WEBVTT - Getting MIDI With It 0:00:04.160 --> 0:00:07.200 Get in text with technology with tech Stuff from half 0:00:07.240 --> 0:00:13.880 stuff works dot com. PA there and welcome to tech Stuff. 0:00:13.920 --> 0:00:17.160 I'm your host, Jonathan Strickland. I'm an executive producer and 0:00:17.320 --> 0:00:20.959 I love all things tech and today I'm going to 0:00:21.040 --> 0:00:25.080 tackle a topic suggested by listener Jesse, So shout out 0:00:25.120 --> 0:00:28.440 to Jesse. Jesse wanted to know more about MIDDI. What 0:00:28.840 --> 0:00:31.280 is middy, how does it work? Why is it nearly 0:00:31.360 --> 0:00:35.320 synonymous with computer music? And is it actually a type 0:00:35.320 --> 0:00:38.479 of music in of itself. I'm going to tackle that 0:00:38.560 --> 0:00:41.000 last question first, because that was just an easy setup. 0:00:41.040 --> 0:00:43.800 While some people use MIDDY as shorthand for computer music, 0:00:43.800 --> 0:00:47.199 the two are not exactly the same thing. MIDDI stands 0:00:47.240 --> 0:00:51.320 for Musical Instrument Digital Interface and it's a protocol, a 0:00:51.360 --> 0:00:55.320 set of rules that allows a synthesizer or MIDI controller 0:00:55.360 --> 0:00:59.040 to send data to a computer or other synthesizers in 0:00:59.080 --> 0:01:02.960 a meaningful way. And in fact, no sound is sent 0:01:03.000 --> 0:01:07.039 through MINI at all, which might seem a little strange. 0:01:07.440 --> 0:01:11.800 To understand MINY and how it works, it first behooves 0:01:11.920 --> 0:01:15.000 us to go into a little history on synthesizers in general, 0:01:15.160 --> 0:01:20.120 starting with analog synthesizers. Now an analog synthesizer is an 0:01:20.120 --> 0:01:24.280 electronic musical instrument that makes use of various components to 0:01:24.360 --> 0:01:28.960 produce and shape sound. These components can be modular. In fact, 0:01:28.959 --> 0:01:33.080 the earliest analog synthesizers were entirely modular. You have to 0:01:33.120 --> 0:01:36.560 get a whole bunch of different components and patch them 0:01:36.600 --> 0:01:40.000 together with cables. This is what we call patches in 0:01:40.120 --> 0:01:42.360 order to make any sort of meaningful sound at all, 0:01:42.760 --> 0:01:45.800 and certain modules are in charge of creating certain effects 0:01:45.959 --> 0:01:51.360 or sounds. Modules can include stuff like oscillators, filters, and 0:01:51.480 --> 0:01:57.400 voltage control amplifiers. Typically, a synthesizer has at minimum three 0:01:57.600 --> 0:02:01.760 basic modules. The first is an oscillator. The oscillator's job 0:02:01.880 --> 0:02:05.280 is to create a base tone. This tone is what 0:02:05.400 --> 0:02:08.240 the rest of the modules can shape to create the 0:02:08.360 --> 0:02:12.280 different pitches and effects that change the shape of the sound. 0:02:13.040 --> 0:02:16.760 And oscillator causes energy to move between two states at 0:02:16.760 --> 0:02:20.280 a particular frequency. Now this is easiest to imagine. I 0:02:20.320 --> 0:02:23.720 think with a physical oscillator like a pendulum. If you 0:02:23.760 --> 0:02:27.320 push a pendulum, it will begin to swing or oscillate, 0:02:28.040 --> 0:02:31.320 and one full swing is one full oscillation. At the 0:02:31.320 --> 0:02:33.560 height of its swing, all of the energy in the 0:02:33.600 --> 0:02:36.799 system is potential energy. Right, it's not moving. It's at 0:02:36.840 --> 0:02:40.720 its highest point that energy converts into kinetic energy as 0:02:40.720 --> 0:02:44.880 gravity takes hold in the pendulum swings downward. This swinging 0:02:45.200 --> 0:02:49.360 is the oscillation. But oscillators will eventually run out of 0:02:49.440 --> 0:02:52.000 energy due to loss in the system. This is that 0:02:52.120 --> 0:02:56.880 law of thermodynamics. In this physical example, friction cuts down 0:02:56.960 --> 0:02:59.639 on the amount of energy within the system. It actually 0:02:59.680 --> 0:03:01.919 means the you're losing energy out of the system due 0:03:01.960 --> 0:03:06.519 to heat in circuits. Oscillators lose energy due to electrical resistance, 0:03:06.520 --> 0:03:09.040 so it's very similar. The point being that unless you 0:03:09.120 --> 0:03:11.880 continue to pour energy back into the system, it will 0:03:11.919 --> 0:03:16.240 eventually run down because it will lose enough energy so 0:03:16.280 --> 0:03:20.240 it doesn't perpetuate itself anymore. Now, again, to the physics 0:03:20.240 --> 0:03:22.440 of oscillators would take up a bit more time, so 0:03:22.600 --> 0:03:24.560 let's just leave it at the idea that there is 0:03:24.560 --> 0:03:29.160 a component within an analog synthesizer that generates a steady 0:03:29.240 --> 0:03:33.000 frequency that serves as the baseline for all other modules 0:03:33.120 --> 0:03:37.000 in the synthesizer. The second component in a typical synthesizer 0:03:37.080 --> 0:03:40.200 is the mechanism for controlling the oscillator, which is usually 0:03:40.200 --> 0:03:43.000 a keyboard similar to one on a piano, you could 0:03:43.040 --> 0:03:45.960 use other means to change the wave form, though, for example, 0:03:46.040 --> 0:03:50.240 Derriman's use fluctuations in the electromagnetic field to affect the 0:03:50.280 --> 0:03:54.120 baseline waveform. Though precise complete control of the signal isn't 0:03:54.240 --> 0:03:57.000 really possible with such an instrument, even under the control 0:03:57.040 --> 0:04:00.800 of a skilled player. The keyboard or pitch wheel or 0:04:00.880 --> 0:04:05.160 whatever can set the oscillators frequency will affect the pitch. 0:04:05.520 --> 0:04:08.000 The frequency of a sound and how we perceive that 0:04:08.120 --> 0:04:13.240 sound are directly related. Lower frequencies produce lower pitch sounds. 0:04:13.600 --> 0:04:17.240 Human hearing ranges from about twenty hurts or twenty cycles 0:04:17.279 --> 0:04:21.200 of a sound wave per second, up to twenty thousand hurts. 0:04:21.720 --> 0:04:25.920 As we get older, like me, those upper ranges start 0:04:25.960 --> 0:04:28.560 to get harder for us to perceive. This is the 0:04:28.600 --> 0:04:33.440 principle behind some anti youth, anti loitering strategy supposedly employed 0:04:33.440 --> 0:04:37.040 by certain convenience store owners, which have reportedly resorted to 0:04:37.120 --> 0:04:41.000 playing very high pitch sounds that adults can't really hear 0:04:41.080 --> 0:04:43.680 because they've lost that ability, they've lost that range of hearing. 0:04:44.080 --> 0:04:47.480 But those lousy kids in that mangi mutt can totally 0:04:47.600 --> 0:04:49.560 hear it, and it irritates the heck out of them 0:04:49.600 --> 0:04:51.560 so they don't stick a range your store for too long. 0:04:51.920 --> 0:04:55.240 The third component found in typical synthesizers would be the 0:04:55.240 --> 0:04:58.320 filters and effects you can apply to the sounds waveform 0:04:58.440 --> 0:05:01.360 to change the nature of the sound the feel of it. 0:05:01.960 --> 0:05:04.640 These filters let you select which elements of the frequency 0:05:04.720 --> 0:05:07.120 can pass through to an amplifier so that it can 0:05:07.200 --> 0:05:10.560 hit a speaker and be heard by gate Keeping elements 0:05:10.600 --> 0:05:13.599 of frequencies, you can change that shape or nature of 0:05:13.640 --> 0:05:16.160 a sound, which is why you can have a synthesizer 0:05:16.200 --> 0:05:19.240 take on many different sounds even though it's starting with 0:05:19.320 --> 0:05:23.800 the same basic wave form. Beyond frequency or pitch and 0:05:23.960 --> 0:05:27.560 amplitude or volume, you can also manipulate the change in 0:05:27.680 --> 0:05:31.400 volume over the lifespan of a sound. So if you 0:05:31.480 --> 0:05:35.040 press down on a piano key quickly and firmly, you'll 0:05:35.080 --> 0:05:38.520 notice that the sound is initially loud and then fades off, 0:05:38.560 --> 0:05:41.200 and when you let up off the piano key, it 0:05:41.240 --> 0:05:44.800 will eventually stop. If you mess with the sustained pedals, 0:05:44.839 --> 0:05:47.000 you can push down that same key with the same 0:05:47.000 --> 0:05:49.160 force and hear it play out a little differently. And 0:05:49.200 --> 0:05:53.000 we describe this process with synthesizers by dividing it up 0:05:53.040 --> 0:05:58.240 into phases, and they're called attack, decay, sustain and release, 0:05:58.560 --> 0:06:02.159 or a D s R. The attack describes the time 0:06:02.160 --> 0:06:04.599 it takes from the press of a key or the 0:06:04.760 --> 0:06:09.280 null sound zero volume to reach the peak of that 0:06:09.440 --> 0:06:12.320 keys volume. The decay is the time it takes to 0:06:12.360 --> 0:06:15.279 go from the peak of the volume to a designated 0:06:15.480 --> 0:06:18.440 sustain level. The sustain is the volume of sound that 0:06:18.480 --> 0:06:21.719 should play until the respective key is released by the player. 0:06:22.200 --> 0:06:24.080 The release time is the amount of time it takes 0:06:24.120 --> 0:06:27.640 for the sustained volume to decay to null again. The 0:06:27.720 --> 0:06:31.400 various effects on synthesizers can change these elements, creating louder 0:06:31.560 --> 0:06:33.680 or softer sustains. You can even have a sustain that 0:06:33.720 --> 0:06:36.520 gets louder than the attack if you wanted to, and 0:06:36.600 --> 0:06:39.640 you could have longer or shorter decay times and tons 0:06:39.720 --> 0:06:42.880 more effects. It helps create a more dynamic experience with 0:06:43.000 --> 0:06:47.960 a synthesized instrument. After the synthesizer manipulates the basic wave 0:06:48.000 --> 0:06:50.920 form based on the keys pressed or however you're controlling 0:06:50.960 --> 0:06:53.880 the pitch and the various filters or effects that are 0:06:53.880 --> 0:06:57.320 in play, it sends the electrical signal to an amplifier. 0:06:57.680 --> 0:07:00.839 The amplifier's job is to control the volume the played sound, 0:07:00.880 --> 0:07:02.960 typically by passing it through a series of what are 0:07:03.000 --> 0:07:05.839 called envelope controls. That goes back to that a D 0:07:06.120 --> 0:07:09.800 s R. I was talking about. Envelope controls are essentially 0:07:09.840 --> 0:07:12.320 tables of data points that describe the nature of the 0:07:12.320 --> 0:07:16.560 sound generated when a key is pressed. Early synthesizers used 0:07:16.600 --> 0:07:20.120 actual physically distinct modules to control all this, like I 0:07:20.120 --> 0:07:22.560 said before, and you would hook all these modules up 0:07:22.560 --> 0:07:25.760 to a keyboard with various patch wires, and you would 0:07:25.800 --> 0:07:28.640 manipulate various switches and knobs to coax the sound you 0:07:28.680 --> 0:07:31.600 wanted out of the synthesizer. And if you didn't get 0:07:31.600 --> 0:07:33.480 the sound you wanted, you might have to add additional 0:07:33.520 --> 0:07:37.560 components to change things up. Now, the history of synthesizers 0:07:38.240 --> 0:07:41.400 is somewhat debatable, and that's because people disagree over what 0:07:41.560 --> 0:07:45.760 actually counts as a synthesizer. Some say that the tell 0:07:45.800 --> 0:07:49.280 harmonium should count. The tell harmonium, also known as the 0:07:49.520 --> 0:07:54.240 dynamo phone, which I swear sounds like something Homer Simpson 0:07:54.280 --> 0:07:59.080 would say, was invented by Thaddeus Hill in the eighteen nineties. 0:07:59.360 --> 0:08:01.520 It was an elect trick organ that could send music 0:08:01.560 --> 0:08:05.520 electronically across telephone networks. Now, his goal was to create 0:08:05.560 --> 0:08:10.320 an instrument capable of creating perfect tones consistently. Physical musical 0:08:10.360 --> 0:08:13.000 instruments need to be tuned, and they can change their 0:08:13.040 --> 0:08:16.720 tones based upon variables like temperature and humidity, not to 0:08:16.760 --> 0:08:19.400 mention the skill of the person playing, But the tell 0:08:19.440 --> 0:08:24.280 Harmonium would harness electricity to create pitch perfect tones over 0:08:24.480 --> 0:08:27.760 and over again, or so is the idea. But the 0:08:27.760 --> 0:08:31.520 tell Harmonium didn't allow the player to put precise controls 0:08:31.560 --> 0:08:35.240 on the quality of a sound, something that some argue 0:08:35.240 --> 0:08:38.240 should be a basic trait of synthesizers. So they say, well, 0:08:38.240 --> 0:08:42.600 it shouldn't count. The Thereman, which came out in nineteen nineteen, 0:08:42.840 --> 0:08:48.040 also fails in this regard. French inventors Eduard Couplau and 0:08:48.200 --> 0:08:52.079 Armand give Lais created the piano with electronic components in 0:08:53.120 --> 0:08:56.440 that comes closer to the definition many except as cannon 0:08:56.559 --> 0:08:59.880 for synthesizers. The first device to actually use the word 0:09:00.040 --> 0:09:03.760 synthesizer appears to have been the r C A Electronic 0:09:03.960 --> 0:09:07.720 music Synthesizer Mark one, which debuted in nineteen fifty six 0:09:08.080 --> 0:09:12.160 and used tuning forks to generate tones It read music 0:09:12.240 --> 0:09:15.000 from a strip of paper tape that had holes punched 0:09:15.000 --> 0:09:17.040 into it, so it's sort of like a player piano. 0:09:17.520 --> 0:09:20.680 But if we're talking modern synthesizers, we got to talk 0:09:20.760 --> 0:09:25.440 about Robert Moge, the genius behind the Mogue synthesizer. I've 0:09:25.440 --> 0:09:28.160 done a full episode about Mogue in the past, so 0:09:28.240 --> 0:09:30.600 I'm not gonna dwell on it too much here. I'll 0:09:30.600 --> 0:09:33.800 just add that he created the first commercial synthesizer by 0:09:33.840 --> 0:09:36.800 modern standards in nineteen sixty four, and it was the 0:09:36.920 --> 0:09:42.839 Moges nine series Modular Systems. One big limitation in most 0:09:42.920 --> 0:09:46.400 analog synthesizers is in the number of notes it can 0:09:46.440 --> 0:09:52.280 play simultaneously. Many analog synthesizers are monophonic, meaning they can 0:09:52.320 --> 0:09:55.440 only produce one tone at a time. If you held 0:09:55.480 --> 0:09:58.760 down two keys, you would not get two tones. If 0:09:58.800 --> 0:10:01.200 you want to create a poly phonic sound the way 0:10:01.240 --> 0:10:03.920 you could with say a piano, you'd have to either 0:10:04.000 --> 0:10:06.960 get a whole bunch of musicians together, each playing one 0:10:07.000 --> 0:10:11.239 section of a polyphonic piece on their own mode synthesizer, 0:10:11.480 --> 0:10:16.000 whichever analog synthesizer they're using, all in time with one another, 0:10:16.800 --> 0:10:19.560 or you'd have to record multiple tracks to fill in 0:10:19.600 --> 0:10:24.679 the tones, so each track would represent a different monophonic melody, 0:10:25.080 --> 0:10:28.400 and played together you would get the polyphonic effect. Eventually, 0:10:28.559 --> 0:10:32.400 some analog synthesizers supported polyphonic tones at a limited level, 0:10:32.480 --> 0:10:36.320 for example four notes played simultaneously, and they tended to 0:10:36.320 --> 0:10:41.679 be incredibly expensive. As for digital synthesizers, which are at 0:10:41.720 --> 0:10:45.360 their hearts computers working with bits as a good old 0:10:45.559 --> 0:10:49.240 zeros and ones of machine language, those trace their history 0:10:49.280 --> 0:10:52.720 back to research in the late fifties, but commercial digital 0:10:52.720 --> 0:10:57.000 synthesizers really got their start in the nineteen eighties next 0:10:57.040 --> 0:11:01.079 craze New Wave. Like analog sent the sizers, they generate 0:11:01.200 --> 0:11:04.760 or modulate waveforms to create sounds. The process from a 0:11:04.960 --> 0:11:07.840 very high level is similar, but the details are different, 0:11:08.120 --> 0:11:11.839 and digital synthesizers can do some things that analog synthesizers 0:11:11.880 --> 0:11:15.560 either cannot do or it cannot do very well. For example, 0:11:15.640 --> 0:11:20.319 one analog synthesizer might be monophonic or have limited polyphonic capabilities. 0:11:20.679 --> 0:11:24.520 A basic digital synthesizer could have a polyphony if you like, 0:11:25.000 --> 0:11:29.640 of sixty four notes being played simultaneously, although I should 0:11:29.840 --> 0:11:32.920 add that that depends also on how many voices you're 0:11:32.920 --> 0:11:36.760 playing on this synthesizer. With each voice, you reduce the 0:11:36.880 --> 0:11:40.360 number of notes that can be played simultaneously, because each 0:11:40.760 --> 0:11:43.880 voice gets a certain number of notes dedicated to it. 0:11:44.280 --> 0:11:46.959 That being said, there's no guarantee that a digital synthesizer 0:11:47.200 --> 0:11:51.360 will sound better than an analog one. It could or 0:11:51.400 --> 0:11:53.840 it might not. It all depends upon build quality of 0:11:53.840 --> 0:11:56.840 the two synthesizers. Sound quality relies on more than just 0:11:56.920 --> 0:11:59.280 the number of options you have when you're shaping sound. 0:12:00.080 --> 0:12:03.000 All right, So that's the basic info on synthesizers. Now 0:12:03.080 --> 0:12:06.760 let's talk about many. In the early nineteen eighties, a 0:12:06.760 --> 0:12:09.760 man named Dave Smith saw the need for a universal 0:12:09.800 --> 0:12:12.880 standard that would allow synthesizers to send data to other 0:12:12.920 --> 0:12:17.600 instruments or to computers. This would give musicians unprecedented options 0:12:17.600 --> 0:12:21.839 when making music, including new ways to manipulate sound. Synthesizers 0:12:21.840 --> 0:12:25.520 were versatile, but no two models were exactly alike, particularly 0:12:25.559 --> 0:12:28.720 from different manufacturers. One model might have a really cool 0:12:28.800 --> 0:12:31.680 feature that other synthesizers lacked, but fall short on a 0:12:31.720 --> 0:12:35.800 completely different feature. A universal protocol could let a musician 0:12:35.880 --> 0:12:40.080 chain together multiple instruments or perform additional processes on sound 0:12:40.200 --> 0:12:43.920 at the computer level. Related to this problem, is one 0:12:44.120 --> 0:12:49.479 of competing proprietary approaches to musical interfaces. Without a standard, 0:12:49.880 --> 0:12:53.840 each synthesizer manufacturer would be compelled to produce its own 0:12:53.960 --> 0:12:58.520 interface with other synthesizers and with computers. In fact, such 0:12:58.559 --> 0:13:01.720 standards did exist, they weren't there weren't really standards, They 0:13:01.720 --> 0:13:06.840 were proprietary approaches that were unique to specific manufacturers like Rowland, 0:13:06.840 --> 0:13:09.640 for example, or Yamaha. Then you would have a bunch 0:13:09.640 --> 0:13:13.080 of competing technologies on the market that more likely than not, 0:13:13.240 --> 0:13:16.480 would be impossible to chain together, so you'd be locked 0:13:16.520 --> 0:13:19.560 into one ecosystem. You would have to be all in 0:13:19.640 --> 0:13:21.960 on Rowland, or all in on Mogue, or all in 0:13:22.000 --> 0:13:25.480 on Yamaha. You couldn't mix and match because they wouldn't 0:13:25.480 --> 0:13:27.200 be able to talk to each other. It's sort of 0:13:27.240 --> 0:13:31.000 like the early days of computing before arpanet came along 0:13:31.400 --> 0:13:33.240 and you had a set of protocols that would let 0:13:33.240 --> 0:13:36.960 computers talk to each other. Same basic problem existed at 0:13:37.240 --> 0:13:40.679 at the early nineteen eighties. It was a huge mess 0:13:40.720 --> 0:13:43.880 for musicians and producers, So a universal standard would set 0:13:43.920 --> 0:13:46.960 a level playing field, give musicians and producers the greatest 0:13:47.040 --> 0:13:50.680 number of options when creating music and avoid fragmentation of 0:13:50.720 --> 0:13:53.839 the market. Dave Smith first proposed such a standard in 0:13:53.960 --> 0:13:57.520 nine one at a meeting of the Audio Engineering Society, 0:13:57.800 --> 0:14:01.440 and he called his first approach the Universal Synthesizer Interface. 0:14:02.080 --> 0:14:05.680 Smith recognized that while manufacturers were able to create systems 0:14:05.720 --> 0:14:08.400 that would allow you to control multiple synthesizers made by 0:14:08.440 --> 0:14:11.560 that manufacturer, there was still no standard that would allow 0:14:11.640 --> 0:14:15.880 for interoperability, and manufacturers were concerned that this issue was 0:14:15.960 --> 0:14:20.640 costing them customers by creating this frustrating environment. Two years later, 0:14:20.840 --> 0:14:23.760 he would release the first version of the MIDI protocols. 0:14:23.800 --> 0:14:27.800 This three, he didn't develop the protocol all by himself. 0:14:27.920 --> 0:14:32.320 Major synthesizer companies like Roland, Yamaha, and several others were 0:14:32.360 --> 0:14:35.400 all involved in designing the set of rules and standards. 0:14:35.720 --> 0:14:38.920 It was a pretty remarkable display of competitors working together 0:14:39.000 --> 0:14:41.920 to create a technology that would benefit the entire industry, 0:14:42.200 --> 0:14:45.840 not just one company within it. The designers decided that 0:14:45.920 --> 0:14:49.040 MIDI would send information as a list of events or 0:14:49.160 --> 0:14:52.880 messages to instruct a device how to make a certain 0:14:52.960 --> 0:14:56.520 type of sound. Now, again, this wasn't a music file 0:14:56.760 --> 0:15:00.240 or any other form of music, but rather direct is 0:15:00.280 --> 0:15:04.560 the recipient would follow to generate the appropriate sound. I'll 0:15:04.560 --> 0:15:07.200 talk about some of the typical MINI messages in the 0:15:07.240 --> 0:15:10.520 next section, but first let's take a quick break to 0:15:10.680 --> 0:15:20.440 thank our sponsor. Here are a few basic messages the 0:15:20.480 --> 0:15:24.720 Mini protocol defined note on. This is a message that 0:15:24.800 --> 0:15:28.360 indicates a note has been initiated, which is pretty self 0:15:28.360 --> 0:15:31.560 explanatory by the name. So on a keyboard, this would 0:15:31.560 --> 0:15:34.760 be when a key has been pressed, but other instruments 0:15:34.760 --> 0:15:36.960 can also have mini ports on them, so it could 0:15:36.960 --> 0:15:39.840 also mean a guitar string is strummed or a clarinet 0:15:39.920 --> 0:15:43.960 has produced a note. The instructions tell the device receiving 0:15:44.400 --> 0:15:48.360 this data which note has been played and the velocity 0:15:48.400 --> 0:15:52.040 of the note. Velocity equals how hard the note was played, 0:15:52.440 --> 0:15:55.240 so with a piano key it relates to volume. For example, 0:15:55.280 --> 0:15:58.640 if you press the key faster, it indicates a harder strike, 0:15:58.760 --> 0:16:01.960 which means that the note should be louder. Not every 0:16:02.040 --> 0:16:05.880 MIDI keyboard is capable of actually recording that information, but 0:16:05.960 --> 0:16:09.160 a lot of them are. They have that velocity sensitive keys, 0:16:09.440 --> 0:16:13.360 so you can actually record that info. Note off is 0:16:13.400 --> 0:16:16.200 a similar message. It tells when the receiving device uh 0:16:16.240 --> 0:16:19.520 that a played note has ended, so it might be 0:16:19.560 --> 0:16:22.920 when you have released a key, or when vibrations stop 0:16:23.000 --> 0:16:25.840 along a string, and that message says, all right, at 0:16:25.840 --> 0:16:29.560 this point, stop playing a note because there's no longer 0:16:30.080 --> 0:16:34.440 a thing that's generating that sound. Polyphonic key pressure is 0:16:34.480 --> 0:16:37.600 another instruction that tells the receiving device how hard a 0:16:37.720 --> 0:16:40.720 key was pressed once it bottoms out in its lowest position. 0:16:41.200 --> 0:16:43.720 Some keyboards use this to add effects to notes, such 0:16:43.720 --> 0:16:46.120 as vibrato. The bar brato. By the way, that's a 0:16:46.240 --> 0:16:50.800 rapid variation in pitch, So you add a quick oscillation 0:16:50.840 --> 0:16:55.960 and pitch to create vibrato. It adds a richness to sound. Also, 0:16:56.240 --> 0:16:58.240 singers use it to cover up the fact that they 0:16:58.240 --> 0:17:01.080 can't hit a note. That's a little shade throwing right there. Also, 0:17:01.160 --> 0:17:04.640 I do this too. So. Control change is a message 0:17:04.680 --> 0:17:07.879 that indicates that some sort of controller has been activated 0:17:08.119 --> 0:17:11.320 to affect the quality of a sound. Controllers can take 0:17:11.400 --> 0:17:14.000 many forms. You could have pedals, you could have knobs. 0:17:14.200 --> 0:17:18.280 Then control change message contains information that indicates which controller 0:17:18.359 --> 0:17:21.280 was used and the signs of value from zero to one, 0:17:21.720 --> 0:17:25.639 seven or one to eight, depending upon the implementation. To 0:17:25.760 --> 0:17:30.479 describe the magnitude of this change, The pitch wheel change 0:17:30.680 --> 0:17:35.800 message records instances of pitch wheel use. That's not useful 0:17:35.840 --> 0:17:38.280 at all, is it. A pitch wheel is a control 0:17:38.320 --> 0:17:40.639 that allows a musician to affect the pitch of a 0:17:40.720 --> 0:17:44.240 played note, and they can control it dynamically. This creates 0:17:44.240 --> 0:17:47.920 the effect of bending a musical note. So if you've 0:17:47.960 --> 0:17:51.760 ever heard musical piece where a note is played and 0:17:51.800 --> 0:17:54.600 then starts to shift to a different pitch without a 0:17:54.680 --> 0:17:57.280 new note being played, that's kind of what a pitch 0:17:57.280 --> 0:17:59.679 wheel is able to do. And then there are system 0:17:59.720 --> 0:18:04.440 ex collusive or six X messages. This allows for custom 0:18:04.520 --> 0:18:08.560 patches and effects. Manufacturers could use these messages to allow 0:18:08.640 --> 0:18:12.000 a mini controller to take advantage of unique features of 0:18:12.359 --> 0:18:16.399 their instruments. For example, so let's say you're a manufacturer 0:18:16.400 --> 0:18:18.600 and you've got a synthesizer that has a new type 0:18:18.640 --> 0:18:22.800 of effect, and no other synthesizers have this. It's proprietary. 0:18:22.840 --> 0:18:25.040 You've got this cool effect that no one else has 0:18:25.080 --> 0:18:27.920 been able to replicate. The sis X feature would allow 0:18:27.960 --> 0:18:30.439 you to designate a method for a mini controller to 0:18:30.520 --> 0:18:34.919 engage that feature without sharing it to everybody else. Otherwise, 0:18:34.960 --> 0:18:37.720 you'd have a keyboard that has a really cool ability, 0:18:37.760 --> 0:18:39.240 but you never be able to use it through a 0:18:39.280 --> 0:18:44.960 MIDI controller because there'd be no way to designate that command. Right, 0:18:45.119 --> 0:18:48.879 you have your own proprietary effect. If you don't create 0:18:48.920 --> 0:18:52.440 a command for it in MIDI, then the controller won't 0:18:52.480 --> 0:18:54.879 have any instructions that can send to the synthesizer to 0:18:54.960 --> 0:18:57.520 replicate it. Now, you could still get the effect by 0:18:57.560 --> 0:19:00.320 working with the synthesizer directly, but you want be able 0:19:00.359 --> 0:19:03.680 to send those MIDI instructions to any other device because 0:19:03.680 --> 0:19:06.800 there wouldn't there wouldn't be language to take care of 0:19:06.840 --> 0:19:11.000 that particular instance. Sis X messages allowed for these exceptions, 0:19:11.040 --> 0:19:15.359 these custom patches. A MIDI file has the extension m 0:19:15.440 --> 0:19:18.640 I D or mid. If you could read these files 0:19:18.680 --> 0:19:21.679 in natural language, like if you were able to translate 0:19:21.760 --> 0:19:24.479 this as a set of instructions, they would seem like 0:19:24.760 --> 0:19:29.320 really detailed instructions on how to play a certain piece 0:19:29.359 --> 0:19:32.160 of music, not just what the notes are, but how 0:19:32.200 --> 0:19:34.200 to play those notes. It would be similar to reading 0:19:34.240 --> 0:19:36.760 sheet music, but only if the sheet music contained all 0:19:36.800 --> 0:19:40.240 sorts of minutia about the performance of the piece. And 0:19:40.280 --> 0:19:43.360 it's not just how that one piece should be performed, 0:19:43.359 --> 0:19:46.680 it's how that piece actually was performed once upon a time, 0:19:47.000 --> 0:19:50.080 So it means you're not just transcribing music. You are 0:19:50.200 --> 0:19:54.600 actually re creating a performance of a musical piece. And 0:19:54.640 --> 0:19:56.679 you could actually create a mid file by playing a 0:19:56.720 --> 0:20:00.359 MIDI enabled musical instrument connected to a computer. It it's 0:20:00.400 --> 0:20:05.240 actually sequencing your playing as you play it, so you 0:20:05.320 --> 0:20:08.800 are creating a precise record on how to play the 0:20:08.840 --> 0:20:11.200 same piece of music the exact same way in the future. 0:20:11.640 --> 0:20:14.080 So again, it's not a recording, it's a set of 0:20:14.080 --> 0:20:16.679 instructions saying, if you want to play what I just 0:20:16.760 --> 0:20:20.879 played the way I played it, follow these instructions exactly, 0:20:21.400 --> 0:20:23.679 and it will be as if I were playing it 0:20:23.760 --> 0:20:26.560 all over again, which is pretty cool all by itself. 0:20:26.560 --> 0:20:29.479 But an additional benefit of the MIDI system is that 0:20:29.520 --> 0:20:34.280 you can modify those instructions in different ways without affecting everything. So, 0:20:34.320 --> 0:20:37.040 for example, if you record a piece of music in 0:20:37.119 --> 0:20:40.080 some sort of conventional format and you then play it 0:20:40.119 --> 0:20:42.879 at a faster speed, you're going to increase the pitch. 0:20:43.359 --> 0:20:46.199 If I recorded a performance onto a physical medium like 0:20:46.240 --> 0:20:48.840 a vinyl record, and then I played the record back 0:20:49.160 --> 0:20:51.280 at a speed that was one and a half times 0:20:51.359 --> 0:20:56.760 faster than what a normal playback would be. But with midfiles, 0:20:56.800 --> 0:20:59.920 you can increase the speed of a playback without effecting 0:21:00.320 --> 0:21:03.960 the pitch. You aren't speeding up a recording of a performance, 0:21:04.080 --> 0:21:07.800 but rather decreasing the amount of time between instructions, and 0:21:07.840 --> 0:21:11.240 so you can change the temple of music easily without 0:21:11.320 --> 0:21:14.840 also changing the pitch of the recording. Or if you 0:21:14.880 --> 0:21:17.200 want to change the pitch, you could do that too. 0:21:17.280 --> 0:21:19.840 You could take the instructions and apply a new instruction 0:21:20.160 --> 0:21:23.399 to shift the playback into a different key of music. 0:21:23.680 --> 0:21:25.880 The tempo would be the same, but the key would 0:21:25.880 --> 0:21:28.320 be completely different. You could take music that was programmed 0:21:28.359 --> 0:21:29.840 in a major key and you can flip it to 0:21:29.880 --> 0:21:32.239 a minor key. Or you could take a song and 0:21:32.280 --> 0:21:35.440 shift the pitch down or up to better suit someone's voice. 0:21:35.920 --> 0:21:39.159 If you've ever gone to karaoke and the karaoke machine 0:21:39.200 --> 0:21:42.320 had an option to change the pitch to pitch something 0:21:42.440 --> 0:21:45.200 up or down so it's closer to your vocal range, 0:21:45.520 --> 0:21:48.720 you've experienced this. The karaoke machine was using mini files 0:21:48.720 --> 0:21:52.800 to recreate a song, and then you can dynamically tell it, Hey, 0:21:52.800 --> 0:21:55.199 I need this pitched up or pitched down so I 0:21:55.240 --> 0:21:58.120 can actually rock out with Hit me with your best 0:21:58.119 --> 0:22:01.520 shot and make sure it's in my vocal range. Another 0:22:01.560 --> 0:22:04.800 big benefit of the mid file format is file size. 0:22:05.200 --> 0:22:08.480 Because there's no recorded media in the file, the file 0:22:08.520 --> 0:22:11.920 sizes are relatively small, so a minute of compressed audio 0:22:12.119 --> 0:22:15.040 like an MP three might end up being about ten 0:22:15.119 --> 0:22:17.920 megabytes of data data. But if you take a MIDI 0:22:17.960 --> 0:22:21.280 file and it represents the exact same amount of sound, 0:22:22.000 --> 0:22:25.080 it's a minute worth of sound. Although again remember there's 0:22:25.080 --> 0:22:28.320 no recorded sound in a MIDI file just represents that 0:22:28.320 --> 0:22:31.840 that would only take up ten kilobytes of space, so 0:22:32.200 --> 0:22:35.400 much smaller file sizes and a lot of data gets 0:22:35.440 --> 0:22:38.919 packed into those small files. The MIDI protocol supports a 0:22:38.960 --> 0:22:43.840 total of one notes, ranging from the C five five 0:22:43.840 --> 0:22:46.679 octaves below middle C all the way up to the 0:22:46.760 --> 0:22:51.159 G ten octaves above middle C up to sixteen. Separate 0:22:51.200 --> 0:22:54.080 devices can be controlled through a single chain or more 0:22:54.119 --> 0:22:56.960 if you want multiple devices to produce the same response. 0:22:57.040 --> 0:22:59.560 So for example, if you want both a piano and 0:22:59.560 --> 0:23:01.880 a claren at to play back the same melody line 0:23:01.920 --> 0:23:03.720 and a piece of music, they could each follow the 0:23:03.720 --> 0:23:06.639 exact same set of instructions and they would count as 0:23:06.680 --> 0:23:09.880 just one channel of data rather than two channels of data. 0:23:09.920 --> 0:23:12.600 You would just put those in serial with each other, 0:23:12.960 --> 0:23:14.800 and you could still divide up the rest of the 0:23:14.880 --> 0:23:17.840 channels among other instruments. In addition to this, you can 0:23:17.880 --> 0:23:20.720 have up to a D eight voice or effect settings 0:23:20.720 --> 0:23:24.080 called programs. These are the various modifiers that can change 0:23:24.119 --> 0:23:27.120 the shape of the sound in various ways to keep 0:23:27.119 --> 0:23:31.240 everything synchronized across multiple instruments. Not to mention other elements 0:23:31.280 --> 0:23:34.159