Rush e is an example of a music subgenre called black midi, it's notable by being generally only able to be played in simulation just by the shear number of notes played at once and it's called black midi because if you tried to print of the sheet music you wouldn't be able to see the staff for the notes
@@matticawood It's less about finesse and actually performing as well as a real singer and more about how impressive the piano's ability to mimic words is. It's like a fun house mirror. It's not about clarity. It's about a spectacle.
So nice to hear this from a professional who plays the piano. That Rush E song, man, it just wasn't meant to be played on a physical piano, whether it had "TAS" programming or not.
I'm pretty sure the "piano catching fire" part is just theatrics. The smoke is perfectly timed and comes from the bottom of the piano, while the part that is the most likely to overheat and produce smoke like that (the solenoids) are placed directly underneath the keys. You'd also see keys fail pretty quickly because that (usually toxic) smoke would be the insulation between coil windings burning up.
As a side note, it should be fully possible to make the solenoids to do pretty much anything that a human can. If the system can fully control the force, then it could wary the force over time, which I don't see being any different from how a human would play. The only differences (I think) would be that it has no sense of how hard it's pushing, which could be managed beforehand. The solenoid sysem could even be engineered to pull the key back up, but it doesn't seem like this piano can do that. The real limitations would be how the software controls the piano and how the keypresses are recorded.
It’s definitely fake smoke just for the good video title 😂 and yes, i think they probably could mimic a real player, the limitation comes from the midi input which only has 127 levels of velocity and it doesn’t account for speed of press. I think the actual solenoides themselves might also have that limitation but I can’t be sure! Il have to look into it 😊
one thing people seem to forget is that there are 2 Rush E's in circulation. The original impossible version and the Possible version that someone created for people to attempt on real pianos. Still very hard, but actually "possible"
whenever i saw this video i was like “oh i’ve never seen this guy what’s his channel called” and when i saw his sub count i thought it was 83000 and i believed it because of how well made the start was and the entirety of the video. keep it up and i bet you could one day reach 83000 subs and way more. great video!
I love that people are so impressed with RUSH E. It's barely testing the waters of what is black midi lol. Great analysis and vid though. I'm always impressed with how much better an actual pianist can make things sound despite all the technology at our fingertips.
Great video! The reason it is not a similar copy is (first of all 127 inputs) but also when calculating the fourier transform they only take the peaks meaning its more of an average than exact values. Kinda like rounding a number down because you have limited decimals you can use.
@@matticawood Fourier transforms are different than the harmonics of the piano but it sounds like you relized that after making the video. Lucas is correct that the reproduction using the piano isn't going to be faithful because he's only taking three "notes" of the fourier series to reproduce each timestep of the vocals. But, Matthew, (unwittingly) mention another aspect of the piano that makes it distinct from human vocals, which is the harmonics of the piano. So the vocal reproduction suffers not simply from the truncated Fourier representation but also the very different harmonics of the piano to the human vocal cords. With that said, you can get human vocal cord to exhibit harmonic behavior. You might very well be aquainted with Tuvan throat singing.
@@matticawood As Nathan says, the Fourier transforms are not equivelent to the harmonics of an instrument. While it can be equivalent if the sound being processed is purely the sound of a singular instrument, the Fourier transform is a far more generalized function of waves in general, not even limited to music. It's a way to decompose a wave into it's component frequencies at a given point in time. And it's not an approximation, it's an exact representation. The approximation comes when using the computed Fourier series to do something. There are a number of reasons for the inaccuracy: 1) The Fourier series becomes better approximation the more of the component frequencies are used, three or even 4-5 isn't really enough to get a good approximation 2) The Fourier series works on pure sine waves of a single frequency, since each note of the piano has it's own harmonics when played combining the notes will never add up completely correctly because they are not pure tones. 3) The piano contains fundamental frequencies each a semitone apart, meaning any tone that requires a frequency between two keys must be rounded to the closest frequency within the tuning method (in the case of the video equal temperament) or use both neighboring keys with differing action, either way it's not the same as if there was a key of that specific frequency. 4) The example shown with decomposing the wave into a combination of other waves zooms way in on the sound wave. Any complex sounds or combinations of sounds will have the frequencies it's composed of change very frequently. You'd need to recalculate the Fourier transform tens of times a second in some cases to get to a point where it's indistinguishable. And considering the limits of the piano in terms of playing and stopping notes extremely fast this means you can only reasonably calculate the Fourier series a handful of times a second. The Fourier transform is only 100% exact at a specific given instant of time along the wave, and when the resolution/fidelity is infinite. In practical usage we use a Fast Fourier Transform (FFT) which is an approximation, since a true, 100% accurate Fourier Transform takes a long time to compute that it's not practical. Even so a FFT is far more than accurate enough to recreate any given wave to a quality where it's indistinguishable from the original. It still has the same limitations as a full Fourier Transform in that it's only accurate at a specific given moment, if you can recreate all necessary possible component frequencies as pure frequencies, and for extremely small periods of time. Which can not be done on a piano for the reasons listed above. The issue is not with the 127 levels of input, as at 127 levels the difference between two adjacent levels is already indistinguishable in person (I guarantee that even the most exacting, precise pianist in the world, when playing the same note in the same song, will naturally vary the velocity by upwards of 3-5 equivalent levels when trying to recreate the same velocity. While it's true that from a physical standpoint a human player has an analog basically infinite range of speeds here, no one is consistent enough to play the key at a specific velocity every single time, there will always be deviance, and such deviances will be larger than a single level of those 127 possible levels), let alone when combines with the limited bit depth of audio samples in digital audio. No one would be able to tell the difference between level 111 and 112 as the change would be so small that the only way you'd be able to tell would be through audio analysis. The human ear is very powerful, but just like vision our hearing cheats a lot, just as there are visual illusions so to are there auditory illusions, because our hearing is not perfect and subtleties are lost almost entirely without other stimulation for comparison against.
The Fourier transformed on small fractions of sound is a mathematical simulation of how the human ear works, with the hundreds of tiny hairs inside the snail, each resonating on a single frequency and activating a nerve for each.
This video has TONS of professionalism put into it, I thought you had well over 500K, but I was shocked to find that you were a smaller channel. You've blown my mind! Keep up the great work and I'll be sure to keep up with what you post!
I am not a self-playing piano expert, but I think playing Rush E on a piano roll would be impossible, as you would need to have so many holes so close together that it would possibly affect the integrity of the roll.
What makes it groundbreaking is that it can very closely mimic a human player using the mechanical action of a real piano. Many pianos like it are similar but… less. They can’t produce as close as that to a human. Also… while the speaking may sound bad, you also need to remember that it requires a very specific set of keys to be pressed at the perfect speed, pressure and timing to even get there. The fact you can still understand the words said, from a piano is pretty amazing. If it were done on an electronic one, that would be nothing special but no, that was on a real, playable by a human still, piano.
I agree, making sure that the right frequencies are played at a velocity to get the quality of sound to hear words is very impressive from an engineering perspective and would be very difficult to do as a human player. I still think there are limitations using midi input because it can’t account for the action (speed of key depression) and it can only work with velocity in 127 steps so from a musical perspective it will always lack the nuances a human can achieve when playing music 😊
@@matticawood fwiw it's completely possible to use other formats to get more than the 127 steps limitation of midi and they only used that for this cause it was simplest with their current tools compared to special software that is written to save even more information. Really to even get that extra information (and they might have already done this) they would need to have another means of recording player input like custom feedback switches for each key on that piano and then it could truly record a player's actions with the most precision possible and replicate the playing almost completely precise give or take the slight changes from the temperature of the room affecting the sound ;) If that was added.. and professional players were hired in to "record" some beautiful music that could forever be replicated to such high precision would be amazing.. just hope for your sake they don't mass produce them ;) lmfao
He was actually a 3.k+ TH-camr and i thought he has 500k or 800k+ subs... Its so mind boggling how underrated he is cause look at the audio quality and the video quality and also the professionalism put into the vid is just *chefs kiss*..
On midi Velocity: It depends on the standard and extension of midi that the device is capable of or listening to. Yamaha Pianos that have midi capability run on Yamaha's XG format which allows for 512 velocity and dynamic levels. And there is an extension that allows for a theoretical 1024 levels of velocity and dynamic range. All of which are theoretically possible with that piano. The general opinion among electronic music composers tends to be that 128 levels is enough anyways. I thought you'd find the extra info interesting.
That is super interesting, I didn’t know there was the possibility of a wider range of velocities. I assumed it might be capped for some technical limitation or something! 😊
"This is a cool take produced really well!" *scrolls down* 800 SUBSCRIBERS!?!? THIS IS CRIMINAL! Keep up the excellent work; I have no doubt this channel will eventually explode!
@@matticawood I have my fingers crossed for you; not to get your hopes up but I honestly feel you're about to get a bump in the algorithm. Subscribing now!
@@razordu30 this video has already done pretty well, I appreciate every impression and every view people decide to give because I know there’s a lot of good videos for people to spend their time on 😊
I would think that traditional selfplaying pianos might have issues with number of simultaneous keys. If it has enough force for all. More than 4 hands is pretty rare I think. Why would they need to make the system have enough force for 30 simultaneous keys. But with MIDI-based solenoid controll it is fairly easy to make the powersupply big enough
You are right. My Grandmother has a player piano and some paper rolls. The piano only has so much air flow to actuate things; too many holes and the keys only get gently pressed without enough force to make the note sound. The other problem with black midi type songs on paper player pianos is that there would be almost no paper left and the roll would be very fragile.
To answer the question of speed. F=ma which means the force equals mass times acceleration. Thus with mass being the same the acceleration must change. To change acceleration we need a change in velocity over time. Thus, the time it takes to reach a position relates to acceleration meaning the speed of movement decides force.
If you got enough pianos together with their own players, would it be possible to play Rush E in its entirety by breaking the score up into manageable chunks?
The reason I think it sound bad when he talks is actually not that it isn't precise enough, but because the Piano doesn't play sine waves, but each note itself also has overtones. So therefor it is just impossible to create the pure sound of a voice as you always have the aditional harmonics
No, he did not do anything truly ground breaking with the creation of the piano. The tech to make this piano has been around for a long time, and has probably already been done. But it would be very expensive, so not a commercial ready product. But, what is great is how he used this expensive piano and a meme to teach. That is his strength! Starting with something relatable and seemingly difficult (but of course it is not difficult for a computer) he taught about the inner workings of the piano and taught some high level math with the Fourier Series. Doing that in a concise and entertaining way is ground breaking.
Time traveler: “Wow a lot of you guys are gonna lose your jobs in my time.” Pianists: “Yeah but out jobs require skill so we don’t have to worry about it, right?” Time traveler: “Uhhhh… Gotta go! Bye!
that version of rush e 100% cannot be played on piano roll. the "genre" of it is called black midi, because if you show the midi files as sheet music the whole thing is black, without being able to see individual notes. so the piano roll, as a physical piece of paper, literally isnt precise enough to store the song, let alone play it
The “impossible version” of Rush E was the original, and people made a possible version of it so that they could play the most popular piano song of that time
It was interesting watching his video. I’m a guy who cuts open pianos and turns them into player pianos lol It would be easier to play the Rush E song on a grand piano than the upright, grands naturally have faster repetition than uprights due to the fact that gravity is helping them reset the hammers into firing position rather than an upright piano which needs additional help as it is traveling horizontally. To answer your question about key velocity it’s a yes and no. Like mark said you have a range of values for how hard the solenoid needs to work to press the key, aka how much power to put behind the strike. You can think of it as velocity. Below a certain threshold the solenoid won’t build up enough force to fire the key and strike the string. As for determining the touch things get tricky. You have to keep in mind that as cool as player pianos are they still run off midi files which don’t always have as much nuanced detail as live performances. This is improving over time though- look into some of the cool stuff the Disklavier is able to do with its optical recording models. This records less “a G5 played at this second/millisecond at this volume” and more “I saw the key back move at this time and hammer butt move at this time so this key is moving in this way.” When they record with motion rather than stop rail data you can record yourself lightly wiggling the keys or pedals without making and actual sound and the piano still picks it up.
That’s super interesting! I think audibly replicating the speed of a key is probably much harder, with midi it does a good job with the 127 values but it would just be interesting to see how human it can really get! 😊
"I don't want to be out of a job" Famous Last Words
ปีที่แล้ว +1
10:52 It depends on what you are comparing. Your version was missing singer's voice played by keys, which was the whole point of that specific demonstration in that segment of Mark's video. In this context your version failed. Your version definitely sounded better, but that was not the point of that demonstration. :D
9:27 i think that it can mimic human touch because velocity is after all momentum when considering mass and change in momentum in a definite span of time would give you the force. The time can be adjusted by the current through the solenoid
Such professional talking for a person with a low sub I see your letting your passion comes first than subs count to show your professionalism You earn a subscribe
I just want to mention that a lot of the notes that Sheet Music Boss puts in their videos aren’t actually played. For example, their Tetris video where they create Tetris shapes without affecting the song being played, which is Type A.
Answering Mathew's question at 8.35 . yes you can increase and decrase velocity of the key just by adjusting the voltage and current across the solenoid.
Percussion is defined as a "musical instrument played by striking with the hand or with a handheld or pedal-operated stick or beater, or by shaking, including drums, cymbals, xylophones, gongs, bells, and rattles."
That’s right! A piano does that, but it is also a string instrument 😊 - “String instruments, stringed instruments, or chordophones are musical instruments that produce sound from vibrating strings when a performer plays or sounds the strings in some manner.”
It depends on what decade/century you were classifying it. Originally it was stringed … then reclassified as percussion …. But it’s current classification is as a keyboard instrument … at least that’s what I was taught at the RCM about 25 years ago
I've owned and played a player piano. Rush e would rapidly outstrip the capacity of the pnumatics. Most are pedal bellows. Some modification may help, but punching holes that close on the roll would mean that the valve slots would not be completely covered. Player pianos have a limited repeat rate. You would have to write an abridged version to suit it.
You are correct in that the piano is technically still a string instrument despite percussive force being applied to make the sound. I know it's not the ONLY classification system for instruments but the Hornbostel-Sachs system classifies the piano as a chordophone, and not an idiophone or membranophone like most percussion instruments are. This is DESPITE the fact that it uses the hammers to make sound, and is represented thus as a 314.122-4-8 in that system's numbering method, or a simple chordophone using a string bearer that is shaped like a board or is the ground that is played with a keyboard and sounded with hammers. Interestingly, this means that a piano is in the same general instrument family as zithers, such as the chinese guqin and the japanese koto. Also, keyboard instruments under that system are not in their own category, but rather being played with a keyboard is a suffix that you add on when representing an instrument in that system. But hey, that's just one of the most common systems that organologists use, it's not the only one as I said so you CAN still say the piano is a string and percussion instrument that also fits neatly into the subcategory of keyboard based instruments.
"Probably because it was mediocre, no offense." (BTW this video is excellent, in case anybody stumbles across this and doesn't know, the line is a reference to a meme argument specifically about his composition "Rush E".)
From a phonetic perspective, the recreation of speech is fascinating. Sound is transformed into impulses in our ear's cochlea by the location of hair cells that only respond to a particular frequency. If these frequencies are in the range of the human voice, then a specific area in the brain is activated that categorizes frequencies into meaningful sounds of speech (phonemes). We are used to fill in the gaps in case the signal is patchy. This is exactly what chopsticks gives, the right triggers. Try holding your hand over the words, and it still kind of works!
Rush e is an example of a music subgenre called black midi, it's notable by being generally only able to be played in simulation just by the shear number of notes played at once and it's called black midi because if you tried to print of the sheet music you wouldn't be able to see the staff for the notes
I didn’t know this, how interesting! 😊
Here have an example th-cam.com/video/NQPfYx0utvM/w-d-xo.html
it says 2 replies for me but i only see matthew's, obviously i meant there was 2 replies before this comment.
@@markpullman785 well it said 3 replies for me before I added my own :)
Its a hard piano meme song
Just so you know, the piano is actually singing Rick Astley. It's not just playing the song, it's also singing it. That's the impressive part.
Yeah, I knew that was what the intention was 😊 I think it still didn’t pull of the same finesse as Rick 😂
dQw4w9WgXcQ moment
Yes
@@matticawood It's less about finesse and actually performing as well as a real singer and more about how impressive the piano's ability to mimic words is. It's like a fun house mirror. It's not about clarity. It's about a spectacle.
This mean also we get rickrolled
Mark rober is possibly one of the only people to ever hear the full version of rush e in person, not through any form of screen
cameraman: -_-
@@abhimanyusagrawal2005
The phrase "one of": -_-
So nice to hear this from a professional who plays the piano. That Rush E song, man, it just wasn't meant to be played on a physical piano, whether it had "TAS" programming or not.
It definitely wasn’t, it’s more of an exercise in programming then anything else 😊
You can do rush e it’s east I can do it with my eyes closed
@@ol_brambo5031 That’s nothing. I can do it with my feet while blindfolded, suspended over a lake.
Yeah but if its played realy slow im sure that someone could do it
Yeah but if its played realy slow im sure that someone could do it
No human pianist on this freaking planet (no matter how professional or prodigy you are) can play exactly what that robot piano just played
Maybe an octopus could? 🐙😂
No *HUMAN* pianist!
Animals don’t count!
@@matticawood maybe 3-4 pianists?
@@danielwols that would probably look like a musical game of twister
I'm pretty sure the "piano catching fire" part is just theatrics. The smoke is perfectly timed and comes from the bottom of the piano, while the part that is the most likely to overheat and produce smoke like that (the solenoids) are placed directly underneath the keys. You'd also see keys fail pretty quickly because that (usually toxic) smoke would be the insulation between coil windings burning up.
As a side note, it should be fully possible to make the solenoids to do pretty much anything that a human can. If the system can fully control the force, then it could wary the force over time, which I don't see being any different from how a human would play. The only differences (I think) would be that it has no sense of how hard it's pushing, which could be managed beforehand. The solenoid sysem could even be engineered to pull the key back up, but it doesn't seem like this piano can do that.
The real limitations would be how the software controls the piano and how the keypresses are recorded.
It’s definitely fake smoke just for the good video title 😂 and yes, i think they probably could mimic a real player, the limitation comes from the midi input which only has 127 levels of velocity and it doesn’t account for speed of press. I think the actual solenoides themselves might also have that limitation but I can’t be sure! Il have to look into it 😊
E
Ye
Ahh understood every word hehe
one thing people seem to forget is that there are 2 Rush E's in circulation. The original impossible version and the Possible version that someone created for people to attempt on real pianos. Still very hard, but actually "possible"
whenever i saw this video i was like “oh i’ve never seen this guy what’s his channel called” and when i saw his sub count i thought it was 83000 and i believed it because of how well made the start was and the entirety of the video. keep it up and i bet you could one day reach 83000 subs and way more. great video!
Thank you!! I appreciate this…I’m glad you think it’s worthy of that! It means a lot! 😊
@@matticawood your welcome! keep up the videos!
Same
how do you mistake 1k subs for 83k subs 💀💀💀
@@zclor9055 if they had 8k subs they prob saw the 8 and the k and just assumed from the quality that it was 80k after a quick glance :P
I love that people are so impressed with RUSH E. It's barely testing the waters of what is black midi lol.
Great analysis and vid though. I'm always impressed with how much better an actual pianist can make things sound despite all the technology at our fingertips.
Thanks!! I’m glad you liked it 😊
because its popular and everything that dies on the internet comes back 6 years later no exceptions
Death Waltz has entered the chat
Great video! The reason it is not a similar copy is (first of all 127 inputs) but also when calculating the fourier transform they only take the peaks meaning its more of an average than exact values. Kinda like rounding a number down because you have limited decimals you can use.
Thanks! I thought it might be something like this, i wasn’t hugely familiar with Fourier transforms before this 😊
@@matticawood Fourier transforms are different than the harmonics of the piano but it sounds like you relized that after making the video. Lucas is correct that the reproduction using the piano isn't going to be faithful because he's only taking three "notes" of the fourier series to reproduce each timestep of the vocals. But, Matthew, (unwittingly) mention another aspect of the piano that makes it distinct from human vocals, which is the harmonics of the piano. So the vocal reproduction suffers not simply from the truncated Fourier representation but also the very different harmonics of the piano to the human vocal cords.
With that said, you can get human vocal cord to exhibit harmonic behavior. You might very well be aquainted with Tuvan throat singing.
@@matticawood As Nathan says, the Fourier transforms are not equivelent to the harmonics of an instrument. While it can be equivalent if the sound being processed is purely the sound of a singular instrument, the Fourier transform is a far more generalized function of waves in general, not even limited to music. It's a way to decompose a wave into it's component frequencies at a given point in time. And it's not an approximation, it's an exact representation. The approximation comes when using the computed Fourier series to do something. There are a number of reasons for the inaccuracy:
1) The Fourier series becomes better approximation the more of the component frequencies are used, three or even 4-5 isn't really enough to get a good approximation
2) The Fourier series works on pure sine waves of a single frequency, since each note of the piano has it's own harmonics when played combining the notes will never add up completely correctly because they are not pure tones.
3) The piano contains fundamental frequencies each a semitone apart, meaning any tone that requires a frequency between two keys must be rounded to the closest frequency within the tuning method (in the case of the video equal temperament) or use both neighboring keys with differing action, either way it's not the same as if there was a key of that specific frequency.
4) The example shown with decomposing the wave into a combination of other waves zooms way in on the sound wave. Any complex sounds or combinations of sounds will have the frequencies it's composed of change very frequently. You'd need to recalculate the Fourier transform tens of times a second in some cases to get to a point where it's indistinguishable. And considering the limits of the piano in terms of playing and stopping notes extremely fast this means you can only reasonably calculate the Fourier series a handful of times a second.
The Fourier transform is only 100% exact at a specific given instant of time along the wave, and when the resolution/fidelity is infinite. In practical usage we use a Fast Fourier Transform (FFT) which is an approximation, since a true, 100% accurate Fourier Transform takes a long time to compute that it's not practical. Even so a FFT is far more than accurate enough to recreate any given wave to a quality where it's indistinguishable from the original. It still has the same limitations as a full Fourier Transform in that it's only accurate at a specific given moment, if you can recreate all necessary possible component frequencies as pure frequencies, and for extremely small periods of time. Which can not be done on a piano for the reasons listed above.
The issue is not with the 127 levels of input, as at 127 levels the difference between two adjacent levels is already indistinguishable in person (I guarantee that even the most exacting, precise pianist in the world, when playing the same note in the same song, will naturally vary the velocity by upwards of 3-5 equivalent levels when trying to recreate the same velocity. While it's true that from a physical standpoint a human player has an analog basically infinite range of speeds here, no one is consistent enough to play the key at a specific velocity every single time, there will always be deviance, and such deviances will be larger than a single level of those 127 possible levels), let alone when combines with the limited bit depth of audio samples in digital audio. No one would be able to tell the difference between level 111 and 112 as the change would be so small that the only way you'd be able to tell would be through audio analysis. The human ear is very powerful, but just like vision our hearing cheats a lot, just as there are visual illusions so to are there auditory illusions, because our hearing is not perfect and subtleties are lost almost entirely without other stimulation for comparison against.
The Fourier transformed on small fractions of sound is a mathematical simulation of how the human ear works, with the hundreds of tiny hairs inside the snail, each resonating on a single frequency and activating a nerve for each.
This video has TONS of professionalism put into it, I thought you had well over 500K, but I was shocked to find that you were a smaller channel. You've blown my mind! Keep up the great work and I'll be sure to keep up with what you post!
That’s super kind! Thank you 😊
omg same!
same
I thought it said like 300k not 3k! Holy moly lol!!
I am not a self-playing piano expert, but I think playing Rush E on a piano roll would be impossible, as you would need to have so many holes so close together that it would possibly affect the integrity of the roll.
What makes it groundbreaking is that it can very closely mimic a human player using the mechanical action of a real piano. Many pianos like it are similar but… less. They can’t produce as close as that to a human. Also… while the speaking may sound bad, you also need to remember that it requires a very specific set of keys to be pressed at the perfect speed, pressure and timing to even get there. The fact you can still understand the words said, from a piano is pretty amazing. If it were done on an electronic one, that would be nothing special but no, that was on a real, playable by a human still, piano.
I agree, making sure that the right frequencies are played at a velocity to get the quality of sound to hear words is very impressive from an engineering perspective and would be very difficult to do as a human player.
I still think there are limitations using midi input because it can’t account for the action (speed of key depression) and it can only work with velocity in 127 steps so from a musical perspective it will always lack the nuances a human can achieve when playing music 😊
@@matticawood Definitely. I think the fact that it can get CLOSE it what makes this amazing.
@@matticawood fwiw it's completely possible to use other formats to get more than the 127 steps limitation of midi and they only used that for this cause it was simplest with their current tools compared to special software that is written to save even more information. Really to even get that extra information (and they might have already done this) they would need to have another means of recording player input like custom feedback switches for each key on that piano and then it could truly record a player's actions with the most precision possible and replicate the playing almost completely precise give or take the slight changes from the temperature of the room affecting the sound ;)
If that was added.. and professional players were hired in to "record" some beautiful music that could forever be replicated to such high precision would be amazing.. just hope for your sake they don't mass produce them ;) lmfao
@Matthew Cawood Your human nuance is ubiquitous compared to the midi input. Talent vs. Extraordinary. Not apples/apples.
This is the kind of reaction video I like. Actually giving input instead of just going "heh", "wow", and "okay".
You've found a new subscriber!
Thanks! I appreciate that, I’m glad you enjoyed my input 😊
He was actually a 3.k+ TH-camr and i thought he has 500k or 800k+ subs... Its so mind boggling how underrated he is cause look at the audio quality and the video quality and also the professionalism put into the vid is just *chefs kiss*..
Wow thanks! I really appreciate that and I’m glad you liked the video!! 😊
@@matticawood your most welcome.
I’m usually not the one to watch very in depth music videos but this video kept my attention the whole time great job
Thanks! I’m glad I could keep you interested! 😊
On midi Velocity: It depends on the standard and extension of midi that the device is capable of or listening to. Yamaha Pianos that have midi capability run on Yamaha's XG format which allows for 512 velocity and dynamic levels. And there is an extension that allows for a theoretical 1024 levels of velocity and dynamic range. All of which are theoretically possible with that piano.
The general opinion among electronic music composers tends to be that 128 levels is enough anyways. I thought you'd find the extra info interesting.
That is super interesting, I didn’t know there was the possibility of a wider range of velocities. I assumed it might be capped for some technical limitation or something! 😊
"This is a cool take produced really well!"
*scrolls down*
800 SUBSCRIBERS!?!? THIS IS CRIMINAL!
Keep up the excellent work; I have no doubt this channel will eventually explode!
Thank you! I really appreciate that and I’m super glad you enjoyed it 😊
@@matticawood I have my fingers crossed for you; not to get your hopes up but I honestly feel you're about to get a bump in the algorithm.
Subscribing now!
@@razordu30 this video has already done pretty well, I appreciate every impression and every view people decide to give because I know there’s a lot of good videos for people to spend their time on 😊
You are right, 2 weeks from now Matt will have double subscribers
Subs have quintupled in 3 months
Just a quick correction for 9:45. There is 128 values you can use. It's 0 - 127 where 0 is equal to the key being released
Yay! The supergeek inside me dies a little every time i hear 127. Thank you!!!
I would think that traditional selfplaying pianos might have issues with number of simultaneous keys. If it has enough force for all.
More than 4 hands is pretty rare I think. Why would they need to make the system have enough force for 30 simultaneous keys. But with MIDI-based solenoid controll it is fairly easy to make the powersupply big enough
You are right! I imagine they probably would have a lot of issues. It’s more of an exercise in pushing the boundaries 😊
You are right. My Grandmother has a player piano and some paper rolls. The piano only has so much air flow to actuate things; too many holes and the keys only get gently pressed without enough force to make the note sound. The other problem with black midi type songs on paper player pianos is that there would be almost no paper left and the roll would be very fragile.
3.68K!? I thought you were in the millions! Keep up the profesional work!
Thanks! Maybe one day 😂
Matt’s a legend and I agree with almost everything he says!
“Almost”!! Well..rude! 😂
@@matticawood what is rude?
@@Raternood_ it was a joke because they said “almost” 😂
Matt’s a legend and i agree with everything he says!
@@BucketOPopcorn almost*
This video is so well edited!, Why this is underrated? Edit: thanks for the 30 like
Thank you! I’m glad you think so! 😊
I know right
Yes this is REALLY underrated channel
Ye
Yeah, I thought this channel had at least 10k subs, but he doesn’t even have 1k
To answer the question of speed. F=ma which means the force equals mass times acceleration. Thus with mass being the same the acceleration must change. To change acceleration we need a change in velocity over time. Thus, the time it takes to reach a position relates to acceleration meaning the speed of movement decides force.
13:35 this is where the piano starts playing Rush E
0:00 this is where the good bit starts! 😂
@@matticawood that actually made me laugh lmao
I like how he has a bunch of pianos in the background
You can never have too many sets of keys 😂
It would be actually interesting to see just how many hands a good pianist would need too play this if they were efficient
It really would! If I had the time to score it out and ask other pianists to play…that would be an excellent video 😂
This is a really neat analysis!
Thanks!! 😊
This was an excellent video! Looking forward rooting through what you've made and seeing what next
Thanks, I appreciate that! 😊
Just saying THIS WONDERFUL CHANEL NEEDS MORE SUBS
If you got enough pianos together with their own players, would it be possible to play Rush E in its entirety by breaking the score up into manageable chunks?
I think it would! There may be a future video there somewhere! 😊
If you had 3 people (your estimated 30 fingers) or four could everyone squeeze in enough to play it???
@@danlhoy if they can play it on time and be comfortable…
Now that you mentioned it, I've been looking for a video of 3-4 people actually playing rush E together on separate pianos. That would be so cool.
Bro spent almost the whole video explaining the physics of a piano-
The reason I think it sound bad when he talks is actually not that it isn't precise enough, but because the Piano doesn't play sine waves, but each note itself also has overtones. So therefor it is just impossible to create the pure sound of a voice as you always have the aditional harmonics
That’s very true! I hadn’t thought of this 😊
i like your funny words magic man
No, he did not do anything truly ground breaking with the creation of the piano. The tech to make this piano has been around for a long time, and has probably already been done. But it would be very expensive, so not a commercial ready product. But, what is great is how he used this expensive piano and a meme to teach. That is his strength! Starting with something relatable and seemingly difficult (but of course it is not difficult for a computer) he taught about the inner workings of the piano and taught some high level math with the Fourier Series. Doing that in a concise and entertaining way is ground breaking.
This video deserves so much more attention.
Thank you! Very kind 😊
Agreed His really Good a Explanations and Reactions kind of Videos
Your video quality is awesome, there's no reason for you to have so few subscribers
Thanks!! I appreciate that! 😊
this man is really good at the piano, i have a keyboard but they're pretty much the same you got a sub
Thanks! There’s a bit of a difference in how you play it, but it has the same pressy things 😂
Time traveler: “Wow a lot of you guys are gonna lose your jobs in my time.”
Pianists: “Yeah but out jobs require skill so we don’t have to worry about it, right?”
Time traveler: “Uhhhh… Gotta go! Bye!
I hope not!! 😂
Chopstix: COME BACK HERE!
Your such a cool person, and I love your talent👍
Thank you!! That’s super nice 😊
It took me 20 minutes to realize I’ve just been Rick rolled by a fricking piano
Matt’s a legend and I agree with nothing he says!
The first part - YES! The second part - NO! 😂
this is what i have been waiting for. Thank you!
You are very welcome 😊
that version of rush e 100% cannot be played on piano roll. the "genre" of it is called black midi, because if you show the midi files as sheet music the whole thing is black, without being able to see individual notes.
so the piano roll, as a physical piece of paper, literally isnt precise enough to store the song, let alone play it
I wonder if there was a piano and paper wife enough if it could technically be done! Maybe that should be Mr Rober’s next venture 😊
5:02 Chopsticks says edelweiss from the movie The Sound Of Music when the captain sings to his kids if I’m correct
This man should be more popular like his setup his editing is just yes he should be more popular!
Thanks! I’m glad you think so 😊
The “impossible version” of Rush E was the original, and people made a possible version of it so that they could play the most popular piano song of that time
It was interesting watching his video. I’m a guy who cuts open pianos and turns them into player pianos lol
It would be easier to play the Rush E song on a grand piano than the upright, grands naturally have faster repetition than uprights due to the fact that gravity is helping them reset the hammers into firing position rather than an upright piano which needs additional help as it is traveling horizontally.
To answer your question about key velocity it’s a yes and no. Like mark said you have a range of values for how hard the solenoid needs to work to press the key, aka how much power to put behind the strike. You can think of it as velocity. Below a certain threshold the solenoid won’t build up enough force to fire the key and strike the string. As for determining the touch things get tricky. You have to keep in mind that as cool as player pianos are they still run off midi files which don’t always have as much nuanced detail as live performances. This is improving over time though- look into some of the cool stuff the Disklavier is able to do with its optical recording models.
This records less “a G5 played at this second/millisecond at this volume” and more “I saw the key back move at this time and hammer butt move at this time so this key is moving in this way.” When they record with motion rather than stop rail data you can record yourself lightly wiggling the keys or pedals without making and actual sound and the piano still picks it up.
That’s super interesting! I think audibly replicating the speed of a key is probably much harder, with midi it does a good job with the 127 values but it would just be interesting to see how human it can really get! 😊
A new Mark Rober video came out while I was watching this, unexpected.
Awesome video, keep up the good work
You did good at rush E
I didn’t think I’d be rickrolled when watching this lol good video Matt.
You are never too far away from Rick 😂 Thanks! I appreciate that 😊
How is this video not blow up is nice
Thanks! That’s super kind! Maybe TH-cam need a strongly worded letter 😂
@@matticawood e lol mabye about something with anthems cus I like them
Woah, you're so underrated. This is a really good video. You earned a new sub.
Thanks! I appreciate that! 😊
Very interesting and well made ! I drop a follow !
Thanks!! I’m glad you liked it 😊
Great insight would love to see you react to how harmoniums are made
Thanks! That sounds like it would be interesting, il have a look into it 😊
Nice video love ur content keep up The work
Thanks!! 😊
10:38 Favorite Part
While I was watching the video:Simply piano is great!
How could anyone dislike this? I don’t see anyone playing it
Probably because it's mediocre, no offense.
@@alephzero1984 Was it? I'd like to see you try. Give us an awe-inspiring performance.
@@aranyabanerjee4520 Oh boy. You will embarrass yourself. I'll show you what a good performance of this piece sounds like.
@@alephzero1984 Bruh I know about Fanchen. I forgot what the hater said. Trying to impersonate him.
@@alephzero1984 you ruined it
This guy neeeeeeds sooooo much more attention bro, im subbing foor sure
Thanks so much! Super kind 😊
You actually played the rickroll song much better than the talking piano.
Thanks! 😊
I think that might be because the piano added other notes to actually sound like it was singing
I just found ur channel and am loving it
Thanks! I’m glad you like my videos 😊
Cool video.
Thanks 🙏
bread
@@godfreymang7928 bread
Very good and very logger
Keep it up
Cuz u deserve more subs
Thanks!! 😊
Matt’s a myth and a legend!
Is it a bird…is it a plane….no…it’s the lesser spotted Matt 😱
You really should react to the original rush e
Nice
Thanks! 😊
"I don't want to be out of a job"
Famous Last Words
10:52 It depends on what you are comparing. Your version was missing singer's voice played by keys, which was the whole point of that specific demonstration in that segment of Mark's video. In this context your version failed. Your version definitely sounded better, but that was not the point of that demonstration. :D
8:39 there are some piston types that can do that but they would take even more circuitry which would cost even more of a fortune.
That’s interesting! Maybe one day it will be inexpensive enough to put me out of a job 😂
This video is amazing, when i first saw it i taught you might have 100k subs... But... You deserve more than 800
Thank you! Maybe some day I will..I appreciate that you think it deserves it! 😊
@@matticawood you totally deserve it... From the quality to talent
BRO UNDERMINDED THE WHOLE THING
I am already pre-congratulating you on 10k or more
Thank you!! Il hopefully get to numbers like that one day! 😊
9:27 i think that it can mimic human touch because velocity is after all momentum when considering mass and change in momentum in a definite span of time would give you the force. The time can be adjusted by the current through the solenoid
Man you really know how to play the keyboard.
I've been learning without any progress
How far have you got? 😊
7:10 It's like butterfly clicking!
Mark robbert makes learning really fun☺️
Such professional talking for a person with a low sub
I see your letting your passion comes first than subs count to show your professionalism
You earn a subscribe
Fanchen didnt try to learn it he did learn it, and PERFECTED IT
He did! 😊
@@matticawood :D
I just want to mention that a lot of the notes that Sheet Music Boss puts in their videos aren’t actually played. For example, their Tetris video where they create Tetris shapes without affecting the song being played, which is Type A.
That’s super interesting! I didn’t know that 😊
Great video you just earned a subscriber
Thank you! 😊
Bro i tought u had like 700k subs bc ur vids are so professional
Thanks!! I’m glad you think so 😊
Everybody let’s get Matthew to 1 million subscribers!!!
I agree! 😂
Answering Mathew's question at 8.35 . yes you can increase and decrase velocity of the key just by adjusting the voltage and current across the solenoid.
the piano just rick rolled us... thats impressive
How did the person use his fingers to use the piano? How did the person that made the rushy song? How did he make it with his fingers?
this guy is so underrated, subbed.
Thanks! I really appreciate that 😊
I don't regret randomly clicking on this video
I’m glad! 😂
Percussion is defined as a "musical instrument played by striking with the hand or with a handheld or pedal-operated stick or beater, or by shaking, including drums, cymbals, xylophones, gongs, bells, and rattles."
That’s right! A piano does that, but it is also a string instrument 😊 - “String instruments, stringed instruments, or chordophones are musical instruments that produce sound from vibrating strings when a performer plays or sounds the strings in some manner.”
It depends on what decade/century you were classifying it. Originally it was stringed … then reclassified as percussion …. But it’s current classification is as a keyboard instrument … at least that’s what I was taught at the RCM about 25 years ago
Underrated dude u need more subs
Thanks! I appreciate that 😊
I've owned and played a player piano. Rush e would rapidly outstrip the capacity of the pnumatics.
Most are pedal bellows.
Some modification may help, but punching holes that close on the roll would mean that the valve slots would not be completely covered. Player pianos have a limited repeat rate.
You would have to write an abridged version to suit it.
You are correct in that the piano is technically still a string instrument despite percussive force being applied to make the sound. I know it's not the ONLY classification system for instruments but the Hornbostel-Sachs system classifies the piano as a chordophone, and not an idiophone or membranophone like most percussion instruments are. This is DESPITE the fact that it uses the hammers to make sound, and is represented thus as a 314.122-4-8 in that system's numbering method, or a simple chordophone using a string bearer that is shaped like a board or is the ground that is played with a keyboard and sounded with hammers. Interestingly, this means that a piano is in the same general instrument family as zithers, such as the chinese guqin and the japanese koto.
Also, keyboard instruments under that system are not in their own category, but rather being played with a keyboard is a suffix that you add on when representing an instrument in that system.
But hey, that's just one of the most common systems that organologists use, it's not the only one as I said so you CAN still say the piano is a string and percussion instrument that also fits neatly into the subcategory of keyboard based instruments.
I didn’t know this! That’s super interesting, I will have to read up on it so I have a fool proof answer next time 😊
WHY IS THIS VIDEO SO UNDERRATED?!?!?!?!
I’m glad you think it is! 😊
"Probably because it was mediocre, no offense."
(BTW this video is excellent, in case anybody stumbles across this and doesn't know, the line is a reference to a meme argument specifically about his composition "Rush E".)
@@razordu30 😂😂😂
I love how matthew just threw rush e
From a phonetic perspective, the recreation of speech is fascinating. Sound is transformed into impulses in our ear's cochlea by the location of hair cells that only respond to a particular frequency. If these frequencies are in the range of the human voice, then a specific area in the brain is activated that categorizes frequencies into meaningful sounds of speech (phonemes). We are used to fill in the gaps in case the signal is patchy. This is exactly what chopsticks gives, the right triggers. Try holding your hand over the words, and it still kind of works!
Watched mark robers video first and had the rickroll and I forgot, so mark, u rickrolled me twice
Sorry about that 😂
This guy deserves a lot more subscribers
Thanks!! 😊
Who spotted a smoke generator behind the piano in the vid
If you did Rush E on a piano roll you would probably tear the sheet in 1000 pieces just by holding it...