[Script Info]
Title: 
[Events]
Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
Dialogue: 0,0:00:00.00,0:00:19.70,Default,,0000,0000,0000,,{\i1}36C3 preroll music{\i0}
Dialogue: 0,0:00:19.70,0:00:27.70,Default,,0000,0000,0000,,Herald-Angel: Good. Ladies and gentlemen,\Nwe have here a talk by Sebastian Staacks.
Dialogue: 0,0:00:27.70,0:00:29.77,Default,,0000,0000,0000,,Do I pronounce this well?\NSebastian Staacks: Yes.
Dialogue: 0,0:00:29.77,0:00:33.16,Default,,0000,0000,0000,,Herald: Yes. Staacks. Staacks. [In German]\NIch musste das mal in Deutsch sagen. And
Dialogue: 0,0:00:33.16,0:00:40.64,Default,,0000,0000,0000,,he's related to the University of Aachen.\NHe did a PhD physics. And he was in a team
Dialogue: 0,0:00:40.64,0:00:47.16,Default,,0000,0000,0000,,that developed a fantastic application, as\NI mentioned earlier on. He developed the
Dialogue: 0,0:00:47.16,0:00:51.62,Default,,0000,0000,0000,,app phyphox. Do I pronounce this well?\NStaacks: I would say phi-phox, physical
Dialogue: 0,0:00:51.62,0:00:55.35,Default,,0000,0000,0000,,phone experiments.\NHerald: Okay. Yep. Of course. I'm sorry.
Dialogue: 0,0:00:55.35,0:01:02.38,Default,,0000,0000,0000,,I'm not in that kind of department. But\Nthis application actually gives you all
Dialogue: 0,0:01:02.38,0:01:07.16,Default,,0000,0000,0000,,the possibilities off your the usage, off\Nyour smart smartphone. Really? Really
Dialogue: 0,0:01:07.16,0:01:12.61,Default,,0000,0000,0000,,extending certain borders, to my opinion.\NSo please give a warm, warm welcome here
Dialogue: 0,0:01:12.61,0:01:14.25,Default,,0000,0000,0000,,to Stefan.
Dialogue: 0,0:01:14.25,0:01:20.79,Default,,0000,0000,0000,,{\i1}Applause{\i0}
Dialogue: 0,0:01:20.79,0:01:24.56,Default,,0000,0000,0000,,Stefan: Thank you. Thank you for the\Nintroduction and welcome everybody to my
Dialogue: 0,0:01:24.56,0:01:31.01,Default,,0000,0000,0000,,talk. Yeah. As you've just heard, I'm a\Nphysicist from the RWTH Aachen university
Dialogue: 0,0:01:31.01,0:01:36.31,Default,,0000,0000,0000,,where I developed the app phyphox. Phyphox\Nis an app for those of you who do not know
Dialogue: 0,0:01:36.31,0:01:41.04,Default,,0000,0000,0000,,it already. That uses the sensors in the\Nsmartphone for physics teaching. So the
Dialogue: 0,0:01:41.04,0:01:46.38,Default,,0000,0000,0000,,idea is that students can use their own\Nphones to do experimentation in class, in
Dialogue: 0,0:01:46.38,0:01:52.11,Default,,0000,0000,0000,,the lecture hall. So for schools and\Nuniversities. I should explain. That in
Dialogue: 0,0:01:52.11,0:01:56.50,Default,,0000,0000,0000,,contrast to some other talks by me. This\None will not be that much about education
Dialogue: 0,0:01:56.50,0:01:59.78,Default,,0000,0000,0000,,because it is the chaos communication\NCongress and this is the hardware track
Dialogue: 0,0:01:59.78,0:02:04.71,Default,,0000,0000,0000,,here. So I tried to tell you a little bit\Nabout the app, a little bit about the
Dialogue: 0,0:02:04.71,0:02:10.16,Default,,0000,0000,0000,,sensors that we have on our phones and.\NYeah. Would we love to get in touch with
Dialogue: 0,0:02:10.16,0:02:15.41,Default,,0000,0000,0000,,some, especially people from maker\Ncommunity and from open source communities
Dialogue: 0,0:02:15.41,0:02:20.84,Default,,0000,0000,0000,,to find some connections, how he can get\Nmany open source projects together?
Dialogue: 0,0:02:20.84,0:02:25.49,Default,,0000,0000,0000,,Because I've got so much feedback from\Nteachers and I think I could also use some
Dialogue: 0,0:02:25.49,0:02:31.91,Default,,0000,0000,0000,,feedback from other developers as well. So\NI would like to start with a short
Dialogue: 0,0:02:31.91,0:02:37.41,Default,,0000,0000,0000,,explanation of what we actually do. So\Nyes, I said I come from a university and
Dialogue: 0,0:02:37.41,0:02:42.31,Default,,0000,0000,0000,,there we have this introductionary lecture\Nfor physics students, which is called
Dialogue: 0,0:02:42.31,0:02:47.36,Default,,0000,0000,0000,,experimental physics one. And it's typical\Nlecture. Looks like this. We have a fancy
Dialogue: 0,0:02:47.36,0:02:52.12,Default,,0000,0000,0000,,new lecture hall by now, but the situation\Nis the same. We've got 300 I think 370
Dialogue: 0,0:02:52.12,0:02:57.15,Default,,0000,0000,0000,,students this year sitting in a lecture\Nhall and doing no experimentation at all.
Dialogue: 0,0:02:57.15,0:03:00.32,Default,,0000,0000,0000,,There's only one guy experimenting and\Nthat's the professor. And the students are
Dialogue: 0,0:03:00.32,0:03:05.57,Default,,0000,0000,0000,,sitting there and enjoying the whole show\Nlike they would enjoy a YouTube video and
Dialogue: 0,0:03:05.57,0:03:09.64,Default,,0000,0000,0000,,maybe they are mildly amused if something\Ngoes wrong. OK. And we thought we could
Dialogue: 0,0:03:09.64,0:03:14.08,Default,,0000,0000,0000,,change this by using the sensors in the\Nsmartphones. We're not the first ones with
Dialogue: 0,0:03:14.08,0:03:18.03,Default,,0000,0000,0000,,the idea to use the sensors there, but for\Nsome reason we decided to write our own
Dialogue: 0,0:03:18.03,0:03:23.02,Default,,0000,0000,0000,,app, which turned out to be quite\Nsuccessful then. So in contrast to the old
Dialogue: 0,0:03:23.02,0:03:26.31,Default,,0000,0000,0000,,version where students just had to look at\Nand I'll get the assignments where they
Dialogue: 0,0:03:26.31,0:03:30.28,Default,,0000,0000,0000,,can do their own experiments with their\Nown measurement devices. And to give you
Dialogue: 0,0:03:30.28,0:03:34.54,Default,,0000,0000,0000,,an idea of what this looks like. I would\Nlike to start with the first experiment.
Dialogue: 0,0:03:34.54,0:03:39.12,Default,,0000,0000,0000,,Which is about centrifugal acceleration or\Ncentripetal acceleration depending on your
Dialogue: 0,0:03:39.12,0:03:45.19,Default,,0000,0000,0000,,preferred frame of reference. So the idea\Nis from a rotation movement, we want to
Dialogue: 0,0:03:45.19,0:03:50.06,Default,,0000,0000,0000,,measure the radial acceleration as a\Nfunction of the angular velocity. So the
Dialogue: 0,0:03:50.06,0:03:55.39,Default,,0000,0000,0000,,rotation rate. To do this we take a\Nregular smartphone, this is an iPhone 8 in
Dialogue: 0,0:03:55.39,0:04:00.02,Default,,0000,0000,0000,,this case and we put it into a salad\Nspinner. Okay. We get some rotation in
Dialogue: 0,0:04:00.02,0:04:09.70,Default,,0000,0000,0000,,there and whoops let me just place it in\Nthere. Sound is not important, but it
Dialogue: 0,0:04:09.70,0:04:17.36,Default,,0000,0000,0000,,sounds nice. I have been told. So here we\Nget the live data from the phone already.
Dialogue: 0,0:04:17.36,0:04:22.95,Default,,0000,0000,0000,,Acceleration on the y axis and angular\Nvelocity on the x axis. If the salad
Dialogue: 0,0:04:22.95,0:04:27.00,Default,,0000,0000,0000,,spinner is actually moving. And what you\Nsee is the faster I rotate the spinner,
Dialogue: 0,0:04:27.00,0:04:31.85,Default,,0000,0000,0000,,the farther on the right you get your data\Nbecause that's angular velocity and also
Dialogue: 0,0:04:31.85,0:04:36.84,Default,,0000,0000,0000,,the radial acceleration increases. If I'm\Nnot going too fast because then I do not
Dialogue: 0,0:04:36.84,0:04:41.69,Default,,0000,0000,0000,,get any data at all anymore. Let's slow\Ndown again and we can fill up the gaps
Dialogue: 0,0:04:41.69,0:04:47.95,Default,,0000,0000,0000,,there by going really slow and filling up\Nthis path. And in the end, if so, who here
Dialogue: 0,0:04:47.95,0:04:54.75,Default,,0000,0000,0000,,has a physics background some more than\Nexpected. Great. Because those of you who
Dialogue: 0,0:04:54.75,0:04:59.14,Default,,0000,0000,0000,,just raised their hands would not be\Nsurprised that we expect a square
Dialogue: 0,0:04:59.14,0:05:03.40,Default,,0000,0000,0000,,relationship between the radial\Naccaleration and angular velocity. Those
Dialogue: 0,0:05:03.40,0:05:07.86,Default,,0000,0000,0000,,of you who do not know will believe me\Nfrom this plot where on the x axis we've
Dialogue: 0,0:05:07.86,0:05:13.11,Default,,0000,0000,0000,,got the angle of velocity squared and on\Nthe y axis the radial acceleration we get
Dialogue: 0,0:05:13.11,0:05:17.70,Default,,0000,0000,0000,,a straight line and that's what you would\Nexpect. So besides the physics, because
Dialogue: 0,0:05:17.70,0:05:21.64,Default,,0000,0000,0000,,this is not that much about the physics.\NThis is a simple experiment all our
Dialogue: 0,0:05:21.64,0:05:25.86,Default,,0000,0000,0000,,students could do and actually they ge, we\Ngave them this assignment. We gave them
Dialogue: 0,0:05:25.86,0:05:31.12,Default,,0000,0000,0000,,also a bonus point if they created a\Nvideo. Don't worry. Their consent to that
Dialogue: 0,0:05:31.12,0:05:34.34,Default,,0000,0000,0000,,we use the video was not related to the\Npoint, they first got the point and then
Dialogue: 0,0:05:34.34,0:05:39.16,Default,,0000,0000,0000,,we asked for their consent to use the\Nvideo. And we learned two things from
Dialogue: 0,0:05:39.16,0:05:43.53,Default,,0000,0000,0000,,these videos. A Our students do not really\Nhave salad spinners. they've got bicycles
Dialogue: 0,0:05:43.53,0:05:48.76,Default,,0000,0000,0000,,and office chairs, but b and that was the\Nmost important thing. It looks like I
Dialogue: 0,0:05:48.76,0:05:54.99,Default,,0000,0000,0000,,mean, these are from this year where we\Ngot almost 100 videos they we actually
Dialogue: 0,0:05:54.99,0:05:58.31,Default,,0000,0000,0000,,could trigger them to go out, search for\Nsomething where they've got the
Dialogue: 0,0:05:58.31,0:06:04.52,Default,,0000,0000,0000,,rotationary movement and they could repeat\Nthis experiment. Ok. Another example which
Dialogue: 0,0:06:04.52,0:06:09.18,Default,,0000,0000,0000,,actually changed just the course of the\Nlecture a little bit is a situation where
Dialogue: 0,0:06:09.18,0:06:14.39,Default,,0000,0000,0000,,we first give the assignment before we\Nactually let them, before we actually
Dialogue: 0,0:06:14.39,0:06:18.97,Default,,0000,0000,0000,,discuss the theory behind this, which\Nmeans in this example, this is a little bit
Dialogue: 0,0:06:18.97,0:06:25.05,Default,,0000,0000,0000,,older because we did not get there yet this\Nyear, we assigned our students to build
Dialogue: 0,0:06:25.05,0:06:30.49,Default,,0000,0000,0000,,string pendulums. They look very similar\Nbecause we were very precise about how
Dialogue: 0,0:06:30.49,0:06:35.27,Default,,0000,0000,0000,,they should build them. And then we had an\Nonline form where they could submit the
Dialogue: 0,0:06:35.27,0:06:39.08,Default,,0000,0000,0000,,length of their pendulum and the frequency\Nthey received from it that they measured
Dialogue: 0,0:06:39.08,0:06:44.48,Default,,0000,0000,0000,,with the pendulum. They should do this for\Nthree different cases. And the idea was
Dialogue: 0,0:06:44.48,0:06:48.92,Default,,0000,0000,0000,,that we did this assignment long before we\Ndiscussed the pendulum in the lecture so
Dialogue: 0,0:06:48.92,0:06:53.13,Default,,0000,0000,0000,,that they have got a little bit of\Nresearch experience. And after we
Dialogue: 0,0:06:53.13,0:06:57.38,Default,,0000,0000,0000,,collected all the data from them, then the\Nlecture would discuss the pendulum. So the
Dialogue: 0,0:06:57.38,0:07:01.63,Default,,0000,0000,0000,,physicists were there now. We do a small\Nangle, approximation solving differential
Dialogue: 0,0:07:01.63,0:07:05.33,Default,,0000,0000,0000,,equation. All this theory stuff. And in\Nthe end we were done, we could tell our
Dialogue: 0,0:07:05.33,0:07:09.41,Default,,0000,0000,0000,,students, well, we do not have to do this\Nexperiment on stage. Now, because all of
Dialogue: 0,0:07:09.41,0:07:13.68,Default,,0000,0000,0000,,you did this experiment and we simply can\Ncompare the theory that we just arrived
Dialogue: 0,0:07:13.68,0:07:19.45,Default,,0000,0000,0000,,with your data. And it worked out quite\Nwell. So you see most of the white points,
Dialogue: 0,0:07:19.45,0:07:23.22,Default,,0000,0000,0000,,which is the data from the students\Nmatches the theory, which is the orange
Dialogue: 0,0:07:23.22,0:07:27.77,Default,,0000,0000,0000,,line, except maybe for those three who\Nshould proceed on a career of theoretical
Dialogue: 0,0:07:27.77,0:07:33.29,Default,,0000,0000,0000,,physics. But yeah, so this is all\Nsomething got nice feedback from and this
Dialogue: 0,0:07:33.29,0:07:37.54,Default,,0000,0000,0000,,is in principle how we use the app and\Nwhat it's designed for. There are also of
Dialogue: 0,0:07:37.54,0:07:41.51,Default,,0000,0000,0000,,course many applications in school by now.\NMore teachers use this in school than we
Dialogue: 0,0:07:41.51,0:07:45.82,Default,,0000,0000,0000,,use it at the university. So we take this\Ninto consideration as well. But that's the
Dialogue: 0,0:07:45.82,0:07:49.45,Default,,0000,0000,0000,,reason that I am standing here talking\Nabout the sensors in the smartphone.
Dialogue: 0,0:07:49.45,0:07:54.72,Default,,0000,0000,0000,,That's the reason that I am trying to\Naccess them. so let's have a look at what
Dialogue: 0,0:07:54.72,0:07:58.92,Default,,0000,0000,0000,,sensors we actually have in our phone. I\Nthink the first one that most of you would
Dialogue: 0,0:07:58.92,0:08:04.12,Default,,0000,0000,0000,,think often talking about sensors besides\Nobvious stuff like the microphone would be
Dialogue: 0,0:08:04.12,0:08:10.19,Default,,0000,0000,0000,,the accelerometer. So I think yeah, I\Nthink I first explain how the
Dialogue: 0,0:08:10.19,0:08:13.90,Default,,0000,0000,0000,,accelerometer works. OK, so the\Naccelerometer in your phone is actually a
Dialogue: 0,0:08:13.90,0:08:18.17,Default,,0000,0000,0000,,so-called MEMS device. MEMS is M E M S\Nstands for Micro Electrical Mechanical
Dialogue: 0,0:08:18.17,0:08:24.01,Default,,0000,0000,0000,,System and it looks roughly like this.\NIt's a simplification. If you search for
Dialogue: 0,0:08:24.01,0:08:28.65,Default,,0000,0000,0000,,actual MEMS devices, simply search for M E\NM S and accelerometer and you find some
Dialogue: 0,0:08:28.65,0:08:31.27,Default,,0000,0000,0000,,pictures. They usually are a little bit\Nmore complicated, although the
Dialogue: 0,0:08:31.27,0:08:35.57,Default,,0000,0000,0000,,accelerometer is not that much more\Ncomplicated. It consists of an orange
Dialogue: 0,0:08:35.57,0:08:40.28,Default,,0000,0000,0000,,case. Yeah, well so far so obvious, but\Nalso two contacts. The blue and the red
Dialogue: 0,0:08:40.28,0:08:45.23,Default,,0000,0000,0000,,one and important part is this silvery\Nstructure here or the metallic structure
Dialogue: 0,0:08:45.23,0:08:49.13,Default,,0000,0000,0000,,which is under etched its bit hard to see\Non this picture, but it's actually
Dialogue: 0,0:08:49.13,0:08:53.21,Default,,0000,0000,0000,,floating. It's only attached to the sides,\Nyou see light in between here. So if you
Dialogue: 0,0:08:53.21,0:08:59.36,Default,,0000,0000,0000,,move around, the accelerometer the inner\Npath, can actually move. So let's do this.
Dialogue: 0,0:08:59.36,0:09:03.58,Default,,0000,0000,0000,,So at each point where the device is extra\Nriding in one direction or the other
Dialogue: 0,0:09:03.58,0:09:11.34,Default,,0000,0000,0000,,direction, due to inertia the that the\Nmetallic part in here is distorted, moved
Dialogue: 0,0:09:11.34,0:09:16.09,Default,,0000,0000,0000,,into one direction and we can measure the\Namount by which it is deflected by this
Dialogue: 0,0:09:16.09,0:09:20.98,Default,,0000,0000,0000,,movement with the two contacts by\Nmeasuring the capacity between these
Dialogue: 0,0:09:20.98,0:09:26.46,Default,,0000,0000,0000,,structures. So that's the principle of the\Naccelerometer. One thing to mention at
Dialogue: 0,0:09:26.46,0:09:32.88,Default,,0000,0000,0000,,this point is that it's in the sense of\Nphysics. It does not really only measure
Dialogue: 0,0:09:32.88,0:09:36.85,Default,,0000,0000,0000,,acceleration. It measures acceleration you\Nsee in this image of the device
Dialogue: 0,0:09:36.85,0:09:41.17,Default,,0000,0000,0000,,accelerating. We get some data, but if you\Nimagine we take this device and rotate it
Dialogue: 0,0:09:41.17,0:09:47.27,Default,,0000,0000,0000,,like this, then of course you also get a\Ndeflection of the of the metallic part by
Dialogue: 0,0:09:47.27,0:09:51.03,Default,,0000,0000,0000,,gravity. So gravity is pulling it down as\Nwell. And that's the main reason the
Dialogue: 0,0:09:51.03,0:09:55.84,Default,,0000,0000,0000,,accelerometer is in there because the\Ndevelopers and manufacturers of the phones
Dialogue: 0,0:09:55.84,0:09:59.95,Default,,0000,0000,0000,,are not really interested in measuring\Nacceleration, at least there aren't that
Dialogue: 0,0:09:59.95,0:10:04.29,Default,,0000,0000,0000,,many use cases for it. But instead, what\Nthey want to have is an indication on
Dialogue: 0,0:10:04.29,0:10:08.45,Default,,0000,0000,0000,,which direction is down or which direction\Nis up. So when you rotate the screen of
Dialogue: 0,0:10:08.45,0:10:12.74,Default,,0000,0000,0000,,your phone, actually they can rotate the\Ncontent of the phone as well or with this
Dialogue: 0,0:10:12.74,0:10:17.69,Default,,0000,0000,0000,,you can also then control video games by\Ntilting your phone and stuff like this.
Dialogue: 0,0:10:17.69,0:10:24.08,Default,,0000,0000,0000,,Because gravity also deflects the\Naccelerometers. Earth's acceleration,
Dialogue: 0,0:10:24.08,0:10:27.09,Default,,0000,0000,0000,,which you try to avoid because from\Ndidactic point of view, this is a
Dialogue: 0,0:10:27.09,0:10:33.91,Default,,0000,0000,0000,,nightmare to distinguish these both. But\Nthe point is that we can detect rotations
Dialogue: 0,0:10:33.91,0:10:39.08,Default,,0000,0000,0000,,like this and this is pretty much in every\Nphone. I mean, this is not really a
Dialogue: 0,0:10:39.08,0:10:43.00,Default,,0000,0000,0000,,statistic. This is just the first pie\Nchart we have about availability. I have
Dialogue: 0,0:10:43.00,0:10:48.40,Default,,0000,0000,0000,,never encountered a single phone or tablet\Nthat does not have an accelerometer. So if
Dialogue: 0,0:10:48.40,0:10:53.73,Default,,0000,0000,0000,,anyone ever encountered some special\Ndevice, some very unique device that
Dialogue: 0,0:10:53.73,0:10:57.27,Default,,0000,0000,0000,,doesn't have one. Let me know because I\Nwould be interested in this at least. I do
Dialogue: 0,0:10:57.27,0:11:00.99,Default,,0000,0000,0000,,not know of any device on which phyphox\Nactually runs, which doesn't have an
Dialogue: 0,0:11:00.99,0:11:06.92,Default,,0000,0000,0000,,accelerometer. A bit more interesting is\Nwhich data rate we can achieve. So most
Dialogue: 0,0:11:06.92,0:11:12.18,Default,,0000,0000,0000,,accelerometers have several hundred\Nsamples per second. Actually the fastest
Dialogue: 0,0:11:12.18,0:11:18.85,Default,,0000,0000,0000,,ones go up to 500 hertz and but there are\Nalso many devices that only do one hundred
Dialogue: 0,0:11:18.85,0:11:23.90,Default,,0000,0000,0000,,hertz That's 100 values per second. These\Nare mostly the cheaper Android devices and
Dialogue: 0,0:11:23.90,0:11:28.83,Default,,0000,0000,0000,,all the iPhones. So I think the internal\Naccelerometer will do more on an iPhone.
Dialogue: 0,0:11:28.83,0:11:34.63,Default,,0000,0000,0000,,But I have to admit, at some point I can\Nunderstand why they might limit this. But
Dialogue: 0,0:11:34.63,0:11:39.00,Default,,0000,0000,0000,,on an iPhone, you get 100 hertz. That's\Nthe limit. From the API, what you can get
Dialogue: 0,0:11:39.00,0:11:46.41,Default,,0000,0000,0000,,there. But this is actually quite a lot. I\Nwill later see what we can do with this.
Dialogue: 0,0:11:46.41,0:11:50.45,Default,,0000,0000,0000,,And one other point about calibration of\Nthis thing. Actually for all the sensors
Dialogue: 0,0:11:50.45,0:11:54.90,Default,,0000,0000,0000,,to get reasonable units from the system so\Nthe acceleration is given in meter per
Dialogue: 0,0:11:54.90,0:12:00.02,Default,,0000,0000,0000,,square second. I just realized that if I\Nget the units, that's something I would
Dialogue: 0,0:12:00.02,0:12:05.58,Default,,0000,0000,0000,,really tell my students. But yeah. So on\Nthe x axis, it's a meter per square second
Dialogue: 0,0:12:05.58,0:12:11.38,Default,,0000,0000,0000,,and you see that as a wide range of values\Nthat you get there. So this data is from
Dialogue: 0,0:12:11.38,0:12:14.61,Default,,0000,0000,0000,,our sensor database. I would mention it\Nlater as well. This is contribution from
Dialogue: 0,0:12:14.61,0:12:19.33,Default,,0000,0000,0000,,our users what data there this. This only\Nabsolute value that we get from resting
Dialogue: 0,0:12:19.33,0:12:23.79,Default,,0000,0000,0000,,phones and we would expect nine point\Neight one meter per square second for
Dialogue: 0,0:12:23.79,0:12:29.03,Default,,0000,0000,0000,,earth acceleration. There are some local\Nvariation, but not on that scale. So do
Dialogue: 0,0:12:29.03,0:12:33.11,Default,,0000,0000,0000,,not expect your sensors to be well\Ncalibrated. Also, if you've got any app
Dialogue: 0,0:12:33.11,0:12:37.80,Default,,0000,0000,0000,,that tells you you can push a single\Nbutton and then calibrates your sensor,
Dialogue: 0,0:12:37.80,0:12:42.91,Default,,0000,0000,0000,,don't trust it. It's not that simple.\NThese sensors may have different errors on
Dialogue: 0,0:12:42.91,0:12:48.51,Default,,0000,0000,0000,,each axis. They're all 3D sensors we've\Ngot an X, Y and z axis. These errors can
Dialogue: 0,0:12:48.51,0:12:52.36,Default,,0000,0000,0000,,be linear errors so you have to multiply a\Ncorrection. It could be an offset. So it
Dialogue: 0,0:12:52.36,0:12:56.68,Default,,0000,0000,0000,,would have to add an correction. And on\Ntop of this, the entire device could be
Dialogue: 0,0:12:56.68,0:13:01.18,Default,,0000,0000,0000,,tilted within your phone. OK. So actually,\Nif you look into the data sheets of the
Dialogue: 0,0:13:01.18,0:13:06.61,Default,,0000,0000,0000,,accelerometer, they have some tolerance on\Nhow much they might already be shifted or
Dialogue: 0,0:13:06.61,0:13:10.81,Default,,0000,0000,0000,,rotated within the package. And when\Nsoldering it into the phone, I would
Dialogue: 0,0:13:10.81,0:13:15.03,Default,,0000,0000,0000,,assume there will be an additional error.\NI've seen so many different errors on
Dialogue: 0,0:13:15.03,0:13:19.57,Default,,0000,0000,0000,,different phones. It's not that easy to\Nsimply calibrate that. But let me give you
Dialogue: 0,0:13:19.57,0:13:24.59,Default,,0000,0000,0000,,an example of what you can do with it. Or\Njust a quick look first. So we see in our
Dialogue: 0,0:13:24.59,0:13:29.36,Default,,0000,0000,0000,,app. Yeah. So this is phyphox. OK. Thank\Nyou. Got this. You have an entry
Dialogue: 0,0:13:29.36,0:13:34.37,Default,,0000,0000,0000,,acceleration with G. That's the extra raw\Ndata from the sensor or as raw as we get
Dialogue: 0,0:13:34.37,0:13:41.47,Default,,0000,0000,0000,,it. If I started you see if I shake it,\Nyou get some readings there. It's fast.
Dialogue: 0,0:13:41.47,0:13:46.45,Default,,0000,0000,0000,,It's already great. You can apply to\Npendulum and measure the acceleration of
Dialogue: 0,0:13:46.45,0:13:51.55,Default,,0000,0000,0000,,the pendulum like this. But something I\Nwant to demonstrate is that we can also
Dialogue: 0,0:13:51.55,0:13:56.37,Default,,0000,0000,0000,,get the frequencies from this data by\Ndoing a fourier transform and calculating
Dialogue: 0,0:13:56.37,0:14:01.66,Default,,0000,0000,0000,,the frequency spectrum of this exploration\Ndata and to demonstrate this I brought a
Dialogue: 0,0:14:01.66,0:14:07.81,Default,,0000,0000,0000,,little device a old hard disk drive. It says\Nit's broken, but it's still rotating and
Dialogue: 0,0:14:07.81,0:14:13.51,Default,,0000,0000,0000,,that's important part for us. So if I\Nplace my phone on top of it, start the
Dialogue: 0,0:14:13.51,0:14:19.78,Default,,0000,0000,0000,,measurement. Turn on the hard disk drive.\NAnd then you see a peak showing up in the
Dialogue: 0,0:14:19.78,0:14:24.74,Default,,0000,0000,0000,,spectrum and it settles at 120 hertz. If\Nyou don't believe me. Unfortunately, we
Dialogue: 0,0:14:24.74,0:14:28.16,Default,,0000,0000,0000,,don't have a camera here right now. You\Ncan later have a look. It's supposed to
Dialogue: 0,0:14:28.16,0:14:34.35,Default,,0000,0000,0000,,run at seven thousand two hundred RPM,\Nwhich is 120 hertz. We can even get a time
Dialogue: 0,0:14:34.35,0:14:38.57,Default,,0000,0000,0000,,resolution of this. So if I turn it off\Nagain, you see how the frequency drops
Dialogue: 0,0:14:38.57,0:14:46.67,Default,,0000,0000,0000,,down. And if I turn it on again. There it\Ncomes up again. OK. So this an example of
Dialogue: 0,0:14:46.67,0:14:49.33,Default,,0000,0000,0000,,what you can do. It's great for students\Nthat can check if the washing machine at
Dialogue: 0,0:14:49.33,0:14:53.64,Default,,0000,0000,0000,,home is working properly or they can\Ncheck other things. But usually I do not
Dialogue: 0,0:14:53.64,0:15:00.01,Default,,0000,0000,0000,,like to bring washing machines to talks.\NSo I used the hard disk drive here. One
Dialogue: 0,0:15:00.01,0:15:04.15,Default,,0000,0000,0000,,other thing you might have noticed before\Nis that we've actually got acceleration
Dialogue: 0,0:15:04.15,0:15:10.44,Default,,0000,0000,0000,,with G and acceleration without G. The\Nsecond one is actually a sensor that
Dialogue: 0,0:15:10.44,0:15:15.92,Default,,0000,0000,0000,,removes Earth's gravity. So if I start the\None with G, you will notice that down here
Dialogue: 0,0:15:15.92,0:15:21.61,Default,,0000,0000,0000,,on the Z, the axis you still have the 9.81\Nmeter per square second, which is great
Dialogue: 0,0:15:21.61,0:15:25.78,Default,,0000,0000,0000,,because if i rotate the phone. This\Ncontribution goes to other axis and we can
Dialogue: 0,0:15:25.78,0:15:29.49,Default,,0000,0000,0000,,determine the orientation of the phone.\NBut this is bad actually for dedactics
Dialogue: 0,0:15:29.49,0:15:33.46,Default,,0000,0000,0000,,because actually the phone is resting.\NIt's not moving at all. There's no
Dialogue: 0,0:15:33.46,0:15:37.33,Default,,0000,0000,0000,,velocity involved. There's no\Nacceleration. So luckily, there's also an
Dialogue: 0,0:15:37.33,0:15:42.58,Default,,0000,0000,0000,,acceleration without G, which gives us\Nroughly 0 an all axis unless I actually
Dialogue: 0,0:15:42.58,0:15:47.75,Default,,0000,0000,0000,,accelerate this thing. Problem with this\Nis this is only a virtual sensor. This is
Dialogue: 0,0:15:47.75,0:15:52.37,Default,,0000,0000,0000,,a sensor that's fusing the data from the\Naccelerometer with an additional sensor
Dialogue: 0,0:15:52.37,0:15:57.06,Default,,0000,0000,0000,,like the gyroscope. So we can actually\Ndistinguish between rotating the phone or
Dialogue: 0,0:15:57.06,0:16:01.76,Default,,0000,0000,0000,,accelerating it in one or the other\Ndirection. Usually you only get
Dialogue: 0,0:16:01.76,0:16:06.76,Default,,0000,0000,0000,,acceleration without G. If you also have a\Ngyroscope in your phone, I've seen two or
Dialogue: 0,0:16:06.76,0:16:10.48,Default,,0000,0000,0000,,three devices that offer you acceleration\Nwithout G, even though they don't have a
Dialogue: 0,0:16:10.48,0:16:16.78,Default,,0000,0000,0000,,gyroscope. This case, don't trust them.\NThis is merely guessing. OK. So it's. They
Dialogue: 0,0:16:16.78,0:16:23.73,Default,,0000,0000,0000,,probably have only low frequency filter\Non top of this, or they're averaging out
Dialogue: 0,0:16:23.73,0:16:29.89,Default,,0000,0000,0000,,your movement and this doesn't really work\Nfor anything. Yeah so that's the
Dialogue: 0,0:16:29.89,0:16:34.30,Default,,0000,0000,0000,,accelerometer or one other thing I want to\Nmention is if you look into the API to
Dialogue: 0,0:16:34.30,0:16:37.75,Default,,0000,0000,0000,,access the sensors yourself for some\Nreason you will notice acceleration
Dialogue: 0,0:16:37.75,0:16:41.68,Default,,0000,0000,0000,,without G is usually called linear\Nacceleration in our app since it's made
Dialogue: 0,0:16:41.68,0:16:45.36,Default,,0000,0000,0000,,for teaching. We decided to call it with\Nand without G. So if you find
Dialogue: 0,0:16:45.36,0:16:49.41,Default,,0000,0000,0000,,accelerometer, that's the one with G and\Nlinear acceleration is the one without G.
Dialogue: 0,0:16:49.41,0:16:54.72,Default,,0000,0000,0000,,If you look at other apps or the API.\NOkay. Next up, I already mentioned this
Dialogue: 0,0:16:54.72,0:17:00.22,Default,,0000,0000,0000,,one is a gyroscope. If you have, some\Nphysics background. Then when you think of
Dialogue: 0,0:17:00.22,0:17:05.19,Default,,0000,0000,0000,,a gyroscope, you're thinking of a device\Nthat's spinning fast so it has some angular
Dialogue: 0,0:17:05.19,0:17:10.36,Default,,0000,0000,0000,,momentum and then usually you want it to\Nbe heavy and to have the weight at the
Dialogue: 0,0:17:10.36,0:17:15.91,Default,,0000,0000,0000,,large radius. We've got a strong moment of\Ninertia so that you get when it's spinning
Dialogue: 0,0:17:15.91,0:17:20.65,Default,,0000,0000,0000,,fast, a strong, angular momentum and due\Nto the conservation of angular momentum.
Dialogue: 0,0:17:20.65,0:17:24.72,Default,,0000,0000,0000,,These spinning devices can keep an axis\Nregardless of rotating the frame around
Dialogue: 0,0:17:24.72,0:17:30.12,Default,,0000,0000,0000,,it. That's what I was thinking about, a\Ngyroscope of what I think is a gyroscope.
Dialogue: 0,0:17:30.12,0:17:34.67,Default,,0000,0000,0000,,When you just give me the term out of\Ncontext, of course, a heavy, huge, fast
Dialogue: 0,0:17:34.67,0:17:38.18,Default,,0000,0000,0000,,spinning device is the last thing you want\Nin your phone. So that's not what's meant
Dialogue: 0,0:17:38.18,0:17:42.66,Default,,0000,0000,0000,,with the gyroscope when people are talking\Nabout gyroscopes in your phone. Instead
Dialogue: 0,0:17:42.66,0:17:48.48,Default,,0000,0000,0000,,there again you have a MEMS device. So\Nagain, micro electromechanical system. You
Dialogue: 0,0:17:48.48,0:17:52.50,Default,,0000,0000,0000,,notice this looks almost exactly like\Naccelerometer. If you look for real
Dialogue: 0,0:17:52.50,0:17:56.13,Default,,0000,0000,0000,,devices, those are actually much more\Ncomplicated because they need some
Dialogue: 0,0:17:56.13,0:18:02.46,Default,,0000,0000,0000,,specific geometry to make sure that they\Ndo not act like an accelerometer. But the
Dialogue: 0,0:18:02.46,0:18:08.29,Default,,0000,0000,0000,,principle is easy to explain with the same\Ngeometry. So we again have this floating
Dialogue: 0,0:18:08.29,0:18:13.89,Default,,0000,0000,0000,,metallic part and we've got 2 contacts. So\Nagain, we've got a part that can wobble in
Dialogue: 0,0:18:13.89,0:18:18.30,Default,,0000,0000,0000,,this direction here. But on top of this,\Nwe've got the motion that's perpendicular
Dialogue: 0,0:18:18.30,0:18:22.54,Default,,0000,0000,0000,,to this. So this is now not depicting the\Nmotion of your phone, but this is
Dialogue: 0,0:18:22.54,0:18:27.15,Default,,0000,0000,0000,,depicting a vibration that the gyroscope\Ndoes by itself all the time. So there are
Dialogue: 0,0:18:27.15,0:18:30.76,Default,,0000,0000,0000,,different ways to build them. Some have a\Nrotary motion, some have this linear
Dialogue: 0,0:18:30.76,0:18:34.99,Default,,0000,0000,0000,,motion. Also, the way to create this\Nmotion makes this device so much more
Dialogue: 0,0:18:34.99,0:18:39.48,Default,,0000,0000,0000,,complicated. But in principle, it's a\Nsimilar structure which is vibrating forth
Dialogue: 0,0:18:39.48,0:18:46.28,Default,,0000,0000,0000,,and back and now if you add rotation to\Nit. It's a little bit hard to see it as
Dialogue: 0,0:18:46.28,0:18:50.89,Default,,0000,0000,0000,,it's rotating the inner part now suddenly\Ngets deflected. That's changed, right?
Dialogue: 0,0:18:50.89,0:18:55.10,Default,,0000,0000,0000,,Frame of reference. So let's get the\Ncamera in sync with this device. What you
Dialogue: 0,0:18:55.10,0:19:00.01,Default,,0000,0000,0000,,now see is that the inner part is moving\Nleft and right, although the device itself
Dialogue: 0,0:19:00.01,0:19:06.26,Default,,0000,0000,0000,,is only moving up and down. And the reason\Nis I don't want to deduce it entirely
Dialogue: 0,0:19:06.26,0:19:09.90,Default,,0000,0000,0000,,here, but most of you probably have heard\Nof it. This is the Coriolis effect. So,
Dialogue: 0,0:19:09.90,0:19:14.65,Default,,0000,0000,0000,,yes, in fact, your phone is determining\Nthe rotation rate of your phone, not the
Dialogue: 0,0:19:14.65,0:19:18.46,Default,,0000,0000,0000,,actual angle, but the rotation rate or\Nangular velocity due to the coriolis
Dialogue: 0,0:19:18.46,0:19:24.71,Default,,0000,0000,0000,,effect, which is just mind blowing if you\Ndo some of the calculations. There are
Dialogue: 0,0:19:24.71,0:19:32.20,Default,,0000,0000,0000,,some manufacturers on the Internet which\Nclaim that they can detect a movement of
Dialogue: 0,0:19:32.20,0:19:36.67,Default,,0000,0000,0000,,the order of magnitude of a single atom.\NAnd I believe them because we use similar
Dialogue: 0,0:19:36.67,0:19:41.80,Default,,0000,0000,0000,,structures in solid state physics. So\Nthat's possible. If you want to try it,
Dialogue: 0,0:19:41.80,0:19:47.08,Default,,0000,0000,0000,,just turn on the gyroscope on your phone.\NAnd do slight rotation like this, which is
Dialogue: 0,0:19:47.08,0:19:52.98,Default,,0000,0000,0000,,about the Z axis, one perpendicular to the\Ndisplay, you can detect really slow
Dialogue: 0,0:19:52.98,0:19:58.09,Default,,0000,0000,0000,,rotations with this. And think about the\Nfact that this is done using the coriolis
Dialogue: 0,0:19:58.09,0:20:04.48,Default,,0000,0000,0000,,effect and it's just mind blowing I think.\NSo this sensor is a bit more available.
Dialogue: 0,0:20:04.48,0:20:10.37,Default,,0000,0000,0000,,Actually, almost 80 percent of the phones\Nhave them. This has become significantly
Dialogue: 0,0:20:10.37,0:20:15.37,Default,,0000,0000,0000,,more since Pokémon GO. The reason is\Nwhen this game came up, suddenly people
Dialogue: 0,0:20:15.37,0:20:19.04,Default,,0000,0000,0000,,noticed that there's a device called the\Ngyroscope. And if it's not present, they
Dialogue: 0,0:20:19.04,0:20:22.54,Default,,0000,0000,0000,,did not have this AR mode where you can\Nactually take pictures of the nice cute
Dialogue: 0,0:20:22.54,0:20:27.65,Default,,0000,0000,0000,,Pokémon and so on. So this is when the\Nmany people noticed it and the
Dialogue: 0,0:20:27.65,0:20:32.32,Default,,0000,0000,0000,,manufacturers decided, OK, let's just\Nthrow in the gyroscope as well, because
Dialogue: 0,0:20:32.32,0:20:35.44,Default,,0000,0000,0000,,it's not that expensive, in fact, usually\Nit's on the same chip as the
Dialogue: 0,0:20:35.44,0:20:39.29,Default,,0000,0000,0000,,accelerometer. Then they're sold as one\Nthing it's an IMU - Inertia Measurement
Dialogue: 0,0:20:39.29,0:20:45.39,Default,,0000,0000,0000,,Unit not important at home, but so it's\Nquite a common thing. And the sensor rates
Dialogue: 0,0:20:45.39,0:20:50.81,Default,,0000,0000,0000,,look pretty much the same. You mostly\Nnotice the dip in the 100 hertz regime
Dialogue: 0,0:20:50.81,0:20:54.61,Default,,0000,0000,0000,,because those are the real cheap phones,\Nwhich then also don't have a gyroscope.
Dialogue: 0,0:20:54.61,0:20:59.62,Default,,0000,0000,0000,,But most of the phones achieve higher\Nrates. Again, since we were laughing
Dialogue: 0,0:20:59.62,0:21:05.51,Default,,0000,0000,0000,,before the iPhones also are here again at\Nthe 100 hertz. Wouldn't make sense to have
Dialogue: 0,0:21:05.51,0:21:10.81,Default,,0000,0000,0000,,the gyroscope faster at this point. Yeah,\Nbut that's it about the gyroscope you've
Dialogue: 0,0:21:10.81,0:21:16.46,Default,,0000,0000,0000,,seen it in action in the salad spinner.\NAnd that's one of the sensors you do not
Dialogue: 0,0:21:16.46,0:21:21.95,Default,,0000,0000,0000,,really see that often directly, but were\Njust mostly there to assist other things
Dialogue: 0,0:21:21.95,0:21:28.46,Default,,0000,0000,0000,,that you do where you need to get smooth\Nmotion like controlling games, AR . And
Dialogue: 0,0:21:28.46,0:21:34.41,Default,,0000,0000,0000,,actually removing the Earth's acceleration\Nfrom the accelerometer. Next up is a
Dialogue: 0,0:21:34.41,0:21:39.40,Default,,0000,0000,0000,,magnetometer, which I think is a more\Nobvious sensor because that's your compass
Dialogue: 0,0:21:39.40,0:21:45.35,Default,,0000,0000,0000,,in your device. So when you're doing\Nnavigation with a GPS in your car, it's a
Dialogue: 0,0:21:45.35,0:21:49.54,Default,,0000,0000,0000,,simple thing. GPS gets a position, you get\Na sequence of position as you going and
Dialogue: 0,0:21:49.54,0:21:52.48,Default,,0000,0000,0000,,from the sequence of the positions you\Nget, the direction you're moving in your
Dialogue: 0,0:21:52.48,0:21:56.53,Default,,0000,0000,0000,,car and your phone is attached to the\Ndashboard at least i hope so. So it's
Dialogue: 0,0:21:56.53,0:21:59.42,Default,,0000,0000,0000,,pointing in the same direction you're\Nmoving, everything's obvious. But if
Dialogue: 0,0:21:59.42,0:22:04.26,Default,,0000,0000,0000,,you're standing on an open space looking\Nfor not sure a train station or anything
Dialogue: 0,0:22:04.26,0:22:07.71,Default,,0000,0000,0000,,and you wondering which direction you want\Nto go from point of view of GPS, it's
Dialogue: 0,0:22:07.71,0:22:11.93,Default,,0000,0000,0000,,always the same position it doesn't get an\Norientation. You need a compass, which is
Dialogue: 0,0:22:11.93,0:22:17.35,Default,,0000,0000,0000,,the magnetometer. How do we get a compass\Non your phone? This is usually a hall
Dialogue: 0,0:22:17.35,0:22:23.21,Default,,0000,0000,0000,,sensor. A hall sensor is in principle just\Na conductor with charge carriers so these
Dialogue: 0,0:22:23.21,0:22:27.76,Default,,0000,0000,0000,,are the nice shiny white balls here\Ndrifting from one side to the other so
Dialogue: 0,0:22:27.76,0:22:31.86,Default,,0000,0000,0000,,it's just an electric current. And if you\Napply a magnetic field to an electric
Dialogue: 0,0:22:31.86,0:22:36.76,Default,,0000,0000,0000,,current or to any electric charge, then\Nthere is an effect. You might know from
Dialogue: 0,0:22:36.76,0:22:41.04,Default,,0000,0000,0000,,school, which is called the lorentz\Neffect. So there is a charge going one
Dialogue: 0,0:22:41.04,0:22:44.71,Default,,0000,0000,0000,,direction, you get the magnetic field\Nperpendicular to this and then the charge
Dialogue: 0,0:22:44.71,0:22:50.58,Default,,0000,0000,0000,,is deflected into a direction\Nperpendicular to the flying direction. And
Dialogue: 0,0:22:50.58,0:22:54.54,Default,,0000,0000,0000,,yeah, that's lorentz effect the older\Nguys, of you would know it from CRTs. If
Dialogue: 0,0:22:54.54,0:22:59.61,Default,,0000,0000,0000,,you bring a magnet close to a CRT, the\Nentire image is messed up due to this
Dialogue: 0,0:22:59.61,0:23:05.98,Default,,0000,0000,0000,,effect. And that's what we're using in\Nhall effect sensor or hall sensor you've
Dialogue: 0,0:23:05.98,0:23:12.06,Default,,0000,0000,0000,,got this electric current and if you bring\Na magnetic field close to it, the charge
Dialogue: 0,0:23:12.06,0:23:17.19,Default,,0000,0000,0000,,carriers are deflected to one side or the\Nother. And therefore, if you're measuring
Dialogue: 0,0:23:17.19,0:23:22.63,Default,,0000,0000,0000,,the voltage perpendicular to the flow of\Nthe count, you get. Yeah. You get an extra
Dialogue: 0,0:23:22.63,0:23:26.23,Default,,0000,0000,0000,,voltage that's proportional to the\Nmagnetic field. That's the hall effect.
Dialogue: 0,0:23:26.23,0:23:32.40,Default,,0000,0000,0000,,That's how your phone is able to determine\Nthe magnetic field. This one is even more
Dialogue: 0,0:23:32.40,0:23:37.30,Default,,0000,0000,0000,,common than the gyroscope simply because\Nit's used for navigation and people start
Dialogue: 0,0:23:37.30,0:23:42.33,Default,,0000,0000,0000,,to notice if it's not. If it's not present\Nand they do not get an orientation in the
Dialogue: 0,0:23:42.33,0:23:49.90,Default,,0000,0000,0000,,navigation software. But the actual rate\Nof the sensors is much slower than for the
Dialogue: 0,0:23:49.90,0:23:54.71,Default,,0000,0000,0000,,accelerometer. Most of them are running at\N100 Hertz. It will be important in two
Dialogue: 0,0:23:54.71,0:24:00.44,Default,,0000,0000,0000,,more slides. Besides that, there's not\Nthat much strange about the availability
Dialogue: 0,0:24:00.44,0:24:05.76,Default,,0000,0000,0000,,of this, but it's extremely sensitive\Nbecause it's supposed to measure Earth's
Dialogue: 0,0:24:05.76,0:24:10.07,Default,,0000,0000,0000,,magnetic field. Earth's magnetic field has\Nthe strength around 50 micro Tesla. This
Dialogue: 0,0:24:10.07,0:24:15.65,Default,,0000,0000,0000,,is not much actually if ever carried the\Nmagnetic magnet with you. Did you fear of
Dialogue: 0,0:24:15.65,0:24:19.92,Default,,0000,0000,0000,,some force from the Earth's magnetic\Nfield? Of course, it didn't need to build
Dialogue: 0,0:24:19.92,0:24:23.11,Default,,0000,0000,0000,,some compass where the needle is floating\Non something like this to actually get a
Dialogue: 0,0:24:23.11,0:24:28.28,Default,,0000,0000,0000,,rotation. It's a very weak field and\Nthat's good news and bad news as well,
Dialogue: 0,0:24:28.28,0:24:32.08,Default,,0000,0000,0000,,because on one hand, it's very sensitive.\Ndownside is it's very sensitive. Which
Dialogue: 0,0:24:32.08,0:24:36.07,Default,,0000,0000,0000,,means it saturates very early. If you want\Nto measure the magnetic field of an actual
Dialogue: 0,0:24:36.07,0:24:41.77,Default,,0000,0000,0000,,magnet. Don't even try it will saturate\Nright away. You do not get anything to
Dialogue: 0,0:24:41.77,0:24:47.75,Default,,0000,0000,0000,,demonstrate how how sensitive this\Nactually is. I've brought a flashlight, so
Dialogue: 0,0:24:47.75,0:24:55.09,Default,,0000,0000,0000,,a very simple one. And I switch to a modus\Nwhere we've got an s.o.s signal. That's
Dialogue: 0,0:24:55.09,0:25:00.68,Default,,0000,0000,0000,,coming up, a point in this direction and I\Nplace it next to the magnetometer in my
Dialogue: 0,0:25:00.68,0:25:05.91,Default,,0000,0000,0000,,phone. And yeah, you see right away so\Nmuch of his seeing the lights are pointing
Dialogue: 0,0:25:05.91,0:25:09.100,Default,,0000,0000,0000,,in this direction. You see the s.o.s\Nsignal popping up in the magnetic field
Dialogue: 0,0:25:09.100,0:25:14.63,Default,,0000,0000,0000,,reading simply because of the current\Ngoing through the LED. So that's what we
Dialogue: 0,0:25:14.63,0:25:17.91,Default,,0000,0000,0000,,call an Oersted-field. This is just the\Ntypical magnetic field you get from any
Dialogue: 0,0:25:17.91,0:25:22.74,Default,,0000,0000,0000,,current flowing. So I stop it. We got a nice\NSOS signal over there. Three short, three
Dialogue: 0,0:25:22.74,0:25:31.09,Default,,0000,0000,0000,,long and three short signals. And it's just\Ncoming from this simple flashlight. And
Dialogue: 0,0:25:31.09,0:25:36.63,Default,,0000,0000,0000,,this is also a good indicator on how\Nsensitive this thing is. I mean, if you
Dialogue: 0,0:25:36.63,0:25:43.58,Default,,0000,0000,0000,,place your phone in a case with a magnetic\Nsome magnetic closing mechanism, compass
Dialogue: 0,0:25:43.58,0:25:47.26,Default,,0000,0000,0000,,wouldn't work anymore. If you're not\Ncareful when paying your clothes and you
Dialogue: 0,0:25:47.26,0:25:51.75,Default,,0000,0000,0000,,place your phone on the big magnet that\Nremoves the theft protection from the
Dialogue: 0,0:25:51.75,0:25:55.86,Default,,0000,0000,0000,,clothes, something in your phone would get\Nmagnetized and would certainly be stronger
Dialogue: 0,0:25:55.86,0:26:01.11,Default,,0000,0000,0000,,than Earth's magnetic field. For the rest\Nof the day, your compass would be pointing
Dialogue: 0,0:26:01.11,0:26:06.74,Default,,0000,0000,0000,,in the wrong direction. Okay. Luckily,\Nusually the phones are able to notice this
Dialogue: 0,0:26:06.74,0:26:11.100,Default,,0000,0000,0000,,and they recalibrate the phone to simply\Nsubtract any constant fears. That again is
Dialogue: 0,0:26:11.100,0:26:15.26,Default,,0000,0000,0000,,bad. If you want to do absolute\Nmeasurements because you have not much
Dialogue: 0,0:26:15.26,0:26:22.70,Default,,0000,0000,0000,,control over the recalibration mechanism,\Nyou can access the raw data value. So if
Dialogue: 0,0:26:22.70,0:26:27.04,Default,,0000,0000,0000,,you folks there's a checkmark where you\Ncan disable the calibration, but then you
Dialogue: 0,0:26:27.04,0:26:30.85,Default,,0000,0000,0000,,have to do everything by hand. You will\Ncertainly have some background that's
Dialogue: 0,0:26:30.85,0:26:34.98,Default,,0000,0000,0000,,annoying. And one other thing, you should\Nalso take care and notice where your
Dialogue: 0,0:26:34.98,0:26:38.93,Default,,0000,0000,0000,,actual magnetometer is because in most\Nphones it's on top left corner, top right
Dialogue: 0,0:26:38.93,0:26:44.10,Default,,0000,0000,0000,,corner, top center. And this Pixel 3 is a\Nvery strange one. It has it on the right
Dialogue: 0,0:26:44.10,0:26:48.92,Default,,0000,0000,0000,,hand side, but it's never dead center. I\Nthink because of all the currents in the
Dialogue: 0,0:26:48.92,0:26:53.36,Default,,0000,0000,0000,,phone, I mean, you're charging your\Nbattery with three amps. How much you
Dialogue: 0,0:26:53.36,0:26:57.33,Default,,0000,0000,0000,,charge them now? This would yield a\Nstronger field than a flashlight and you
Dialogue: 0,0:26:57.33,0:27:03.12,Default,,0000,0000,0000,,would see it in the magnetometer again.\NNow for what you can do with this. So as
Dialogue: 0,0:27:03.12,0:27:07.49,Default,,0000,0000,0000,,little homework for all of you who came by\Ntrain yesterday, when I came here on the
Dialogue: 0,0:27:07.49,0:27:13.38,Default,,0000,0000,0000,,ICE, I turned on the magnetic spectrum,\Nthe same thing as the acceleration
Dialogue: 0,0:27:13.38,0:27:17.28,Default,,0000,0000,0000,,spectrum you just seen. And when you're\Ndoing it on train, you would see a peak at
Dialogue: 0,0:27:17.28,0:27:24.45,Default,,0000,0000,0000,,16.6 hertz. It might depend on your actual\Nseat. You might move it around a little
Dialogue: 0,0:27:24.45,0:27:31.19,Default,,0000,0000,0000,,bit. But so far I usually always saw this\Npeak. This is the electrification
Dialogue: 0,0:27:31.19,0:27:36.48,Default,,0000,0000,0000,,frequency of the German railway. So you\Ncan simply check if it's working properly.
Dialogue: 0,0:27:36.48,0:27:43.18,Default,,0000,0000,0000,,You should see 16.7 hertz. Okay. One other\Nthing that some of you might get in your
Dialogue: 0,0:27:43.18,0:27:46.35,Default,,0000,0000,0000,,head right now, that you could do this\Nwith simple electrical outlets. There you
Dialogue: 0,0:27:46.35,0:27:49.59,Default,,0000,0000,0000,,would get a problem with the rate. So\Nthat's what I mentioned, that the rate of
Dialogue: 0,0:27:49.59,0:27:55.80,Default,,0000,0000,0000,,the sensor is quite important. I also got\Nsomething via Twitter yesterday. Just as a
Dialogue: 0,0:27:55.80,0:27:58.29,Default,,0000,0000,0000,,response to the other one, I thought,\Nwell, I was looking for an example like
Dialogue: 0,0:27:58.29,0:28:02.49,Default,,0000,0000,0000,,this for this talk talk, so I just put it\Nin. This is a measurement of an American
Dialogue: 0,0:28:02.49,0:28:07.92,Default,,0000,0000,0000,,power outlet which is run at 60 hertz. But\Nthis guy is seeing 40 hertz and he was
Dialogue: 0,0:28:07.92,0:28:15.99,Default,,0000,0000,0000,,wondering about this. That's what's called\Naliasing. So the alias effect, sort of you
Dialogue: 0,0:28:15.99,0:28:19.83,Default,,0000,0000,0000,,might notice this from computer games.\NThey usually use it in slightly different
Dialogue: 0,0:28:19.83,0:28:25.62,Default,,0000,0000,0000,,context. The idea is if you're measuring a\Nfrequency that's higher than half of the
Dialogue: 0,0:28:25.62,0:28:29.57,Default,,0000,0000,0000,,data acquisition rate of your sensor. So\Nthis one is runningat 100 hertz like most
Dialogue: 0,0:28:29.57,0:28:35.76,Default,,0000,0000,0000,,of the phones do. Then half of this\Nfrequency is what's called the Nyquist-
Dialogue: 0,0:28:35.76,0:28:39.56,Default,,0000,0000,0000,,frequency. And you notice that the\Nspectrum goes from zero to the Nyquist-
Dialogue: 0,0:28:39.56,0:28:45.50,Default,,0000,0000,0000,,frequency. This is simple math, not simple\Nmath, but its maths. The roots of the
Dialogue: 0,0:28:45.50,0:28:49.31,Default,,0000,0000,0000,,fourier-transformation, you could say so.\NAnd if you try to detect a frequency
Dialogue: 0,0:28:49.31,0:28:53.77,Default,,0000,0000,0000,,that's higher than this, so an American\Npower outlet with 60 hertz, actually the
Dialogue: 0,0:28:53.77,0:29:00.32,Default,,0000,0000,0000,,higher frequency is showing up as on the\Nother side of this upper limit at 40
Dialogue: 0,0:29:00.32,0:29:03.31,Default,,0000,0000,0000,,hertz, even if you go to a higher\Nfrequency, it would shift down further and
Dialogue: 0,0:29:03.31,0:29:06.42,Default,,0000,0000,0000,,further until reaching zero and then it\Nwould shift up again. So if you're
Dialogue: 0,0:29:06.42,0:29:13.45,Default,,0000,0000,0000,,interested in this. Check out some\Narticles about aliasing. If you're not
Dialogue: 0,0:29:13.45,0:29:16.65,Default,,0000,0000,0000,,that interested in this. Just keep in\Nmind, if you're measuring frequencies that
Dialogue: 0,0:29:16.65,0:29:20.34,Default,,0000,0000,0000,,are higher than half your data acquisition\Nrate, you will not see the correct
Dialogue: 0,0:29:20.34,0:29:26.84,Default,,0000,0000,0000,,frequency. OK. Then one of my favorite\Nsensors, the pressure sensor for this one
Dialogue: 0,0:29:26.84,0:29:32.96,Default,,0000,0000,0000,,I need. Again, the phone. That's not on a\Nwire. Let me before before I show
Dialogue: 0,0:29:32.96,0:29:37.56,Default,,0000,0000,0000,,anything. Let me demonstrate what it can\Ndo, because that's something I find quite
Dialogue: 0,0:29:37.56,0:29:42.25,Default,,0000,0000,0000,,surprising. Let's turn on the measurement.\NBy the way, those who are wondering how
Dialogue: 0,0:29:42.25,0:29:47.97,Default,,0000,0000,0000,,this works. There's a function in phyphox,\Nwe call it remote access. It's basically a
Dialogue: 0,0:29:47.97,0:29:53.47,Default,,0000,0000,0000,,web server running in the app which\Nprovides the data so we can simply access
Dialogue: 0,0:29:53.47,0:29:59.05,Default,,0000,0000,0000,,the data on the phone to demonstrate or to\Ncontrol the measurement. And now here we
Dialogue: 0,0:29:59.05,0:30:04.05,Default,,0000,0000,0000,,see the pressure sensor. Right now, just\Nmostly noise or what I do now is I hold it
Dialogue: 0,0:30:04.05,0:30:07.77,Default,,0000,0000,0000,,up. And if we wait a few seconds, you\Nwould see that the pressure's actually
Dialogue: 0,0:30:07.77,0:30:15.45,Default,,0000,0000,0000,,dropping. It has dropped far enough. Then\NI place it on the ground and the pressure
Dialogue: 0,0:30:15.45,0:30:20.19,Default,,0000,0000,0000,,is rising again. So actually, your phone,\Nif it has a pressure sensor, has a
Dialogue: 0,0:30:20.19,0:30:25.85,Default,,0000,0000,0000,,pressure sensor that's sensitive enough.\NSo we turn it off to measure a change of
Dialogue: 0,0:30:25.85,0:30:33.28,Default,,0000,0000,0000,,pressure of a distance like this. OK. And\Nthat's again, when I first tried this, I
Dialogue: 0,0:30:33.28,0:30:39.53,Default,,0000,0000,0000,,repeated this test several times before,\Nbelieved it was just not by accident. And
Dialogue: 0,0:30:39.53,0:30:43.51,Default,,0000,0000,0000,,how do they do this? You have got another\Ndevice that actually has a cavity. So
Dialogue: 0,0:30:43.51,0:30:47.65,Default,,0000,0000,0000,,below the bluish gray part, there's a\Ncavity in there which is covered by a
Dialogue: 0,0:30:47.65,0:30:52.52,Default,,0000,0000,0000,,silicon membrane, which is the bluish\Npart. And if you change the pressure this
Dialogue: 0,0:30:52.52,0:30:57.53,Default,,0000,0000,0000,,simply moves it like you would expect from\Na membrane just in small. And to detect
Dialogue: 0,0:30:57.53,0:31:02.78,Default,,0000,0000,0000,,this movement, here is some material on\Ntop of this which changes its resistance.
Dialogue: 0,0:31:02.78,0:31:11.95,Default,,0000,0000,0000,,Or resistivity depending on the strain\Ncreated by morphing, dismembering. And
Dialogue: 0,0:31:11.95,0:31:15.58,Default,,0000,0000,0000,,unfortunately, this sensor is not that\Nmuch available. So about a third of the
Dialogue: 0,0:31:15.58,0:31:19.87,Default,,0000,0000,0000,,devices that we know of have the sensor.\NOf course, there's some bias in there from
Dialogue: 0,0:31:19.87,0:31:27.43,Default,,0000,0000,0000,,the users that submit data to us. This\Nmeans that, yes, these are usually the
Dialogue: 0,0:31:27.43,0:31:31.74,Default,,0000,0000,0000,,more expensive devices. So my rule of\Nthumb is if it's an iPhone, they usually
Dialogue: 0,0:31:31.74,0:31:36.16,Default,,0000,0000,0000,,have the pressure sensor except for the\NiPhone SE or some older models. If it's an
Dialogue: 0,0:31:36.16,0:31:39.70,Default,,0000,0000,0000,,Android, if you payed half as much as you\Npaid for an iPhone, then you have a good
Dialogue: 0,0:31:39.70,0:31:47.23,Default,,0000,0000,0000,,chance that you have to pressure sensor as\Nwell. But OK, that data rates? Yeah.
Dialogue: 0,0:31:47.23,0:31:54.58,Default,,0000,0000,0000,,Varies a lot. So the iPhones, like you\Njust saw the rate of about 1 Hertz. Most
Dialogue: 0,0:31:54.58,0:31:59.44,Default,,0000,0000,0000,,Android phones are on five, ten or twenty\Nfive hertz. I've never had a device like
Dialogue: 0,0:31:59.44,0:32:02.62,Default,,0000,0000,0000,,this in my hand. It does 100 hertz. I\Ndon't really believe that this makes sense
Dialogue: 0,0:32:02.62,0:32:09.72,Default,,0000,0000,0000,,because I already noticed on my phone that\NI think it does 25 hertz. Just handing it
Dialogue: 0,0:32:09.72,0:32:15.74,Default,,0000,0000,0000,,because of the sealed casing introduces\Nmore noise than you can actually use, at
Dialogue: 0,0:32:15.74,0:32:20.92,Default,,0000,0000,0000,,least for these small distances that I use\Nit for. But you can do other funny things
Dialogue: 0,0:32:20.92,0:32:26.49,Default,,0000,0000,0000,,with this. So this is something I received\Nby Dianna Cowern. You might know her as a
Dialogue: 0,0:32:26.49,0:32:32.26,Default,,0000,0000,0000,,YouTuber called "The Physics Girl". She\Nused a pressure measurement on the flight.
Dialogue: 0,0:32:32.26,0:32:35.51,Default,,0000,0000,0000,,It's something you should do anyways,\Nbecause that's the way you can figure out
Dialogue: 0,0:32:35.51,0:32:39.14,Default,,0000,0000,0000,,how much air you get to breathe up there.\NIt's much lower than you might expect.
Dialogue: 0,0:32:39.14,0:32:43.66,Default,,0000,0000,0000,,But she saw something else. So at some\Npoint she saw the drop in the pressure and
Dialogue: 0,0:32:43.66,0:32:48.35,Default,,0000,0000,0000,,increase again. And she asked her\Nfollowers, what could this be? And I'm not
Dialogue: 0,0:32:48.35,0:32:51.91,Default,,0000,0000,0000,,asking the audience right now. I just give\Nyou the solution. She wasn't lavatory and
Dialogue: 0,0:32:51.91,0:32:56.10,Default,,0000,0000,0000,,she flushed the toilet. So when water and\Nair gets sucked out, you can actually
Dialogue: 0,0:32:56.10,0:33:01.48,Default,,0000,0000,0000,,measure this. And then about a month ago,\NI found someone else who allowed me to use
Dialogue: 0,0:33:01.48,0:33:09.35,Default,,0000,0000,0000,,his measurement. So this guy, Phillip\NSmith, was on an airplane again. But he
Dialogue: 0,0:33:09.35,0:33:12.82,Default,,0000,0000,0000,,did not actually go to the lavatory. He\Nstayed on his seat and he just checked
Dialogue: 0,0:33:12.82,0:33:19.02,Default,,0000,0000,0000,,when people were flushing the toilet. So\Nas he sat, there was there were
Dialogue: 0,0:33:19.02,0:33:21.61,Default,,0000,0000,0000,,turbulence. So they couldn't go for a\Nwhile. And then there was the rush while
Dialogue: 0,0:33:21.61,0:33:27.30,Default,,0000,0000,0000,,the toilet and he was plotting it. So just\Nfor those of you that came here by plane,
Dialogue: 0,0:33:27.30,0:33:31.72,Default,,0000,0000,0000,,just a hint as a conversation starter next time,\Nwhen the guy next to you goes to the toilet and
Dialogue: 0,0:33:31.72,0:33:36.57,Default,,0000,0000,0000,,he comes back, tell him exactly all of the\Nhead to flush the toilet and ask him why.
Dialogue: 0,0:33:36.57,0:33:41.95,Default,,0000,0000,0000,,Okay. And you would enjoy the rest of the\Nflight. Some other example that we
Dialogue: 0,0:33:41.95,0:33:47.52,Default,,0000,0000,0000,,actually use is measuring the movement of\Nan elevator. So this is a lift in Aachen.
Dialogue: 0,0:33:47.52,0:33:52.25,Default,,0000,0000,0000,,We have the accelometer which measures the\Nacceleration of this thing, gets the total
Dialogue: 0,0:33:52.25,0:33:56.42,Default,,0000,0000,0000,,height difference of the elevator from the\Nagain, from the pressure sensor, a
Dialogue: 0,0:33:56.42,0:34:01.07,Default,,0000,0000,0000,,barometer. That's a pressure sensor. And\Nthe velocity of the elevator as well from
Dialogue: 0,0:34:01.07,0:34:07.47,Default,,0000,0000,0000,,the change in height. OK, so next time you\Nenter an elevator, I want to see you all
Dialogue: 0,0:34:07.47,0:34:11.61,Default,,0000,0000,0000,,to take out your phones and measure the\Ndistance that the elevator is traveling
Dialogue: 0,0:34:11.61,0:34:17.32,Default,,0000,0000,0000,,and the velocity at which it does so. OK.\NSo these are, in my opinion, most
Dialogue: 0,0:34:17.32,0:34:22.20,Default,,0000,0000,0000,,important sensors, some honorable\Nmentions. Almost all phones have a light
Dialogue: 0,0:34:22.20,0:34:26.45,Default,,0000,0000,0000,,sensor as well, which controls the display\Nbrightness depending on the ambient light.
Dialogue: 0,0:34:26.45,0:34:32.06,Default,,0000,0000,0000,,Unfortunately, there is no API on IOS\Nto access this. So if there are apps
Dialogue: 0,0:34:32.06,0:34:35.84,Default,,0000,0000,0000,,that seem to access a sensor like this,\Nthey usually use the camera instead, which
Dialogue: 0,0:34:35.84,0:34:39.31,Default,,0000,0000,0000,,is which also works quite well. But it's\Nslightly different since the difference
Dialogue: 0,0:34:39.31,0:34:44.23,Default,,0000,0000,0000,,between illuminance and luminance, which I\Ndo not want to go into detail here. And on
Dialogue: 0,0:34:44.23,0:34:48.83,Default,,0000,0000,0000,,most Android phones, they are badly\Ncalibrated or do this so much difference
Dialogue: 0,0:34:48.83,0:34:52.59,Default,,0000,0000,0000,,in the quality of the sensors. We have to\Ncheck it on your own phone if it's worth
Dialogue: 0,0:34:52.59,0:34:57.38,Default,,0000,0000,0000,,anything. But it's a bit difficult. This\Nproximity sensor, which is the one that
Dialogue: 0,0:34:57.38,0:35:01.14,Default,,0000,0000,0000,,turns off the screen when you hold the\Nphone to your ear when you're
Dialogue: 0,0:35:01.14,0:35:04.94,Default,,0000,0000,0000,,actually doing your call. Sounds\Ninteresting, but unfortunately it only
Dialogue: 0,0:35:04.94,0:35:09.13,Default,,0000,0000,0000,,distinguishes or has I know it\Ndistinguishes between between the near and
Dialogue: 0,0:35:09.13,0:35:13.91,Default,,0000,0000,0000,,far value, which is the difference between\Nfive centimeters. So I do not have that
Dialogue: 0,0:35:13.91,0:35:22.18,Default,,0000,0000,0000,,much use for it. There is the temperature sensor,\Nmaybe if they are officially there, then
Dialogue: 0,0:35:22.18,0:35:26.55,Default,,0000,0000,0000,,they usually come along with the humidity\Nsensor, but that's the sensors in your
Dialogue: 0,0:35:26.55,0:35:30.88,Default,,0000,0000,0000,,phone. So you should be a little bit\Nskeptical about this. You're mostly
Dialogue: 0,0:35:30.88,0:35:34.65,Default,,0000,0000,0000,,measuring the heat from your battery or\Nfrom your device. They tried to compensate
Dialogue: 0,0:35:34.65,0:35:38.40,Default,,0000,0000,0000,,for this, but that's a difficult thing to\Ndo. So if you actually, one, need a
Dialogue: 0,0:35:38.40,0:35:44.07,Default,,0000,0000,0000,,thermometer, take a thermometer. They're\Nnot that expensive. OK. You might see some
Dialogue: 0,0:35:44.07,0:35:49.57,Default,,0000,0000,0000,,temperature sensors that are not official.\NWhich phyphhox can pick up. Those are usually
Dialogue: 0,0:35:49.57,0:35:54.30,Default,,0000,0000,0000,,temperature sensors that are part of the\Npressure sensor to compensate for
Dialogue: 0,0:35:54.30,0:35:58.79,Default,,0000,0000,0000,,temperature effects. So they're not even\Ndesigned to get an outside temperature.
Dialogue: 0,0:35:58.79,0:36:06.06,Default,,0000,0000,0000,,OK. So I wanted to mention this. While the\Ninformation about where we got the
Dialogue: 0,0:36:06.06,0:36:10.31,Default,,0000,0000,0000,,information about the sensors from, so in our\NApp at the very bottom, does this entry
Dialogue: 0,0:36:10.31,0:36:17.51,Default,,0000,0000,0000,,submit to a sensor database which tells\Nyou to leave the phone resting on a table?
Dialogue: 0,0:36:17.51,0:36:21.18,Default,,0000,0000,0000,,It also checks if you're actually doing\Nthis, doesn't let you submit it before it
Dialogue: 0,0:36:21.18,0:36:26.92,Default,,0000,0000,0000,,is happy about the error rate or the\Nstandard deviation of the accelometer. And
Dialogue: 0,0:36:26.92,0:36:31.38,Default,,0000,0000,0000,,if you submit it, we collect the data on\Nphyphox.org/sensordb and that's
Dialogue: 0,0:36:31.38,0:36:35.72,Default,,0000,0000,0000,,where I got the statistics from so far. So\Nif you're interested in what a new phone
Dialogue: 0,0:36:35.72,0:36:40.75,Default,,0000,0000,0000,,that you're about to buy can actually do.\NOf course we don't give you any guarantee,
Dialogue: 0,0:36:40.75,0:36:46.42,Default,,0000,0000,0000,,but you can check up or check out all the\Ndata, all the phones. At least those that
Dialogue: 0,0:36:46.42,0:36:49.50,Default,,0000,0000,0000,,are already in our database. And of\Ncourse, I'm happy if you contribute
Dialogue: 0,0:36:49.50,0:36:54.32,Default,,0000,0000,0000,,statistics about the census in your phone\Nas well. So you might want to play with
Dialogue: 0,0:36:54.32,0:37:02.09,Default,,0000,0000,0000,,this later. And then finally, the last\Nthing to finally conclude is some
Dialogue: 0,0:37:02.09,0:37:05.73,Default,,0000,0000,0000,,information on how you can access the\Nsensors. Of course you can write your own
Dialogue: 0,0:37:05.73,0:37:10.40,Default,,0000,0000,0000,,APP. I think here quite a few who can do\Nthis. Just have a look if you can write an
Dialogue: 0,0:37:10.40,0:37:14.13,Default,,0000,0000,0000,,App. Have a look at the API. They're not\Ntoo complicated. It's easy to access the
Dialogue: 0,0:37:14.13,0:37:18.29,Default,,0000,0000,0000,,sensor data. If you're not interested in\Ndesigning your own app, but you want to
Dialogue: 0,0:37:18.29,0:37:22.22,Default,,0000,0000,0000,,include sensor data in some other\Nprojects, there are three ways you can use
Dialogue: 0,0:37:22.22,0:37:27.13,Default,,0000,0000,0000,,Phyfox for this, which I want to\Nintroduce, because that's something that's
Dialogue: 0,0:37:27.13,0:37:32.28,Default,,0000,0000,0000,,one of the reasons I wanted to connect\Nhere. Don't hesitate. Phyfox is free. You
Dialogue: 0,0:37:32.28,0:37:36.21,Default,,0000,0000,0000,,can get for free on Google Play and on the\NAppstore. And when I say it's free, I mean
Dialogue: 0,0:37:36.21,0:37:41.85,Default,,0000,0000,0000,,it's really free. So it's open source. The\NGPL and you can also get an afterwards we
Dialogue: 0,0:37:41.85,0:37:48.28,Default,,0000,0000,0000,,assured of code running on your phone\Nis the code that you see. And we have
Dialogue: 0,0:37:48.28,0:37:53.74,Default,,0000,0000,0000,,three versions how you can.. At least they\Nare categorized into three versions. How
Dialogue: 0,0:37:53.74,0:37:57.41,Default,,0000,0000,0000,,you can access the sensor data. First\Nthing is you can implement something in
Dialogue: 0,0:37:57.41,0:38:01.80,Default,,0000,0000,0000,,Phyfox yourself. So I've got this editor,\Nvisual editor of all file format, which
Dialogue: 0,0:38:01.80,0:38:06.49,Default,,0000,0000,0000,,allows you to take a sensor, place on\Nmathematics. So this is just adding stuff,
Dialogue: 0,0:38:06.49,0:38:11.75,Default,,0000,0000,0000,,but you can apply a Fourier transform or\Nanything and then assign it to a graph.
Dialogue: 0,0:38:11.75,0:38:16.81,Default,,0000,0000,0000,,Alternatively, and of course a bit more\Npowerful. You can have a look at our XML
Dialogue: 0,0:38:16.81,0:38:21.37,Default,,0000,0000,0000,,format, which defines all the experiments.\NSo actually all experiments to see in
Dialogue: 0,0:38:21.37,0:38:25.91,Default,,0000,0000,0000,,Phyfox are not hardcoded, but they are defined\Nin our own file format you can edit any of
Dialogue: 0,0:38:25.91,0:38:32.64,Default,,0000,0000,0000,,them to your needs. And when you're done\Nyou can transfer your data with the QR
Dialogue: 0,0:38:32.64,0:38:36.74,Default,,0000,0000,0000,,code. Do not try to scan this QR code just\Nfrom your QR code app. You have to scan
Dialogue: 0,0:38:36.74,0:38:42.38,Default,,0000,0000,0000,,it from within Phyfox and if you do,\Nyou'll find a nice little experiment which
Dialogue: 0,0:38:42.38,0:38:47.70,Default,,0000,0000,0000,,uses our file formats to implement a Turing\Nmachine that's counting binary up to 256.
Dialogue: 0,0:38:47.70,0:38:51.28,Default,,0000,0000,0000,,So this is the proof that all file format\Nactually is Turing complete. So you can do
Dialogue: 0,0:38:51.28,0:38:55.62,Default,,0000,0000,0000,,a lot with it. Okay. I'm not suggesting\Nthat you're trying to implement doom on it
Dialogue: 0,0:38:55.62,0:38:59.98,Default,,0000,0000,0000,,or something like this because you won't\Nbe able to. It's not efficient that way.
Dialogue: 0,0:38:59.98,0:39:04.13,Default,,0000,0000,0000,,It's not designed to be Turing complete.\NIt just happens to be Turing complete. So
Dialogue: 0,0:39:04.13,0:39:08.46,Default,,0000,0000,0000,,if you want to do something more, you can\Nconnect to Phyfox via a network. You've
Dialogue: 0,0:39:08.46,0:39:11.73,Default,,0000,0000,0000,,seen one example with the salad spinner.\NWhen I said that there is a
Dialogue: 0,0:39:11.73,0:39:17.18,Default,,0000,0000,0000,,a web server running on the App. You can use\Nthis to access the data directly from your
Dialogue: 0,0:39:17.18,0:39:21.57,Default,,0000,0000,0000,,preferred programing language. There's an\Nexample where I'm using Python to read out
Dialogue: 0,0:39:21.57,0:39:26.78,Default,,0000,0000,0000,,the sensor data and control a synthesizer.\NSo what's running on the web server is
Dialogue: 0,0:39:26.78,0:39:31.15,Default,,0000,0000,0000,,basically a rest API. So yeah. Just visit\Nour website and learn how to do this. So
Dialogue: 0,0:39:31.15,0:39:34.85,Default,,0000,0000,0000,,you can read out the sensor data of a\Nnetwork and control your project with it.
Dialogue: 0,0:39:34.85,0:39:38.82,Default,,0000,0000,0000,,An alternative to this is a new network\Ninterface that we have, which is more on
Dialogue: 0,0:39:38.82,0:39:45.40,Default,,0000,0000,0000,,this XML side or the design of our\Nexperiment configurations, which is meant
Dialogue: 0,0:39:45.40,0:39:51.45,Default,,0000,0000,0000,,to collect data from many users and not\Nlife data. So we had this lecture. So this is
Dialogue: 0,0:39:51.45,0:39:55.86,Default,,0000,0000,0000,,the new lecture hall, by the way. So we\Nhad a lecture where every student got a
Dialogue: 0,0:39:55.86,0:40:00.67,Default,,0000,0000,0000,,spring from us and there was supposed to\Nbuild a spring pendulum and we collected
Dialogue: 0,0:40:00.67,0:40:05.76,Default,,0000,0000,0000,,the data from all students and the lecture\Nhall in realtime on the big screen to
Dialogue: 0,0:40:05.76,0:40:13.75,Default,,0000,0000,0000,,determine the dependency of the frequency\Nfrom the mass of the pendulum. And another
Dialogue: 0,0:40:13.75,0:40:17.40,Default,,0000,0000,0000,,example. Just a few days ago, we during\Nthe winter solstice, we asked our
Dialogue: 0,0:40:17.40,0:40:22.37,Default,,0000,0000,0000,,international users to point their phone\Nat the sun. So we get an angle for the
Dialogue: 0,0:40:22.37,0:40:27.52,Default,,0000,0000,0000,,elevation of the sun and the azimuth from\Nthe magnetometer with a compass. And this
Dialogue: 0,0:40:27.52,0:40:31.12,Default,,0000,0000,0000,,way we could trace the path of the sun\Nacross the earth from all the users. What
Dialogue: 0,0:40:31.12,0:40:36.04,Default,,0000,0000,0000,,each black point with the line is a\Ncontribution from a user. So, yeah, from
Dialogue: 0,0:40:36.04,0:40:40.90,Default,,0000,0000,0000,,this we could, for example, determine the\Ntilted angle of the earth's axis. OK, so
Dialogue: 0,0:40:40.90,0:40:43.81,Default,,0000,0000,0000,,just example, what you can do is this\Nnetwork interface, as long as we're able
Dialogue: 0,0:40:43.81,0:40:47.66,Default,,0000,0000,0000,,to set up some server to receive the data,\Nyou can use this network interface. We're
Dialogue: 0,0:40:47.66,0:40:52.13,Default,,0000,0000,0000,,still working on this network interface.\NSo far it can only do HTTP requests, get
Dialogue: 0,0:40:52.13,0:40:58.24,Default,,0000,0000,0000,,or post. But we are also planning on\Nimplementing Mqtt and other protocols like
Dialogue: 0,0:40:58.24,0:41:02.98,Default,,0000,0000,0000,,this. And the third option is a Bluetooth\Nconnection, which is mostly designed for
Dialogue: 0,0:41:02.98,0:41:07.36,Default,,0000,0000,0000,,sensors. So if you want. If you have some\NBluetooth low energy sensor that you want
Dialogue: 0,0:41:07.36,0:41:11.51,Default,,0000,0000,0000,,to read out, you can use Phyfox. So\Nthere's an example of a Texas Instruments
Dialogue: 0,0:41:11.51,0:41:15.39,Default,,0000,0000,0000,,sensor tech, which has a software which is\Nnot designed for Phyfox. But our file
Dialogue: 0,0:41:15.39,0:41:20.33,Default,,0000,0000,0000,,format is flexible enough to\Nsimply tell Phyfox how to read all the
Dialogue: 0,0:41:20.33,0:41:25.92,Default,,0000,0000,0000,,data and suddenly we've got the sensor\Nthat can run independently from the phone.
Dialogue: 0,0:41:25.92,0:41:30.33,Default,,0000,0000,0000,,And of course you can include your own\Nprojects like this. So there is an example from
Dialogue: 0,0:41:30.33,0:41:35.55,Default,,0000,0000,0000,,actually my institute, because originally I'm in\Nsolid state physicist. So we're working a
Dialogue: 0,0:41:35.55,0:41:39.69,Default,,0000,0000,0000,,lot with graphene and this is a\Ndemonstrated we create that was an ESP 32.
Dialogue: 0,0:41:39.69,0:41:46.00,Default,,0000,0000,0000,,So this is another version of an Arduino, or\NArduino compatibel. What we're doing here.
Dialogue: 0,0:41:46.00,0:41:50.46,Default,,0000,0000,0000,,We're reading out a graphene Hallsensor\Nand so. It's all similar to the holecenter
Dialogue: 0,0:41:50.46,0:41:55.59,Default,,0000,0000,0000,,of phone, but based on graphene and we can\Nget life measurements in Phyfox with this.
Dialogue: 0,0:41:55.59,0:41:59.59,Default,,0000,0000,0000,,And so if you have an Arduino project with\Nwhich you want to.. from which you want
Dialogue: 0,0:41:59.59,0:42:04.28,Default,,0000,0000,0000,,to send data that is plotted in Phyfox,\Nyou can do it with a bluetooth low
Dialogue: 0,0:42:04.28,0:42:08.84,Default,,0000,0000,0000,,energy interface. But if you have some\Npatients and maybe wait two more months,
Dialogue: 0,0:42:08.84,0:42:12.89,Default,,0000,0000,0000,,we are working on Arduino library to make\Nthis simpler. So this the entire code, you
Dialogue: 0,0:42:12.89,0:42:17.87,Default,,0000,0000,0000,,would need to read out the analog input\Nfrom an Arduino and send it to Phyfox to
Dialogue: 0,0:42:17.87,0:42:22.61,Default,,0000,0000,0000,,be plotted. OK, so this is working right\Nnow. If you cannot wait, you can check it
Dialogue: 0,0:42:22.61,0:42:25.31,Default,,0000,0000,0000,,out on our website. So this is already\Navailable, although it's a work in
Dialogue: 0,0:42:25.31,0:42:29.59,Default,,0000,0000,0000,,progress. The interface will change a bit\Nstill. I would prefer if you want to
Dialogue: 0,0:42:29.59,0:42:33.20,Default,,0000,0000,0000,,start right now, if you contact me so we\Ncan get some feedback and maybe even
Dialogue: 0,0:42:33.20,0:42:39.78,Default,,0000,0000,0000,,design the library also to your needs. So that\Nwe get an idea. So with this, I'm about to
Dialogue: 0,0:42:39.78,0:42:47.11,Default,,0000,0000,0000,,finish. So just a short summary what I'm\Nhoping I can trigger. Yeah. So if you were
Dialogue: 0,0:42:47.11,0:42:52.02,Default,,0000,0000,0000,,mildly amused, mightily entertained by\Nthis by this talk, check out our Web site
Dialogue: 0,0:42:52.02,0:42:55.67,Default,,0000,0000,0000,,or check out our YouTube channel or\NTwitter. We can get some more examples,
Dialogue: 0,0:42:55.67,0:43:03.29,Default,,0000,0000,0000,,what we do with the sensors in the phone.\NIf you are a teacher, are teachers here?
Dialogue: 0,0:43:03.29,0:43:09.13,Default,,0000,0000,0000,,Quite a few. That's great! And if you want\Nto use this in class or in a lecture,
Dialogue: 0,0:43:09.13,0:43:12.12,Default,,0000,0000,0000,,check out our Web site phyfox.org. We've\Ngot a database of experiments that you can
Dialogue: 0,0:43:12.12,0:43:15.09,Default,,0000,0000,0000,,do: phyfox.org/experiments .That's then\Nactually about physics and less about the
Dialogue: 0,0:43:15.09,0:43:19.60,Default,,0000,0000,0000,,hardware where we also demonstrate the\Nexperiments and how they work. If you are
Dialogue: 0,0:43:19.60,0:43:24.66,Default,,0000,0000,0000,,a teacher and has a specific project in\Nmind. Check out our editor to design your
Dialogue: 0,0:43:24.66,0:43:31.61,Default,,0000,0000,0000,,own set up with which you can do something\Nspecific for a very specific experiment.
Dialogue: 0,0:43:31.61,0:43:36.15,Default,,0000,0000,0000,,phyfox.org/editor. Then if you are working\Non arduino project and want to plot
Dialogue: 0,0:43:36.15,0:43:40.08,Default,,0000,0000,0000,,something, you can visit Phyfox.org/arduino,\Nwhere you already can access
Dialogue: 0,0:43:40.08,0:43:45.01,Default,,0000,0000,0000,,our library. Although it's not complete as\NI said. So maybe wait a little bit or
Dialogue: 0,0:43:45.01,0:43:50.29,Default,,0000,0000,0000,,contact me first. If you have a Bluetooth\Nlow energy device that you want to use or
Dialogue: 0,0:43:50.29,0:43:56.10,Default,,0000,0000,0000,,integrate. You can visit phyfox.org/ble.\NIf it's about a device that you did not
Dialogue: 0,0:43:56.10,0:44:00.06,Default,,0000,0000,0000,,design yourself, you probably need some\Nbackground information about bluetooth low
Dialogue: 0,0:44:00.06,0:44:04.87,Default,,0000,0000,0000,,energy. Should know what a GATT server is\Nand how characteristics and services, new
Dialogue: 0,0:44:04.87,0:44:08.26,Default,,0000,0000,0000,,ideas and all this stuff and bluetooth\Nenergy works. And it's good to get some
Dialogue: 0,0:44:08.26,0:44:13.13,Default,,0000,0000,0000,,documentation or to be good as reverse\Nengineering, but in principle I haven't
Dialogue: 0,0:44:13.13,0:44:19.98,Default,,0000,0000,0000,,seen many devices so far which could not\Nwork with phyfox easily. Then if you want
Dialogue: 0,0:44:19.98,0:44:26.46,Default,,0000,0000,0000,,to read the values for another project via\Nnetwork, visit our website, the wiki on
Dialogue: 0,0:44:26.46,0:44:30.83,Default,,0000,0000,0000,,our website. phyfox.org/wiki, where you\Ncan get information about the rest API and
Dialogue: 0,0:44:30.83,0:44:34.87,Default,,0000,0000,0000,,on your network interface. And finally,\Nsomething I would really love if you want
Dialogue: 0,0:44:34.87,0:44:39.82,Default,,0000,0000,0000,,to contribute. If you can write some apps,\NI mean you can use a lot of things. The iOS
Dialogue: 0,0:44:39.82,0:44:45.51,Default,,0000,0000,0000,,app is written in swift. The Android\Nversion is written in Java. Our webserver,
Dialogue: 0,0:44:45.51,0:44:50.32,Default,,0000,0000,0000,,of course, has web development and Html in\NJavaScript. So if you want to contribute
Dialogue: 0,0:44:50.32,0:44:55.80,Default,,0000,0000,0000,,there. Visit our Web site at a\Nphyfox.org/source. And we would love to
Dialogue: 0,0:44:55.80,0:45:01.80,Default,,0000,0000,0000,,see some help in development. With this I\Nfinish my talk and I'm looking forward to
Dialogue: 0,0:45:01.80,0:45:05.26,Default,,0000,0000,0000,,any exchange we will have later and any\Nquestions. And I'm just thankful that it
Dialogue: 0,0:45:05.26,0:45:09.63,Default,,0000,0000,0000,,was allowed to talk here and\Nget so much attention. Thank you.
Dialogue: 0,0:45:09.63,0:45:21.31,Default,,0000,0000,0000,,{\i1}Applause{\i0}
Dialogue: 0,0:45:21.31,0:45:27.03,Default,,0000,0000,0000,,Oh, by the way, since it is up there. One\Nbad news, unfortunately, I can only be
Dialogue: 0,0:45:27.03,0:45:32.27,Default,,0000,0000,0000,,here today. So if you want to talk to me,\Ntry to catch me today. You can also call
Dialogue: 0,0:45:32.27,0:45:36.80,Default,,0000,0000,0000,,me. I actually brought a DECT phone,\Nbut, uh, sorry, only today.
Dialogue: 0,0:45:36.80,0:45:42.50,Default,,0000,0000,0000,,Herald: Oh, my God. So quickly, though, we\Nhave questions now, 15 minutes, then 15
Dialogue: 0,0:45:42.50,0:45:47.25,Default,,0000,0000,0000,,minutes, I think. And then afterwards, you\Nhave to find him and catch him. Thank you,
Dialogue: 0,0:45:47.25,0:45:56.41,Default,,0000,0000,0000,,Sebastian. Questions. Shoot. There is one.\NQuestion: You mentioned aliasing affect
Dialogue: 0,0:45:56.41,0:46:02.21,Default,,0000,0000,0000,,during.. Is it possible to change or modulate\Nthe sampling frequency to actually find
Dialogue: 0,0:46:02.21,0:46:08.58,Default,,0000,0000,0000,,out our frequency above the sampling frequency?\NSebastian: Yeah, that's that's a good
Dialogue: 0,0:46:08.58,0:46:13.58,Default,,0000,0000,0000,,question. Not only because of the of the\Nalias affect, but also because some
Dialogue: 0,0:46:13.58,0:46:17.18,Default,,0000,0000,0000,,projects also want to reduce the sampling\Nfrequency. It's a little bit tricky
Dialogue: 0,0:46:17.18,0:46:22.06,Default,,0000,0000,0000,,because on both APIs and both IOS and\NAndroid, you cannot specify a target
Dialogue: 0,0:46:22.06,0:46:26.19,Default,,0000,0000,0000,,frequency, you can only specify a\Nfrequency that specific for certain use
Dialogue: 0,0:46:26.19,0:46:31.81,Default,,0000,0000,0000,,case. So for example, you say I need the\Naccelerometer data, that's which at a rate
Dialogue: 0,0:46:31.81,0:46:36.11,Default,,0000,0000,0000,,that's reasonable for UI changes or at a\Nrate that's reasonable for games. Right.
Dialogue: 0,0:46:36.11,0:46:40.89,Default,,0000,0000,0000,,Right. That's as fast as possible. So if\Nyou do it for UI, you get something like
Dialogue: 0,0:46:40.89,0:46:44.45,Default,,0000,0000,0000,,let's say two three hertz. We heard something like\Nthis or you doesn't waiting ages before
Dialogue: 0,0:46:44.45,0:46:48.46,Default,,0000,0000,0000,,the screen rotates for games. It's 25 50\Nhertz something like this. So we can
Dialogue: 0,0:46:48.46,0:46:53.27,Default,,0000,0000,0000,,control the game and fastest is the\Ndata I've just plotted. And Phyfox always
Dialogue: 0,0:46:53.27,0:46:57.47,Default,,0000,0000,0000,,request the fastest we can see and in\NPhyfox we have a setting, we can limit the
Dialogue: 0,0:46:57.47,0:47:03.02,Default,,0000,0000,0000,,frequency. Unfortunately, if your\Nfrequency is not simply a multiple, no,
Dialogue: 0,0:47:03.02,0:47:06.82,Default,,0000,0000,0000,,the other way around is. The frequency\Ngiven by the device, is not a multiple of the
Dialogue: 0,0:47:06.82,0:47:10.54,Default,,0000,0000,0000,,frequency that you gave. It's not easy to\Nbreak it down to the target frequency. So
Dialogue: 0,0:47:10.54,0:47:18.20,Default,,0000,0000,0000,,you usually see some odd cases where\NPhyfox tells to group the sensor events
Dialogue: 0,0:47:18.20,0:47:24.20,Default,,0000,0000,0000,,along this to get near this frequency. So\Nit might not work that well. And
Dialogue: 0,0:47:24.20,0:47:27.54,Default,,0000,0000,0000,,especially if you're looking for the alias\Neffect. This might really mess up their
Dialogue: 0,0:47:27.54,0:47:31.80,Default,,0000,0000,0000,,alias effect, so you might need to try a\Nlittle bit which frequency looks good to
Dialogue: 0,0:47:31.80,0:47:35.50,Default,,0000,0000,0000,,do this. But of course in principle you\Ncan average about multiple values in this
Dialogue: 0,0:47:35.50,0:47:42.10,Default,,0000,0000,0000,,way or simply pick only every end value.\NAnd this way we'd use the frequency. And
Dialogue: 0,0:47:42.10,0:47:46.01,Default,,0000,0000,0000,,yeah, this can be done to our editor or to\Nthe main screen. There's a plus button
Dialogue: 0,0:47:46.01,0:47:49.95,Default,,0000,0000,0000,,with which you can simply expand which\Nalready allows you to set this simple
Dialogue: 0,0:47:49.95,0:47:53.53,Default,,0000,0000,0000,,frequency. Just keep in mind that you\Ncannot really always get to the
Dialogue: 0,0:47:53.53,0:47:57.74,Default,,0000,0000,0000,,target frequency, right?\NHerald: Right. There is another question.
Dialogue: 0,0:47:57.74,0:48:01.55,Default,,0000,0000,0000,,No? Yes. Please.\NQuestion: Hi. Thanks for the cool task.
Dialogue: 0,0:48:01.55,0:48:05.93,Default,,0000,0000,0000,,It's a great app. I love using it in\Nschool. I was wondering if those cool
Dialogue: 0,0:48:05.93,0:48:09.79,Default,,0000,0000,0000,,animations how to sensor types of working\Nare available.
Dialogue: 0,0:48:09.79,0:48:15.39,Default,,0000,0000,0000,,Sebastian: Sorry. The animation scene.\NYeah. I think I wonder how to do this
Dialogue: 0,0:48:15.39,0:48:21.10,Default,,0000,0000,0000,,best. Before that, I was already thinking\Nabout sharing the slides. Actually, my
Dialogue: 0,0:48:21.10,0:48:27.35,Default,,0000,0000,0000,,talk is space it's just written in HTML in Javascript\Nit's not easy to control for everyone. That's
Dialogue: 0,0:48:27.35,0:48:33.17,Default,,0000,0000,0000,,why I did not simply upload it. I would if\NI would check later, if I can, upload the entire
Dialogue: 0,0:48:33.17,0:48:37.26,Default,,0000,0000,0000,,talk in some way that makes sense either\Non our website. I'm not sure if it makes
Dialogue: 0,0:48:37.26,0:48:42.49,Default,,0000,0000,0000,,sense to upload it to the system of the\Nconference. Still, after the talk, I would
Dialogue: 0,0:48:42.49,0:48:49.85,Default,,0000,0000,0000,,check it, but I am not... I want to share\Nthe slides, but I probably need to add
Dialogue: 0,0:48:49.85,0:48:53.75,Default,,0000,0000,0000,,some documentation on how to use them\Nbecause they are not Power point PDF or
Dialogue: 0,0:48:53.75,0:48:58.45,Default,,0000,0000,0000,,Latex generated PDF.\NIt's handmade.
Dialogue: 0,0:48:58.45,0:49:05.06,Default,,0000,0000,0000,,Herald: You can always cut them out of the\Nvideo getting streamed and La la la la la.
Dialogue: 0,0:49:05.06,0:49:08.54,Default,,0000,0000,0000,,Yeah, right.\NQuestion: Just a quick question of the the
Dialogue: 0,0:49:08.54,0:49:12.69,Default,,0000,0000,0000,,axis of the phone. They're like like that\Nand that distorts us.
Dialogue: 0,0:49:12.69,0:49:19.97,Default,,0000,0000,0000,,Sebastian: So it's not for most phones.\NThe X-axis is reading directlon. The
Dialogue: 0,0:49:19.97,0:49:26.21,Default,,0000,0000,0000,,Y-axis is upwards along the screen and Z\Naccess, Z-axis depending on your dialect is
Dialogue: 0,0:49:26.21,0:49:32.46,Default,,0000,0000,0000,,perpendicular to the screen. I'd say in\Nmost cases because officially the X-Axis
Dialogue: 0,0:49:32.46,0:49:36.00,Default,,0000,0000,0000,,at least I think I've written this\Ndocumentation for Android is along the
Dialogue: 0,0:49:36.00,0:49:40.15,Default,,0000,0000,0000,,natural reading direction of the device.\NSo if you've got a huge tablet which you
Dialogue: 0,0:49:40.15,0:49:45.99,Default,,0000,0000,0000,,naturally would put in horizontal\Nalignment, not portrait mode, it might be
Dialogue: 0,0:49:45.99,0:49:51.49,Default,,0000,0000,0000,,that the X-Axis is the long Axis. I have never\Nseen this myself, but I'm a little bit
Dialogue: 0,0:49:51.49,0:49:55.46,Default,,0000,0000,0000,,careful to say that all these devices have\Nthe same axis, but Z is definitely always
Dialogue: 0,0:49:55.46,0:49:59.81,Default,,0000,0000,0000,,perpendicular to the screen and X and Y are\Nthan the other ones and they are fixed and
Dialogue: 0,0:49:59.81,0:50:05.01,Default,,0000,0000,0000,,usually the short side is the X-Axis.\NHerald: Ok. There is one more question
Dialogue: 0,0:50:05.01,0:50:10.83,Default,,0000,0000,0000,,there, please, sir. Take the microphone.\NIt's next to you. You got it off the
Dialogue: 0,0:50:10.83,0:50:13.86,Default,,0000,0000,0000,,...\NQuestion: Hi, you mentioned the necessity
Dialogue: 0,0:50:13.86,0:50:20.76,Default,,0000,0000,0000,,of the magnetic sensor to to determine the\Ncontent orientation. Can you not use past
Dialogue: 0,0:50:20.76,0:50:24.51,Default,,0000,0000,0000,,G.P.S. data and then integrate over the\Ngyroscope data to get the current
Dialogue: 0,0:50:24.51,0:50:27.48,Default,,0000,0000,0000,,orientation?\NSebastian: {\i1}Lauthing{\i0} Mathematically, your
Dialogue: 0,0:50:27.48,0:50:34.96,Default,,0000,0000,0000,,correct, problem is integrating sensor\Ndata is not as simple. I'm often
Dialogue: 0,0:50:34.96,0:50:42.08,Default,,0000,0000,0000,,surprised on what some software\Ncan actually do. If you do it naively
Dialogue: 0,0:50:42.08,0:50:45.74,Default,,0000,0000,0000,,right now I only have an example in mind\Nfor the accelerometer cause it could also
Dialogue: 0,0:50:45.74,0:50:49.47,Default,,0000,0000,0000,,say you can integrate the accelometer data\Nto get velocity. You can integrate the
Dialogue: 0,0:50:49.47,0:50:54.55,Default,,0000,0000,0000,,velocity to get the displacement of the\Nphone of the location. If you do this,
Dialogue: 0,0:50:54.55,0:50:58.04,Default,,0000,0000,0000,,we've got a very simple example in our\Nwiki. Very naiv even one without any
Dialogue: 0,0:50:58.04,0:51:03.12,Default,,0000,0000,0000,,filtering, then just the noise means\Nthat's if there's little arrow, you summit
Dialogue: 0,0:51:03.12,0:51:08.27,Default,,0000,0000,0000,,up integrations, nothing else but\Nsuming up in small steps. You get an
Dialogue: 0,0:51:08.27,0:51:13.67,Default,,0000,0000,0000,,offset error in the velocity. If you\Nintegrate this again, you get an error in
Dialogue: 0,0:51:13.67,0:51:17.73,Default,,0000,0000,0000,,the location with which is growing with\Nthe square of the time. So if you do this
Dialogue: 0,0:51:17.73,0:51:22.42,Default,,0000,0000,0000,,for location and try it out with our naive\Napproach your phone is supposed to
Dialogue: 0,0:51:22.42,0:51:27.42,Default,,0000,0000,0000,,be 100 meters upwards after about 10\Nseconds. If you do this for the gyroscope,
Dialogue: 0,0:51:27.42,0:51:31.68,Default,,0000,0000,0000,,it's a little easier because you only\Nwant integration. But still there will be
Dialogue: 0,0:51:31.68,0:51:37.21,Default,,0000,0000,0000,,some drift. I'm not sure about all the\Ntechniques the manufacturers imployed to
Dialogue: 0,0:51:37.21,0:51:41.40,Default,,0000,0000,0000,,filter out any errors. I mean, obviously\Nthe gyroscope is self calibrating
Dialogue: 0,0:51:41.40,0:51:45.37,Default,,0000,0000,0000,,otherwise, it would be pointing in\Ndifferent direction all the time. And on
Dialogue: 0,0:51:45.37,0:51:50.32,Default,,0000,0000,0000,,some phones I've seen it jumping\Nwhen it recalibrates. But if you simply
Dialogue: 0,0:51:50.32,0:51:54.90,Default,,0000,0000,0000,,integrate this, you will certainly get\Ndrift, there's no way that you can get
Dialogue: 0,0:51:54.90,0:51:59.13,Default,,0000,0000,0000,,a fixed position. What I think what they\Nprobably do for most cases, they use the
Dialogue: 0,0:51:59.13,0:52:06.29,Default,,0000,0000,0000,,gyroscope to immediate direct\Nrotation and then try to fuze it
Dialogue: 0,0:52:06.29,0:52:10.18,Default,,0000,0000,0000,,in some way with the magnetometer\Ninformation to keep it fixed so that at
Dialogue: 0,0:52:10.18,0:52:13.80,Default,,0000,0000,0000,,the end you're not pointing the wrong\Ndirection. But the gyroscope itself, only
Dialogue: 0,0:52:13.80,0:52:17.66,Default,,0000,0000,0000,,on its own, is unfortunately only giving\Nyou the rotation rate, not the absolute
Dialogue: 0,0:52:17.66,0:52:22.55,Default,,0000,0000,0000,,rotation in contrast to an actual\Ngyroscope. The big one that's rotating. So
Dialogue: 0,0:52:22.55,0:52:27.53,Default,,0000,0000,0000,,it's at least not that easy.\NThat's all I can say.
Dialogue: 0,0:52:27.53,0:52:32.76,Default,,0000,0000,0000,,Herald. Whow? What the bunch of\Ninformation, Sebastian? I really love the
Dialogue: 0,0:52:32.76,0:52:35.96,Default,,0000,0000,0000,,.... There is someone else with a\Nquestion. I really love your replication.
Dialogue: 0,0:52:35.96,0:52:39.25,Default,,0000,0000,0000,,Actually, it was really immediately fun\Nto, go.
Dialogue: 0,0:52:39.25,0:52:44.01,Default,,0000,0000,0000,,Question: Thank you so much for a great\Napplication. And my question is, just very
Dialogue: 0,0:52:44.01,0:52:50.01,Default,,0000,0000,0000,,short. Can you also integrate external\Nsensors through Wi-Fi or is it only to be
Dialogue: 0,0:52:50.01,0:52:52.83,Default,,0000,0000,0000,,early?\NSebastian: No. That's what I meant with
Dialogue: 0,0:52:52.83,0:52:59.58,Default,,0000,0000,0000,,the network connection. Network usually\Nhas Wi-Fi in this case, I'm not sure if it
Dialogue: 0,0:52:59.58,0:53:05.23,Default,,0000,0000,0000,,would work on a conference like this into\Nthe cable. So now you can get the data
Dialogue: 0,0:53:05.23,0:53:12.68,Default,,0000,0000,0000,,through our REST API. Might not be the\Nfastest thing. Maybe we will add to our
Dialogue: 0,0:53:12.68,0:53:19.81,Default,,0000,0000,0000,,network, our new network functionality,\Nsomething that will keep open apart and
Dialogue: 0,0:53:19.81,0:53:24.64,Default,,0000,0000,0000,,push the data in there, so far the best\Nthing to go is with our rest API.
Dialogue: 0,0:53:24.64,0:53:27.39,Default,,0000,0000,0000,,Question: I was just thinking about the\Nexternal sensor connection.
Dialogue: 0,0:53:27.39,0:53:32.96,Default,,0000,0000,0000,,Sebastian: So external? Sorry, I was\Nthinking a different direction. Actually,
Dialogue: 0,0:53:32.96,0:53:35.59,Default,,0000,0000,0000,,that's a good question. That reminds me of\Nthat, that there's something I wanted to
Dialogue: 0,0:53:35.59,0:53:41.70,Default,,0000,0000,0000,,add. You can use the REST API in theory to\Npush data in there, but that's only a
Dialogue: 0,0:53:41.70,0:53:46.17,Default,,0000,0000,0000,,parameter in the Url. It's simple a Get/\Npush off a single value which doesn't get
Dialogue: 0,0:53:46.17,0:53:51.69,Default,,0000,0000,0000,,get you far and which is quite\Ninefficient. However, within you network
Dialogue: 0,0:53:51.69,0:53:57.05,Default,,0000,0000,0000,,interface you can do requests to other\Ndevices so you can GET request and already
Dialogue: 0,0:53:57.05,0:54:01.70,Default,,0000,0000,0000,,is able to respa Json packet as a\Nresponse us to interpret the adjacent
Dialogue: 0,0:54:01.70,0:54:05.38,Default,,0000,0000,0000,,packet as a response. And that's where\Nadding Mqtt and stuff like this, this is
Dialogue: 0,0:54:05.38,0:54:10.00,Default,,0000,0000,0000,,supposed to go in both directions. But\Nthis is really new. So if you've got
Dialogue: 0,0:54:10.00,0:54:14.49,Default,,0000,0000,0000,,something specific, try if it works or\Ncontact me if it's not working, if you
Dialogue: 0,0:54:14.49,0:54:19.77,Default,,0000,0000,0000,,need some help, if you find the bug. So but\Nit's supposed to work on your network
Dialogue: 0,0:54:19.77,0:54:22.96,Default,,0000,0000,0000,,stuff. That's there in the configuration. So\Nthe idea of the workflow of all this
Dialogue: 0,0:54:22.96,0:54:27.78,Default,,0000,0000,0000,,connection with specific devices have\Nsomething set up like this. You create a
Dialogue: 0,0:54:27.78,0:54:32.59,Default,,0000,0000,0000,,configuration for Phyfox, which in the end\Nis supplied to the QR code. For example,
Dialogue: 0,0:54:32.59,0:54:36.44,Default,,0000,0000,0000,,the user scans the QR code. And this all\Nthe information, how to communicate with
Dialogue: 0,0:54:36.44,0:54:41.61,Default,,0000,0000,0000,,the device is already supplied. You can\Nalso do this for Bluetooth. That the
Dialogue: 0,0:54:41.61,0:54:45.05,Default,,0000,0000,0000,,device itself provides it to Phyphox.\NBut in the end it's these configurations
Dialogue: 0,0:54:45.05,0:54:50.59,Default,,0000,0000,0000,,and for the new network interface, it can\Nalso receive data from the network. But so
Dialogue: 0,0:54:50.59,0:54:55.79,Default,,0000,0000,0000,,far only via HTTP.\NQuestion: OK. Thank you.
Dialogue: 0,0:54:55.79,0:55:01.96,Default,,0000,0000,0000,,Herald: I have maybe a last question if no\None else has. What's the next step? What
Dialogue: 0,0:55:01.96,0:55:06.37,Default,,0000,0000,0000,,is your next goal? Because this is a\Ntremendous successful thing. And you see
Dialogue: 0,0:55:06.37,0:55:12.24,Default,,0000,0000,0000,,the educational purposes. So that's\Nfantastic, actually, isn't it? It's not
Dialogue: 0,0:55:12.24,0:55:17.60,Default,,0000,0000,0000,,only on university level if you're using\Nit, that's all around in Germany.
Dialogue: 0,0:55:17.60,0:55:20.14,Default,,0000,0000,0000,,Sebastian: That's not in Germany. It's by\Nthe way another thing you could
Dialogue: 0,0:55:20.14,0:55:24.10,Default,,0000,0000,0000,,contribute. If you're speaking a language\Nthat has been translated into Phyfox is
Dialogue: 0,0:55:24.10,0:55:29.31,Default,,0000,0000,0000,,translated by volunteers and it's already\Navailable, I think in 2010 and 2012, 2013,
Dialogue: 0,0:55:29.31,0:55:34.31,Default,,0000,0000,0000,,14 languages, something around this. So\Nyeah, but next step I think will be using
Dialogue: 0,0:55:34.31,0:55:38.42,Default,,0000,0000,0000,,the camera because that's another sensor,\Nbroadly speaking, which we are not using
Dialogue: 0,0:55:38.42,0:55:43.73,Default,,0000,0000,0000,,at all, which can do a lot, but we haven't\Nyet started on this. So lot to do in this
Dialogue: 0,0:55:43.73,0:55:45.86,Default,,0000,0000,0000,,project.\NHerald: Super. I'm looking forward to see
Dialogue: 0,0:55:45.86,0:55:55.08,Default,,0000,0000,0000,,you next year then. {\i1}Laughing, Applause{\i0}\NSebastian Starks, thank you very much. An
Dialogue: 0,0:55:55.08,0:55:57.93,Default,,0000,0000,0000,,honor and a pleasure to have you.
Dialogue: 0,0:55:57.93,0:56:02.34,Default,,0000,0000,0000,,{\i1}Postroll music{\i0}
Dialogue: 0,0:56:02.34,0:56:24.00,Default,,0000,0000,0000,,Subtitles created by c3subtitles.de\Nin the year 2020. Join, and help us!