understand
[21:20.620 --> 21:22.460]  so that people didn't have to
[21:22.460 --> 21:25.140]  be geniuses to figure out how it works.
[21:25.540 --> 21:27.180]  And from that,
[21:27.180 --> 21:28.920]  when we come to this badge,
[21:28.920 --> 21:30.400]  this actually also has different
[21:30.400 --> 21:32.100]  levels, I feel,
[21:32.100 --> 21:34.220]  where it can be
[21:34.220 --> 21:36.340]  as simple as helping each other figure out how
[21:36.340 --> 21:38.560]  not to suffocate, to
[21:39.200 --> 21:40.820]  how you can
[21:42.160 --> 21:42.720]  jump
[21:42.720 --> 21:44.980]  from your badge to another person's badge,
[21:44.980 --> 21:46.760]  and then to the point where
[21:46.760 --> 21:48.820]  multiple people in one person's
[21:48.820 --> 21:50.980]  badge can talk to each other.
[21:51.260 --> 21:52.820]  So there were different levels
[21:52.820 --> 21:54.740]  that we wanted to try to hit
[21:54.740 --> 21:56.200]  for different reactions
[21:57.240 --> 21:58.280]  so that when people
[21:58.980 --> 22:00.640]  got to that reach goal,
[22:00.640 --> 22:02.100]  they're like, dudes.
[22:02.460 --> 22:04.440]  And that was motivation
[22:04.440 --> 22:06.560]  for us to keep feature
[22:06.560 --> 22:07.840]  creeping in some of these
[22:09.460 --> 22:10.420]  different aspects
[22:10.420 --> 22:12.360]  of the badge. So hopefully
[22:12.360 --> 22:14.920]  you guys have gotten to relay with each other on the badge.
[22:14.920 --> 22:16.540]  If you haven't, it is
[22:16.540 --> 22:17.420]  possible.
[22:17.820 --> 22:20.540]  So speaking of feature creep,
[22:20.540 --> 22:22.580]  how do you decide what to
[22:22.580 --> 22:24.480]  keep and when to cut off
[22:24.480 --> 22:27.100]  new ideas and such like that?
[22:27.600 --> 22:28.620]  And tell us stories
[22:28.620 --> 22:30.400]  about when you came up with an idea
[22:30.400 --> 22:32.360]  that was just too good to not
[22:32.360 --> 22:34.520]  do and was worth staying
[22:34.520 --> 22:36.380]  up all night for a few weeks to do.
[22:36.420 --> 22:38.860]  It's Tuesday night. The con's on Thursday.
[22:43.240 --> 22:44.860]  Charles can't answer this question.
[22:44.860 --> 22:46.060]  He does not know.
[22:47.840 --> 22:49.960]  I was going to say, Tuesday night?
[22:50.060 --> 22:51.560]  That's your cut off?
[22:53.080 --> 22:55.040]  This one it was.
[22:56.240 --> 22:57.160]  I think
[22:58.220 --> 22:59.840]  last year we had one that
[23:01.840 --> 23:02.800]  they're delivering
[23:02.800 --> 23:04.320]  the badges to the con in
[23:04.320 --> 23:05.960]  12 hours.
[23:06.840 --> 23:07.960]  Yes.
[23:08.520 --> 23:09.980]  I remember that night.
[23:13.340 --> 23:14.700]  So anyone who's
[23:14.700 --> 23:16.540]  interacted with this one knows that
[23:16.540 --> 23:17.680]  we didn't
[23:18.820 --> 23:20.800]  stop at any point with our feature
[23:20.800 --> 23:22.800]  creep. We just kept going and going
[23:22.800 --> 23:23.760]  and going.
[23:24.420 --> 23:26.800]  It has two processors in it,
[23:26.800 --> 23:28.780]  which is reasons,
[23:28.780 --> 23:30.800]  right? The PIC hardware
[23:30.800 --> 23:32.960]  can do the IRDA stack
[23:32.960 --> 23:34.880]  and USB stack better
[23:34.880 --> 23:36.380]  than the PROP could have.
[23:36.420 --> 23:38.540]  But the PROP can do all sorts of neat things
[23:38.540 --> 23:40.980]  that the PIC wouldn't be very good at.
[23:40.980 --> 23:42.340]  So we just used both.
[23:42.820 --> 23:44.760]  The problem with the PROP
[23:44.760 --> 23:46.800]  is it only has 8k longs
[23:46.800 --> 23:48.340]  of programming space.
[23:48.380 --> 23:51.020]  And I needed to implement a
[23:51.020 --> 23:53.620]  timesharing operating system,
[23:53.620 --> 23:55.340]  an LED driver, and
[23:55.340 --> 23:57.080]  a text adventure, etc.
[23:57.080 --> 23:58.640]  So I kept running out of longs.
[23:58.640 --> 24:00.660]  Every two days,
[24:00.660 --> 24:02.560]  there would be like 40 longs left.
[24:02.560 --> 24:04.620]  And I would have to go back over the whole codebase
[24:05.380 --> 24:06.760]  and figure out ways
[24:06.760 --> 24:08.360]  to make it smaller and smaller and smaller
[24:08.360 --> 24:10.560]  so I could just have enough room to be able to
[24:10.560 --> 24:12.520]  implement the next feature that these guys
[24:12.520 --> 24:14.480]  were asking for without
[24:15.160 --> 24:16.540]  overflowing my stacks and
[24:16.540 --> 24:18.560]  the processor crashing and puking
[24:18.560 --> 24:20.380]  things all over the screen, etc.
[24:20.980 --> 24:22.180]  And then we found
[24:22.180 --> 24:24.460]  an idea that just let us go
[24:24.460 --> 24:26.040]  absolutely crazy.
[24:26.040 --> 24:28.100]  We took all of the strings
[24:28.100 --> 24:30.000]  in my source code and
[24:30.000 --> 24:32.720]  put them on the EEPROM in the PIC chip.
[24:32.960 --> 24:34.220]  Now I can just send
[24:34.740 --> 24:35.960]  a long to the PIC chip
[24:35.960 --> 24:37.960]  and the PIC chip loads up the string
[24:37.960 --> 24:39.660]  and sends it on to the user. So now I have
[24:39.660 --> 24:42.260]  8k longs of all code
[24:42.260 --> 24:43.320]  space.
[24:43.780 --> 24:46.040]  So we were able to do things like implement the payphone
[24:46.040 --> 24:48.360]  and implement frequency generation on the fly
[24:48.360 --> 24:49.160]  for all the
[24:50.120 --> 24:51.960]  blue box and red box tone
[24:51.960 --> 24:54.140]  freaking, you know, all that stuff.
[24:54.140 --> 24:56.240]  So we basically just never
[24:56.240 --> 24:58.160]  said no. We just kept going and
[24:58.160 --> 25:00.800]  going and figured out how to fit it in.
[25:02.380 --> 25:04.240]  That seems to be a theme. How about you guys
[25:04.240 --> 25:05.500]  with the vendor badge?
[25:06.860 --> 25:08.120]  So the way we picked
[25:08.120 --> 25:09.980]  features, we had maybe
[25:09.980 --> 25:12.020]  40 different ideas in our spreadsheet
[25:12.020 --> 25:14.220]  that the team contributed to.
[25:14.540 --> 25:15.980]  If you saw my talk, I mentioned
[25:16.480 --> 25:18.120]  FPGA. There was really no
[25:18.120 --> 25:20.760]  reason why other than to have an FPGA.
[25:21.280 --> 25:22.120]  So what we tried
[25:22.120 --> 25:23.840]  to do was just map,
[25:23.840 --> 25:25.740]  we have this cool hardware,
[25:25.740 --> 25:27.280]  what will it actually do?
[25:27.440 --> 25:29.180]  What's the person going to use it for?
[25:29.180 --> 25:30.800]  And then try to find commonality
[25:30.800 --> 25:33.280]  between those ideas.
[25:33.460 --> 25:35.780]  So one of the things we removed from this
[25:35.780 --> 25:37.380]  year's or this upcoming badge
[25:37.380 --> 25:39.200]  is accelerometer.
[25:39.580 --> 25:41.320]  We couldn't find anything to do with it
[25:41.320 --> 25:42.960]  other than to tilt the badge.
[25:42.960 --> 25:45.000]  So it was a huge waste of money
[25:45.000 --> 25:47.540]  just to put this QFN part on there.
[25:49.300 --> 25:51.180]  I've got to think of another one.
[25:51.640 --> 25:52.440]  As far as
[25:52.440 --> 25:54.200]  coding and feature creep,
[25:54.200 --> 25:56.420]  that is how it goes for us.
[25:56.420 --> 25:58.300]  The same thing as the toy makers.
[25:58.500 --> 26:00.340]  Our issue tracker this year
[26:00.340 --> 26:02.580]  already is up to 300 issues
[26:02.580 --> 26:04.200]  or 330 issues.
[26:04.200 --> 26:07.300]  That's well beyond anything we did last year.
[26:07.360 --> 26:08.140]  We're up to
[26:08.140 --> 26:10.400]  170k of flash used.
[26:10.400 --> 26:12.880]  We only used 108 last year.
[26:13.360 --> 26:14.440]  So we're
[26:14.440 --> 26:16.240]  just above everything
[26:16.240 --> 26:17.220]  we did.
[26:17.220 --> 26:18.380]  And I don't think we're going to stop
[26:18.380 --> 26:19.440]  until right before the con
[26:19.440 --> 26:21.060]  where we have to flash these things.
[26:21.260 --> 26:23.460]  The issue is really stability.
[26:23.460 --> 26:25.060]  So every time I introduce a new feature
[26:25.060 --> 26:27.540]  into the code, what am I breaking?
[26:27.540 --> 26:28.860]  I have to go back and regression test
[26:28.860 --> 26:30.600]  everything that's already in there.
[26:30.600 --> 26:32.140]  And where I may have added a timer
[26:32.140 --> 26:34.380]  over here to drive an LED,
[26:34.380 --> 26:36.180]  I just affected the display driver
[26:36.180 --> 26:38.880]  over here on some animation or some game.
[26:38.880 --> 26:40.760]  So it's really difficult to ferret
[26:40.760 --> 26:42.260]  all those bugs out.
[26:42.460 --> 26:44.880]  Especially when you have so many things going on at once.
[26:46.280 --> 26:46.980]  And so I'm
[26:46.980 --> 26:48.600]  an expert at the
[26:50.060 --> 26:50.940]  game inside
[26:50.940 --> 26:52.820]  the Texas Adventure game because I'm the
[26:52.820 --> 26:54.300]  alpha tester for them.
[26:54.560 --> 26:56.580]  So I just have to know, I have to
[26:56.580 --> 26:58.580]  first be able to test all the deaths,
[26:58.580 --> 27:00.820]  to make sure I can get through the system,
[27:00.820 --> 27:02.700]  that all the actions and the looks
[27:02.700 --> 27:04.100]  are correct.
[27:04.820 --> 27:06.660]  So that's probably why I think
[27:06.660 --> 27:07.700]  it's so easy.
[27:08.340 --> 27:09.460]  Yeah, so this year,
[27:09.460 --> 27:11.100]  I'll throw it out, so Hyren
[27:11.100 --> 27:13.240]  is one of our new team members.
[27:13.460 --> 27:15.140]  He's been furiously testing the badge.
[27:15.140 --> 27:17.860]  Every time I push something to get, he's flashing onto his,
[27:17.860 --> 27:19.560]  he's doing some sort of test, and he's dumping
[27:19.560 --> 27:21.560]  about ten different issues into the
[27:21.560 --> 27:23.680]  tracker. So I close one thing, he opens
[27:23.680 --> 27:25.960]  ten. And it's usually some sort of frustrating
[27:25.960 --> 27:27.600]  chat log between us going,
[27:27.600 --> 27:30.000]  what just happened and why did this happen?
[27:30.180 --> 27:31.780]  So, yeah, I don't really have the same
[27:31.780 --> 27:33.700]  experience, but Hyren does.
[27:35.700 --> 27:36.660]  Again,
[27:36.660 --> 27:37.940]  masochism.
[27:43.300 --> 27:44.680]  Alright, let's start another
[27:44.680 --> 27:46.820]  Flame War. Language
[27:46.820 --> 27:48.820]  to write your code in.
[27:52.360 --> 27:55.020]  So, one of our
[27:55.020 --> 27:56.840]  friends that we hang out
[27:56.840 --> 27:59.320]  in IRC with is a microchip rep.
[27:59.320 --> 28:00.800]  And these guys originally said
[28:00.800 --> 28:02.600]  we want to do USB
[28:02.600 --> 28:04.640]  serial and IRDA.
[28:04.640 --> 28:07.160]  So we need two different chips to do this.
[28:07.160 --> 28:08.560]  He's like, oh, you can just
[28:08.560 --> 28:09.860]  do that in a PIC.
[28:10.300 --> 28:12.360]  And that's where I got involved, because I've done
[28:13.100 --> 28:14.240]  a lot of PIC code.
[28:14.660 --> 28:16.420]  Unfortunately, with the time frame we had
[28:16.420 --> 28:18.620]  and the chip they picked, they forced
[28:18.620 --> 28:19.800]  me to use C.
[28:20.160 --> 28:22.560]  So I have a ton of experience
[28:22.560 --> 28:24.220]  in assembly, because my
[28:24.220 --> 28:26.360]  previous jobs, I had to do
[28:26.360 --> 28:27.780]  some really tight optimization
[28:29.220 --> 28:30.540]  for work.
[28:30.620 --> 28:32.340]  And that just, you know, I learned
[28:32.340 --> 28:34.260]  in assembly, they wanted it done
[28:34.260 --> 28:36.260]  in assembly, because that's the only way we could optimize
[28:36.260 --> 28:37.880]  for certain cases.
[28:38.260 --> 28:40.740]  And then these guys come along like, hey, can you do this?
[28:40.740 --> 28:42.500]  There's this whole code thing over here, you just paste
[28:42.500 --> 28:43.740]  it in. So
[28:44.260 --> 28:47.020]  it was a real experience.
[28:47.020 --> 28:48.280]  You know, I knew C.
[28:48.300 --> 28:50.280]  I just didn't use it for
[28:50.280 --> 28:52.460]  programming microcontrollers until this last
[28:52.460 --> 28:53.440]  go-around.
[28:54.320 --> 28:56.200]  So, my preference is
[28:56.200 --> 28:58.240]  assembly. I'm a hardware guy.
[28:58.240 --> 29:00.300]  If I do it in assembly, I know exactly
[29:00.300 --> 29:01.840]  what the hardware is going to do.
[29:01.840 --> 29:05.000]  The downside is, I have to optimize everything.
[29:05.080 --> 29:06.740]  So it's totally a trade-off.
[29:06.980 --> 29:08.220]  If you trust your
[29:09.600 --> 29:10.520]  compiler,
[29:10.520 --> 29:12.340]  and you know it's going to do
[29:12.420 --> 29:14.320]  a good job, go with a compiled
[29:14.320 --> 29:16.260]  language. If you're
[29:16.260 --> 29:18.160]  doing something a little odd, or
[29:18.160 --> 29:20.280]  really odd in my case, then
[29:20.280 --> 29:22.120]  assembly might be the better way to go.
[29:25.960 --> 29:27.380]  I spend most of my time
[29:27.380 --> 29:28.560]  in C,
[29:29.620 --> 29:31.280]  although it should be pointed out
[29:31.280 --> 29:33.440]  that you can embed assembly in C,
[29:33.440 --> 29:35.260]  and I've had to do that on a few occasions
[29:35.260 --> 29:37.360]  when the compiler just couldn't figure
[29:37.360 --> 29:39.320]  out exactly what I was trying to do.
[29:39.860 --> 29:40.820]  The compiler
[29:40.820 --> 29:43.320]  also can usually do all kinds of
[29:43.320 --> 29:45.740]  amazing things that you would never think of.
[29:45.860 --> 29:47.280]  And I also find it very
[29:47.280 --> 29:49.520]  useful. The compiler, or
[29:49.520 --> 29:50.780]  one of the object
[29:51.280 --> 29:53.360]  manipulation things
[29:53.360 --> 29:55.280]  in the toolchain, will
[29:55.280 --> 29:56.980]  spit out an assembly listing
[29:56.980 --> 29:59.400]  of the code that is compiled.
[29:59.400 --> 30:01.280]  And if you really need to optimize something,
[30:01.280 --> 30:03.640]  you really need to figure out
[30:03.640 --> 30:05.160]  why you just lost
[30:05.160 --> 30:06.880]  2K of program space
[30:07.380 --> 30:09.320]  when you added something that seemed really
[30:09.320 --> 30:11.280]  simple, you can go back to that
[30:11.280 --> 30:13.540]  listing and see what the compiler actually did
[30:13.540 --> 30:15.420]  and figure out where it might
[30:15.420 --> 30:16.880]  be doing something stupid.
[30:18.940 --> 30:19.420]  So
[30:19.420 --> 30:21.260]  first of all, Arduino sucks.
[30:22.380 --> 30:23.100]  I did
[30:23.100 --> 30:25.580]  the Bender badge last year in Arduino.
[30:25.580 --> 30:28.540]  If you look at our code
[30:28.540 --> 30:30.520]  on GitHub, we had about 30
[30:30.520 --> 30:32.600]  different files. I had to name the file
[30:32.600 --> 30:34.200]  and give them the shortest name possible
[30:34.200 --> 30:36.820]  so it would fit on those tabs across the top.
[30:37.200 --> 30:38.300]  This year we're
[30:38.300 --> 30:40.360]  C, using Nordic.
[30:40.360 --> 30:42.760]  I actually kind of like the Nordic SDK.
[30:42.760 --> 30:44.400]  It's pretty nice to work with.
[30:44.400 --> 30:46.040]  As long as you don't use their high-level
[30:46.500 --> 30:48.760]  libraries too much, you're pretty much okay.
[30:51.720 --> 30:52.200]  So
[30:53.060 --> 30:54.640]  contrary to basically
[30:54.640 --> 30:56.340]  every Hackaday comment ever
[30:56.340 --> 30:58.500]  about the Parallax Propeller, you can
[30:58.500 --> 31:00.440]  use C on it.
[31:01.200 --> 31:02.460]  I don't,
[31:02.460 --> 31:04.340]  because I've been using Spin for a long time
[31:04.340 --> 31:06.260]  and I'm pretty good with it, and I
[31:06.260 --> 31:08.320]  treat it more like you'd approach
[31:08.320 --> 31:09.280]  assembly.
[31:10.160 --> 31:12.380]  I'm mostly shifting things in and out of
[31:12.380 --> 31:13.840]  registers all the time.
[31:13.840 --> 31:15.240]  To me, it's more about
[31:16.280 --> 31:18.160]  how I'm using memory than it is about
[31:18.160 --> 31:19.580]  what language I'm using.
[31:19.720 --> 31:21.660]  All this stuff is super low-level.
[31:22.200 --> 31:23.880]  But again, this is a weird
[31:23.880 --> 31:25.740]  one because not only was I
[31:25.740 --> 31:27.720]  having to choose a programming language
[31:27.720 --> 31:29.900]  to work in, that made sense for my
[31:29.900 --> 31:30.920]  platform.
[31:31.240 --> 31:33.420]  There's also a programming language
[31:33.420 --> 31:35.080]  interpreter built into this.
[31:35.080 --> 31:37.860]  So I had to design a programming language
[31:37.860 --> 31:39.300]  for it.
[31:39.620 --> 31:41.240]  So, I don't know.
[31:44.780 --> 31:45.820]  Alrighty.
[31:46.560 --> 31:48.080]  Here's the last fun one
[31:48.080 --> 31:49.620]  we got from Twitter.
[31:49.980 --> 31:51.620]  And I will add a couple of details to it
[31:51.620 --> 31:53.760]  because it's kind of short.
[31:53.920 --> 31:55.760]  Would you rather hand solder
[31:55.760 --> 31:58.120]  1,000 duck-sized horse badges
[31:58.120 --> 32:00.020]  or one horse-sized
[32:00.020 --> 32:00.920]  duck badge?
[32:00.940 --> 32:02.860]  Asking for a friend.
[32:02.860 --> 32:05.580]  And I'm going to say same component count
[32:05.580 --> 32:07.320]  for both scenarios.
[32:07.320 --> 32:08.800]  How many components?
[32:09.220 --> 32:10.520]  The same.
[32:11.960 --> 32:13.820]  I see what you mean.
[32:13.820 --> 32:16.340]  How big are they?
[32:16.500 --> 32:17.480]  Well, there are
[32:17.480 --> 32:19.280]  1,000 duck-sized horse badges
[32:19.280 --> 32:21.840]  and one horse-sized duck badge.
[32:21.840 --> 32:23.060]  Yeah, but the components can be
[32:23.060 --> 32:25.360]  kind of small on them, or they could be like
[32:25.360 --> 32:26.720]  you know, whatever.
[32:26.720 --> 32:27.760]  Your choice.
[32:28.320 --> 32:30.680]  Does it weigh the same as a duck?
[32:32.660 --> 32:34.220]  Does it quack like a duck?
[32:34.220 --> 32:34.740]  Yes.
[32:39.360 --> 32:40.740]  1,000.
[32:41.320 --> 32:42.840]  1,000 ducks.
[32:43.520 --> 32:45.680]  Wait, horse-sized ducks?
[32:45.700 --> 32:46.860]  Duck-sized horses.
[32:47.480 --> 32:48.580]  I don't know.
[32:48.580 --> 32:49.300]  Whatever.
[32:49.460 --> 32:52.960]  The 1,000 options, because you have reach goals.
[32:53.300 --> 32:55.860]  So like, when you're soldering,
[32:55.860 --> 32:56.680]  you know that
[32:56.680 --> 32:58.780]  if you get this 10 done,
[32:58.780 --> 33:01.080]  like this group of 10 done,
[33:01.080 --> 33:03.000]  you're 10 closer to the end.
[33:03.040 --> 33:05.060]  And you can look at your little stack
[33:05.060 --> 33:07.160]  of done piles and you're like, look what I did.
[33:07.560 --> 33:08.180]  That's so nice.
[33:08.180 --> 33:10.860]  But with the horse-sized duck,
[33:11.620 --> 33:13.540]  then you're just waiting forever.
[33:13.980 --> 33:14.940]  You know, and you're still
[33:14.940 --> 33:17.520]  pasting, pasting, pasting.
[33:17.660 --> 33:18.780]  And, yeah.
[33:18.780 --> 33:20.140]  So I think I'd do the 1,000.
[33:20.420 --> 33:22.800]  I mean, I did 2,400, so...
[33:22.800 --> 33:23.480]  whatever.
[33:25.720 --> 33:26.860]  I'll agree with the 1,000
[33:26.860 --> 33:28.360]  because it becomes repetitive, right?
[33:28.360 --> 33:30.400]  That was kind of our pick and place last year.
[33:31.380 --> 33:32.860]  It's really easy to line up
[33:32.860 --> 33:34.660]  four badges and then drop the same component
[33:34.660 --> 33:37.260]  on all four and then put those into the oven.
[33:37.740 --> 33:38.860]  If we did the
[33:39.260 --> 33:40.840]  horse-sized duck, I'd just make
[33:40.840 --> 33:43.080]  Jorge do that one because he's really good at soldering
[33:43.600 --> 33:45.060]  a bunch of different components.
[33:45.980 --> 33:47.080]  Now the other thing is
[33:47.080 --> 33:49.340]  if you have like a quilting club, you know,
[33:49.340 --> 33:51.320]  you just set everybody around
[33:51.580 --> 33:52.980]  a lazy Susan or something
[33:52.980 --> 33:55.100]  and then they can take care of a part
[33:55.100 --> 33:56.880]  of the horse. So I guess there's that, but
[33:56.880 --> 33:59.280]  if you can't rely on, you know,
[33:59.280 --> 34:01.040]  everybody soldering like
[34:01.280 --> 34:03.300]  a masochist, then, yeah,
[34:03.300 --> 34:04.980]  1,000. I'll echo the
[34:04.980 --> 34:07.120]  1,000 for the assembly.
[34:07.200 --> 34:08.980]  You know, you can, you know,
[34:08.980 --> 34:10.680]  do assembly line style. You can
[34:10.680 --> 34:13.720]  get into the right rhythm.
[34:15.180 --> 34:16.400]  Can we make
[34:16.400 --> 34:18.740]  the large duck out of the smaller
[34:18.740 --> 34:21.480]  horses? Why not?
[34:21.580 --> 34:22.700]  Because then
[34:22.700 --> 34:25.520]  it's basically the same thing.
[34:29.870 --> 34:31.310]  I would just send mine to
[34:31.610 --> 34:33.270]  Macrofab for assembly.
[34:37.650 --> 34:39.970]  I'd send it to my picking place.
[34:40.030 --> 34:41.250]  That would be me.
[34:43.890 --> 34:45.670]  All right, I'm out of Twitter questions,
[34:45.670 --> 34:48.370]  so it's time for audience questions. Who's got a question?
[34:49.330 --> 34:51.070]  All right, you get first.
[34:53.190 --> 34:56.070]  So what feature of your badge
[34:56.750 --> 34:58.170]  do you hope
[34:58.170 --> 35:00.130]  and pray somebody finds?
[35:00.130 --> 35:01.790]  An Easter egg or
[35:01.790 --> 35:03.590]  some kind of hidden component?
[35:03.930 --> 35:06.430]  I'm asking for myself.
[35:08.050 --> 35:09.310]  The woods?
[35:09.310 --> 35:10.350]  The woods?
[35:11.310 --> 35:12.630]  For you?
[35:12.630 --> 35:14.330]  Probably. For Peter,
[35:14.330 --> 35:15.850]  it'd be the woods.
[35:16.090 --> 35:18.250]  If you find the woods.
[35:18.730 --> 35:20.470]  One person has found the
[35:20.470 --> 35:22.310]  woods, and they're like this
[35:22.310 --> 35:23.750]  close to getting in.
[35:26.090 --> 35:28.410]  No one has come anywhere near
[35:28.410 --> 35:30.030]  locating the panopticon
[35:30.030 --> 35:32.870]  that is in the center of the missile silo.
[35:32.990 --> 35:34.450]  So, you know,
[35:34.450 --> 35:35.850]  get busy, guys.
[35:38.510 --> 35:40.270]  So next year's badge will be
[35:40.270 --> 35:42.270]  trawlable like last year's.
[35:42.310 --> 35:44.270]  I had a little device last year that was
[35:44.270 --> 35:46.110]  sending out the rickroll packet
[35:46.110 --> 35:47.910]  every second, and we were rickrolling
[35:47.910 --> 35:49.530]  everyone we walked by.
[35:49.690 --> 35:51.930]  So of course we're going to do that next year.
[35:52.050 --> 35:54.190]  And if you're really good at Bluetooth,
[35:54.190 --> 35:56.250]  you may be able to figure out how to rickroll
[35:56.250 --> 35:58.130]  or Hypnotoad or
[35:58.130 --> 35:59.570]  whatever you want, all other
[35:59.570 --> 36:02.130]  vendor badges. And you won't need a badge to
[36:02.130 --> 36:04.030]  do it. You can do it with your phone.
[36:04.210 --> 36:05.730]  I remember I was like
[36:05.730 --> 36:08.030]  two drinks in, and all of a sudden I looked down
[36:08.030 --> 36:10.230]  and I'm getting rickrolled. And I'm like, where are they?
[36:13.330 --> 36:13.770]  So
[36:14.730 --> 36:16.170]  a story that I can tell
[36:16.170 --> 36:18.270]  is back
[36:18.270 --> 36:20.190]  when I was a noob at going to
[36:20.190 --> 36:22.330]  cons and nobody knew or cared who I was
[36:22.810 --> 36:24.110]  I'd put puzzles
[36:24.110 --> 36:26.010]  on my shirt or something
[36:27.150 --> 36:28.030]  and nobody
[36:28.030 --> 36:29.350]  cared and nobody would
[36:29.350 --> 36:31.490]  try to solve it.
[36:32.350 --> 36:32.430]  So
[36:34.530 --> 36:35.090]  for...
[36:35.090 --> 36:37.350]  and I want to echo what was
[36:37.350 --> 36:39.350]  said earlier about basically if you
[36:39.350 --> 36:41.370]  find anything that we've put
[36:41.370 --> 36:42.110]  into these badges
[36:43.470 --> 36:45.370]  that's just amazing for us.
[36:45.370 --> 36:47.290]  That someone actually took the time to find it
[36:47.290 --> 36:49.370]  and mess around with it.
[36:50.890 --> 36:53.090]  And I think actually that goes like
[36:53.090 --> 36:55.090]  last year's badge with encryption
[36:55.090 --> 36:57.270]  we had a single encryption
[36:58.010 --> 36:59.410]  sentence, I guess, on the back
[36:59.410 --> 37:00.970]  of the badge. And
[37:00.970 --> 37:03.450]  of 225 people
[37:03.450 --> 37:05.370]  I think, or 250 people who came to the
[37:05.370 --> 37:07.450]  conference, one team
[37:07.450 --> 37:09.370]  solved it. And it was students.
[37:09.370 --> 37:11.550]  So we love students.
[37:11.950 --> 37:13.490]  Because they roll over
[37:13.490 --> 37:15.430]  to our table. Okay, does
[37:15.430 --> 37:17.030]  it work like this?
[37:17.290 --> 37:19.330]  Oh no! And then they roll
[37:19.330 --> 37:21.170]  back and they work on it. So the
[37:21.170 --> 37:23.410]  working on it is really like
[37:23.410 --> 37:24.950]  what drives us and what
[37:24.950 --> 37:26.910]  we're going to be talking about on the
[37:26.910 --> 37:28.770]  trip home, to be honest.
[37:28.970 --> 37:30.910]  Yes, those students that cracked it
[37:30.910 --> 37:33.110]  last year, you may see them in the hallway.
[37:33.110 --> 37:35.050]  They've taken over an alcove. They've found
[37:35.050 --> 37:36.950]  an outlet. They have disassembled the
[37:36.950 --> 37:39.030]  badge, hooked logic analyzers
[37:39.030 --> 37:40.910]  up to it, broken
[37:40.910 --> 37:42.890]  basically everything. They're
[37:42.890 --> 37:45.010]  going through it every step.
[37:45.010 --> 37:46.890]  And that is
[37:46.890 --> 37:49.270]  worth more than whatever Getsman paid us.
[37:49.270 --> 37:50.830]  That feels really good.
[37:53.110 --> 37:54.290]  You know,
[37:54.290 --> 37:55.150]  continue to
[37:55.150 --> 37:56.970]  play with the badge. Just because the
[37:56.970 --> 37:59.290]  con's over, you can still play with it.
[37:59.290 --> 38:00.950]  And if you do find stuff,
[38:00.950 --> 38:02.810]  tweet it out at us. I mean, it's
[38:02.810 --> 38:04.790]  cool to see that, hey, two
[38:04.790 --> 38:07.250]  years from now someone tweets you and is like, hey,
[38:07.250 --> 38:08.930]  on the CypherCon 2.0 badge
[38:08.930 --> 38:10.610]  I found this. Just
[38:10.610 --> 38:12.770]  know that it's still interesting
[38:12.770 --> 38:14.830]  to someone. They just didn't all get thrown out at
[38:14.830 --> 38:15.890]  the end of the con.
[38:16.430 --> 38:18.910]  And for me personally, there are two
[38:18.910 --> 38:21.410]  things that I want to see done.
[38:21.650 --> 38:23.070]  So one, I've already seen
[38:23.070 --> 38:24.950]  kind of because people wanted points, like the
[38:24.950 --> 38:26.810]  QR code points, which was relay
[38:26.810 --> 38:28.350]  people chatting with each other.
[38:28.490 --> 38:30.490]  That blew my mind when
[38:30.490 --> 38:32.990]  Whisker implemented it. They're like, hey,
[38:32.990 --> 38:34.710]  sit down. And I'm like,
[38:34.710 --> 38:36.950]  okay, whatever. They're like, relay me.
[38:36.950 --> 38:38.750]  And I'm like, okay, blah. And then he starts
[38:38.750 --> 38:41.150]  talking to me, and I'm like,
[38:41.150 --> 38:42.730]  what is this?
[38:42.810 --> 38:44.990]  Right? I was amazed. And I love
[38:44.990 --> 38:46.850]  that, like, chat with each other on
[38:47.010 --> 38:48.230]  a badge.
[38:49.430 --> 38:50.630]  The other thing
[38:50.630 --> 38:53.350]  is when I was testing,
[38:53.350 --> 38:54.730]  I never managed to get
[38:54.730 --> 38:56.390]  more than seven hops
[38:57.370 --> 38:58.710]  through the badges. Seven
[38:58.710 --> 39:01.010]  to eight, maybe, hops
[39:01.010 --> 39:02.710]  before it got way too slow
[39:02.710 --> 39:04.510]  for me to talk or to
[39:04.510 --> 39:06.990]  telnet to the next badge. So I want to see
[39:06.990 --> 39:08.790]  how far somebody
[39:08.790 --> 39:10.710]  can telnet, which may require
[39:10.710 --> 39:12.430]  getting a bunch of people together in a circle
[39:12.430 --> 39:14.710]  and keeping track of
[39:14.710 --> 39:16.710]  where you are in the chain to make sure
[39:16.710 --> 39:19.090]  you're not doubling back on yourself.
[39:19.430 --> 39:21.270]  So, yeah.
[39:21.430 --> 39:22.770]  So one thing that we did at the
[39:22.770 --> 39:24.650]  last minute is I put
[39:24.650 --> 39:26.730]  these coins together that we've been
[39:26.730 --> 39:28.750]  handing out a few of them here and there.
[39:29.170 --> 39:30.770]  There's a couple of things on
[39:30.770 --> 39:32.990]  here that you can try to figure out.
[39:34.390 --> 39:35.270]  If you
[39:35.270 --> 39:36.770]  go to a lot of these cons, like
[39:36.770 --> 39:38.370]  DefCon, DerbyCon,
[39:38.370 --> 39:40.810]  we're going to try to hook that
[39:40.810 --> 39:43.010]  into something that we bring to those.
[39:43.010 --> 39:45.090]  So, yeah.
[39:45.090 --> 39:46.910]  I would be really
[39:46.910 --> 39:48.930]  happy if someone actually makes that connection
[39:48.930 --> 39:50.950]  and does that. Especially since I don't
[39:50.950 --> 39:53.130]  know what exactly
[39:53.130 --> 39:55.270]  it's going to do yet because I haven't written it.
[39:55.690 --> 39:56.370]  Okay.
[40:00.690 --> 40:01.470]  How much
[40:01.470 --> 40:03.330]  time is spent on research and
[40:03.330 --> 40:05.270]  development compared to production
[40:05.270 --> 40:06.350]  of the badges?
[40:09.850 --> 40:11.410]  So, a lot of that
[40:12.030 --> 40:13.510]  come... oh, sorry.
[40:13.790 --> 40:14.790]  Can I go first?
[40:14.790 --> 40:16.070]  Go ahead.
[40:16.470 --> 40:18.070]  A lot of that comes down to
[40:18.510 --> 40:21.030]  what you're doing and have you done it before.
[40:21.630 --> 40:22.950]  So, this year
[40:22.950 --> 40:24.730]  with the USB stack and the
[40:24.730 --> 40:26.870]  IRDA stack, that was a lot
[40:27.430 --> 40:28.490]  of... I mean,
[40:28.490 --> 40:30.710]  it wasn't so much research and development, but it was
[40:30.710 --> 40:32.030]  just a lot of learning.
[40:32.550 --> 40:34.890]  I knew this part existed here, I knew that part
[40:34.890 --> 40:37.150]  existed there, but I never put them together.
[40:37.790 --> 40:39.290]  So, what do you say?
[40:39.550 --> 40:41.070]  40-60 split?
[40:42.950 --> 40:44.030]  Yeah.
[40:44.750 --> 40:44.970]  For
[40:44.970 --> 40:47.130]  the hardware portion, it
[40:47.130 --> 40:49.130]  also depends on your parts. So,
[40:49.130 --> 40:50.830]  we sourced through a broker in China
[40:51.510 --> 40:52.050]  and
[40:53.410 --> 40:55.110]  there are ups and downs to
[40:55.110 --> 40:57.210]  that. One of the downs
[40:57.210 --> 40:58.950]  was that the part that I
[40:58.950 --> 40:59.970]  asked him to get
[41:00.990 --> 41:03.090]  was completely opposite from the part
[41:03.090 --> 41:05.090]  that we got. So, we had
[41:05.090 --> 41:07.070]  like 2100
[41:08.090 --> 41:09.170]  LEDs that
[41:09.170 --> 41:11.190]  were completely backwards.
[41:11.710 --> 41:13.250]  And so, what that meant was I had to
[41:13.250 --> 41:15.190]  re-spin all the boards, like
[41:15.190 --> 41:16.650]  all the sides,
[41:17.170 --> 41:19.210]  and make sure that those prototypes worked
[41:19.210 --> 41:20.930]  with the new LED
[41:20.930 --> 41:22.810]  that we actually got sourced.
[41:23.210 --> 41:25.190]  And then the other thing is partially
[41:25.190 --> 41:27.170]  feature creep.
[41:27.290 --> 41:29.850]  So, if you look at your badges,
[41:29.850 --> 41:31.270]  there are, on the sides
[41:31.270 --> 41:33.430]  and the top, there are absolutely
[41:33.430 --> 41:34.930]  no vias.
[41:35.370 --> 41:37.170]  So, there's nothing...
[41:37.170 --> 41:39.730]  Yeah, okay, I'm really proud of that.
[41:39.950 --> 41:41.150]  There are no vias
[41:41.150 --> 41:43.190]  on these boards. And so, if you look
[41:43.190 --> 41:45.130]  at the rose pattern
[41:45.130 --> 41:47.030]  inside for these
[41:47.030 --> 41:49.490]  16 LEDs, and you've
[41:49.490 --> 41:51.150]  done trace work before, like
[41:51.730 --> 41:53.190]  that's... I'd like to think
[41:53.190 --> 41:54.350]  that's impressive.
[41:55.050 --> 41:57.170]  So, the amount of time that was spent
[41:57.170 --> 41:58.630]  doing that trace work
[41:59.170 --> 42:01.190]  was a little more than
[42:01.190 --> 42:03.810]  your normal board.
[42:04.330 --> 42:04.890]  But,
[42:05.730 --> 42:06.470]  yeah, so
[42:06.470 --> 42:08.890]  at least for me, that part
[42:08.890 --> 42:10.710]  depended on the parts and
[42:10.710 --> 42:12.790]  whether it worked or not.
[42:13.970 --> 42:14.970]  So, for us, it's
[42:14.970 --> 42:17.330]  about 30-40%
[42:17.330 --> 42:19.770]  of the time is spent just researching,
[42:19.770 --> 42:21.350]  buying sample parts,
[42:21.350 --> 42:22.950]  prototyping, writing drivers,
[42:22.950 --> 42:24.950]  and just kind of figuring out all
[42:24.950 --> 42:26.830]  those ideas that we have, which ones will
[42:26.830 --> 42:27.990]  actually work.
[42:28.690 --> 42:30.550]  The production part is...
[42:30.550 --> 42:33.030]  it's a furious activity. At least last year
[42:33.030 --> 42:35.170]  it was. It's about 6 weeks long.
[42:35.230 --> 42:37.270]  Not counting sourcing any of the parts.
[42:37.270 --> 42:39.170]  One thing I've learned, especially dealing with China,
[42:39.170 --> 42:41.090]  is buy the parts early,
[42:41.090 --> 42:43.410]  get them on hand, and then finalize
[42:43.410 --> 42:45.370]  the design based upon that.
[42:45.370 --> 42:47.170]  So, our displays last year, I think
[42:47.170 --> 42:49.430]  I had those on hand in March, and we didn't finalize
[42:49.430 --> 42:50.950]  until May.
[42:51.350 --> 42:53.050]  And that was just because I wasn't sure what pins
[42:53.050 --> 42:54.450]  I would actually end up with.
[42:57.510 --> 42:59.270]  Any more questions from the audience?
[42:59.870 --> 43:00.930]  Don't be shy.
[43:02.490 --> 43:03.490]  Can we
[43:03.490 --> 43:05.230]  make up more questions?
[43:06.290 --> 43:06.890]  Any
[43:06.890 --> 43:09.250]  duck-sized dogs or horse-sized
[43:09.250 --> 43:10.670]  elephants?
[43:17.600 --> 43:19.140]  Is there any
[43:19.140 --> 43:21.800]  way you can control
[43:21.800 --> 43:23.820]  things like TVs with the
[43:23.820 --> 43:25.000]  badges?
[43:29.780 --> 43:30.660]  The
[43:30.660 --> 43:32.480]  IRDA stack
[43:32.480 --> 43:34.480]  relies on a hardware IRDA
[43:34.480 --> 43:36.220]  engine in the PicMicro.
[43:36.560 --> 43:38.300]  And the way that IRDA
[43:38.300 --> 43:40.320]  works is whenever you...
[43:40.320 --> 43:42.280]  We're using SIR, which is
[43:42.280 --> 43:44.220]  simple IR. IRDA
[43:44.220 --> 43:46.440]  mode. So basically, it acts
[43:46.440 --> 43:48.200]  like a glorified UARP.
[43:48.240 --> 43:50.260]  The way it does it, though, is it takes
[43:50.260 --> 43:52.500]  for any time you send a zero,
[43:52.500 --> 43:54.480]  it takes the bit time,
[43:54.480 --> 43:55.860]  divides it by 8, and sends
[43:55.860 --> 43:58.220]  one high pulse for that
[43:58.220 --> 44:00.280]  bit time. And then on the other end,
[44:00.280 --> 44:02.560]  it uses that to sync things back up
[44:02.560 --> 44:04.620]  because they're going to be totally out of sequence
[44:04.620 --> 44:06.480]  and also due to reflections
[44:06.480 --> 44:08.720]  and other things, if it sees two
[44:08.720 --> 44:10.760]  pulses in that same window because
[44:10.760 --> 44:12.660]  of reflections, it
[44:12.660 --> 44:14.680]  ignores the second one. So it helps it
[44:14.680 --> 44:16.180]  recover the signal.
[44:16.680 --> 44:18.380]  So in theory, no.
[44:18.640 --> 44:20.360]  But as soon as I say that, someone's going to be like,
[44:20.360 --> 44:22.420]  hey, I just turned off my TV. So
[44:22.420 --> 44:24.780]  if you get the right baud rate
[44:24.780 --> 44:26.680]  coming out of IRDA and the bits
[44:26.680 --> 44:28.960]  just happen to line up in the right spots,
[44:28.960 --> 44:30.940]  it might be possible.
[44:30.940 --> 44:32.080]  So
[44:33.600 --> 44:34.240]  the
[44:34.940 --> 44:37.000]  TVVigon is a really neat little
[44:37.000 --> 44:38.860]  mischief making device.
[44:38.980 --> 44:40.900]  And it's completely open source
[44:40.900 --> 44:42.840]  so you can get the plans off of
[44:42.840 --> 44:44.660]  wherever.
[44:44.700 --> 44:46.900]  Adafruit has one version of it.
[44:46.920 --> 44:48.500]  It runs with an
[44:48.500 --> 44:51.660]  ATTiny85, a couple of transistors,
[44:51.660 --> 44:53.260]  and infrared LEDs.
[44:53.640 --> 44:54.720]  And, well, crystal
[44:54.720 --> 44:57.080]  power filtering capacitors, all that.
[44:57.480 --> 44:58.760]  Boilerplate.
[44:58.760 --> 45:00.160]  And that
[45:00.560 --> 45:02.860]  has been and will be rolled into
[45:04.380 --> 45:05.660]  badges.
[45:06.060 --> 45:06.700]  So
[45:06.700 --> 45:08.980]  the first IRDA part
[45:08.980 --> 45:10.560]  that I picked had
[45:11.260 --> 45:12.440]  a RR remote mode
[45:12.440 --> 45:14.680]  built into it. But unfortunately,
[45:14.680 --> 45:16.760]  when that part just turned out to be
[45:17.920 --> 45:19.180]  junk,
[45:19.180 --> 45:20.660]  the new part didn't
[45:20.660 --> 45:22.900]  have that, so that feature got dropped.
[45:23.920 --> 45:24.760]  But it would have
[45:24.760 --> 45:26.320]  been a good feature for you.
[45:26.320 --> 45:28.400]  Yeah, I mean, we were totally planning
[45:28.400 --> 45:30.600]  with the original part to support
[45:31.800 --> 45:33.200]  there's a
[45:33.200 --> 45:34.940]  Linux open source project
[45:34.940 --> 45:36.780]  for controlling remote TVs
[45:36.780 --> 45:38.720]  and whatnot, where they have a standard
[45:38.720 --> 45:40.820]  format of libraries for IR codes.
[45:40.820 --> 45:42.840]  We were planning on supporting that format
[45:42.840 --> 45:44.600]  so you could just type it in and your badge
[45:44.600 --> 45:46.780]  would do it. But unfortunately, switching
[45:46.780 --> 45:48.700]  to the other IRDA module
[45:48.700 --> 45:52.020]  sort of killed that feature.
[45:52.020 --> 45:52.680]  I will say
[45:53.560 --> 45:54.440]  that Charles and I have been
[45:54.440 --> 45:56.280]  told not to provide batteries
[45:56.280 --> 45:58.100]  for certain things.
[45:58.100 --> 46:00.320]  You can do that at certain conventions
[46:00.320 --> 46:01.700]  by the hotel staff.
[46:02.220 --> 46:04.320]  Pro tip, some of our TV
[46:04.320 --> 46:06.360]  BeGons and some of our other soldering kits
[46:06.360 --> 46:08.220]  take the same kind of battery.
[46:10.320 --> 46:10.680]  So
[46:10.680 --> 46:12.440]  that was actually one of the ideas
[46:12.440 --> 46:14.740]  we dumped. We had a TV BeGon idea.
[46:14.740 --> 46:16.440]  We wanted to do IR and do some interesting
[46:16.440 --> 46:18.280]  badge-to-badge comms with it.
[46:18.280 --> 46:20.640]  We didn't continue with that idea.
[46:20.640 --> 46:22.440]  But if you have the Defcon 20 or
[46:22.440 --> 46:24.500]  Defcon 22 badge, those are easy
[46:24.500 --> 46:26.440]  to hack to do
[46:26.440 --> 46:28.440]  any Sony IR remote control code
[46:28.440 --> 46:30.500]  you want to do. We did it and
[46:30.500 --> 46:32.360]  we're actually able to basically DOS
[46:32.360 --> 46:34.420]  any IR signal around us
[46:34.420 --> 46:36.520]  with the Defcon 20 badge.
[46:36.520 --> 46:39.240]  So if you have that on hand, go look at it.
[46:39.800 --> 46:40.600]  So what
[46:40.600 --> 46:42.440]  dream feature would you guys like
[46:42.440 --> 46:44.320]  to add to a badge if you had
[46:44.320 --> 46:46.320]  the budget and the time
[46:46.320 --> 46:48.440]  and all of the stars aligned?
[46:48.480 --> 46:49.600]  Defray!
[46:49.600 --> 46:51.180]  Defray or deathray?
[46:52.560 --> 46:53.140]  Let's go ahead and
[46:53.140 --> 46:55.640]  add a deathray.
[46:55.680 --> 46:57.380]  We might have a winner there.
[47:05.880 --> 47:08.020]  Fully fledged relay computer?
[47:09.140 --> 47:11.520]  Clack, clack, clack, clack, clack, clack, clack, clack, clack.
[47:11.520 --> 47:12.760]  Clack, clack, clack, clack, clack, clack, clack, clack, clack.
[47:12.780 --> 47:13.700]  Oh!
[47:13.700 --> 47:15.380]  Yeah, flip dots or something.
[47:15.380 --> 47:17.240]  Flip dots.
[47:17.240 --> 47:18.300]  Flip dots.
[47:21.680 --> 47:23.260]  Alright, so the first thing
[47:23.260 --> 47:24.940]  I'm going to do with mine when I get it home
[47:26.520 --> 47:27.000]  is
[47:27.000 --> 47:29.420]  tap into the UART
[47:29.420 --> 47:31.280]  that's on the bottom and
[47:31.980 --> 47:33.240]  hook it up to my acoustic
[47:33.240 --> 47:35.420]  coupler because we have a PBX
[47:35.420 --> 47:36.820]  in our house because we're
[47:36.820 --> 47:38.420]  those kind of people.
[47:38.780 --> 47:41.280]  And our own ISP on it
[47:41.280 --> 47:43.300]  because I want to go
[47:43.300 --> 47:45.020]  in IR to the badge
[47:45.020 --> 47:46.780]  and then out
[47:46.780 --> 47:48.500]  to the acoustic coupler
[47:48.770 --> 47:50.920]  and dial through to our ISP
[47:50.920 --> 47:52.900]  to get online that way.
[47:53.380 --> 47:55.220]  But I don't think that's really
[47:55.220 --> 47:56.960]  the sort of thing you would do
[47:56.960 --> 47:58.980]  for an actual conference day of
[47:58.980 --> 48:00.900]  rollout thing unless it was just for a demo
[48:00.900 --> 48:03.260]  or something, but it would be a cool feature.
[48:04.060 --> 48:05.100]  Because reasons
[48:05.100 --> 48:07.460]  is perfectly valid for me.
[48:07.460 --> 48:08.980]  Wait, so it's possible?
[48:08.980 --> 48:09.800]  Yes.
[48:11.320 --> 48:13.140]  So actually we probably could still
[48:13.140 --> 48:14.980]  talk to you guys
[48:14.980 --> 48:16.680]  through the internets
[48:16.680 --> 48:18.480]  with the badge.
[48:18.700 --> 48:21.360]  But very, very slowly.
[48:22.620 --> 48:23.340]  So I'm not
[48:23.340 --> 48:25.200]  going to talk about our dream features because that's
[48:25.200 --> 48:27.260]  what I'm actually working on back in the hotel
[48:27.260 --> 48:29.520]  room. I know.
[48:29.520 --> 48:30.960]  I'm sure you won't.
[48:31.000 --> 48:33.000]  It's been obscured on the badge.
[48:33.780 --> 48:35.240]  Sorry, you'll have to wait.
[48:36.860 --> 48:37.340]  Any more
[48:37.340 --> 48:39.160]  questions from the audience? I kind of want that
[48:39.160 --> 48:41.440]  LED matrix up there to have a capacitive
[48:41.440 --> 48:43.020]  touch matrix under it.
[48:44.760 --> 48:45.660]  If budget
[48:45.660 --> 48:47.580]  were not an issue, I think I would have put
[48:47.580 --> 48:49.540]  corner... in time. I think I
[48:49.540 --> 48:51.980]  would have put corner LEDs in everybody's badge.
[48:51.980 --> 48:53.120]  And that was the initial
[48:53.120 --> 48:55.460]  intention, because the corner
[48:55.460 --> 48:57.040]  LEDs are pretty epic.
[48:57.860 --> 48:59.440]  I think I will provide
[48:59.440 --> 49:01.480]  the part number, though, for you guys. If you
[49:01.480 --> 49:03.560]  choose at your own risk to take
[49:03.560 --> 49:05.020]  your badge apart and to
[49:05.020 --> 49:07.300]  put the corner LEDs in.
[49:07.580 --> 49:09.400]  Because when you play
[49:09.400 --> 49:11.240]  the game, the corner LEDs
[49:11.240 --> 49:13.650]  act as a map for the game.
[49:14.200 --> 49:14.740]  So,
[49:15.480 --> 49:16.820]  I think that would have been
[49:17.820 --> 49:18.360]  my
[49:19.000 --> 49:20.500]  dream feature.
[49:21.140 --> 49:22.800]  I definitely want to do that.
[49:24.020 --> 49:26.120]  Any more questions from the audience?
[49:27.300 --> 49:28.160]  Bueller?
[49:28.520 --> 49:29.460]  Bueller?
[49:29.980 --> 49:32.020]  Nobody? Let me see here.
[49:32.660 --> 49:34.080]  Oh, there we go.
[49:40.160 --> 49:41.920]  I know you guys integrated
[49:42.560 --> 49:43.960]  a game or a story into
[49:43.960 --> 49:45.880]  your badge, and I was just
[49:45.880 --> 49:47.920]  wondering, how common is
[49:47.920 --> 49:49.000]  that, or is this
[49:50.080 --> 49:51.860]  a really unique approach?
[49:55.160 --> 49:55.920]  Yeah.
[49:55.920 --> 49:57.980]  These guys have games in theirs.
[49:59.920 --> 50:01.840]  Do you mean like the story
[50:01.840 --> 50:03.780]  or the game, or both?
[50:04.520 --> 50:06.180]  Okay. I think...
[50:06.180 --> 50:07.000]  Oh, sorry.
[50:07.000 --> 50:10.060]  Yeah, I've played with theirs a little bit,
[50:10.060 --> 50:12.120]  and they've got some pretty cool games in there.
[50:12.120 --> 50:14.360]  Is there anything that's really like story-driven
[50:14.900 --> 50:16.120]  on those?
[50:16.280 --> 50:17.380]  No. Nothing.
[50:17.380 --> 50:19.520]  Next year? We'll see.
[50:19.520 --> 50:21.660]  I'm trying to get him to tell us stuff.
[50:23.720 --> 50:25.540]  We did Flappy Bird and
[50:26.000 --> 50:27.840]  SkiFree Asteroids.
[50:28.060 --> 50:29.840]  That's only because I had some
[50:29.840 --> 50:31.580]  back problems, and so I couldn't move,
[50:31.580 --> 50:32.840]  so I just sat there coding.
[50:33.720 --> 50:35.520]  I know you got Snake on there.
[50:35.640 --> 50:37.940]  On last year's, there's Snake.
[50:37.940 --> 50:40.080]  And then on next year's,
[50:40.080 --> 50:41.900]  we've implemented a Chip8 emulator,
[50:41.900 --> 50:43.020]  and so there's 64
[50:43.760 --> 50:45.720]  emulated games in there that we've
[50:45.720 --> 50:46.980]  put on.
[50:47.520 --> 50:50.000]  But nothing story-based. The only other badge
[50:50.000 --> 50:51.940]  I know that even attempted that,
[50:51.940 --> 50:53.920]  the Whiskey Pirates tried one last year,
[50:53.920 --> 50:56.020]  but they had some issues with their hardware.
[50:57.220 --> 50:59.680]  I will say that the SkyDogCon badge
[50:59.680 --> 51:02.020]  that I'm wearing was designed by Red,
[51:02.020 --> 51:03.860]  and it had an interactive
[51:03.860 --> 51:05.700]  game of tank that all of the participants
[51:05.700 --> 51:08.040]  in the con could play versus each other.
[51:08.040 --> 51:10.040]  They had that running off a server
[51:10.040 --> 51:11.840]  over the wireless without
[51:12.780 --> 51:14.460]  completely killing the wireless,
[51:14.460 --> 51:15.760]  which was pretty
[51:15.760 --> 51:17.540]  amazing in of itself, and
[51:17.540 --> 51:19.440]  that was pretty impressive.
[51:20.220 --> 51:21.820]  Yeah. And speaking of badge
[51:21.820 --> 51:24.020]  life, if I can tell a story,
[51:24.920 --> 51:26.040]  Charles and I arrived
[51:26.040 --> 51:27.900]  at SkyDogCon going, yay, a con!
[51:27.900 --> 51:29.800]  We won't be up all night soldering badges
[51:29.800 --> 51:31.600]  before, and we'll actually get sleep.
[51:31.600 --> 51:33.340]  And then we met Red, and he was
[51:33.340 --> 51:34.820]  still putting together the last
[51:35.340 --> 51:37.180]  finishing touches on these, and
[51:37.180 --> 51:39.280]  we were talking badge stuff
[51:39.280 --> 51:40.960]  and soldering, and
[51:41.840 --> 51:43.260]  one way or another the challenge
[51:43.260 --> 51:45.060]  was issued, and so Charles and I
[51:45.060 --> 51:47.180]  stayed up until I think one or two in the morning
[51:47.180 --> 51:48.400]  soldering our own badges
[51:50.200 --> 51:51.520]  because masochism.
[51:54.080 --> 51:55.460]  Is that time?
[51:56.040 --> 51:57.080]  Yeah, that's about
[51:57.720 --> 51:59.520]  it. So, thank you all very much
[51:59.520 --> 52:01.240]  for coming, and for the
[52:01.240 --> 52:02.740]  panelists for talking to us all, and
[52:02.740 --> 52:05.360]  thank you very much for inviting us out.
[52:05.780 --> 52:06.180]  applause