0 00:00:00,000 --> 00:00:30,000 Dear viewer, these subtitles were generated by a machine via the service Trint and therefore are (very) buggy. If you are capable, please help us to create good quality subtitles: https://c3subtitles.de/talk/438 Thanks! 1 00:00:11,070 --> 00:00:13,169 All right, good morning, everyone, thank 2 00:00:13,170 --> 00:00:15,809 you for the nice introduction. 3 00:00:15,810 --> 00:00:18,029 I'm Polygon and I 4 00:00:18,030 --> 00:00:20,339 want to talk to you about low 5 00:00:20,340 --> 00:00:22,619 cost, high speed photography. 6 00:00:22,620 --> 00:00:24,749 So I started doing photography 7 00:00:24,750 --> 00:00:26,879 as kind of a hobby about 8 00:00:26,880 --> 00:00:28,169 eight years ago. 9 00:00:28,170 --> 00:00:30,179 And after going through the usual phase 10 00:00:30,180 --> 00:00:32,339 of doing everything from, I don't 11 00:00:32,340 --> 00:00:34,469 know, Panorama's to macro's, 12 00:00:34,470 --> 00:00:36,809 I kind of concentrated on 13 00:00:36,810 --> 00:00:39,089 studio photography and 14 00:00:39,090 --> 00:00:41,219 being a student with 15 00:00:41,220 --> 00:00:42,869 limited room available. 16 00:00:42,870 --> 00:00:45,059 I kind of improvised a lot 17 00:00:45,060 --> 00:00:46,060 with what was there 18 00:00:47,310 --> 00:00:49,349 and around at that time. 19 00:00:49,350 --> 00:00:51,749 I also watched a lot of YouTube 20 00:00:51,750 --> 00:00:54,419 and Vimeo videos which were partially 21 00:00:54,420 --> 00:00:56,609 high speed, high 22 00:00:56,610 --> 00:00:58,769 speed action, and I 23 00:00:58,770 --> 00:01:00,719 wanted to do this myself. 24 00:01:00,720 --> 00:01:02,999 And since I didn't have twenty thousand 25 00:01:03,000 --> 00:01:05,159 euros to spend on a phantom flex, 26 00:01:05,160 --> 00:01:07,409 high speed camera, I started 27 00:01:07,410 --> 00:01:09,539 looking around to try 28 00:01:09,540 --> 00:01:11,909 to find out what I can do with 29 00:01:11,910 --> 00:01:14,129 my camera and the other equipment 30 00:01:14,130 --> 00:01:15,659 I already had. 31 00:01:15,660 --> 00:01:18,299 And so this talk is 32 00:01:18,300 --> 00:01:20,699 basically a mixture of a tutorial. 33 00:01:20,700 --> 00:01:22,349 So I did a lot of research on the 34 00:01:22,350 --> 00:01:23,279 Internet. 35 00:01:23,280 --> 00:01:25,379 And so here is what the final result 36 00:01:25,380 --> 00:01:27,719 looked like as well as 37 00:01:27,720 --> 00:01:29,729 like a slideshow. 38 00:01:29,730 --> 00:01:31,080 So to give you an idea, 39 00:01:32,220 --> 00:01:33,359 what is this all about? 40 00:01:33,360 --> 00:01:35,609 Here's two pictures. 41 00:01:35,610 --> 00:01:37,920 So the first one, this is 42 00:01:39,150 --> 00:01:41,699 called chalk being smashed 43 00:01:41,700 --> 00:01:44,489 by a mousetrap. 44 00:01:44,490 --> 00:01:46,919 And it gives this kind of 45 00:01:46,920 --> 00:01:49,109 exploding rainbow effect, if 46 00:01:49,110 --> 00:01:50,339 you will. 47 00:01:50,340 --> 00:01:52,110 And another example 48 00:01:53,130 --> 00:01:54,689 is this. 49 00:01:54,690 --> 00:01:56,939 This one, this is a water 50 00:01:56,940 --> 00:01:59,609 balloon right at the moment 51 00:01:59,610 --> 00:02:00,689 where it pops. 52 00:02:00,690 --> 00:02:02,729 So you can see in the lower right corner, 53 00:02:02,730 --> 00:02:05,039 the rubber skin 54 00:02:05,040 --> 00:02:06,599 is still partially intact, 55 00:02:07,980 --> 00:02:10,288 whereas on other places 56 00:02:10,289 --> 00:02:12,839 the waters are the water is already 57 00:02:12,840 --> 00:02:14,129 basically splashing out. 58 00:02:16,020 --> 00:02:17,020 So. 59 00:02:18,190 --> 00:02:20,590 Just to formulate the requirements, 60 00:02:22,120 --> 00:02:24,549 just talk about freezing some very 61 00:02:24,550 --> 00:02:26,649 fast action and the action is fast in 62 00:02:26,650 --> 00:02:27,849 two senses. 63 00:02:27,850 --> 00:02:30,009 So the first thing, it just happens 64 00:02:30,010 --> 00:02:31,149 very fast. 65 00:02:31,150 --> 00:02:33,279 So you need a very, very 66 00:02:33,280 --> 00:02:35,619 short exposure time of 67 00:02:35,620 --> 00:02:38,049 roughly 150, thousands to 110, 68 00:02:38,050 --> 00:02:40,749 thousands of a second. 69 00:02:40,750 --> 00:02:43,029 And since these things look 70 00:02:43,030 --> 00:02:45,159 good only for, let's say, a couple 71 00:02:45,160 --> 00:02:47,289 of milliseconds afterwards, 72 00:02:47,290 --> 00:02:49,339 the water has splashed out a lot. 73 00:02:49,340 --> 00:02:51,489 So it's no longer interesting. 74 00:02:51,490 --> 00:02:53,739 You also need to time the exposure very 75 00:02:53,740 --> 00:02:56,079 precisely. So roughly 76 00:02:56,080 --> 00:02:58,149 with the precision of one millisecond 77 00:02:58,150 --> 00:03:00,069 is desired. 78 00:03:00,070 --> 00:03:01,070 Now. 79 00:03:02,040 --> 00:03:04,589 Common digital cameras, 80 00:03:04,590 --> 00:03:07,379 they fail both of these requirements. 81 00:03:07,380 --> 00:03:10,229 So even if you look at high end 82 00:03:10,230 --> 00:03:12,449 laws, I think 83 00:03:12,450 --> 00:03:14,849 I've seen one over a thousand 84 00:03:14,850 --> 00:03:16,439 of a second. 85 00:03:16,440 --> 00:03:17,440 Um. 86 00:03:18,870 --> 00:03:21,239 But they cheat, so the way a camera 87 00:03:21,240 --> 00:03:23,699 shutter works in eight years, it has 88 00:03:23,700 --> 00:03:24,959 to curtains. 89 00:03:24,960 --> 00:03:26,279 So basically, when you start your 90 00:03:26,280 --> 00:03:28,139 exposure at the first curtain moves down, 91 00:03:28,140 --> 00:03:30,419 it exposes DICENZO to the light 92 00:03:30,420 --> 00:03:32,219 and at the end of the exposure, the 93 00:03:32,220 --> 00:03:34,349 second curtain moves down, blocking the 94 00:03:34,350 --> 00:03:36,419 light. Now if you take a 95 00:03:36,420 --> 00:03:38,699 picture faster than roughly one five 96 00:03:38,700 --> 00:03:40,859 hundredths of a second, they start 97 00:03:40,860 --> 00:03:42,359 moving the second curtain down. 98 00:03:42,360 --> 00:03:44,309 Before the first has opened, there was 99 00:03:44,310 --> 00:03:46,499 basically a slit moving over 100 00:03:46,500 --> 00:03:48,569 the sensor. So every part 101 00:03:48,570 --> 00:03:50,999 gets one eight seconds 102 00:03:51,000 --> 00:03:51,959 of light. 103 00:03:51,960 --> 00:03:54,059 But basically the top of the 104 00:03:54,060 --> 00:03:56,099 image will show you something earlier 105 00:03:56,100 --> 00:03:57,689 than the bottom of the image. 106 00:03:58,920 --> 00:04:01,139 So essentially the camera will 107 00:04:01,140 --> 00:04:03,629 be taken completely out of this. 108 00:04:03,630 --> 00:04:06,689 And an alternative 109 00:04:06,690 --> 00:04:08,789 is fleshless. 110 00:04:08,790 --> 00:04:11,159 No flushers, most commonly 111 00:04:11,160 --> 00:04:12,160 known in this 112 00:04:13,470 --> 00:04:15,849 form. So they have the flashbulb 113 00:04:15,850 --> 00:04:18,179 on top. Then there is usually 114 00:04:18,180 --> 00:04:20,609 some kind of settings panel and 115 00:04:20,610 --> 00:04:22,769 on the bottom there is the hot new 116 00:04:22,770 --> 00:04:23,770 value slided 117 00:04:25,320 --> 00:04:26,219 on the camera. 118 00:04:26,220 --> 00:04:28,319 So what can it do 119 00:04:28,320 --> 00:04:30,839 for us to help in 120 00:04:30,840 --> 00:04:32,909 taking these pictures? 121 00:04:32,910 --> 00:04:35,039 So let's go back to the very basics and 122 00:04:35,040 --> 00:04:37,289 compare flesh with an ordinary 123 00:04:37,290 --> 00:04:40,049 lamp. So if you have an ordinary lamp, 124 00:04:40,050 --> 00:04:42,389 it's basically characterized by giving 125 00:04:42,390 --> 00:04:45,479 off a certain amount of light 126 00:04:45,480 --> 00:04:46,679 per time. 127 00:04:46,680 --> 00:04:48,929 So the amount of light you capture 128 00:04:48,930 --> 00:04:51,089 in the frame basically depends 129 00:04:51,090 --> 00:04:52,739 on how long your shutter is open. 130 00:04:52,740 --> 00:04:55,019 That's, I guess, not 131 00:04:55,020 --> 00:04:56,999 a big surprise to anyone. 132 00:04:57,000 --> 00:04:59,309 Now, if you compare this with 133 00:04:59,310 --> 00:05:00,310 a flash, 134 00:05:01,800 --> 00:05:04,079 a flash, just basically a capacitor 135 00:05:04,080 --> 00:05:06,389 and a flashbulb and a capacitor 136 00:05:06,390 --> 00:05:08,459 as charged, and then it discharges over 137 00:05:08,460 --> 00:05:10,639 the flashbulb, creating 138 00:05:10,640 --> 00:05:13,259 insane amount of power, several kilowatts 139 00:05:13,260 --> 00:05:15,269 for a very short time, giving off these 140 00:05:15,270 --> 00:05:17,519 very fast pulses of light. 141 00:05:17,520 --> 00:05:19,709 So basically, 142 00:05:19,710 --> 00:05:21,869 as long as your shutter is 143 00:05:21,870 --> 00:05:24,449 open, when the flash fires, 144 00:05:24,450 --> 00:05:26,249 you will always get the same result. 145 00:05:26,250 --> 00:05:28,529 No matter if you are exposed for like one 146 00:05:28,530 --> 00:05:30,329 tenth of a second or 10 seconds. 147 00:05:30,330 --> 00:05:32,429 The only thing important is your shutter 148 00:05:32,430 --> 00:05:34,859 is open when the flash fires. 149 00:05:34,860 --> 00:05:37,049 And this is what I'm 150 00:05:37,050 --> 00:05:39,749 going to use to my advantage. 151 00:05:39,750 --> 00:05:41,999 So putting this together, 152 00:05:42,000 --> 00:05:43,859 if you take a picture in complete 153 00:05:43,860 --> 00:05:45,539 darkness, so 154 00:05:46,560 --> 00:05:48,629 only the only light source is 155 00:05:48,630 --> 00:05:51,269 the flash, then 156 00:05:51,270 --> 00:05:53,279 the actual exposure you are going to get 157 00:05:53,280 --> 00:05:55,349 will be independent of whatever you set 158 00:05:55,350 --> 00:05:57,569 on your camera and 159 00:05:57,570 --> 00:05:58,919 your effective exposure. 160 00:05:58,920 --> 00:06:00,989 Time of the image becomes 161 00:06:00,990 --> 00:06:02,729 equal to the flash time. 162 00:06:02,730 --> 00:06:05,249 So basically how long the flash 163 00:06:05,250 --> 00:06:06,250 is on. 164 00:06:07,290 --> 00:06:09,179 Now I. 165 00:06:09,180 --> 00:06:11,579 I searched Google to find 166 00:06:11,580 --> 00:06:13,649 the flash burn time for common 167 00:06:13,650 --> 00:06:15,989 flushers and it turns out 168 00:06:15,990 --> 00:06:17,339 it's in your manual. 169 00:06:17,340 --> 00:06:19,829 So this is from my flash. 170 00:06:19,830 --> 00:06:23,219 It's a Nikon SB 900. 171 00:06:23,220 --> 00:06:25,679 And so you can see that the flash 172 00:06:25,680 --> 00:06:27,779 duration or the burn time 173 00:06:27,780 --> 00:06:29,919 depends basically on the power settings. 174 00:06:29,920 --> 00:06:32,189 So if you fire the flash at full power, 175 00:06:32,190 --> 00:06:34,589 you'll be roughly Gane one thousands 176 00:06:34,590 --> 00:06:35,699 of a second. 177 00:06:35,700 --> 00:06:37,769 And if you tune it down to two, the 178 00:06:37,770 --> 00:06:40,019 lowest power setting, you see 179 00:06:41,190 --> 00:06:43,619 a flash duration of roughly 140 180 00:06:43,620 --> 00:06:45,779 thousands of a second. 181 00:06:45,780 --> 00:06:48,329 And so that's pretty good. 182 00:06:48,330 --> 00:06:50,999 That's actually within the desired 183 00:06:51,000 --> 00:06:53,099 exposure times that I was aiming 184 00:06:53,100 --> 00:06:54,029 for. 185 00:06:54,030 --> 00:06:56,219 And what is also very nice by 186 00:06:56,220 --> 00:06:57,630 changing the power settings 187 00:06:58,700 --> 00:07:01,319 excuse me, we actually have some control 188 00:07:01,320 --> 00:07:03,089 over the exposure. 189 00:07:03,090 --> 00:07:05,279 So that solves problem one, the very 190 00:07:05,280 --> 00:07:07,709 short exposure time that is required 191 00:07:07,710 --> 00:07:10,439 to basically freeze the image in place. 192 00:07:10,440 --> 00:07:11,939 Now, just to give you an idea 193 00:07:13,830 --> 00:07:15,929 of what a high power versus a low power 194 00:07:15,930 --> 00:07:17,759 flash actually looks like. 195 00:07:17,760 --> 00:07:20,279 So on the left, you see a picture taken 196 00:07:20,280 --> 00:07:22,529 with almost full power 197 00:07:22,530 --> 00:07:24,509 and you see these small water droplets 198 00:07:24,510 --> 00:07:26,549 flying out. They become streaks. 199 00:07:26,550 --> 00:07:28,709 That is because the exposure lasts 200 00:07:28,710 --> 00:07:30,509 almost for a millisecond. 201 00:07:30,510 --> 00:07:32,639 So you get this very dynamic 202 00:07:32,640 --> 00:07:34,769 image, whereas on the right side you 203 00:07:34,770 --> 00:07:36,929 see a picture taken with a very low power 204 00:07:36,930 --> 00:07:39,329 flash. So everything is really frozen 205 00:07:39,330 --> 00:07:41,399 in place and 206 00:07:41,400 --> 00:07:43,069 it's basically, yeah. 207 00:07:44,130 --> 00:07:45,029 Frozen at the moment. 208 00:07:45,030 --> 00:07:47,309 So there is, of course, 209 00:07:47,310 --> 00:07:49,769 no absolute truth here. 210 00:07:49,770 --> 00:07:51,989 It depends on what picture 211 00:07:51,990 --> 00:07:52,990 you want to take, 212 00:07:55,260 --> 00:07:57,269 how you select your your flash power, 213 00:07:57,270 --> 00:07:58,270 basically. 214 00:07:59,800 --> 00:08:02,019 So now, having solved this problem, I 215 00:08:02,020 --> 00:08:04,299 already told you that 216 00:08:04,300 --> 00:08:06,669 in digital cameras are too slow. 217 00:08:06,670 --> 00:08:08,979 They also have quite a long delay 218 00:08:08,980 --> 00:08:10,989 when you file them of at least one 219 00:08:10,990 --> 00:08:13,839 hundred milliseconds, which is a 220 00:08:13,840 --> 00:08:15,039 much too long. 221 00:08:15,040 --> 00:08:17,319 So how can you trigger the flash 222 00:08:17,320 --> 00:08:18,320 without a camera? 223 00:08:20,770 --> 00:08:22,659 And the first way it may sound 224 00:08:22,660 --> 00:08:25,219 ridiculous, but you can do it manually, 225 00:08:25,220 --> 00:08:27,369 so most flashers have a button 226 00:08:27,370 --> 00:08:28,719 on the back and when you press the 227 00:08:28,720 --> 00:08:31,119 button, the flash will fire with 228 00:08:31,120 --> 00:08:33,548 whatever power setting you currently 229 00:08:33,549 --> 00:08:34,538 had. 230 00:08:34,539 --> 00:08:36,369 No, of course, this is very inaccurate, 231 00:08:36,370 --> 00:08:39,339 but it's also very easy to do. 232 00:08:39,340 --> 00:08:41,529 And I actually took some pictures that 233 00:08:41,530 --> 00:08:44,559 way, which I am going to show you later. 234 00:08:44,560 --> 00:08:46,629 But of course, at some point 235 00:08:46,630 --> 00:08:48,369 you want to move to bring a 236 00:08:48,370 --> 00:08:50,679 microcontrollers into the picture. 237 00:08:50,680 --> 00:08:52,839 So the most obvious way would 238 00:08:52,840 --> 00:08:55,389 be to interface with the flat flesh 239 00:08:55,390 --> 00:08:57,549 you. So this is how a flash 240 00:08:57,550 --> 00:08:59,439 looks from below on the left side. 241 00:08:59,440 --> 00:09:00,669 It's quite a new flash. 242 00:09:00,670 --> 00:09:03,699 It has five contacts 243 00:09:03,700 --> 00:09:06,009 for like digital 244 00:09:06,010 --> 00:09:08,109 communications, for measurements, for 245 00:09:08,110 --> 00:09:09,639 several signals. 246 00:09:09,640 --> 00:09:11,319 So modern fleshless have lots of 247 00:09:11,320 --> 00:09:13,450 functionality, which we don't need 248 00:09:15,310 --> 00:09:16,310 to. Right. 249 00:09:16,990 --> 00:09:19,149 What you basically show is a very old 250 00:09:19,150 --> 00:09:21,219 flash. It has just one contact. 251 00:09:21,220 --> 00:09:23,949 And this contact is called 252 00:09:23,950 --> 00:09:26,529 Hexing for experiencing 253 00:09:26,530 --> 00:09:28,690 because SINNEN is what's in the bulb. 254 00:09:29,830 --> 00:09:31,929 And the only thing that I'm going 255 00:09:31,930 --> 00:09:34,629 to need is this xanthan. 256 00:09:34,630 --> 00:09:36,279 And as you can see, if you compare the 257 00:09:36,280 --> 00:09:38,199 two pictures, it's also there on the left 258 00:09:38,200 --> 00:09:40,449 side. That's because 259 00:09:40,450 --> 00:09:42,429 there wasn't a standard for this and it 260 00:09:42,430 --> 00:09:43,719 defines the pin. 261 00:09:43,720 --> 00:09:45,489 So all the flashes that were made in the 262 00:09:45,490 --> 00:09:47,949 last 25 years and therefore what you 263 00:09:47,950 --> 00:09:49,109 will probably work. 264 00:09:49,110 --> 00:09:51,339 So you have really a lot of flashes 265 00:09:51,340 --> 00:09:52,960 to shop for, which 266 00:09:53,980 --> 00:09:55,299 probably makes it quite cheap. 267 00:09:56,800 --> 00:09:58,989 Now, unfortunately, this hot issue is 268 00:09:58,990 --> 00:10:00,579 very tedious to work with. 269 00:10:00,580 --> 00:10:02,889 So those contacts are basically spring 270 00:10:02,890 --> 00:10:05,049 contacts. They will slide back in 271 00:10:05,050 --> 00:10:06,939 when you push them. 272 00:10:06,940 --> 00:10:09,459 And so attaching something to it is 273 00:10:09,460 --> 00:10:11,049 this quite tedious. 274 00:10:11,050 --> 00:10:13,029 So there is another method which is 275 00:10:13,030 --> 00:10:14,309 called PC. 276 00:10:15,520 --> 00:10:18,399 This was actually how you fired the Flash 277 00:10:18,400 --> 00:10:19,850 before. That was the hot issue. 278 00:10:21,040 --> 00:10:22,959 So you basically had these cables that 279 00:10:22,960 --> 00:10:25,029 you see on the left with this 280 00:10:25,030 --> 00:10:26,859 weird connector there. 281 00:10:26,860 --> 00:10:29,049 And it's still being 282 00:10:29,050 --> 00:10:31,389 used in studio flushers and 283 00:10:31,390 --> 00:10:33,729 in also a high end 284 00:10:33,730 --> 00:10:34,929 on camera flushers. 285 00:10:34,930 --> 00:10:37,089 However, the cheapest models, they 286 00:10:37,090 --> 00:10:38,529 only have to hot you. 287 00:10:38,530 --> 00:10:40,509 But luckily you can get these these 288 00:10:40,510 --> 00:10:42,939 adapters for like 10 euros 289 00:10:42,940 --> 00:10:45,999 and they basically will convert 290 00:10:46,000 --> 00:10:48,339 your Hjörtur to a PC using 291 00:10:48,340 --> 00:10:50,799 cable. And you can get the cables with 292 00:10:50,800 --> 00:10:52,989 a standard headphone 293 00:10:52,990 --> 00:10:54,969 jack, which is very easy to use in 294 00:10:54,970 --> 00:10:56,949 microcontroller circuits. 295 00:10:56,950 --> 00:10:59,259 So how do you actually find the flash? 296 00:10:59,260 --> 00:11:00,459 That's very easy. 297 00:11:00,460 --> 00:11:03,309 You connect the accent to ground. 298 00:11:03,310 --> 00:11:05,619 So ground on the issue is basically 299 00:11:05,620 --> 00:11:07,899 the metal layer outside and 300 00:11:07,900 --> 00:11:10,059 on the PC. So it's one of 301 00:11:10,060 --> 00:11:12,219 the signals carried with it. 302 00:11:12,220 --> 00:11:14,199 And when you do that, the flesh will 303 00:11:14,200 --> 00:11:15,200 fire. 304 00:11:16,450 --> 00:11:18,759 So that brings me to the last question 305 00:11:18,760 --> 00:11:20,739 regarding flushers, and that is, how 306 00:11:20,740 --> 00:11:22,749 accurately can you fire them or what is 307 00:11:22,750 --> 00:11:24,129 the delay? 308 00:11:24,130 --> 00:11:26,199 So I did not find a lot of 309 00:11:26,200 --> 00:11:28,269 information, so I decided to measure it 310 00:11:28,270 --> 00:11:29,559 myself. 311 00:11:29,560 --> 00:11:31,159 So you can see here, um, 312 00:11:32,230 --> 00:11:34,539 basically a scope measurement of the 313 00:11:34,540 --> 00:11:36,819 signal shown in green starting 314 00:11:36,820 --> 00:11:38,289 on the top left. 315 00:11:38,290 --> 00:11:40,569 And the yellow signal is 316 00:11:40,570 --> 00:11:42,729 the response of a photo diet 317 00:11:42,730 --> 00:11:45,459 which was placed in front of the flash. 318 00:11:45,460 --> 00:11:47,589 And so you can see the green signal 319 00:11:47,590 --> 00:11:49,899 basically goes to ground, which triggers 320 00:11:49,900 --> 00:11:51,219 the flash to fire. 321 00:11:51,220 --> 00:11:54,579 And then in the second quarter 322 00:11:54,580 --> 00:11:56,859 of the after picture, 323 00:11:56,860 --> 00:11:58,929 the photo diad responds and I don't 324 00:11:58,930 --> 00:12:00,189 know if you can see it. 325 00:12:00,190 --> 00:12:02,259 There is a time delay 326 00:12:02,260 --> 00:12:05,079 of thirty nine microseconds 327 00:12:05,080 --> 00:12:07,089 between these two events. 328 00:12:07,090 --> 00:12:09,249 So let's say roughly 100 329 00:12:09,250 --> 00:12:11,079 microseconds. And I repeated this 330 00:12:11,080 --> 00:12:12,080 measurement. 331 00:12:12,950 --> 00:12:15,019 Probably 10 times, and 332 00:12:15,020 --> 00:12:17,119 it's very accurate, so the variation I've 333 00:12:17,120 --> 00:12:19,219 seen is within two 334 00:12:19,220 --> 00:12:21,409 microseconds, so basically 335 00:12:21,410 --> 00:12:22,410 negligible. 336 00:12:23,400 --> 00:12:25,559 So that solves the problem, too. 337 00:12:25,560 --> 00:12:27,809 So apparently we can file this flesh 338 00:12:27,810 --> 00:12:30,149 in a fraction of a millisecond, 339 00:12:30,150 --> 00:12:32,339 which is very good. 340 00:12:33,960 --> 00:12:36,149 So how would, you know, go and 341 00:12:36,150 --> 00:12:38,819 connect your flesh to 342 00:12:38,820 --> 00:12:39,989 a microcontroller? 343 00:12:39,990 --> 00:12:42,149 Basically, the circuitry is as simple as 344 00:12:42,150 --> 00:12:43,169 it gets. 345 00:12:43,170 --> 00:12:45,509 So you connect to 346 00:12:45,510 --> 00:12:47,909 hexing support to some 347 00:12:47,910 --> 00:12:49,440 random digital input 348 00:12:51,210 --> 00:12:52,349 of your microcontroller. 349 00:12:52,350 --> 00:12:54,659 You can use our one as a current 350 00:12:54,660 --> 00:12:56,580 limiter. But to be honest, I 351 00:12:57,600 --> 00:12:58,559 didn't bother with it. 352 00:12:58,560 --> 00:13:00,539 It just worked and. 353 00:13:02,630 --> 00:13:03,980 There's two things to 354 00:13:05,570 --> 00:13:07,549 to do correctly here, because otherwise 355 00:13:07,550 --> 00:13:09,319 I've seen a very overt behavior from your 356 00:13:09,320 --> 00:13:10,459 flesh. 357 00:13:10,460 --> 00:13:12,679 So the way this works is the support from 358 00:13:12,680 --> 00:13:15,379 the flesh. It carries a voltage of 359 00:13:15,380 --> 00:13:17,629 three point three or five watts, 360 00:13:17,630 --> 00:13:19,879 and it's basically 361 00:13:19,880 --> 00:13:22,429 being put up internally by the flesh 362 00:13:22,430 --> 00:13:25,099 and it sends us when some external source 363 00:13:25,100 --> 00:13:26,779 pulls this pin low. 364 00:13:26,780 --> 00:13:28,939 So you do not want to drive 365 00:13:28,940 --> 00:13:31,729 extra high from your microcontroller, 366 00:13:31,730 --> 00:13:33,919 so don't use any pull up resistors. 367 00:13:33,920 --> 00:13:36,319 And even worse, don't configure 368 00:13:36,320 --> 00:13:37,879 dependance, output and pull it. 369 00:13:37,880 --> 00:13:39,949 I've seen very strange behavior 370 00:13:39,950 --> 00:13:42,799 happening when I did this. 371 00:13:42,800 --> 00:13:45,079 And the second thing, when you 372 00:13:45,080 --> 00:13:47,149 pull the pin low, only do 373 00:13:47,150 --> 00:13:48,799 so for a short time, like one or two 374 00:13:48,800 --> 00:13:50,899 milliseconds and then basically 375 00:13:50,900 --> 00:13:52,279 release it again. 376 00:13:52,280 --> 00:13:54,709 Otherwise your flesh 377 00:13:54,710 --> 00:13:56,779 will also start to behave 378 00:13:56,780 --> 00:13:57,889 strangely minded. 379 00:13:57,890 --> 00:13:59,929 So it's stopped firing for a while. 380 00:13:59,930 --> 00:14:01,279 You had to take out the batteries. 381 00:14:02,780 --> 00:14:04,939 Apparently you can confuse them quite 382 00:14:04,940 --> 00:14:07,009 a lot by by keeping the 383 00:14:07,010 --> 00:14:08,299 pin low for too long. 384 00:14:10,040 --> 00:14:12,049 Now, one very important thing to make 385 00:14:12,050 --> 00:14:14,809 sure is you should check the voltage 386 00:14:14,810 --> 00:14:17,329 that your flesh gives you on the pin. 387 00:14:17,330 --> 00:14:19,279 This is not standardized. 388 00:14:19,280 --> 00:14:21,479 So I've seen flashes worth three point 389 00:14:21,480 --> 00:14:23,719 three and five volts, which is very 390 00:14:23,720 --> 00:14:25,969 convenient for microcontrollers. 391 00:14:25,970 --> 00:14:28,309 But I've also seen flashes 392 00:14:28,310 --> 00:14:31,099 with 12 or even 24 worlds 393 00:14:31,100 --> 00:14:33,259 where your microcontroller may actually 394 00:14:33,260 --> 00:14:35,569 be damaged. So in that case, 395 00:14:35,570 --> 00:14:38,179 you may need a little bit more 396 00:14:38,180 --> 00:14:40,309 and elaborated circuit, maybe use up 397 00:14:40,310 --> 00:14:42,139 to a couple of years to kind of 398 00:14:42,140 --> 00:14:44,359 completely separate the two circuits 399 00:14:44,360 --> 00:14:45,829 electronically. 400 00:14:45,830 --> 00:14:48,020 And if you have a very, very old flash, 401 00:14:49,170 --> 00:14:51,379 you should actually make sure 402 00:14:51,380 --> 00:14:53,689 that they don't basically 403 00:14:53,690 --> 00:14:55,549 put the actual flash circuit on there. 404 00:14:55,550 --> 00:14:57,649 So in the very first flushers, 405 00:14:57,650 --> 00:14:59,479 you basically had to 400 volts. 406 00:14:59,480 --> 00:15:01,759 That is on the flash capacitor right 407 00:15:01,760 --> 00:15:02,749 on this pin. 408 00:15:02,750 --> 00:15:04,309 And so the camera would close the 409 00:15:04,310 --> 00:15:05,329 circuit. 410 00:15:05,330 --> 00:15:07,489 And really in a flash 411 00:15:07,490 --> 00:15:09,619 capacitor, there's enough energy to 412 00:15:09,620 --> 00:15:11,159 to knock you out or even kill you. 413 00:15:11,160 --> 00:15:13,519 So if your flash is like twenty 414 00:15:13,520 --> 00:15:14,899 five years old, please check that. 415 00:15:14,900 --> 00:15:15,919 Please don't die. 416 00:15:15,920 --> 00:15:16,920 It's not worth. 417 00:15:25,300 --> 00:15:27,489 All right, so just as 418 00:15:27,490 --> 00:15:29,709 a short recapitulation of 419 00:15:29,710 --> 00:15:31,809 what the flesh does for us, it 420 00:15:31,810 --> 00:15:33,759 gives us a very short exposure time, 421 00:15:33,760 --> 00:15:36,129 roughly 140 thousands of a second. 422 00:15:36,130 --> 00:15:37,609 It's super easy to control with a 423 00:15:37,610 --> 00:15:39,669 microcontroller. So you don't need 424 00:15:39,670 --> 00:15:40,970 to be an electronics expert. 425 00:15:41,980 --> 00:15:44,109 It has a very short firing delay 426 00:15:44,110 --> 00:15:46,599 of 100 microseconds and it basically 427 00:15:46,600 --> 00:15:48,039 fires a super accurately. 428 00:15:49,390 --> 00:15:51,639 So with that being out of the way, 429 00:15:51,640 --> 00:15:53,529 here's how you take a high speed picture 430 00:15:53,530 --> 00:15:55,719 in seven easy steps. 431 00:15:55,720 --> 00:15:57,759 So the first step is you build your set 432 00:15:57,760 --> 00:15:59,889 up so kind of, 433 00:15:59,890 --> 00:16:02,019 I don't know, put up reflectors, put up 434 00:16:02,020 --> 00:16:04,329 a background place, whatever 435 00:16:04,330 --> 00:16:05,830 action you want to take a picture of. 436 00:16:07,120 --> 00:16:09,279 And when you're done that, you're 437 00:16:09,280 --> 00:16:11,289 talking to room and you start a long term 438 00:16:11,290 --> 00:16:13,239 exposure on your camera. 439 00:16:13,240 --> 00:16:14,619 Now, this may sound a little bit 440 00:16:14,620 --> 00:16:16,059 counterintuitive for high speed 441 00:16:16,060 --> 00:16:18,309 photography, but 442 00:16:18,310 --> 00:16:20,439 as I said earlier, since the room is 443 00:16:20,440 --> 00:16:22,539 dark, the exposure 444 00:16:22,540 --> 00:16:24,039 time when your camera doesn't matter at 445 00:16:24,040 --> 00:16:26,289 all. So you might as well as well make 446 00:16:26,290 --> 00:16:28,419 it a long term exposure to kind of give 447 00:16:28,420 --> 00:16:30,489 you enough time to work with. 448 00:16:30,490 --> 00:16:32,679 Now, during this long term exposure, 449 00:16:32,680 --> 00:16:35,049 you start your high speed action. 450 00:16:35,050 --> 00:16:37,059 So I don't know, you fire your crossbow 451 00:16:37,060 --> 00:16:39,249 at an unfortunate episode or whatever. 452 00:16:40,270 --> 00:16:41,270 Um. 453 00:16:42,300 --> 00:16:44,459 You have a microcontroller somewhere 454 00:16:44,460 --> 00:16:46,469 in there and. 455 00:16:46,470 --> 00:16:48,779 It was sense that the action 456 00:16:48,780 --> 00:16:51,149 has started, so maybe it is 457 00:16:51,150 --> 00:16:52,739 sensitive to sound. 458 00:16:52,740 --> 00:16:54,569 There may be a light barrier, something 459 00:16:54,570 --> 00:16:56,789 is flying through or just some 460 00:16:56,790 --> 00:16:59,129 simple improvised switching 461 00:16:59,130 --> 00:17:00,130 going on. 462 00:17:01,500 --> 00:17:03,389 And after the microcontroller senses the 463 00:17:03,390 --> 00:17:05,489 action, it basically it's in a 464 00:17:05,490 --> 00:17:07,559 delay. And after delay delay, 465 00:17:07,560 --> 00:17:08,640 it fails to flash. 466 00:17:09,690 --> 00:17:11,818 So then you wait for your camera 467 00:17:11,819 --> 00:17:13,140 to finish your exposure 468 00:17:14,609 --> 00:17:17,249 and then you will well review 469 00:17:17,250 --> 00:17:18,568 your result. 470 00:17:18,569 --> 00:17:19,739 And most certainly 471 00:17:21,119 --> 00:17:23,189 probably 99 of 100 cases 472 00:17:23,190 --> 00:17:24,689 you will not be satisfied. 473 00:17:24,690 --> 00:17:26,699 So you improve your setup and then you 474 00:17:26,700 --> 00:17:27,719 repeat it all again. 475 00:17:29,850 --> 00:17:31,949 Now, there are two little points I didn't 476 00:17:31,950 --> 00:17:34,169 really talk about yet, and that is how 477 00:17:34,170 --> 00:17:36,659 to build a set, how to do this, 478 00:17:36,660 --> 00:17:39,029 and also how to let the microcontroller 479 00:17:39,030 --> 00:17:41,789 sense whatever action you are doing. 480 00:17:41,790 --> 00:17:44,279 And that is because, 481 00:17:44,280 --> 00:17:47,189 unfortunately, these two parts are 482 00:17:47,190 --> 00:17:49,259 very specific to what you are trying to 483 00:17:49,260 --> 00:17:52,079 do. So instead of giving you more theory, 484 00:17:52,080 --> 00:17:54,209 what I will do is I will explain 485 00:17:54,210 --> 00:17:56,489 to set ups that I did. 486 00:17:57,630 --> 00:17:59,669 I will explain how I built the sets, how 487 00:17:59,670 --> 00:18:01,379 I made the microcontroller sense, the 488 00:18:01,380 --> 00:18:04,109 action to hopefully give you an idea 489 00:18:04,110 --> 00:18:05,909 that you then can build your own setups 490 00:18:05,910 --> 00:18:06,910 on. 491 00:18:07,860 --> 00:18:09,180 So let me start with 492 00:18:10,200 --> 00:18:11,099 the first one. 493 00:18:11,100 --> 00:18:13,349 This is about poking water 494 00:18:13,350 --> 00:18:15,539 balloons and taking 495 00:18:15,540 --> 00:18:17,609 pictures of how they 496 00:18:17,610 --> 00:18:19,460 pop up. 497 00:18:20,700 --> 00:18:22,199 This is the setup. 498 00:18:22,200 --> 00:18:23,200 This is my 499 00:18:24,540 --> 00:18:26,789 bathroom back when I was living in 500 00:18:26,790 --> 00:18:27,809 a dormitory. 501 00:18:27,810 --> 00:18:30,179 And as you can see, it's quite 502 00:18:30,180 --> 00:18:31,349 ticked up. 503 00:18:31,350 --> 00:18:32,640 So there is the camera 504 00:18:34,140 --> 00:18:35,099 on a tripod. 505 00:18:35,100 --> 00:18:37,199 You can see all those cables 506 00:18:37,200 --> 00:18:38,489 around the toilet where the 507 00:18:38,490 --> 00:18:39,839 microcontroller is sitting. 508 00:18:40,840 --> 00:18:41,840 And 509 00:18:43,800 --> 00:18:45,269 that's probably not taking it. 510 00:18:45,270 --> 00:18:47,669 At least I hope the camera's 511 00:18:47,670 --> 00:18:49,769 pointing to the shower 512 00:18:49,770 --> 00:18:52,349 where there is a black curtain 513 00:18:52,350 --> 00:18:54,539 and there is to 514 00:18:54,540 --> 00:18:56,729 basically kuruvilla pods combined 515 00:18:56,730 --> 00:18:58,799 into a function called Alapatt, holding 516 00:18:58,800 --> 00:19:00,749 this scary looking device with a lot of 517 00:19:00,750 --> 00:19:02,099 duct tape. 518 00:19:02,100 --> 00:19:04,499 And you have a flash 519 00:19:04,500 --> 00:19:07,229 on another tripod with a 520 00:19:07,230 --> 00:19:09,329 self-made soft box. 521 00:19:09,330 --> 00:19:11,699 Now you notice the walls. 522 00:19:11,700 --> 00:19:13,679 They are white, which is very lucky. 523 00:19:13,680 --> 00:19:15,389 Otherwise, you would want to put white 524 00:19:15,390 --> 00:19:16,619 fabric there. 525 00:19:16,620 --> 00:19:19,019 And the reason is photographic lighting. 526 00:19:19,020 --> 00:19:20,460 So the only light source 527 00:19:21,720 --> 00:19:24,209 there is is basically 528 00:19:24,210 --> 00:19:25,379 the flash. 529 00:19:25,380 --> 00:19:27,629 And if you notice, typical 530 00:19:27,630 --> 00:19:29,909 flash pictures, they have very hot 531 00:19:29,910 --> 00:19:32,399 shadows. And basically 532 00:19:32,400 --> 00:19:33,899 what you want to have is you want to have 533 00:19:33,900 --> 00:19:36,149 lots of reflection coming from the sides, 534 00:19:36,150 --> 00:19:38,009 from the top, from the bottom, just to 535 00:19:38,010 --> 00:19:40,619 give you a more smooth lighting. 536 00:19:40,620 --> 00:19:42,839 So that's why everything that is 537 00:19:42,840 --> 00:19:44,999 not the background should be 538 00:19:45,000 --> 00:19:47,099 as bright as possible and 539 00:19:47,100 --> 00:19:48,100 ideally white. 540 00:19:50,280 --> 00:19:53,159 However, before, um, 541 00:19:53,160 --> 00:19:55,619 starting to take pictures with this 542 00:19:55,620 --> 00:19:57,630 pokey stick, um. 543 00:19:58,900 --> 00:20:01,059 I did some pictures manually because 544 00:20:01,060 --> 00:20:03,189 I, uh, I took most of these pictures 545 00:20:03,190 --> 00:20:05,469 together with a friend, Chris, 546 00:20:05,470 --> 00:20:08,769 who unfortunately is for this Congress, 547 00:20:08,770 --> 00:20:11,019 and he was very skeptical about 548 00:20:11,020 --> 00:20:13,269 the whole idea. So he said, before I 549 00:20:13,270 --> 00:20:15,549 start programing microcontrollers, 550 00:20:15,550 --> 00:20:17,589 I want to know that this at least can 551 00:20:17,590 --> 00:20:18,609 work. 552 00:20:18,610 --> 00:20:20,789 So I said, OK, we try 553 00:20:20,790 --> 00:20:22,959 manually. And apparently he trusted the 554 00:20:22,960 --> 00:20:25,089 idea enough that 555 00:20:25,090 --> 00:20:27,429 he let me set up a water 556 00:20:27,430 --> 00:20:30,099 balloon that he was holding in his hand 557 00:20:30,100 --> 00:20:32,319 with a knife in complete darkness 558 00:20:32,320 --> 00:20:33,909 while triggering the flash with my other 559 00:20:33,910 --> 00:20:34,910 hand. 560 00:20:42,950 --> 00:20:45,319 So I know it always takes 561 00:20:45,320 --> 00:20:47,539 a bit of magic out of the whole process, 562 00:20:47,540 --> 00:20:49,129 but just to give you an idea of what you 563 00:20:49,130 --> 00:20:51,679 can expect, this is how the picture 564 00:20:51,680 --> 00:20:54,619 came out of the camera and 565 00:20:54,620 --> 00:20:56,989 a little bit of magic later. 566 00:20:56,990 --> 00:20:59,329 It's actually not a lot. 567 00:20:59,330 --> 00:21:01,219 You don't end up with this. 568 00:21:01,220 --> 00:21:03,409 I've already shown it as 569 00:21:03,410 --> 00:21:05,659 an example, 570 00:21:05,660 --> 00:21:07,909 but it's not another long 571 00:21:07,910 --> 00:21:09,589 way. It's a little bit of contrast, a 572 00:21:09,590 --> 00:21:10,699 little bit of sharpening. 573 00:21:10,700 --> 00:21:12,649 And you basically got this. 574 00:21:13,940 --> 00:21:16,099 So the second example would be 575 00:21:16,100 --> 00:21:18,139 this picture. I have also shown it in the 576 00:21:18,140 --> 00:21:20,359 beginning. But just that you see 577 00:21:20,360 --> 00:21:22,459 again how it comes out of camera and 578 00:21:22,460 --> 00:21:24,170 how close the knife was to his fingers. 579 00:21:27,380 --> 00:21:29,449 So and again, the end 580 00:21:29,450 --> 00:21:31,609 result that you've already seen. 581 00:21:33,860 --> 00:21:36,559 So that we are 582 00:21:36,560 --> 00:21:38,689 two lucky shot and 583 00:21:38,690 --> 00:21:40,909 in between there are a lot of shots 584 00:21:40,910 --> 00:21:42,859 that look like this, so if you do it 585 00:21:42,860 --> 00:21:45,349 manually, you will get 586 00:21:45,350 --> 00:21:47,269 lots of crap, basically. 587 00:21:47,270 --> 00:21:49,699 So you may fire the flash before 588 00:21:49,700 --> 00:21:50,839 you may fire the flash. 589 00:21:50,840 --> 00:21:53,059 Way too late, but also you can 590 00:21:53,060 --> 00:21:55,180 see in the top right arm. 591 00:21:57,050 --> 00:21:59,149 This is basically not 592 00:21:59,150 --> 00:22:02,119 much too late when the Flash fired, 593 00:22:02,120 --> 00:22:04,099 but you can see the water droplets, they 594 00:22:04,100 --> 00:22:06,829 already spread out all over the frame. 595 00:22:06,830 --> 00:22:08,989 The water itself has moved to this 596 00:22:08,990 --> 00:22:10,279 strange mess. 597 00:22:10,280 --> 00:22:12,409 So really, this process 598 00:22:12,410 --> 00:22:14,509 looks only good for the first 599 00:22:14,510 --> 00:22:15,589 few milliseconds. 600 00:22:15,590 --> 00:22:17,869 And afterwards, at least in my opinion, 601 00:22:17,870 --> 00:22:19,729 it becomes largely uninteresting. 602 00:22:19,730 --> 00:22:21,739 So let's finally bring microcontrollers 603 00:22:21,740 --> 00:22:22,740 in the picture. 604 00:22:23,660 --> 00:22:25,849 So this is the water 605 00:22:25,850 --> 00:22:28,129 balloon pop sensor in big. 606 00:22:28,130 --> 00:22:31,309 It's pretty simple, really. 607 00:22:31,310 --> 00:22:33,289 So basically it does a wire connected to 608 00:22:33,290 --> 00:22:35,419 the body using this crew 609 00:22:35,420 --> 00:22:36,319 there. 610 00:22:36,320 --> 00:22:38,330 And there is a second open wire 611 00:22:39,410 --> 00:22:41,509 very close to the 612 00:22:41,510 --> 00:22:43,109 poky end of it. 613 00:22:43,110 --> 00:22:45,139 And the idea is the water balloon falls 614 00:22:45,140 --> 00:22:47,299 under the water, creates a connection, 615 00:22:47,300 --> 00:22:48,499 and this is what turns our 616 00:22:48,500 --> 00:22:49,399 microcontroller. 617 00:22:49,400 --> 00:22:51,499 Now it's time to to trigger 618 00:22:51,500 --> 00:22:52,500 the exposure. 619 00:22:53,450 --> 00:22:56,109 So here's a bit of. 620 00:22:56,110 --> 00:22:58,509 The circuit I attended 621 00:22:58,510 --> 00:22:59,680 in the more famous symbol, 622 00:23:00,790 --> 00:23:01,790 so. 623 00:23:07,590 --> 00:23:09,839 Again, it's not rocket science, 624 00:23:09,840 --> 00:23:12,239 you have to sense, which is the digital 625 00:23:12,240 --> 00:23:14,429 inputs, it's being put 626 00:23:14,430 --> 00:23:16,649 with a pull up resistor to 627 00:23:16,650 --> 00:23:18,809 five watts and it connects to the 628 00:23:18,810 --> 00:23:19,810 open end. 629 00:23:20,520 --> 00:23:23,099 And actually the ground contact 630 00:23:23,100 --> 00:23:24,750 does connect to the body 631 00:23:25,890 --> 00:23:27,869 of this work quick. 632 00:23:27,870 --> 00:23:29,969 And so once the water drops 633 00:23:29,970 --> 00:23:32,039 in there, this little 634 00:23:32,040 --> 00:23:34,200 dashed circuit, they are closer's 635 00:23:35,850 --> 00:23:38,069 pulling descends. Pillow's with a 636 00:23:38,070 --> 00:23:39,779 microcontroller, sees a falling edge to 637 00:23:39,780 --> 00:23:42,269 interrupt Fragos and the explosion 638 00:23:42,270 --> 00:23:44,069 starts. So that's really as simple as it 639 00:23:44,070 --> 00:23:45,070 gets. 640 00:23:45,660 --> 00:23:47,279 So, yes, a little bit of code. 641 00:23:47,280 --> 00:23:49,619 It's the actual interrupt service routine 642 00:23:51,090 --> 00:23:53,189 I used to not, you know, because 643 00:23:53,190 --> 00:23:54,419 it was there and it worked. 644 00:23:54,420 --> 00:23:56,459 So use whatever you want. 645 00:23:56,460 --> 00:23:58,829 Um, I hope there is not a microcontroller 646 00:23:58,830 --> 00:24:01,049 that does not have two digital Io's 647 00:24:01,050 --> 00:24:01,979 and one interrupt. 648 00:24:01,980 --> 00:24:03,720 Otherwise it probably sucks pretty hard. 649 00:24:06,660 --> 00:24:08,639 So there's basically four things going on 650 00:24:08,640 --> 00:24:10,949 here. The first thing is, of course, 651 00:24:10,950 --> 00:24:13,289 disabling to interrupt is very important. 652 00:24:13,290 --> 00:24:15,389 Then there is a delay loop 653 00:24:15,390 --> 00:24:17,369 and the two lines in the middle, this is 654 00:24:17,370 --> 00:24:19,739 basically a deep bouncing. 655 00:24:19,740 --> 00:24:21,479 So in that case, um. 656 00:24:22,810 --> 00:24:25,149 If there was if the signal was going 657 00:24:25,150 --> 00:24:27,369 high again while you were waiting, 658 00:24:27,370 --> 00:24:29,679 then the waiting loop would reset, 659 00:24:29,680 --> 00:24:31,959 I think, for the water balloon shot. 660 00:24:31,960 --> 00:24:33,999 I actually commented this out because you 661 00:24:34,000 --> 00:24:36,369 want to trigger on the very first signal 662 00:24:38,080 --> 00:24:39,339 for the most webshots. 663 00:24:39,340 --> 00:24:41,259 I am going to show you later. 664 00:24:41,260 --> 00:24:42,589 You probably want to have this. 665 00:24:42,590 --> 00:24:43,809 So decide what works. 666 00:24:43,810 --> 00:24:45,549 Try it out then. 667 00:24:45,550 --> 00:24:47,499 Here, the flash is fired. 668 00:24:48,610 --> 00:24:50,769 So not by riding low to the output that 669 00:24:50,770 --> 00:24:53,019 was already there, but the port 670 00:24:53,020 --> 00:24:55,749 was previously configured as an input. 671 00:24:55,750 --> 00:24:58,089 So to the Flash, it will look like a very 672 00:24:58,090 --> 00:25:00,379 high register to ground. 673 00:25:00,380 --> 00:25:02,229 So basically like an open pin. 674 00:25:02,230 --> 00:25:03,819 And that's exactly what the Flash wants 675 00:25:03,820 --> 00:25:04,749 to have. 676 00:25:04,750 --> 00:25:07,089 So it wants to drive its spin and 677 00:25:07,090 --> 00:25:08,769 not be bothered with it. 678 00:25:08,770 --> 00:25:11,319 And you only pull it to ground for 679 00:25:11,320 --> 00:25:12,969 like a very short time to trigger the 680 00:25:12,970 --> 00:25:14,199 flash. 681 00:25:14,200 --> 00:25:16,389 Then this part is another delay. 682 00:25:16,390 --> 00:25:19,599 It waits for roughly two milliseconds 683 00:25:19,600 --> 00:25:21,999 and then basically reconfigures the pin 684 00:25:22,000 --> 00:25:24,549 as an input, which tells the Flash, 685 00:25:24,550 --> 00:25:26,769 get ready for the next shot. 686 00:25:26,770 --> 00:25:29,069 And then finally the system 687 00:25:29,070 --> 00:25:31,389 goes disarmed and it also turns 688 00:25:31,390 --> 00:25:32,589 off an allergy. 689 00:25:32,590 --> 00:25:34,059 And to start it again, you press the 690 00:25:34,060 --> 00:25:35,939 reset. So that's not really a UFO. 691 00:25:35,940 --> 00:25:38,019 Now, if you want to 692 00:25:38,020 --> 00:25:39,879 call it, it's on its own GitHub. 693 00:25:39,880 --> 00:25:41,469 It's really very simple. 694 00:25:41,470 --> 00:25:43,509 So if you want to improve on that, be my 695 00:25:43,510 --> 00:25:44,510 guest. 696 00:25:45,660 --> 00:25:47,849 So testing this out, this is one 697 00:25:47,850 --> 00:25:50,219 of the first pictures taking 698 00:25:50,220 --> 00:25:51,890 now using this poking device, 699 00:25:52,950 --> 00:25:55,619 and you can see it hits the moment 700 00:25:55,620 --> 00:25:56,879 quite well. 701 00:25:56,880 --> 00:25:58,739 And again, a little bit of magic. 702 00:25:58,740 --> 00:26:00,839 Later, you end up with 703 00:26:00,840 --> 00:26:02,729 this picture, which is actually one of my 704 00:26:02,730 --> 00:26:04,619 all time favorites of the whole series. 705 00:26:07,560 --> 00:26:09,809 And you can actually remove this poking 706 00:26:09,810 --> 00:26:10,899 device quite easily. 707 00:26:10,900 --> 00:26:13,289 So I didn't spend two hours removing 708 00:26:13,290 --> 00:26:15,419 that. It was more like, I don't know, 15 709 00:26:15,420 --> 00:26:16,420 minutes. 710 00:26:17,010 --> 00:26:19,050 So now what you are going to see is. 711 00:26:20,470 --> 00:26:22,479 Basically, you get the timing right after 712 00:26:22,480 --> 00:26:24,759 a little bit of tweaking every time, 713 00:26:24,760 --> 00:26:26,979 so, no, you can just go ahead by a 100 714 00:26:26,980 --> 00:26:29,019 water balloons, fill them up and just go 715 00:26:29,020 --> 00:26:31,209 ahead, pop them every time. 716 00:26:31,210 --> 00:26:32,829 Because just because you get the timing 717 00:26:32,830 --> 00:26:34,179 right doesn't mean you will get good 718 00:26:34,180 --> 00:26:35,769 pictures. So there are still a lot of 719 00:26:35,770 --> 00:26:37,839 random stuff going on where 720 00:26:37,840 --> 00:26:39,639 the droplets are going, where the rubber 721 00:26:39,640 --> 00:26:41,499 is ending up. All of this is still 722 00:26:41,500 --> 00:26:42,579 random. 723 00:26:42,580 --> 00:26:44,889 So you still want to take basically 724 00:26:44,890 --> 00:26:47,379 a lot of pictures, but at least 725 00:26:47,380 --> 00:26:49,179 the timing is now, right? 726 00:26:49,180 --> 00:26:51,069 Almost every time. 727 00:26:51,070 --> 00:26:53,859 So now you just keep going, 728 00:26:53,860 --> 00:26:56,319 take 40 pictures, 100 pictures, 729 00:26:56,320 --> 00:26:58,419 and then select the ones you like and and 730 00:26:58,420 --> 00:26:59,420 post-process them. 731 00:27:00,610 --> 00:27:02,409 So, yeah, here's the last one. 732 00:27:02,410 --> 00:27:04,509 One of my favorites, because it reminds 733 00:27:04,510 --> 00:27:06,489 me of a sudden eruption. 734 00:27:06,490 --> 00:27:08,649 So whatever 735 00:27:08,650 --> 00:27:10,719 you like, that's then basically 736 00:27:10,720 --> 00:27:12,359 a matter of taste. 737 00:27:13,930 --> 00:27:16,179 So coming 738 00:27:16,180 --> 00:27:18,279 to the second set up, I want 739 00:27:18,280 --> 00:27:20,469 to show this is about 740 00:27:20,470 --> 00:27:21,909 exploding chalk. 741 00:27:21,910 --> 00:27:23,349 I have also shown this 742 00:27:24,610 --> 00:27:27,069 in the very beginning and 743 00:27:27,070 --> 00:27:29,949 the basic technique that this uses is 744 00:27:29,950 --> 00:27:31,639 mousetraps. 745 00:27:31,640 --> 00:27:33,339 So here's a mousetrap. 746 00:27:33,340 --> 00:27:35,410 Basically, the way it works, it's a 747 00:27:36,430 --> 00:27:38,329 one of those deathtraps, actually. 748 00:27:38,330 --> 00:27:40,749 So it has the spring-loaded bar, 749 00:27:40,750 --> 00:27:42,489 it gets pulled over. 750 00:27:42,490 --> 00:27:44,679 And this then holds in place 751 00:27:44,680 --> 00:27:46,689 by this copper rod that you can see and 752 00:27:46,690 --> 00:27:48,129 the copper rod slides under the 753 00:27:48,130 --> 00:27:49,839 triggering mechanism. 754 00:27:49,840 --> 00:27:52,329 So it keeps the bar from 755 00:27:52,330 --> 00:27:53,859 you are basically snapping over 756 00:27:55,060 --> 00:27:57,189 and you would place debate on the trigger 757 00:27:57,190 --> 00:27:58,459 and then the mouse goes there. 758 00:27:58,460 --> 00:28:00,729 It will move debate the the rod 759 00:28:00,730 --> 00:28:02,899 will break free the Bava 760 00:28:02,900 --> 00:28:05,109 snap and it will basically 761 00:28:05,110 --> 00:28:06,110 kill the mouse. 762 00:28:07,060 --> 00:28:09,159 Now, these are very good to 763 00:28:09,160 --> 00:28:11,259 give a very fast and forceful 764 00:28:11,260 --> 00:28:13,479 impact. So to kind of let high 765 00:28:13,480 --> 00:28:15,159 speed stuff happen. 766 00:28:15,160 --> 00:28:17,259 But the triggering mechanism 767 00:28:17,260 --> 00:28:19,569 is not optimal. 768 00:28:19,570 --> 00:28:21,729 So basically your your 769 00:28:21,730 --> 00:28:23,859 zone where all your action is going on, 770 00:28:23,860 --> 00:28:25,629 what will be in your frame is is to the 771 00:28:25,630 --> 00:28:27,939 right, right where the trigger 772 00:28:27,940 --> 00:28:29,769 is. So you have to trigger in the frame. 773 00:28:29,770 --> 00:28:32,109 You need to somehow maybe attach 774 00:28:32,110 --> 00:28:34,179 a threat to it to kind of trigger it from 775 00:28:34,180 --> 00:28:36,159 far away. And all of this stuff will be 776 00:28:36,160 --> 00:28:38,409 in your image giving you big problems 777 00:28:38,410 --> 00:28:39,480 with post-processing. 778 00:28:40,510 --> 00:28:42,609 So what you wanna do is you've 779 00:28:42,610 --> 00:28:45,159 got the mousetrap and 780 00:28:45,160 --> 00:28:47,739 the solution that I ended up was 781 00:28:47,740 --> 00:28:49,029 looks like this. 782 00:28:49,030 --> 00:28:51,429 So you basically bend the rod 783 00:28:51,430 --> 00:28:53,859 into this s like shape 784 00:28:53,860 --> 00:28:56,289 so you can put it on top and it basically 785 00:28:56,290 --> 00:28:58,779 holds itself in place through 786 00:28:58,780 --> 00:29:01,119 the mechanical forces 787 00:29:01,120 --> 00:29:02,120 and. 788 00:29:03,260 --> 00:29:05,509 Basically, you can slide this road off 789 00:29:05,510 --> 00:29:08,119 to the top by maybe attaching a threat 790 00:29:08,120 --> 00:29:09,120 to the small, 791 00:29:10,250 --> 00:29:13,099 to the small hole on the top, and then 792 00:29:13,100 --> 00:29:15,379 it will basically break free and 793 00:29:15,380 --> 00:29:17,209 trigger the exposure. 794 00:29:17,210 --> 00:29:18,919 So now you have all the triggering 795 00:29:18,920 --> 00:29:20,059 mechanism out of frame. 796 00:29:20,060 --> 00:29:21,739 You can just remove it. 797 00:29:21,740 --> 00:29:23,689 And it's also a lot safer to start 798 00:29:23,690 --> 00:29:26,059 because you don't need your fingers 799 00:29:26,060 --> 00:29:28,279 or something else in the zone that 800 00:29:28,280 --> 00:29:29,929 this bar is going to go. 801 00:29:29,930 --> 00:29:32,539 Now, I took a hit with a mousetrap. 802 00:29:32,540 --> 00:29:34,129 It's nothing that you find very long 803 00:29:34,130 --> 00:29:36,019 about, but it's also not something you 804 00:29:36,020 --> 00:29:37,020 really want to have. 805 00:29:38,540 --> 00:29:40,699 So now he has 806 00:29:40,700 --> 00:29:42,559 to set up again. 807 00:29:42,560 --> 00:29:44,689 That's in my living 808 00:29:44,690 --> 00:29:46,909 room. And it's basically an 809 00:29:46,910 --> 00:29:49,249 improvised light tent 810 00:29:49,250 --> 00:29:51,469 built inside of a chair. 811 00:29:54,020 --> 00:29:56,329 So, again, the background and everything 812 00:29:56,330 --> 00:29:58,999 which will be in the frame is 813 00:29:59,000 --> 00:30:01,159 in black and all 814 00:30:01,160 --> 00:30:03,619 the walls and the ceiling is covered 815 00:30:03,620 --> 00:30:05,569 in white fabric. 816 00:30:05,570 --> 00:30:07,729 The reason is the same as before to 817 00:30:07,730 --> 00:30:09,679 get lots of light, to bounce back from 818 00:30:09,680 --> 00:30:11,779 all direction to the scene and give you a 819 00:30:11,780 --> 00:30:13,939 very even illumination. 820 00:30:15,980 --> 00:30:17,899 So you can see the microcontroller there 821 00:30:17,900 --> 00:30:19,789 and the left does lots of wiring going 822 00:30:19,790 --> 00:30:21,919 on. And you can see that the trigger 823 00:30:21,920 --> 00:30:23,449 mechanism on the right. 824 00:30:23,450 --> 00:30:25,549 So there's this crocodile 825 00:30:25,550 --> 00:30:27,889 clip, the white one, which basically 826 00:30:27,890 --> 00:30:30,079 bites to the bar and is 827 00:30:30,080 --> 00:30:32,989 duct taped to the floor. 828 00:30:32,990 --> 00:30:34,309 So you're going to have a lot of duct 829 00:30:34,310 --> 00:30:36,559 tape for this, by the way, and. 830 00:30:38,130 --> 00:30:40,409 As soon as you release the mechanism, 831 00:30:40,410 --> 00:30:42,719 uh, the crocodile clip will 832 00:30:42,720 --> 00:30:44,999 rip off and this 833 00:30:45,000 --> 00:30:47,759 will tell the microcontroller to fire. 834 00:30:47,760 --> 00:30:49,919 So here is another wiring 835 00:30:49,920 --> 00:30:51,479 diagram. 836 00:30:51,480 --> 00:30:53,819 So I decided to put a five 837 00:30:53,820 --> 00:30:56,909 volt signal on the white crocodile clip. 838 00:30:56,910 --> 00:30:57,910 And 839 00:30:58,980 --> 00:31:01,049 the other clip connects 840 00:31:01,050 --> 00:31:04,169 to the input of the microcontroller 841 00:31:04,170 --> 00:31:06,119 and why they are connected. 842 00:31:06,120 --> 00:31:08,009 It will read the high input and as soon 843 00:31:08,010 --> 00:31:10,439 as a disconnect descends, 844 00:31:10,440 --> 00:31:12,349 input will get put down to ground over 845 00:31:12,350 --> 00:31:14,549 our one, which results 846 00:31:14,550 --> 00:31:17,099 in a falling edge on the microcontroller. 847 00:31:17,100 --> 00:31:18,749 And that is what triggers the interrupt 848 00:31:18,750 --> 00:31:20,519 and allows you to feel the flash. 849 00:31:20,520 --> 00:31:22,679 So basically very 850 00:31:22,680 --> 00:31:24,839 similar to last time and 851 00:31:24,840 --> 00:31:25,980 not very complicated. 852 00:31:27,600 --> 00:31:29,819 Now, before you smash chalk, which is 853 00:31:29,820 --> 00:31:32,399 a very messy business, especially 854 00:31:32,400 --> 00:31:34,499 if you decide to do it in your living 855 00:31:34,500 --> 00:31:35,500 room, 856 00:31:37,770 --> 00:31:40,529 I would advise you to start 857 00:31:40,530 --> 00:31:41,579 easy. 858 00:31:41,580 --> 00:31:43,739 So where we had 859 00:31:43,740 --> 00:31:46,109 some bottle caps ready for various 860 00:31:46,110 --> 00:31:47,110 reasons. 861 00:31:49,130 --> 00:31:51,130 So, uh. 862 00:31:57,830 --> 00:31:59,989 You can put some some 863 00:31:59,990 --> 00:32:01,189 drops of water in there 864 00:32:02,420 --> 00:32:04,549 and actually use this for, let's say, 865 00:32:04,550 --> 00:32:06,859 the first 50 shots that you try 866 00:32:06,860 --> 00:32:09,199 to understand, where are the variables 867 00:32:09,200 --> 00:32:10,669 in this process? 868 00:32:10,670 --> 00:32:12,979 So to make it more controlled, 869 00:32:12,980 --> 00:32:15,309 to get the timing right every time, 870 00:32:15,310 --> 00:32:17,539 but especially in this case, took 871 00:32:17,540 --> 00:32:19,789 probably 100 pictures on its own. 872 00:32:19,790 --> 00:32:22,699 And you don't want to smash 100 times. 873 00:32:22,700 --> 00:32:24,199 Not a good idea. 874 00:32:24,200 --> 00:32:25,200 So. 875 00:32:25,960 --> 00:32:27,399 This works pretty well, it dries of 876 00:32:27,400 --> 00:32:29,199 quickly, and it's not really much 877 00:32:29,200 --> 00:32:31,269 cleaning work, and if you 878 00:32:31,270 --> 00:32:34,089 are lucky, you may even get 879 00:32:34,090 --> 00:32:36,219 quite, uh, quite a nice picture out 880 00:32:36,220 --> 00:32:37,809 of it, like like this one. 881 00:32:39,340 --> 00:32:41,499 And so finally, you get your timing 882 00:32:41,500 --> 00:32:42,519 right. 883 00:32:42,520 --> 00:32:44,979 You put the chalk in and 884 00:32:44,980 --> 00:32:47,079 well, this happens. 885 00:32:47,080 --> 00:32:48,080 So 886 00:32:51,410 --> 00:32:53,769 a mouse trap 887 00:32:53,770 --> 00:32:54,639 just doesn't cut it. 888 00:32:54,640 --> 00:32:56,949 So you can see it 889 00:32:56,950 --> 00:32:58,659 puts a little dent in there and some 890 00:32:58,660 --> 00:33:02,329 small pieces fly off, but ultimately 891 00:33:02,330 --> 00:33:04,089 not too exciting. 892 00:33:04,090 --> 00:33:06,129 But as the old saying goes, when the 893 00:33:06,130 --> 00:33:08,259 mouse trap is not strong enough, 894 00:33:08,260 --> 00:33:09,579 you need a rat trap. 895 00:33:09,580 --> 00:33:10,580 So. 896 00:33:18,810 --> 00:33:20,879 So this is basically scared 897 00:33:20,880 --> 00:33:22,049 up and 898 00:33:23,280 --> 00:33:25,259 everything else is the same. 899 00:33:25,260 --> 00:33:27,329 I did not take a hit to 900 00:33:27,330 --> 00:33:30,209 my fingers with this, but 901 00:33:30,210 --> 00:33:32,249 I wanted to see what it does to a metal 902 00:33:32,250 --> 00:33:33,149 setting. 903 00:33:33,150 --> 00:33:35,309 And after seeing 904 00:33:35,310 --> 00:33:37,469 what it did, unfortunately, I don't 905 00:33:37,470 --> 00:33:39,719 have a picture, but I'm fairly certain 906 00:33:39,720 --> 00:33:40,739 it will break your fingers. 907 00:33:40,740 --> 00:33:42,740 So please be careful if you try this. 908 00:33:46,230 --> 00:33:48,389 Then I decided to to Tumut 909 00:33:48,390 --> 00:33:50,579 the trap again, so you can 910 00:33:50,580 --> 00:33:52,319 see it escaped 911 00:33:53,460 --> 00:33:55,859 the escaped wrath again 912 00:33:55,860 --> 00:33:58,709 to to kind of get the trigger away. 913 00:33:58,710 --> 00:34:01,619 I also decided to to cover 914 00:34:01,620 --> 00:34:03,269 the front part in 915 00:34:04,560 --> 00:34:06,959 this black cardboard, which 916 00:34:06,960 --> 00:34:08,488 you will see later. Why? 917 00:34:08,489 --> 00:34:10,408 And you also want to strap this to 918 00:34:10,409 --> 00:34:12,238 something heavy. So this is a left over 919 00:34:12,239 --> 00:34:14,859 piece from my from my kitchen countertop. 920 00:34:14,860 --> 00:34:16,888 So it's a really heavy piece of wood 921 00:34:16,889 --> 00:34:18,988 because these things, well, 922 00:34:18,989 --> 00:34:20,819 they tend to take off when they fire. 923 00:34:20,820 --> 00:34:22,769 So maybe you have seen some of those slow 924 00:34:22,770 --> 00:34:24,928 motion pictures of mousetraps firing 925 00:34:24,929 --> 00:34:26,789 those things fly all over the place. 926 00:34:26,790 --> 00:34:28,259 That is also what happens here. 927 00:34:28,260 --> 00:34:30,539 So attached to something 928 00:34:30,540 --> 00:34:31,540 heavy. 929 00:34:33,190 --> 00:34:34,988 So obviously, those things have enough 930 00:34:34,989 --> 00:34:37,238 power to obliterate chalk, 931 00:34:37,239 --> 00:34:38,239 basically, 932 00:34:39,400 --> 00:34:40,840 and you can also see 933 00:34:41,889 --> 00:34:44,019 why I did this modifications to the 934 00:34:44,020 --> 00:34:46,988 trap. So if you don't, you get this, 935 00:34:46,989 --> 00:34:49,089 uh, this Brightwood in your 936 00:34:49,090 --> 00:34:51,879 frame, which really is very distracting 937 00:34:51,880 --> 00:34:52,899 for the eye. 938 00:34:52,900 --> 00:34:55,059 And you can also see the trap 939 00:34:55,060 --> 00:34:56,919 basically took off with the back to the 940 00:34:56,920 --> 00:34:59,409 side of the frame, which makes it 941 00:34:59,410 --> 00:35:02,259 very difficult to get the focus right 942 00:35:02,260 --> 00:35:04,329 and also to get the framing right of 943 00:35:04,330 --> 00:35:06,729 your picture. So if you 944 00:35:06,730 --> 00:35:09,039 attach it to something, you will again 945 00:35:09,040 --> 00:35:11,049 take a lot of randomness out of the whole 946 00:35:11,050 --> 00:35:13,059 process, and that's what you want to do. 947 00:35:14,440 --> 00:35:17,079 So now, after improving the process, 948 00:35:17,080 --> 00:35:18,969 this is what you may get. 949 00:35:18,970 --> 00:35:21,279 So at some point I started 950 00:35:21,280 --> 00:35:23,529 playing around with was 951 00:35:23,530 --> 00:35:25,599 putting several small pieces of colored 952 00:35:25,600 --> 00:35:26,919 chalk together. 953 00:35:26,920 --> 00:35:29,079 And I really liked, um, 954 00:35:29,080 --> 00:35:31,449 where this is going, so. 955 00:35:33,820 --> 00:35:35,769 Still improving the process. 956 00:35:35,770 --> 00:35:37,839 This is basically the final 957 00:35:37,840 --> 00:35:40,059 form, if you want, I don't know 958 00:35:40,060 --> 00:35:42,309 how many of these towers I built, but, 959 00:35:42,310 --> 00:35:44,439 um, it takes probably five or 960 00:35:44,440 --> 00:35:45,549 six minutes. 961 00:35:45,550 --> 00:35:47,169 So you're only able to bring this into 962 00:35:47,170 --> 00:35:48,909 play when you got your process down, 963 00:35:48,910 --> 00:35:50,709 because otherwise you would be dancing, 964 00:35:50,710 --> 00:35:53,109 doing nothing but, um, 965 00:35:53,110 --> 00:35:55,359 super gluing choc pieces together. 966 00:35:55,360 --> 00:35:57,819 So that's 967 00:35:57,820 --> 00:35:58,820 that's not so good. 968 00:35:59,920 --> 00:36:02,199 So, yeah, then you just try 969 00:36:02,200 --> 00:36:04,449 if you got your process done 970 00:36:04,450 --> 00:36:06,589 and at some point. 971 00:36:06,590 --> 00:36:09,529 You get this picture where everything 972 00:36:09,530 --> 00:36:10,879 is looking pretty good 973 00:36:11,900 --> 00:36:14,149 and then you just have to take it 974 00:36:14,150 --> 00:36:15,439 to the gym. 975 00:36:15,440 --> 00:36:17,869 And finally, after 976 00:36:17,870 --> 00:36:20,149 a lot more post-processing, this 977 00:36:20,150 --> 00:36:22,339 is what what I ended 978 00:36:22,340 --> 00:36:23,569 up with. 979 00:36:23,570 --> 00:36:24,979 Um. 980 00:36:24,980 --> 00:36:25,980 Thank you. 981 00:36:32,690 --> 00:36:35,059 So, yes, exploding rainbows, 982 00:36:36,140 --> 00:36:38,359 so I hope that that gave you an idea 983 00:36:38,360 --> 00:36:39,869 on. 984 00:36:39,870 --> 00:36:42,419 And how you can go about building a set 985 00:36:42,420 --> 00:36:45,179 and also how to to trigger the flash, 986 00:36:45,180 --> 00:36:47,939 so that basically ends the tutorial part 987 00:36:47,940 --> 00:36:49,109 and now for. 988 00:36:50,190 --> 00:36:53,249 The last couple of minutes, 989 00:36:53,250 --> 00:36:55,289 I want to talk a bit about the 990 00:36:55,290 --> 00:36:58,109 requirements of Yogya to show you 991 00:36:58,110 --> 00:36:59,609 that if you are a little bit into 992 00:36:59,610 --> 00:37:01,979 photography, you probably already 993 00:37:01,980 --> 00:37:03,959 have everything that you are going to 994 00:37:03,960 --> 00:37:04,960 need. 995 00:37:05,540 --> 00:37:07,549 So let's talk about cameras. 996 00:37:07,550 --> 00:37:09,739 Um, your camera 997 00:37:09,740 --> 00:37:12,319 needs to be digital, 998 00:37:12,320 --> 00:37:14,479 so you really want to have 999 00:37:14,480 --> 00:37:16,549 this this 1000 00:37:16,550 --> 00:37:18,909 feedback, especially when 1001 00:37:18,910 --> 00:37:21,079 when calibrating your parameters, 1002 00:37:21,080 --> 00:37:23,299 you don't want to wait a few days 1003 00:37:23,300 --> 00:37:25,369 for your film to come back 1004 00:37:25,370 --> 00:37:27,499 until you take the next change in, I 1005 00:37:27,500 --> 00:37:29,739 don't know, your flash delay. 1006 00:37:29,740 --> 00:37:32,179 So. However, of course, since the camera 1007 00:37:32,180 --> 00:37:34,249 is not doing anything except opening the 1008 00:37:34,250 --> 00:37:36,409 shutter, you can of course, put several 1009 00:37:36,410 --> 00:37:38,719 cameras there and one 1010 00:37:38,720 --> 00:37:40,669 of them can as well be a film camera. 1011 00:37:40,670 --> 00:37:43,249 So if you like film, you are not 1012 00:37:43,250 --> 00:37:44,250 a completely 1013 00:37:45,820 --> 00:37:48,019 you're not completely out of options. 1014 00:37:48,020 --> 00:37:50,419 Now, your camera needs to do 1015 00:37:50,420 --> 00:37:52,579 a long term exposure, so I recommend 1016 00:37:52,580 --> 00:37:54,649 five seconds or more. 1017 00:37:54,650 --> 00:37:56,779 This probably removes 1018 00:37:56,780 --> 00:37:58,159 lots of mobile phones. 1019 00:37:58,160 --> 00:38:00,169 I am honestly not sure if those phones 1020 00:38:00,170 --> 00:38:02,269 that can take long term 1021 00:38:02,270 --> 00:38:03,980 exposures. I know mine cannot, 1022 00:38:05,000 --> 00:38:07,159 but this is this is very practical 1023 00:38:07,160 --> 00:38:09,379 to give you some time to to start 1024 00:38:09,380 --> 00:38:10,380 the action. 1025 00:38:11,670 --> 00:38:14,099 I strongly recommend manual 1026 00:38:14,100 --> 00:38:16,319 focusing or at the very 1027 00:38:16,320 --> 00:38:18,509 least, you should be able 1028 00:38:18,510 --> 00:38:20,549 to set your camera, that it's not 1029 00:38:20,550 --> 00:38:22,619 focusing when you start the 1030 00:38:22,620 --> 00:38:25,170 exposure because you want to pre focus 1031 00:38:26,250 --> 00:38:28,589 excuse me, which you can do 1032 00:38:28,590 --> 00:38:30,869 using the autofocus, but 1033 00:38:30,870 --> 00:38:33,119 then you want to turn off the light 1034 00:38:33,120 --> 00:38:34,619 and then you start the exposure. 1035 00:38:34,620 --> 00:38:37,049 And if it tries to autofocus, it will 1036 00:38:37,050 --> 00:38:38,699 it will probably completely 1037 00:38:40,440 --> 00:38:42,029 get your focus wrong. 1038 00:38:42,030 --> 00:38:44,459 But actually I recommend manual focusing, 1039 00:38:44,460 --> 00:38:47,459 especially in the last steps, because, 1040 00:38:47,460 --> 00:38:49,529 as you know, you have a basically 1041 00:38:49,530 --> 00:38:50,879 depth of field. 1042 00:38:50,880 --> 00:38:52,979 And so the objects that 1043 00:38:52,980 --> 00:38:55,139 are in focus are in a certain 1044 00:38:55,140 --> 00:38:57,659 range, basically 1045 00:38:57,660 --> 00:38:59,759 starting a few centimeters away from your 1046 00:38:59,760 --> 00:39:02,609 camera and ending a little bit more. 1047 00:39:02,610 --> 00:39:03,610 And 1048 00:39:04,680 --> 00:39:06,659 you can basically focus predictively. 1049 00:39:06,660 --> 00:39:09,339 So maybe you want to make sure that. 1050 00:39:09,340 --> 00:39:11,829 The area of sharp objects is 1051 00:39:11,830 --> 00:39:13,899 towards your camera, so that's something 1052 00:39:13,900 --> 00:39:16,329 that during the high speed 1053 00:39:16,330 --> 00:39:18,969 start flying towards your camera 1054 00:39:18,970 --> 00:39:20,589 is still in focus. 1055 00:39:20,590 --> 00:39:22,779 But that's like advanced 1056 00:39:22,780 --> 00:39:24,879 stuff when you when you finally got 1057 00:39:24,880 --> 00:39:27,249 the basics covered and 1058 00:39:27,250 --> 00:39:29,409 talking about stuff flying at your 1059 00:39:29,410 --> 00:39:31,689 camera and 1060 00:39:31,690 --> 00:39:33,789 you want to have a long lens. 1061 00:39:33,790 --> 00:39:36,129 So I was using, uh, 1062 00:39:36,130 --> 00:39:38,559 100 millimeter lens because that's 1063 00:39:38,560 --> 00:39:40,659 the longest I got and also the 1064 00:39:40,660 --> 00:39:41,660 best. 1065 00:39:42,220 --> 00:39:44,409 If you have more, I recommend more 1066 00:39:44,410 --> 00:39:45,819 for two reasons. 1067 00:39:45,820 --> 00:39:48,039 If your lens is longer so 1068 00:39:49,840 --> 00:39:52,089 you can be further away from 1069 00:39:52,090 --> 00:39:54,339 your object and still get the same 1070 00:39:54,340 --> 00:39:56,469 size and having your expensive 1071 00:39:56,470 --> 00:39:59,229 camera further away from 1072 00:39:59,230 --> 00:40:01,749 splashing water or even worse, exploding 1073 00:40:01,750 --> 00:40:03,909 chalk is usually 1074 00:40:03,910 --> 00:40:04,910 a good idea. 1075 00:40:06,010 --> 00:40:08,649 And also, if you're using a 1076 00:40:08,650 --> 00:40:10,929 telephoto lens, you basically 1077 00:40:10,930 --> 00:40:13,569 have a narrower angle of view 1078 00:40:13,570 --> 00:40:15,969 and so basically you will get less 1079 00:40:15,970 --> 00:40:18,369 background into the picture, 1080 00:40:18,370 --> 00:40:20,229 whereas if you use a wide angle lens, you 1081 00:40:20,230 --> 00:40:22,209 have a lot of background in your picture. 1082 00:40:22,210 --> 00:40:23,529 And so you need to make all that 1083 00:40:23,530 --> 00:40:25,809 background look good, whereas with the 1084 00:40:25,810 --> 00:40:28,119 lens, it's just a very small area. 1085 00:40:28,120 --> 00:40:30,279 So it makes building your set a 1086 00:40:30,280 --> 00:40:31,280 lot easier. 1087 00:40:33,710 --> 00:40:35,270 All right, um. 1088 00:40:36,770 --> 00:40:39,019 Let's talk about Flash, 1089 00:40:41,240 --> 00:40:44,269 as I said earlier, you can 1090 00:40:44,270 --> 00:40:45,349 basically shop. 1091 00:40:46,510 --> 00:40:48,849 A wide area of letters from last 1092 00:40:48,850 --> 00:40:49,869 25 years 1093 00:40:51,490 --> 00:40:53,380 when the standardized Hotaru 1094 00:40:54,910 --> 00:40:56,170 came into existence. 1095 00:40:59,410 --> 00:41:01,389 There's a couple of flushers that use a 1096 00:41:01,390 --> 00:41:03,339 proprietary hot shower, you should make 1097 00:41:03,340 --> 00:41:05,409 sure to get one with a standard 1098 00:41:05,410 --> 00:41:07,749 one so you can you can be sure 1099 00:41:07,750 --> 00:41:09,789 that you have the accent contact at the 1100 00:41:09,790 --> 00:41:12,309 right place to use with the commonly 1101 00:41:12,310 --> 00:41:13,629 available adapters. 1102 00:41:13,630 --> 00:41:15,729 Otherwise, stuff may get a little 1103 00:41:15,730 --> 00:41:17,529 bit more expensive. 1104 00:41:17,530 --> 00:41:19,749 The flesh needs to have an excellent port 1105 00:41:19,750 --> 00:41:20,770 and they all have. 1106 00:41:22,410 --> 00:41:24,569 You can also go with studio flushers, 1107 00:41:24,570 --> 00:41:26,849 then you will already have the PC sync 1108 00:41:26,850 --> 00:41:28,979 port, which makes the wiring 1109 00:41:28,980 --> 00:41:29,980 a bit easier. 1110 00:41:31,550 --> 00:41:33,199 And it has also shown to be very 1111 00:41:33,200 --> 00:41:35,269 practical to shop for a flash 1112 00:41:35,270 --> 00:41:37,519 that actually supports different 1113 00:41:37,520 --> 00:41:40,009 power settings, because, 1114 00:41:40,010 --> 00:41:42,319 as I said earlier, this allows you 1115 00:41:42,320 --> 00:41:44,659 to set your effective exposure 1116 00:41:44,660 --> 00:41:46,729 time. So there is some some very 1117 00:41:46,730 --> 00:41:47,869 cheap flashes. 1118 00:41:47,870 --> 00:41:50,029 They don't have any controls except of an 1119 00:41:50,030 --> 00:41:51,709 on off switch. 1120 00:41:51,710 --> 00:41:52,609 So you plug them on. 1121 00:41:52,610 --> 00:41:54,829 They will always fire with full power, 1122 00:41:54,830 --> 00:41:56,809 which is probably not optimal. 1123 00:41:56,810 --> 00:41:58,909 You can start with that, but you 1124 00:41:58,910 --> 00:42:01,849 will lose a lot of control 1125 00:42:01,850 --> 00:42:02,909 over your process. 1126 00:42:05,060 --> 00:42:07,159 So some other stuff 1127 00:42:07,160 --> 00:42:09,469 that you may need, um, well, 1128 00:42:09,470 --> 00:42:11,299 a microcontroller. 1129 00:42:11,300 --> 00:42:13,369 So, for example, an auto, 1130 00:42:13,370 --> 00:42:15,499 you know, or whatever you have 1131 00:42:15,500 --> 00:42:16,819 lying around. 1132 00:42:16,820 --> 00:42:18,949 As I said, any microcontroller should 1133 00:42:18,950 --> 00:42:21,379 have to a digital Io's 1134 00:42:21,380 --> 00:42:23,179 and at least one interrupt. 1135 00:42:23,180 --> 00:42:26,719 Otherwise it is hardly a microcontroller. 1136 00:42:26,720 --> 00:42:28,969 Now, I recommend the tripod 1137 00:42:28,970 --> 00:42:31,099 to put your camera on, 1138 00:42:31,100 --> 00:42:32,839 maybe also a flash. 1139 00:42:32,840 --> 00:42:35,479 So if you do like photography 1140 00:42:35,480 --> 00:42:37,219 a little more seriously, you probably 1141 00:42:37,220 --> 00:42:39,049 already have one. 1142 00:42:39,050 --> 00:42:40,669 It doesn't need to be a fancy one. 1143 00:42:40,670 --> 00:42:42,769 So you don't need to 1144 00:42:42,770 --> 00:42:45,049 spend like a thousand euros on a tripod. 1145 00:42:45,050 --> 00:42:46,759 You can get a cheap one or even a used 1146 00:42:46,760 --> 00:42:48,979 one, um, 1147 00:42:48,980 --> 00:42:50,689 because the requirements are not very 1148 00:42:50,690 --> 00:42:53,089 high. So especially for stability 1149 00:42:53,090 --> 00:42:54,889 at long as long as your camera isn't 1150 00:42:54,890 --> 00:42:56,299 tipping over. 1151 00:42:56,300 --> 00:42:58,519 It probably works because 150 1152 00:42:58,520 --> 00:43:00,439 thousands of a second won't be much 1153 00:43:00,440 --> 00:43:02,149 movement in your tripod. 1154 00:43:02,150 --> 00:43:03,799 Um, yeah. 1155 00:43:03,800 --> 00:43:05,179 For the flash you can get really 1156 00:43:05,180 --> 00:43:06,919 creative. You can duct tape it to 1157 00:43:06,920 --> 00:43:09,379 something. You can put it on this small, 1158 00:43:09,380 --> 00:43:11,689 uh, Kuryla pots or 1159 00:43:11,690 --> 00:43:13,759 the Chinese copies of it, 1160 00:43:13,760 --> 00:43:15,649 which also work pretty well. 1161 00:43:15,650 --> 00:43:18,319 Um, I recommend getting some 1162 00:43:18,320 --> 00:43:20,389 black and white fabric 1163 00:43:20,390 --> 00:43:22,819 so you can use this a set 1164 00:43:22,820 --> 00:43:25,099 of fabric, molton 1165 00:43:25,100 --> 00:43:26,149 or something else. 1166 00:43:26,150 --> 00:43:28,129 I have some fabric which is black on one 1167 00:43:28,130 --> 00:43:30,289 side and white on the other, which kind 1168 00:43:30,290 --> 00:43:31,910 of saves on fabric. 1169 00:43:32,930 --> 00:43:35,539 I recommend having a stack of 1170 00:43:35,540 --> 00:43:37,699 cardboard also in the colors of 1171 00:43:37,700 --> 00:43:39,079 black and white. 1172 00:43:39,080 --> 00:43:41,269 They are super practical to build 1173 00:43:41,270 --> 00:43:43,429 like light formost to block of light 1174 00:43:43,430 --> 00:43:45,169 from areas where you don't want to have 1175 00:43:45,170 --> 00:43:48,049 it. And yeah, go to any 1176 00:43:48,050 --> 00:43:50,419 store that sells autist equipment, 1177 00:43:50,420 --> 00:43:52,099 you can get them for like ten cents 1178 00:43:52,100 --> 00:43:54,289 apiece. So not very 1179 00:43:54,290 --> 00:43:56,509 expensive, some 1180 00:43:56,510 --> 00:43:59,719 really, really entry level electronics. 1181 00:43:59,720 --> 00:44:00,949 So maybe it's altering. 1182 00:44:00,950 --> 00:44:03,229 Iren, some, some twiddles, 1183 00:44:03,230 --> 00:44:05,329 some players and maybe a few resistors 1184 00:44:05,330 --> 00:44:07,699 is really all you're going to need. 1185 00:44:07,700 --> 00:44:10,009 You can also very easily use 1186 00:44:10,010 --> 00:44:11,359 a breadboard for it. 1187 00:44:11,360 --> 00:44:13,399 It's almost perfect. 1188 00:44:13,400 --> 00:44:15,589 And duct tape, lots of duct 1189 00:44:15,590 --> 00:44:17,869 tape. So that's the most 1190 00:44:17,870 --> 00:44:19,159 important thing. 1191 00:44:19,160 --> 00:44:20,160 Um. 1192 00:44:21,260 --> 00:44:23,419 So that is some some 1193 00:44:23,420 --> 00:44:25,759 final tips, which also 1194 00:44:25,760 --> 00:44:27,799 is a little bit opinionated, but that's 1195 00:44:27,800 --> 00:44:28,800 how I. 1196 00:44:29,640 --> 00:44:31,319 That's the things that I find that work 1197 00:44:31,320 --> 00:44:32,349 best. 1198 00:44:32,350 --> 00:44:34,559 So before putting up 1199 00:44:34,560 --> 00:44:36,809 a very elaborate set up 1200 00:44:36,810 --> 00:44:39,209 and smashing towers 1201 00:44:39,210 --> 00:44:41,489 of chalk that require you like 10 1202 00:44:41,490 --> 00:44:43,589 minutes of building 1203 00:44:43,590 --> 00:44:46,259 a first start with something easy, 1204 00:44:46,260 --> 00:44:48,659 um, to basically learn the process 1205 00:44:48,660 --> 00:44:51,029 of how your mousetrap snaps, the delays 1206 00:44:51,030 --> 00:44:52,349 that you need to set. 1207 00:44:52,350 --> 00:44:54,839 There will be a lot of misfires there. 1208 00:44:54,840 --> 00:44:55,840 Um. 1209 00:44:57,180 --> 00:44:59,279 Really don't start to 1210 00:44:59,280 --> 00:45:01,079 complicated it, it will just slow you 1211 00:45:01,080 --> 00:45:03,269 down and frustrate you once you got your 1212 00:45:03,270 --> 00:45:04,439 process done. 1213 00:45:04,440 --> 00:45:06,509 Then please go crazy and make 1214 00:45:06,510 --> 00:45:07,619 it perfect, so. 1215 00:45:09,360 --> 00:45:11,849 Yes, you should expect 1216 00:45:11,850 --> 00:45:14,009 a lot of failures when doing this, 1217 00:45:14,010 --> 00:45:16,439 so you can probably expect to and a 100 1218 00:45:16,440 --> 00:45:19,469 pictures to be relatively decent 1219 00:45:19,470 --> 00:45:21,449 and you just need to keep going. 1220 00:45:21,450 --> 00:45:23,189 So there's lots of randomness in this 1221 00:45:23,190 --> 00:45:25,379 process. And if you want to get 1222 00:45:25,380 --> 00:45:27,269 a good picture out of it, you just need 1223 00:45:27,270 --> 00:45:29,399 to sample by taking a 1224 00:45:29,400 --> 00:45:31,589 lot of pictures and eventually being 1225 00:45:31,590 --> 00:45:33,150 lucky and getting a good one. 1226 00:45:35,210 --> 00:45:37,309 Now, what I also 1227 00:45:37,310 --> 00:45:39,619 think is very important is to 1228 00:45:39,620 --> 00:45:41,869 actually build your set to 1229 00:45:41,870 --> 00:45:44,179 HQ in post-processing and also 1230 00:45:44,180 --> 00:45:46,549 to prevent post-processing. 1231 00:45:46,550 --> 00:45:49,139 So as I said, if you have this 1232 00:45:49,140 --> 00:45:51,049 this Brightwood, for example, from The 1233 00:45:51,050 --> 00:45:53,179 Mousetrap in the pictures, not 1234 00:45:53,180 --> 00:45:55,549 only does it look very bad, 1235 00:45:55,550 --> 00:45:57,679 it also is very, very difficult to 1236 00:45:57,680 --> 00:45:59,449 post-process because it's right where the 1237 00:45:59,450 --> 00:46:01,729 action is. So removing this is 1238 00:46:01,730 --> 00:46:03,529 virtually impossible and it will give you 1239 00:46:03,530 --> 00:46:05,659 a lot of work, whereas if you cover 1240 00:46:05,660 --> 00:46:07,849 it with black 1241 00:46:07,850 --> 00:46:10,009 cardboard, it will basically start to 1242 00:46:10,010 --> 00:46:12,079 vanish as soon as you turn 1243 00:46:12,080 --> 00:46:13,369 up the contrast. 1244 00:46:13,370 --> 00:46:15,499 And the black cardboard can even 1245 00:46:15,500 --> 00:46:18,259 be used as a picture element, 1246 00:46:18,260 --> 00:46:20,509 which is not distracting to the eyes, but 1247 00:46:20,510 --> 00:46:21,510 just gives like 1248 00:46:23,030 --> 00:46:25,189 a nice natural framing to 1249 00:46:25,190 --> 00:46:26,610 the bottom part of the picture. 1250 00:46:28,040 --> 00:46:30,199 Um, you should make sure 1251 00:46:30,200 --> 00:46:32,359 to to clean 1252 00:46:32,360 --> 00:46:34,489 and really prepare the set after every 1253 00:46:34,490 --> 00:46:36,619 shot. So I actually used 1254 00:46:36,620 --> 00:46:38,899 a vacuum cleaner to to clean 1255 00:46:38,900 --> 00:46:41,239 up the chalk set up for every 1256 00:46:41,240 --> 00:46:42,499 after every shot. 1257 00:46:42,500 --> 00:46:44,839 Because if you don't do this, then small 1258 00:46:44,840 --> 00:46:46,969 pieces of chalk will be lying 1259 00:46:46,970 --> 00:46:47,869 around everywhere. 1260 00:46:47,870 --> 00:46:49,969 And to smash from 1261 00:46:49,970 --> 00:46:52,039 the rattrap is really, really strong 1262 00:46:52,040 --> 00:46:53,809 through all of these small particles will 1263 00:46:53,810 --> 00:46:56,659 fly up and give you kind of a 1264 00:46:56,660 --> 00:46:58,909 a mist in your in your image, 1265 00:46:58,910 --> 00:47:01,189 which destroys, 1266 00:47:01,190 --> 00:47:02,689 um, the cleanest. 1267 00:47:03,830 --> 00:47:06,019 And finally, if there is some 1268 00:47:06,020 --> 00:47:08,929 elements like the bars of the mousetrap 1269 00:47:08,930 --> 00:47:10,969 that you cannot really mask and you 1270 00:47:10,970 --> 00:47:12,559 cannot really prevent from getting in 1271 00:47:12,560 --> 00:47:14,719 your image, then you should just embrace 1272 00:47:14,720 --> 00:47:16,909 them. So make them 1273 00:47:16,910 --> 00:47:19,099 part of the image and don't try to 1274 00:47:19,100 --> 00:47:21,439 hide them like Hofstee, because 1275 00:47:21,440 --> 00:47:24,379 that is that is never going to look good. 1276 00:47:24,380 --> 00:47:26,959 So most important thing, of course, is 1277 00:47:26,960 --> 00:47:29,809 to have fun and enjoy doing this. 1278 00:47:29,810 --> 00:47:31,429 So here are some resources. 1279 00:47:31,430 --> 00:47:33,529 Um, again, the coat 1280 00:47:33,530 --> 00:47:35,719 on on GitHub, 1281 00:47:35,720 --> 00:47:38,029 um, and some reads 1282 00:47:38,030 --> 00:47:39,829 that I that I find interesting. 1283 00:47:39,830 --> 00:47:42,379 So the first thing is stroboscope. 1284 00:47:42,380 --> 00:47:44,929 It started basically from this 1285 00:47:44,930 --> 00:47:47,119 this blog stroboscope 1286 00:47:47,120 --> 00:47:48,120 and. 1287 00:47:50,980 --> 00:47:53,049 It became a little bit of a 1288 00:47:53,050 --> 00:47:54,519 movement in photography, 1289 00:47:55,630 --> 00:47:57,849 so it's show business, basically, 1290 00:47:57,850 --> 00:48:00,549 people using flares, 1291 00:48:00,550 --> 00:48:02,649 sometimes lots of flashes 1292 00:48:02,650 --> 00:48:04,959 in a creative way to get good 1293 00:48:04,960 --> 00:48:07,059 pictures. So you can find lots of 1294 00:48:07,060 --> 00:48:09,819 tutorials on how to use a flash 1295 00:48:09,820 --> 00:48:12,279 to give you good photographic lightning. 1296 00:48:12,280 --> 00:48:13,460 I learned a lot from them. 1297 00:48:14,860 --> 00:48:17,109 You can actually look up high 1298 00:48:17,110 --> 00:48:19,359 speed photography tutorials, 1299 00:48:19,360 --> 00:48:21,279 seriously put it like this on Google. 1300 00:48:21,280 --> 00:48:23,499 You'll find tons of results. 1301 00:48:23,500 --> 00:48:24,500 Um. 1302 00:48:25,260 --> 00:48:27,359 If you are more a friend of 1303 00:48:27,360 --> 00:48:29,549 books, one very 1304 00:48:29,550 --> 00:48:31,679 good read I had was the book 1305 00:48:31,680 --> 00:48:33,599 Light, Science and Magic, which is 1306 00:48:33,600 --> 00:48:35,279 basically a book about photographic 1307 00:48:35,280 --> 00:48:36,299 lightning. 1308 00:48:36,300 --> 00:48:38,429 So if you feel comfortable with 1309 00:48:38,430 --> 00:48:40,649 your camera, I 1310 00:48:40,650 --> 00:48:42,659 can heavily recommend it. 1311 00:48:42,660 --> 00:48:44,729 I am not associated with them, 1312 00:48:44,730 --> 00:48:46,559 um, just to make that clear. 1313 00:48:46,560 --> 00:48:47,939 But it's a really good read. 1314 00:48:47,940 --> 00:48:49,709 And you learn a lot about how lightning 1315 00:48:49,710 --> 00:48:51,779 works in the context of 1316 00:48:51,780 --> 00:48:52,780 photography. 1317 00:48:53,940 --> 00:48:56,309 So if you want to contact me, that's 1318 00:48:56,310 --> 00:48:58,149 my email address. 1319 00:48:58,150 --> 00:49:00,569 This is my Flickr account, 1320 00:49:00,570 --> 00:49:01,570 Pouilly. 1321 00:49:03,300 --> 00:49:05,369 I put lots of or actually not so 1322 00:49:05,370 --> 00:49:07,409 many, but all the pictures that I find we 1323 00:49:07,410 --> 00:49:09,449 are showing over the years, so it's not 1324 00:49:09,450 --> 00:49:11,460 just high speed stuff, but also. 1325 00:49:13,000 --> 00:49:15,159 Other studio pictures like the lamps that 1326 00:49:15,160 --> 00:49:17,439 we are there in the beginning and 1327 00:49:17,440 --> 00:49:19,509 since I used a lot of Creative 1328 00:49:19,510 --> 00:49:21,819 Commons pictures for this presentation, 1329 00:49:21,820 --> 00:49:23,649 I finally got around to change the 1330 00:49:23,650 --> 00:49:25,089 license to Creative Commons. 1331 00:49:25,090 --> 00:49:27,249 So if you want to use a picture, be 1332 00:49:27,250 --> 00:49:28,250 my guest. 1333 00:49:35,550 --> 00:49:36,869 So also for 1334 00:49:38,070 --> 00:49:40,169 a little less complicated 1335 00:49:40,170 --> 00:49:42,539 context as compared to email, I created 1336 00:49:42,540 --> 00:49:44,969 a Twitter account actually two days ago, 1337 00:49:44,970 --> 00:49:47,039 so I 1338 00:49:47,040 --> 00:49:49,079 will post a lot of these pictures over 1339 00:49:49,080 --> 00:49:51,779 the next few days and then. 1340 00:49:51,780 --> 00:49:53,399 Well, I see. If I like it or not. 1341 00:49:53,400 --> 00:49:55,139 So see where it goes from there. 1342 00:49:56,580 --> 00:49:58,649 So that's basically the end of my talk. 1343 00:49:58,650 --> 00:50:00,719 I would really love if any one 1344 00:50:00,720 --> 00:50:03,059 of you tries to please contact me through 1345 00:50:03,060 --> 00:50:05,219 any of this media and show me what 1346 00:50:05,220 --> 00:50:07,349 you did. I I like looking 1347 00:50:07,350 --> 00:50:09,569 at pictures almost as much as I like 1348 00:50:09,570 --> 00:50:11,489 taking them and. 1349 00:50:12,990 --> 00:50:15,210 I do have some ideas for. 1350 00:50:16,720 --> 00:50:18,969 High speed photography, which are 1351 00:50:18,970 --> 00:50:21,399 not really suitable for a living room, 1352 00:50:21,400 --> 00:50:23,439 so if you happen to live in Drayson and 1353 00:50:23,440 --> 00:50:25,569 have a room, that doesn't matter if it 1354 00:50:25,570 --> 00:50:28,299 gets a bit messier, um, 1355 00:50:28,300 --> 00:50:30,519 I would really like you to to contact 1356 00:50:30,520 --> 00:50:32,709 me and you may get a 1357 00:50:32,710 --> 00:50:34,869 print out of it. So thank you for your 1358 00:50:34,870 --> 00:50:36,010 attention and. 1359 00:50:45,640 --> 00:50:47,589 OK, thank you, Polygon, for this very 1360 00:50:47,590 --> 00:50:48,789 interesting talk. 1361 00:50:48,790 --> 00:50:50,979 Um, we still have quite a few minutes 1362 00:50:50,980 --> 00:50:52,029 for questions. 1363 00:50:52,030 --> 00:50:54,479 We have microphones in the room. 1364 00:50:54,480 --> 00:50:57,249 You can line up behind the microphones. 1365 00:50:57,250 --> 00:50:58,729 If you really have to leave the room, 1366 00:50:58,730 --> 00:50:59,859 please be quiet. 1367 00:50:59,860 --> 00:51:01,099 That means no talking. 1368 00:51:01,100 --> 00:51:02,079 Thank you. 1369 00:51:02,080 --> 00:51:03,759 And we're going to start with a question 1370 00:51:03,760 --> 00:51:05,309 from the Internet. 1371 00:51:05,310 --> 00:51:07,269 OK, there has been some discussion about 1372 00:51:07,270 --> 00:51:09,159 doing multiple, multiple photos, like a 1373 00:51:09,160 --> 00:51:10,509 series. 1374 00:51:10,510 --> 00:51:13,299 Do you think it is possible to do them 1375 00:51:13,300 --> 00:51:16,149 with the movie mode and open the shutter 1376 00:51:16,150 --> 00:51:18,219 or even using multiple cameras in a 1377 00:51:18,220 --> 00:51:20,079 flash setup with multiple flashes? 1378 00:51:21,870 --> 00:51:23,949 So the first answer is 1379 00:51:25,090 --> 00:51:27,159 we tried this using 1380 00:51:27,160 --> 00:51:28,160 the movie mode 1381 00:51:29,230 --> 00:51:30,250 and actually. 1382 00:51:31,960 --> 00:51:34,359 Set the flesh to kind of a stroke 1383 00:51:34,360 --> 00:51:35,360 mode. 1384 00:51:38,850 --> 00:51:41,069 So it basically fires 30 1385 00:51:41,070 --> 00:51:43,169 flashes a second, so my flash 1386 00:51:43,170 --> 00:51:45,329 can do that, but unfortunately, 1387 00:51:45,330 --> 00:51:46,889 even if you do the movie mode. 1388 00:51:49,100 --> 00:51:51,429 The picture that is taken from the image 1389 00:51:51,430 --> 00:51:53,769 there is small breaks in between, 1390 00:51:53,770 --> 00:51:55,959 so we never really got it to 1391 00:51:55,960 --> 00:51:58,119 a point where the whole picture 1392 00:51:58,120 --> 00:52:00,639 was illuminated by each flash. 1393 00:52:00,640 --> 00:52:02,769 So basically, you you get 1394 00:52:02,770 --> 00:52:06,069 lots of missing pictures or 1395 00:52:06,070 --> 00:52:08,139 partial exposures, 1396 00:52:08,140 --> 00:52:10,359 probably because the way the flash 1397 00:52:10,360 --> 00:52:12,489 is read out by the camera 1398 00:52:12,490 --> 00:52:13,780 and on the other hand, 1399 00:52:15,580 --> 00:52:17,739 30 frames per second, it's 1400 00:52:17,740 --> 00:52:20,249 not really much. So that is, uh. 1401 00:52:21,250 --> 00:52:23,259 Quite a lot of milliseconds between each 1402 00:52:23,260 --> 00:52:25,239 frame, and so you probably have one 1403 00:52:25,240 --> 00:52:27,339 frame, which is good, and 1404 00:52:27,340 --> 00:52:29,469 the rest of the frames, which will be 1405 00:52:29,470 --> 00:52:31,559 very boring. So I try 1406 00:52:31,560 --> 00:52:33,729 to in the movie and it didn't 1407 00:52:33,730 --> 00:52:35,829 work to put it simply, and to build 1408 00:52:35,830 --> 00:52:37,419 multi cameras and multi flash. 1409 00:52:37,420 --> 00:52:39,249 Do you think there's a way to get that 1410 00:52:39,250 --> 00:52:40,250 work? 1411 00:52:40,600 --> 00:52:43,569 If you want to use multiple flushers, 1412 00:52:43,570 --> 00:52:45,399 you can try it. 1413 00:52:45,400 --> 00:52:47,619 Um, you should 1414 00:52:47,620 --> 00:52:49,749 probably use flashes 1415 00:52:49,750 --> 00:52:52,149 of the same build, probably ideally 1416 00:52:52,150 --> 00:52:54,219 even of the same charge. 1417 00:52:54,220 --> 00:52:56,049 So you have exactly the same firing 1418 00:52:56,050 --> 00:52:57,279 delay. 1419 00:52:57,280 --> 00:52:59,979 Then you can probably use several flashes 1420 00:52:59,980 --> 00:53:02,289 if you use flashes from different 1421 00:53:02,290 --> 00:53:04,059 manufacturers. 1422 00:53:04,060 --> 00:53:06,339 I tried this and you will get double 1423 00:53:06,340 --> 00:53:08,409 exposures. So you see there 1424 00:53:08,410 --> 00:53:10,869 is basically two images now, multiple 1425 00:53:10,870 --> 00:53:12,429 cameras. No problem. 1426 00:53:12,430 --> 00:53:14,349 You could point 10 cameras at it. 1427 00:53:14,350 --> 00:53:15,999 As long as they are not in the way of 1428 00:53:16,000 --> 00:53:18,219 each other, you can use as many of them 1429 00:53:18,220 --> 00:53:20,229 as you want. So if any one of you has 1430 00:53:20,230 --> 00:53:22,359 like 20 cameras to spare and 1431 00:53:22,360 --> 00:53:24,399 wants to build a ring to kind of make a 1432 00:53:24,400 --> 00:53:26,139 Matrix style animation, it would be 1433 00:53:26,140 --> 00:53:27,140 really cool. 1434 00:53:31,980 --> 00:53:33,719 Microphone one, please. 1435 00:53:33,720 --> 00:53:36,599 Um, I have a remark about the flashes, 1436 00:53:36,600 --> 00:53:38,639 if you use studio flashes, they have a 1437 00:53:38,640 --> 00:53:40,949 quite long flash tube and 1438 00:53:40,950 --> 00:53:43,259 the flash duration or 1439 00:53:43,260 --> 00:53:45,779 upper end side German 1440 00:53:45,780 --> 00:53:47,579 might then be a problem for high speed 1441 00:53:47,580 --> 00:53:48,719 photography. 1442 00:53:48,720 --> 00:53:50,579 It's when you take a pictures of your 1443 00:53:50,580 --> 00:53:53,729 model and start throwing water or 1444 00:53:53,730 --> 00:53:55,889 colorful powder on your model and 1445 00:53:55,890 --> 00:53:56,940 you want to have it flying, 1446 00:53:58,170 --> 00:54:00,329 then you run into problems 1447 00:54:00,330 --> 00:54:01,769 with cheap flashes. 1448 00:54:01,770 --> 00:54:04,199 Let's say everything under 1000 1449 00:54:04,200 --> 00:54:06,809 euros per flash is then a problem 1450 00:54:06,810 --> 00:54:08,759 with a high speed movement. 1451 00:54:08,760 --> 00:54:11,099 So if you do use studio flashes, yes, 1452 00:54:11,100 --> 00:54:13,469 if you use studio flash as a small 1453 00:54:13,470 --> 00:54:15,779 type of camera, flashes generally don't 1454 00:54:15,780 --> 00:54:17,279 have that problem. 1455 00:54:17,280 --> 00:54:18,359 OK, very good to know. 1456 00:54:18,360 --> 00:54:20,819 So don't use studio flushers use 1457 00:54:20,820 --> 00:54:22,199 on camera flash. Thank you. 1458 00:54:22,200 --> 00:54:23,200 I didn't try them yet. 1459 00:54:25,240 --> 00:54:26,469 Microphone to please 1460 00:54:27,700 --> 00:54:29,319 on the on these cameras, there's like 1461 00:54:29,320 --> 00:54:31,389 these wire shutters, have you 1462 00:54:31,390 --> 00:54:33,549 tried firing the camera, using the 1463 00:54:33,550 --> 00:54:35,699 microchip and 1464 00:54:35,700 --> 00:54:37,929 not yet because it 1465 00:54:37,930 --> 00:54:39,609 would be an accurate. 1466 00:54:39,610 --> 00:54:40,610 So. 1467 00:54:41,570 --> 00:54:43,909 If you have a DSL or they have to 1468 00:54:43,910 --> 00:54:46,219 put up tomorrow, they have to 1469 00:54:46,220 --> 00:54:48,349 prepare the shot like, I don't know, 1470 00:54:48,350 --> 00:54:50,629 clear out the sensor or start to shutter, 1471 00:54:50,630 --> 00:54:52,699 all of this takes probably 1472 00:54:52,700 --> 00:54:55,669 several tens of milliseconds of time. 1473 00:54:55,670 --> 00:54:58,099 And since the trigger pulse only comes 1474 00:54:58,100 --> 00:55:00,349 like really right before, 1475 00:55:00,350 --> 00:55:02,089 you also need to have your exposure 1476 00:55:02,090 --> 00:55:03,169 already. 1477 00:55:03,170 --> 00:55:05,269 Um, this is generally 1478 00:55:05,270 --> 00:55:07,399 too long, so I didn't even try 1479 00:55:07,400 --> 00:55:09,679 it. There is some tutorials 1480 00:55:09,680 --> 00:55:11,030 on the Internet for 1481 00:55:12,620 --> 00:55:13,759 for other scenarios. 1482 00:55:13,760 --> 00:55:15,259 So let's say if you photograph a 1483 00:55:15,260 --> 00:55:17,509 fountain, that doesn't really matter 1484 00:55:17,510 --> 00:55:19,459 when you take the picture, just that you 1485 00:55:19,460 --> 00:55:21,529 freeze the action, then you you 1486 00:55:21,530 --> 00:55:23,599 can probably use 1487 00:55:23,600 --> 00:55:25,819 this cable, this cable trigger. 1488 00:55:25,820 --> 00:55:26,820 Yeah. 1489 00:55:29,150 --> 00:55:30,949 OK, do we have another question from the 1490 00:55:30,950 --> 00:55:31,950 Internet? 1491 00:55:32,780 --> 00:55:34,159 Um, yes, 1492 00:55:35,160 --> 00:55:37,849 um, there was a question about, um, 1493 00:55:37,850 --> 00:55:39,799 what lengths you are using and what if. 1494 00:55:39,800 --> 00:55:41,629 No and did you have any problem with 1495 00:55:41,630 --> 00:55:42,630 focusing? 1496 00:55:43,820 --> 00:55:46,159 Um, so the lens I was using 1497 00:55:46,160 --> 00:55:48,409 was a Tokino 1498 00:55:48,410 --> 00:55:49,639 100. Mm. 1499 00:55:49,640 --> 00:55:52,429 Microland with a 1500 00:55:52,430 --> 00:55:54,499 I think it's an F 2.8, but that 1501 00:55:54,500 --> 00:55:55,969 doesn't really matter because you want to 1502 00:55:55,970 --> 00:55:57,739 close your aperture. 1503 00:55:57,740 --> 00:56:00,529 I think I closed it down to 1504 00:56:00,530 --> 00:56:02,029 f 16 or something. 1505 00:56:02,030 --> 00:56:03,889 So right before you run into problems 1506 00:56:03,890 --> 00:56:06,059 with uh, with uh 1507 00:56:06,060 --> 00:56:08,149 basically deflection because of a two 1508 00:56:08,150 --> 00:56:09,559 small aperture. 1509 00:56:09,560 --> 00:56:11,629 And yes, focusing is a huge problem 1510 00:56:11,630 --> 00:56:13,399 because those objects are very small. 1511 00:56:13,400 --> 00:56:15,079 So this is not very different from a 1512 00:56:15,080 --> 00:56:16,459 macro shot. 1513 00:56:16,460 --> 00:56:18,259 And since you don't really know where 1514 00:56:18,260 --> 00:56:20,509 your stuff ends up, um, during 1515 00:56:20,510 --> 00:56:22,849 your high speed action, um, 1516 00:56:22,850 --> 00:56:24,799 it's a huge source of frustration when 1517 00:56:24,800 --> 00:56:25,909 you get the perfect shot. 1518 00:56:25,910 --> 00:56:28,159 But then your main part is slightly 1519 00:56:28,160 --> 00:56:29,329 out of focus. 1520 00:56:29,330 --> 00:56:31,399 So focusing you will spend 1521 00:56:31,400 --> 00:56:33,469 a lot of time on doing, 1522 00:56:33,470 --> 00:56:35,420 uh, when you try this 1523 00:56:36,500 --> 00:56:38,089 and you will also have lots of failures 1524 00:56:38,090 --> 00:56:39,090 because of focusing. 1525 00:56:42,660 --> 00:56:44,459 Microphone three, please. 1526 00:56:44,460 --> 00:56:46,709 Hi, thanks for the wonderful 1527 00:56:46,710 --> 00:56:47,999 presentation. 1528 00:56:48,000 --> 00:56:50,429 How did you actually create the smoke 1529 00:56:50,430 --> 00:56:51,629 inside lightbulbs? 1530 00:56:51,630 --> 00:56:53,699 Did you run a lot of Gurren through 1531 00:56:53,700 --> 00:56:54,869 the filament? 1532 00:56:54,870 --> 00:56:55,769 Yes. 1533 00:56:55,770 --> 00:56:57,839 So maybe 1534 00:56:57,840 --> 00:56:59,669 as you can see it here, there's light 1535 00:56:59,670 --> 00:57:01,619 coming out of the top. 1536 00:57:01,620 --> 00:57:03,899 So the idea here is I used 1537 00:57:03,900 --> 00:57:06,779 a kind of, um, 1538 00:57:06,780 --> 00:57:08,939 diamond to a like a dentist has 1539 00:57:08,940 --> 00:57:11,189 to put two bulbs, add two holes 1540 00:57:11,190 --> 00:57:12,190 in the. 1541 00:57:12,870 --> 00:57:14,939 In the glass of the light, and then 1542 00:57:14,940 --> 00:57:17,039 you blow in some oxygen to get out of the 1543 00:57:17,040 --> 00:57:19,379 protective gas, and then you just attach 1544 00:57:19,380 --> 00:57:21,719 it to a normal outlet and 1545 00:57:21,720 --> 00:57:22,709 that's going to happen. 1546 00:57:22,710 --> 00:57:23,729 OK, thank you. 1547 00:57:23,730 --> 00:57:24,730 I'll try it. 1548 00:57:30,820 --> 00:57:33,009 And microphone one, hi. 1549 00:57:33,010 --> 00:57:35,379 Have you ever tried to shoot with 1550 00:57:35,380 --> 00:57:37,659 a crossbow in a complete 1551 00:57:37,660 --> 00:57:38,660 darkness? 1552 00:57:42,930 --> 00:57:45,149 No, I didn't, but if I try, I 1553 00:57:45,150 --> 00:57:47,519 would probably put it on something 1554 00:57:47,520 --> 00:57:50,039 that takes aming out of the picture 1555 00:57:50,040 --> 00:57:51,839 and I wouldn't hold the apple with my 1556 00:57:51,840 --> 00:57:52,840 hands. 1557 00:57:56,200 --> 00:57:58,399 Um, another question from 1558 00:57:58,400 --> 00:58:00,499 the Internet, um, yes, 1559 00:58:00,500 --> 00:58:03,889 um, there is a question about 1560 00:58:03,890 --> 00:58:05,239 the way you are triggering. 1561 00:58:05,240 --> 00:58:07,519 You can either sense the moment the trap 1562 00:58:07,520 --> 00:58:09,739 goes off and release the trap 1563 00:58:09,740 --> 00:58:11,749 trap at a specific time for the mouse 1564 00:58:11,750 --> 00:58:12,469 trap. 1565 00:58:12,470 --> 00:58:14,839 Um, so do you think there's any merit in 1566 00:58:14,840 --> 00:58:17,449 timing the process rather than sending 1567 00:58:17,450 --> 00:58:18,949 a sensing when the trap releases? 1568 00:58:21,120 --> 00:58:23,389 Um, I found that generally to work 1569 00:58:23,390 --> 00:58:25,849 better if 1570 00:58:25,850 --> 00:58:28,039 you actually get 1571 00:58:28,040 --> 00:58:29,040 your. 1572 00:58:29,680 --> 00:58:31,779 Interrupt before the 1573 00:58:31,780 --> 00:58:34,209 actual action happens, so when 1574 00:58:34,210 --> 00:58:36,309 the bar breaks loose, of 1575 00:58:36,310 --> 00:58:38,250 course, you should get, um, 1576 00:58:39,430 --> 00:58:40,959 you should get your signal, then it's 1577 00:58:40,960 --> 00:58:43,389 already quite deterministic 1578 00:58:43,390 --> 00:58:45,549 when the actual action will happen so 1579 00:58:45,550 --> 00:58:48,219 that you can configure the delay. 1580 00:58:48,220 --> 00:58:49,689 And there is some idea. 1581 00:58:49,690 --> 00:58:51,279 So, for example, when you have and you're 1582 00:58:51,280 --> 00:58:53,499 firing, let's say, a crossbow or a bullet 1583 00:58:53,500 --> 00:58:55,599 at something, you can also put 1584 00:58:55,600 --> 00:58:58,299 the mechanism behind it maybe to, 1585 00:58:58,300 --> 00:59:00,639 uh, aluminum foil, which gets smashed 1586 00:59:00,640 --> 00:59:03,549 together by the impacting bullet. 1587 00:59:03,550 --> 00:59:05,829 But then you really you're basically 1588 00:59:05,830 --> 00:59:08,109 too late. So then you don't have control. 1589 00:59:08,110 --> 00:59:10,449 You basically want to instantly 1590 00:59:10,450 --> 00:59:12,129 when the interrupt comes, you want to 1591 00:59:12,130 --> 00:59:14,019 fire the flash so 1592 00:59:15,310 --> 00:59:17,199 you lose control. So I found it to work 1593 00:59:17,200 --> 00:59:19,769 much better when, um. 1594 00:59:22,070 --> 00:59:24,109 When I did have the information before, 1595 00:59:24,110 --> 00:59:26,299 so I think I even tried to trigger 1596 00:59:26,300 --> 00:59:28,639 on the bar smashing down on something, 1597 00:59:28,640 --> 00:59:30,109 because obviously you can pull the 1598 00:59:30,110 --> 00:59:31,879 trigger there as well. 1599 00:59:31,880 --> 00:59:33,619 But it didn't work that good because you 1600 00:59:33,620 --> 00:59:36,139 have to small 100 microsecond delay 1601 00:59:36,140 --> 00:59:38,329 before the flash fires and 1602 00:59:38,330 --> 00:59:40,159 that can actually be too much. 1603 00:59:40,160 --> 00:59:42,349 So I find it to work better, getting 1604 00:59:42,350 --> 00:59:44,839 to know before finding 1605 00:59:44,840 --> 00:59:46,969 out a good delay and and then 1606 00:59:46,970 --> 00:59:48,349 firing the flash based on that. 1607 00:59:49,860 --> 00:59:52,049 OK, we have time for one more 1608 00:59:52,050 --> 00:59:53,909 short question, so microphone two, 1609 00:59:53,910 --> 00:59:55,229 please. 1610 00:59:55,230 --> 00:59:57,479 How do you how do you complete 1611 00:59:57,480 --> 00:59:58,560 darkness room? 1612 01:00:00,520 --> 01:00:02,879 Well, take a picture at night, 1613 01:00:02,880 --> 01:00:04,769 cover the windows and turn off the 1614 01:00:04,770 --> 01:00:05,159 lights. 1615 01:00:05,160 --> 01:00:06,160 So, 1616 01:00:09,810 --> 01:00:11,339 OK, thank you. 1617 01:00:11,340 --> 01:00:13,409 Once again, putting on and please give 1618 01:00:13,410 --> 01:00:14,729 them a big round of applause.