Debunking implausible gravity claims in pianism/boxing

[Andrew Thayer]

Hi,

Firstly, I'm primarily a musician with an interest in science. I'm particularly keen on using proper scientific methods to discredit a quasi-religious tendency for pianists to radically overhype the role of gravity in pianism. Here's one article I've written on the subject (which also debunks a load of nonsense from Jack Dempsey about gravity in boxing).

http://pianoscience.blogspot.co.uk/2015/11/piano-technique-weight-in-motion-boxing.html

Currently, I'm particularly interested in how theory could be used to take an approximate measure of how quickly literal use of falling weight could depress multiple keys one by one. I'm already certain that it's hogwash to try to explain any significant speeds through falling weight. I give a demonstration in the post of how dropping a small object by a cm or so and picking it up as fast as possible is incredibly limited. Three drops or so per second already seems rather ambitious to me. While it gives some degree of practical illustration of the limits of falling weight, obviously it's not terribly precise or scientific. Any thoughts on how to get an approximate theoretical measure of a reasonable estimate for the fastest reasonable speed of repetition that could be expected through free fall? I want to go further with an incontrovertible proof about quite how relatively slow repetition of free fall actually is, compared to actively instigated movements.All thoughts are welcome, but in a sense I'm more interested in views of complete non-pianists than of musicians. I've encountered very high level pianists who also studied physics at upper levels and, suprisingly, I've often found them among the least productive to discuss such issues with. It's as if they simply can't bring themselves to reference the two things fully. I got a sense that they weren't prepared to deal with the cognitive dissonance of having to reconcile what they were taught at the piano with what they know about mechanics. They seemed to prefer having a cast iron partition in their minds, and were surprisingly closed minded about applying objectivity to analysis of pianistic mechanics. I want my ideas on both to match up, so I'm looking to go further still in debunking gravity ideas that cannot add up. In many ways the colder and more abstract the scientific viewpoint, on this, the more interested I would be.

 

[Fervent Freyja]

Oh, my... those are nice hands. I would be glad to help you.

@Evo, you must watch & listen! I re-played it 3 times, it's lovely.


Some references addressing some of the physics behind finger movement:
Physiology Of Piano Playing - Probability Theory As Extended Logic
The Physics and Metaphysics of Piano Playing: Twelve Fundamental Principles
Hearing the Pianist's Fingers: The Importance of Touch in Piano Music

Here are a couple of studies incorporating physics and biology (probably what you want to have in order to prove your theory or at least set a maximum speed that could be expected in finger-tapping). These studies should give you an idea of what sort of set-up you need or the various methods you could try to repeat (software,hardware, video, etc.). You may even be able to find some maximum speeds already set in their data, that could be good enough for what you are seeking.
Neuromuscular patterns of finger movements during piano playing. Definition of an experimental protocol.
Finger-tapping ability in male and female pianists and nonmusician controls.
Comparison of the relation between timing and force control during finger-tapping sequences by pianists and non pianists.
Independence of timing and force control during finger-tapping sequences by pianists.
Cognitive and biomechanical influences in pianists' finger tapping.

 

[Andrew Thayer]

Oh, my... those are nice hands. I would be glad to help you.

@Evo, you must watch & listen! I re-played it 3 times, it's lovely.


Some references addressing some of the physics behind finger movement:
Physiology Of Piano Playing - Probability Theory As Extended Logic
The Physics and Metaphysics of Piano Playing: Twelve Fundamental Principles
Hearing the Pianist's Fingers: The Importance of Touch in Piano Music

Here are a couple of studies incorporating physics and biology (probably what you want to have in order to prove your theory or at least set a maximum speed that could be expected in finger-tapping). These studies should give you an idea of what sort of set-up you need or the various methods you could try to repeat (software,hardware, video, etc.). You may even be able to find some maximum speeds already set in their data, that could be good enough for what you are seeking.
Neuromuscular patterns of finger movements during piano playing. Definition of an experimental protocol.
Finger-tapping ability in male and female pianists and nonmusician controls.
Comparison of the relation between timing and force control during finger-tapping sequences by pianists and non pianists.
Independence of timing and force control during finger-tapping sequences by pianists.
Cognitive and biomechanical influences in pianists' finger tapping.

Cheers for the links they're definitely of interest. I should clarify though, it's specifically the maximum potential frequency of free falling weight that I'm looking at here. I have no doubt at all that pianists can play radically quicker through finger movements than by dropping the arm on each finger. However, it's amazing how prevalent this explanation is. I'm looking for ideas about how a probable maximum could either be calculated through theory or experimentally tested. Obvious a human is a poor variable, so while I'm confident that my experiment of trying to quickly drop and catch an object gives a meaningful practical yardstick to a pianist, I'd like to have something more concrete to reference gravity claims against.

 

[mfb]

Some numbers to consider:

In free fall for 0.3 seconds, things fall down by 1/2 * 9.81 m/s^2 * (0.3s)^2 = 0.44 meters. Falling down 10 cm takes just 0.14 = 1/7 seconds, and falling down 1 cm takes 0.045 = 1/22 seconds. The free-fall time is of the same order, or shorter, than the time between key presses, so gravity is certainly relevant.

How much energy does it need to press a key? This website suggests a force of about 1 N, which means an energy of 0.01 J assuming a distance of 1 cm. A human hand has a mass of about 500g according to http://www.clbme.bas.bg/projects/motco/data/massinertial.html , which gives it a gravitational force of 5 N. Just putting down a hand (without arm) should be able to push 5 keys. Is that reasonable? Using the additional gravitational force from the lower arm (1kg) and some momentum from a height of a few centimeters are sufficient to generate a much faster key press than just resting the hand on the keys.

Does gravity help you playing? It reduces the downwards force you have to apply (notably!), but it increases the upwards force. You just need different muscles in different intensity. It helps if any action requires the maximal possible downwards acceleration you can produce, and if at the same time no action requires the maximal upwards force.

 

[Andrew Thayer]

Also, thanks for the reopening the thread. I should clarify in response to the issues that saw it temporarily closed. Firstly, there are scientific studies out there that have made similar suggestions about falling weight in pianism. It's not simply a matter of needing to tackle pseudoscience in piano methodologies. Obviously that is a personal goal of mine, but I'm equally concerned with refuting flawed scientific conclusions (that only help add fuel to the fire of flawed methods) for the sake of science. It's in that spirit that I posted here.

I'll link a scientific paper at some point soon with my criticisms of their conclusions, but I'd really like to know how others might consider making an attempt at a reasonable estimate of the maximum possible frequency of dead weight. If we start from the key, obviously the acceleration won't actually equal 9.81 m/s2 owing to the resistance of the mechanism. But how severely might this be reduced in a reasonable estimate? Even if we suppose for now that you could reset infinitely fast, it would interesting to start with an approximate calculation of how many times you could fall on the key per second. Obviously the true figure (including the reset) would be radically lower, but that alone might already be enough to bring into question attributions to gravity (given quite how many notes per second are verifiably possible in skilled pianism).

 

[Andrew Thayer]

In free fall for 0.3 seconds, things fall down by 1/2 * 9.81 m/s^2 * (0.3s)^2 = 0.44 meters. Falling down 10 cm takes just 0.14 = 1/7 seconds, and falling down 1 cm takes 0.045 = 1/22 seconds. The free-fall time is of the same order, or shorter, than the time between key presses, so gravity is certainly relevant.

Absolutely, to be clear, my issue is not with it having some degree of objective effect. My point would be the fact that it cannot fail to have an effect. That doesn't make it the primary point of interest within an explanation. An example I'm using at present is the crack of a whip. Obviously, anyone who was restricting use of gravity via muscular repression would be guilty of very poor technique. However, it would be lunacy to conclude that the secret to good whip technique is using gravity "instead of" muscular acceleration. The gravity is present only in as much as it cannot fail to be, unless you are repressing its action by directly working against it. Your muscles do virtually all of the meaningful stuff associated with cracking that whip.

http://sites.google.com/site/auditorymotor/FuruyaNeuroscience2009final.pdf

For me, the above paper falls fouls of a comparably illogical interpretation of data. I should stress that I have much respect for those scientists in general and they've done some excellent work. However, I cannot see how their conclusions fit the data. As with whipping, gravity will most certainly contribute. But their conclusions are misleading due to absence of greater context. Whipping effectively involves more gravity release than whipping against repression of gravity. But in no way does it follow that the way to whip well is to consciously aim for more gravity and less muscular acceleration. It seems to me that they haven't properly contextualised the data, by distinguishing between effective muscular actions and ineffective ones. A good swimmer probably uses far smaller muscular efforts when swimming at the same slowish speed as a very bad swimmer. People with poor skills often work their muscles harder as part of poor technique, regardless of whether gravity could help. It's an issue of largely superficial correlation, not significant causation. It doesn't follow that the answer is to aim to reduce muscle contractions and allow gravity to take over, in order to improve- either in pianism or swimming. Its an issue of the quality of technique being applied. Muscular activation is a lot more complex than something that you just remember to pour the right amount of out of a bottle. You don't just remember to pour a bit less out and then magically succeed. For me, a good pianist whips all the more actively via muscles than a bad one. It's more deliberate, not more passive. They're just more efficient and better with their technique- thus they can apply it lightly rather than coarsely, via brute force. It no more follows that gravity is the primary explanation, than in actual use of a whip.

To learn do this (skip to 6:25 in):



I had to become all the more active about using my muscles to generate whipping movements. Yes, gravity is in there, because I'm not fighting it. But I sure as hell couldn't get to that point when I was trying to drop at such a speed instead of using muscular actions. I can't recall if the issue arises in that paper, but the same scientists also did a paper in which they showed that all accomplished pianists were landing octaves with a slight forward movement of the upper arm- which clearly contradicts a free fall landing and very obviously indicates an important muscular action. I feel they haven't given the whole picture, before stressing the role of gravity. Even if the facts are accurate (which I don't doubt), the context is missing.

How much energy does it need to press a key? This website suggests a force of about 1 N, which means an energy of 0.01 J assuming a distance of 1 cm. A human hand has a mass of about 500g according to http://www.clbme.bas.bg/projects/motco/data/massinertial.html , which gives it a gravitational force of 5 N. Just putting down a hand (without arm) should be able to push 5 keys. Is that reasonable? Using the additional gravitational force from the lower arm (1kg) and some momentum from a height of a few centimeters are sufficient to generate a much faster key press than just resting the hand on the keys.

It's a bit difficult here as there are a lot of factors regarding how weight can manifest itself. It's hard to separate hand from arm, without literally cutting it off. However, the typical ideas are based on arm-weight rather than merely on consideration of the hand's mass. To get a model of this, I'd probably remove the human factor at first. Taking a moderately heavy object and seeing how quickly it can depress one key might be easiest way to get a reference. Is there any cheap timing equipment available for such purposes? If so, it might be easier to measure such things, rather than estimate them? I think one problem is that key resistance is variable (it increases in response to a larger force, despite starting at around 1 N). I don't know if it's likely that reasonable measure of the true acceleration rate could be made?

It helps if any action requires the maximal possible downwards acceleration you can produce, and if at the same time no action requires the maximal upwards force.

The problem here is that models tend to treat the whole thing as if the human hand is a rigid body. It's impossible to get anything meaningful on such an assumption. If merely falling, the knuckles will collapse inwards- thus carrying a higher velocity than both the fingertips (which lose potential speed due to key resistance) and the keys. The only way to apply potential gravitational speed in full is to either have a rigid hand, or for the hand to actively expand during contact (thus accelerating the fingertip to a faster speed than the falling arm, while pushing back up at the knuckles and reducing their capacity to fall down).

 

[Andrew Thayer]

falling down 1 cm takes 0.045 = 1/22 seconds.

Sorry, I didn't address that point before. I have to admit that I'm surprised that it's quite so quick. To explain the practical limitations, presumably the true rate of acceleration must be significantly limited by the resistance of the piano key, or it must be more to do with the reversals of direction? I wonder, even if forgetting the piano key for now and having unimpeded free fall, what would be the quickest rate that an object could be whisked back up and allowed to fall the cm again (using some kind of device rather than fallible human hands)? I think adding the upward aspect will make even hypothetical calculations more difficult- although it seems that this is necessary in order to explain why repeated dropping is actually rather slow in practise.

Perhaps a lot of this lies in momentum issues when reversing direction? I give an example in my post of how if you throw a golf ball to around eye level, it virtually seems to "hang" in the air briefly, while changing direction. Obviously this doesn't happen when you simply do a dead drop, as the slowing down of an ascent and the beginning of the descent are notably more visible when combined together. Maybe it's not so much the limits of a dead drop which is to be blamed here, but rather the slowness with which gravity negatively accelerates an object which has been whisked back up (before it can even begin generating downward motion)?

 

[A.T.]

People with poor skills often work their muscles harder as part of poor technique ...For me, a good pianist whips all the more actively via muscles than a bad one.

Your theories sound confusing. But the length of your posts definitely indicates practical skill.

 

[Andrew Thayer]

Your theories sound confusing. But the length of your posts definitely indicates practical skill.

Well, at the root of it all is a very simple issue. I don't believe that it's possible to lift and drop an object repeatedly at a notable frequency, assuming that the descent is left to gravity alone (without the assistance of "whipping" style muscular activations). Using appropriate scientific methods to gain a measure of the approximate limits is certainly a lot more confusing, but I'm hoping there may be some way. It needn't even be piano specific. I suspect whatever the hypothetical limit would be for any object being dropped, it would be vastly slower than high speed piano playing- and thus adequate to disprove anything which overhypes gravity.

 

[A.T.]

I suspect whatever the hypothetical limit would be for any object being dropped, it would be vastly slower than high speed piano playing

Why speculate, instead of calculating, as mfb already did in post #4 ?

 

[Andrew Thayer]

Why speculate, instead of calculating, as mfb already did in post #4 ?

As I already stated, piano keys have resistance. The figure he gives for 1cm of falling is a valuable point of reference, but 9.81m/s2 will not literally apply when a key is in the way. Also, see the points I made about the reset, and consider the problems with jerking the mass of the arm quickly upwards ready to fall again. A finger can reset very quickly (the piano springs will do a great deal to push such a small mass back up, especially when it moves explosively and thus causes a significant upward reaction force). An arm that has fallen as dead weight isn't likely to bounce far and would have to be reset. However, raising a whole arm quickly means there are significant inertia issues before it can begin to fall again. As I said in my reply, I was surprised by how quick a 1cm drop could be, but there must be an explanation of why the practical reality of trying to repeat the process quickly doesn't work. Rest your hand on a surface and try to lift your wrist by a cm and drop again as fast as you can. Even when transparently using muscular assistance, it's not especially easy to do it fast. When trying hard to merely fall, rather than be propelled back down, it's very slow indeed. Not even close to the rate I play octaves at in that film (and there are plenty of virtuosi who put my pace to shame). It's so much slower, that it can't reasonably be blamed on faulty human coordination alone- (especially not if trusting the theory which says what I do at the piano must be owed to my ability to use falling weight).

Anyway, his figure clarifies for me that it's not a fall in isolation that demands the most attention- but the process of having to get from down to up and then wait for momentum to dissipate before you can begin to fall again (due to being passive, rather than using muscular acceleration to quicken the turnaround). Not easy to analyse, but it seems clear to me now that this is the area to focus on, when trying to ascertain a feasible limit for the maximum speed.

 

[A.T.]

I was surprised by how quick a 1cm drop could be, but there must be an explanation of why the practical reality of trying to repeat the process quickly doesn't work.

The explanation are the physiological limits of muscle activation speed. The muscles cannot switch very fast and are the bottleneck that limits the frequency of repeated drops from such a small height.

 

[PeroK]

@Andrew Thayer

I think you're wrong about Jack Dempsey. If you went a few rounds with a heavyweight, you might understand better how they get their weight into a punch!

 

[Andrew Thayer]

The explanation are the physiological limits of muscle activation speed. The muscles cannot switch very fast and are the bottleneck that limits the frequency of repeated drops from such a small height, not gravity.

Well, no. Because when I use my muscles to both accelerate the turnaround and descent it's definitely faster, beyond any doubt. The process is faster when switching, than when only being active in a single direction and using passivity to come back, so it doesn't add up as being so simple. There would have to be some kind of experiment in which a machine were designed to whisk an object upwards before dropping it again. But I'm not convinced it could be all that much quicker than a human version, with gravity as the solitary provider of downward acceleration. I believe there's something innate to the turnaround of mass, rather than something that is purely down to muscular limits.

 

[A.T.]

Well, no. Because when I use my muscles to both accelerate the turnaround and descent it's definitely faster, beyond any doubt. The process is faster when switching, than when only being active in a single direction and using passivity to come back, so it doesn't add up as being so simple.

Being active in one direction has the same requirement on muscle switching frequency as being active in both directions (for the same movement frequency). The difference is that in the two-way-push you can co-contract the antagonistic muscles, and then just vary their levels of force. This might be faster than activating a relaxed muscle and then relaxing it again (one-way-push). It also makes sense to have the tendons tensioned all the time, as tensioning them creates a delay.

 

[Andrew Thayer]

Being active in one direction has the same requirement on muscle switching frequency as being active in both directions (for the same movement frequency). The difference is that in the two-way-push you can co-contract the antagonistic muscles, and then just vary their levels of force. This might be faster than activating a relaxed muscle and then relaxing it again (one-way-push). It also makes sense to have the tendons tensioned all the time, as tensioning them creates a delay.

I take your point, but I think it still stands that its faster when very much down to muscular acceleration, than when letting gravity take on half of the movement role. I'm not sure that any internal details can really override the fact that gravity is definably more limited than the muscles, in the context of an arm.

To be honest, in pianistic terms all that matters is that it's definitely not fast enough to be a credible explanation. However, on a scientific level it does intrigue me. Imagine something akin to a pneumatic drill, where there were very active retraction followed by a release. I'm sure we could agree that it would be nowhere near as fast as a proper drill (and also utterly uselessly for drilling), but it would interesting to know what limit would exist there. My personal suspicion is that it would be slightly faster than dropping the arm but probably not as fast as whipping the arm via muscular acceleration. Either way, the hypothetical limit that exists for an object when the down is left to passive forces would be an important yardstick- against which to reference the severity of the human factor.

 

[Andrew Thayer]

@Andrew Thayer

I think you're wrong about Jack Dempsey. If you went a few rounds with a heavyweight, you might understand better how they get their weight into a punch!

I'm sure he had plenty of power, but the guy simply didn't understand the difference between traveling mass and falling weight. There may have been some reason why dropping his torso loosened something up to help him hit harder, but he clearly wasn't redirecting acceleration due to gravity.

 

[A.T.]

gravity is definably more limited than the muscles, in the context of an arm.

More limited in what sense? 1/22s for a 1cm drop (half cycle) corresponds to 11 full cycles per second. Can humans move their fingers/hands faster than at 11Hz over 1cm, using their muscles alone (for example in the horizontal direction).

 

[mfb]

As I already stated, piano keys have resistance.

I took this into account in my post already. If the force value I found is reasonable, the resistance is negligible unless you want to press 5 or more keys at once.

 

[Andrew Thayer]

I specified in the context of an arm, but I'm still highly doubtful than even a specially designed machine would cope well with the problem of reversal. Dropping frequency can't be abstracted from the reset for another drop. Sure, the frequency of the gravity drop itself would be high if it could be reset by magic, but I'm talking within the practical reality of having to perform those resets. It seems me that this larger cycle is the issue. As you can't depend on falling without the rest of the cycle, we can't draw too much into the 1/22 (especially given the resistance of the piano and the imperfection of acceleration when slowing down, ready to begin a reversal of direction).

To blame it on the muscles with certainty, we'd have to prove that eliminating muscles in favour of an better means of reset would be substantially faster. It would be difficult to prove either way, but I'm not convinced. Eliminating the muscular whipping in favour of gravity alone for the down part seems like a bigger compromise. Whatever is truly to blame, the only way to actually establish it would be to uncover the limit for dropping and resetting with an external object.

More limited in what sense? 1/22s for a 1cm drop (half cycle) corresponds to 11 full cycles per second. Can humans move their fingers/hands faster than at 11Hz over 1cm, using their muscles alone (for example in the horizontal direction).

 

[Andrew Thayer]

The resistance is more significant when playing louder (which is where people particularly stress gravity as the provider of energy). With more significant acceleration, the mechanism will put up notably more resistance. The 1N that you quote is only enough to put a key down silently or near silently. I wouldn't like to try to make an estimate, but it's certainly a lot larger for real playing. For instance if I rest my phone on a key and release, the tone is particularly flimsy. It's probably 5N or so, but one key slows it significantly compared to an unimpeded fall. I'd need multiple times that to play loud.

I took this into account in my post already. If the force value I found is reasonable, the resistance is negligible unless you want to press 5 or more keys at once.

 

[Andrew Thayer]

There would be a lot of force to overcome in this style of playing, in order to avoid substantial slowing down due to key resistance.



And at this level of volume the forces applied could easily be 50N and beyond

I wouldn't like to try estimate the level of resistance from the keys, but 1N is a figure that won't apply in practise. The resistance would definitely be a significant limiting factor when using gravity over a small distance, rather than with a run up to accumulate energy.

 

[A.T.]

I'm talking within the practical reality of having to perform those resets.

Practically it's about the physiological speed limits and the effect of pre-tension on them.

 

[Andrew Thayer]

Practically it's about the physiological speed limits and the effect of pre-tension on them.

Well, I don't think we can casually state that as fact without building a mechanism to perform a rapid turnaround frequency- thus proving it can be done. We haven't verified that it's about the limits of physiology and not simply about the general difficulty of quickly resetting for another drop in general (whether through muscles or some kind of robotic device). Given that I can play faster with my muscles (when doing more with them, rather than trying for less), I don't think it seems logical to blame the physiology so much as the limits of a gravitational concept. My physiology is capable of far more when not trying to limit itself to what gravity offers. It doesn't make sense to blame that element, given both that it's capable of superior operation than when dropping and that we haven't proved superior frequency via gravity in any independent setup.

 

[PeroK]

I'm sure he had plenty of power, but the guy simply didn't understand the difference between traveling mass and falling weight. There may have been some reason why dropping his torso loosened something up to help him hit harder, but he clearly wasn't redirecting acceleration due to gravity.

By dropping his weight and pivoting, that is exactly how a boxer could transfer his weight into a punch. The normal shot put technique involves this as well. At the beginning, they go right up on their toes, then drop their weight. That's why shot putters are so big, compared to javelin throwers, where fast arm speed is the key.

 

[A.T.]

the general difficulty of quickly resetting for another drop in general

No idea what that means.

 

[Andrew Thayer]

Removing the human element and seeing if an object in general can achieve a higher frequency, despite the problem of having to be raised up before it can fall again. You're blaming our physiology but there's no evidence it can be done faster without that variable. And it CAN be done faster when we use our physiology better, with less gravity dependence and more active acceleration.

No idea what that means.

 

[Andrew Thayer]

By dropping his weight and pivoting, that is exactly how a boxer could transfer his weight into a punch. The normal shot put technique involves this as well. At the beginning, they go right up on their toes, then drop their weight. That's why shot putters are so big, compared to javelin throwers, where fast arm speed is the key.

But that would be mass, not weight. I'm not normally a pedant on the distinction but it's of key importance here. Particularly with those bizarre diagrams about turning downward speed into horizontal speed. There's no mechanism to do so. You can throw the mass of your body into a punch for added momentum, but gravitational acceleration is a different kettle of fish.

As I showed in the post, to accelerate the speed of a fist to world record speed requires a a free fall of some 140 or so feet. There's no issue with throwing mass into a punch, but when we're talking about weight (as linked to gravitational acceleration) it's rationally unsupportable. I wouldn't speculate too much on the reason for shot putters dropping weight, but it surely has more to do with finding an advantageous position from which to provide muscular acceleration? I can't see how a sense of falling could literally be redirected into propelling the shot putt to a significant degree (especially as it would be on a downward trajectory, while the body pivots). . With Dempsey too, I'm not saying he had bad technique. I'm saying he got the explanation wrong of why he found it useful.

 

[Andrew Thayer]

One thing that's springing to mind is a bouncing ball. Obviously it's a lot bouncier than a human arm, but there must a way to get some kind of yardstick from there? On a hypothetical basis of perfect energy conservation (to reflect the fact that the amplitude cannot afford to decrease when moving a piano key), I'm struggling to picture a ball bouncing more than four or five times per second- without significant loss of amplitude to ramp up the frequency. Obviously this would be a higher frequency than any conceivable reality but could it be calculated as a point of reference?

 

[PeroK]

But that would be mass, not weight. I'm not normally a pedant on the distinction but it's of key importance here. Particularly with those bizarre diagrams about turning downward speed into horizontal speed. There's no mechanism to do so. You can throw the mass of your body into a punch for added momentum, but gravitational acceleration is a different kettle of fish.
.

Perhaps one day some clever engineer will invent a machine that turns downward speed into horizontal speed.

I did think that a pole vaulter manages to turn horizontal speed into vertical speed.

It's funny, though, that that is possible but not the other way round.

Perhaps if the vaulter landed on an angled trampoline, that would fire him forward, turning his downward speed into horizontal speed again. That would be bizarre!

 

[Andrew Thayer]

Perhaps one day some clever engineer will invent a machine that turns downward speed into horizontal speed.

I did think that a pole vaulter manages to turn horizontal speed into vertical speed.

It's funny, though, that that is possible but not the other way round.

Perhaps if the vaulter landed on an angled trampoline, that would fire him forward, turning his downward speed into horizontal speed again. That would be bizarre!

It's perfectly possible in demspey's sledge example. I'm just baffled by how it can be done to any meaningful degree in a punch.

 

[PeroK]

It's perfectly possible in demspey's sledge example. I'm just baffled by how it can be done to any meaningful degree in a punch.

Actually, I just remembered something I heard recently. I was watching the tennis and heard a tip on volleying at the net. You take a little jump. You have to time the jump, so that you land just as your opponent is hitting the shot, and you have to make a decision whether to move left or right. The little jump allows you to spring left or right faster than from a normal standing position.

Perhaps that's just more sports pseudo science!

 

[Andrew Thayer]

Actually, I just remembered something I heard recently. I was watching the tennis and heard a tip on volleying at the net. You take a little jump. You have to time the jump, so that you land just as your opponent is hitting the shot, and you have to make a decision whether to move left or right. The little jump allows you to spring left or right faster than from a normal standing position.

Perhaps that's just more sports pseudo science!

Sounds more credible to me. Leaning doesn't generate phenomenal speed, but it certainly prepares you to move. There's some bizarre running method called the falling rod. The idea is that you "fall" forwards continuously like a leaning rod on your hand that balances as long as as you keep moving. Obviously it's untrue and you don't actually fall. The sense you're about o fall just keeps you pushing on, to stop you falling.

I heard a theory from feldenkrais that we evolved to two feet because you're ready to lean in any direction and instantly start moving that way. It doesn't work on all fours. Probably impossible to prove but an interesting idea. Creating instability by leaning slightly with gravity is certainly an excellent way to trigger powerful movement (even if dubious when portrayed as a power source, rather than a trigger).

 

[A.T.]

You're blaming our physiology

For setting the limit on how fast humans can move? Of course.

but there's no evidence it can be done faster without that variable.

Are seriously doubting that a machine could do it faster?

 

[Andrew Thayer]

For setting the limit on how fast humans can move? Of course.Are seriously doubting that a machine could do it faster?

Yes, I'm doubting it would be a whole lot faster when limited to waiting for gravity to both decelerate the upward part and generate the whole downward part.
My muscles are radically quicker than gravity. I can whip an arm up even directly against gravity far quicker than it will fall. Looking at the true amplitude of octave playing, we really ought to be doing calculations for at least three centimeters of movement, rather than the one cm I originally gave for key movement. I don't think it's unreasonable to think that I might perform faster under muscular reversal than a machine that is limited to upward acceleration alone. Take a look at the speed of Argerich's octaves. I don't believe a machine could reach that speed without active downward impulses. I don't doubt in the least that she's faster than a machine that is powerless in any direction other than up. It's only through ACTIVE downward acceleration beyond gravity that a machine becomes superior to a human.