dwelltime

tt4me wrote: AgentHEX wrote: tt4me wrote: haggisv wrote: It's simply another subjective term, just like 'throw', that's meaningless as an actual number, No it isn't!!!! One can get a high speed camera a count the frames the ball is in contact with the paddle.  It is simple. Yet this has little to do with what people are referring to as "dwell". The ball doesn't care about what people feel.  The trajectory of the ball is not influenced by what people feel.  Only what actually happens affects the ball.

Sure, but people care about what people feel, and how people feel potentially influence their play.

Recall this is confusing due to two words/concepts that happen to be spelled the same. Show me where dwell is a medical term used to describe how long a ball is in contact with the paddle.

Dwell as people define it seems to be some mix of various items discussed above. Note this is not the same definition as you use. Surely you can understand the concept of one spelling with different meanings given your reasonable proficiency with the english language.

You also don't need a high speed camera to measure dwell, just a reasonably accurate one. An accelerometer would do if the sample rate is high enough.  What else would you propose? Most are probably wondering WTF am I talking about.

My proposal was already mentioned above. With two sample points for an object, one can determine velocity and the rest is the classic trivial two trains meeting in the middle problem. As long as the samples are sufficiently accurate, it should be possible to determine exact time when both the ball contacts and leaves the bat. This is probably most easily done with fixed bat and camera, in which case only 4 total samples (frames) on ball are necessary, or 6 to account for likely insignificant aero if you want to be really retentive. Not necessarily perfectly accurate on rebound since we need to assume reconstitution of rubber, but certainly good enough for relative comparisons.


Edited by AgentHEX - 08/30/2013 at 3:35am

tt4me wrote: I agree that people should use what they feel comfortable with but ignoring what is actually happening is putting your head in the sand.

What's actually happening is quite useless to a player in this case, but how a strike feels so you can predict its path is quite useful.

Show me in a dictionary.  Show me that medical people use the term dwell the way Baal uses it.

Language is defined via usage, not by diktat. That's basic linguistics, and quite obvious given that language existed well before dictionaries.

Also, Baal isn't "defining" the term, he's just guessing (however accurately or not) at what others feel/mean when they use the word.

Show me.  It isn't that simple.

What? You want a lesson on how to solve the train problem?

OK, I agree but why hasn't anybody done it?

Maybe they already had ready access to the fast camera, and that's more definitive given less assumptions of how the rubber moves along with host of information about impact other than dwell (magnitude of deformity, etc).

i agree again to a point.  Why hasn't anybody done it?  If it were easy then someone should have made some high speed videos by now and calculated the dwell time.    I know why.  There are few people that can fit a system of differential equations to data.

These are linear equations. The train problem is a "dummy" math joke for people really bad at math.

Smart people probably didn't bother with dwell time much because they know it's a pain in the ass regardless to collect for all range of speeds times range of angles, times #rubber times #blade combos for something that doesn't matter.


Edited by AgentHEX - 08/30/2013 at 4:12am

AgentHEX is right about duration vs reaction time.  The conduction velocities simply mean that you cannot possibly react in time to make a change in what you are doing while the ball is still on your blade. Another measure of interest is that the maximum frequency of a vibrational stimulus that most people can detect with their hands is about 300 Hz (and still discern it as a "buzz").

mercuur wrote: http://ooakforum.com/viewtopic.php?f=38&t=14333 This is where my 1000 fps came from. Skipping seconds with the mouse over the runbar shows a filmic effect on itself. That can,t be from one ms frame or two offcourse. But also shows the filmic effect as much shorter then another second for watching the same part on the video. This must have the shorter filmic effect spread over this second. To me this said and says it has more slow-mo technicques combined inside the camera (as a real camera and as a projektor to a screen, two funktions it combines). Even with celluloid film there are different aspects influencing movietime (adapting images/meter celluloid, projektor runspeed as frequency and diameter of the rolled up film ....all combined). So this nothing spectacular on itself for this part. 10 fps display for this camera is plain japanese technical information. It is what it is. Slower or faster to a clock or realtime, whatever that is (or not realtime would have to be), all doesn,t apply to these 10 fps because it simply would be less or more fps then (or the manual lies with this value). No matter how or how much steps  it clearly has another slow mow rate of factor 100 before or after as 1000=10*100 (no units) or 1/10 * 1 /100 = 1/1000 (any clock-unit for physics). Otherwise it has  0,01 sec/sec for science-fiction.

How do I translate this into english? Google translate w/ auto-detect is of no help.

Bet nobody anticipated it would be mercuur who goes and parodies mercuur.

JacekGM wrote: Maybe from a different angle: do different rackets have consistently different dwell times? Now, be careful how you answer this. If you say no, you are in trouble.

Baal wrote: tt4me wrote: I can do the math well enough to make a pretty good estimation of the dwell time.  It seems that these threads go round and round because no body does the math nor does anybody know if they can feel the difference.   Ok.  Show us.

Everytime I have told you that this is pnatchwey, you have always said this cannot be him.  Believe me now?

NextLevel wrote: Baal wrote: tt4me wrote: I can do the math well enough to make a pretty good estimation of the dwell time.  It seems that these threads go round and round because no body does the math nor does anybody know if they can feel the difference. Ok.  Show us. Everytime I have told you that this is pnatchwey, you have always said this cannot be him.  Believe me now?

Unfortunately it takes some level of introspection to play the satirical archetype game well. Consider the former "fundamentals" for the sport of mocking people.

Pnacthway it is, except that he has lost his medications and has gotten worse.

Dwell time  is the length of time the ball stays in contact with the blade and it depends on the nature of the collision between the ball and the blade, so it is a question for engineers and physicists and their cameras and equations and whatnot, and I am sure they can do this.  The one paper I have found where somebody actually looked at the collision (using a camera at 4500 frames per second and a blade hooked up to a golf swing robot) showed that it is far too short for anyone to do anything about it in the course of a single shot. 

Then there is the complex sensation that is evoked by the ball hitting the blade, which varies with equipment (and the shot being made too, obviously).  One aspect of that sensation is sometimes described as "dwell" by players.  If this sensation is reproducible, it would be useful to know what players are feeling (presumably vibrations in the handle and blade) and one would like to know the capabilities of player's sensory systems to actually discriminate between small changes in the stimulus.  I tried to mention some of those sensory capabilities, and it is clearly relevant to the discussion, even if tt4me doesn't think so.   

AgentHEX used the analogy of cars to talk about how when people talk about "fast" cars.  He  mentions it is not clear if they are talking about HP, torque, top speed, etc.  But those things are often interelated, and there is no doubt that anyone who has driven would reliably report that there is a difference between a BMW M3 and a Ford Fiesta.  Skilled drivers on a test track would discern the difference between the BMW M3 and the BMW 335, and what they report would be related in some way to HP and torque, and even topspeed if they pushed the cars.  Of course, the sense of acceleration is very different from what one perceives when one is driving really fast at a constant speed, so it is easy to feel those differences.  Table tennis sensations may be more subtle.

So, when players report a difference in "dwell" what are they actually measuring and is it a reliable measure?  When someone talks about dwell (personally I never do) is that different from a setup feeling "hard" or "soft" or "flexible" or "stiff"?  I agree with AgentHEX when he complains about it not being clear at all what people mean. 

Of course, then there is the completely different question of whether it matters.  Just like for getting to and from work in rush hour, you get there at exactly the same time if you drive the M3 or the 335 -- and the Fiesta too, if you don't ever have to floor it to make it through a yellow light.  Of course, there are times when it would matter.  Are you driving across the country?  Are you racing cars?  Trying to pick up chicks? Concerned about gas mileage?

I actually think that making changes between broadly similar classes of equipment has no meaningful effect on how you play once you get used to the different feel.  So if you switch from one composite blade to another -- from an ALC to a ZLC or from Tenergy to Bluefire -- it really won't matter.  You will feel the difference, you will get used to it, and if you continue to use one kind of equipment, this is the thing you will eventually like best. 

Making big changes does matter.  So when AgentHEX says that there is probably a 2200 player who can go from inverted to LP and only drop 100 points in level? Or to a clipboard??? I wouldn't want to say that such a player does not exist on this earth (for the inverted to LP change) but this person would have to play with both kinds of equipment frequently to keep those levels that similar, and even then it would be extremely unusual.  The implication of AgentHEX's statement that he can do it is that he is a very low level player who likes to call people noobs.  (Or alternatively a very very very high level player which seems less likely).  But he obviously knows a lot about engineering, so a lot of the other stuff he says is probably useful.

Edited by Baal - 08/31/2013 at 5:43pm

No problem for me to parodie myself if necessary..
I could keep doing that until someone explains otherwise how  a camera with (only) 1000 fps can give a decent filmic illusion of dwellperiod when this has only 1 or 2 msec.
I'm all open for it.

Analogic a bladder book with pictures of donald duck.
Bladder the book at slower or faster rate. Donald could start running now. As slow or fast as someone would want or please. 
But what time would a picture represent ? The time it took to draw them by hand ?

Fine .... but the time for X pictures drawing between two pictures stays the time for drawing x pictures. The bladderrate turning pages won,t affect that. 
The timerelation is simple that there is no relation. It,s called illusion or film.

The amount of frames matters more but even the camera frequency (or they could use more then one ?) is not relevant when the open time for a lenseshutter for each frame is less then 0,001 sec or when the frames are not timecontinuous otherwise.

So I would keep an eye on not confusing time with frequency.








Edited by mercuur - 08/31/2013 at 6:33pm

JacekGM wrote: FINALLY!  I just stopped believing the kind of low speeds suggested by my respectable predecessors here... and I actually googled "superfast camera". Here is the first hit I got: http://www.techimaging.com/photron . Well, as I had expected, if we wanted to use a camera for our experiment, we can.

Who are we?  Do you know how much those phroton cameras cost?  They are in the $100K range.  That will buy a few sheets of T05.

Maybe Baal can spare some of his grant money.

Then you must also spend even more money on lighting.  Without lighting one can't record at very short shutter times and if the shutter times are not short the images will blur.  All that lighting generates heat.  I wouldn't want to stand in that heat for too long.  Then there is the issue of even seeing the ball to swing at it when staring into all that lighting.

The available speeds are muuuuuch higher (like 1000 000 frames per second..., yes that's in the area of microsecond shutter time) than the 1000 fps suggested by some. And this is just the first hit I looked at.

Photron is top of the line.  That is what they use on TV programs.

Come on, these days things are available out there...

Yes, I have know that but they cost too much just to record some TT videos.  There would need to be some other reason to justify spending that kind of money.

I haven't googled superfast balance yet .

You don't want a balance.  You want an accelerometer.   A very high speed accelerometer if you want to capture data 100K times a second.  Then you must have a way to get all that data to a PC quickly. 

NextLevel wrote: Baal wrote: tt4me wrote: I can do the math well enough to make a pretty good estimation of the dwell time.  It seems that these threads go round and round because no body does the math nor does anybody know if they can feel the difference.   Ok.  Show us. Everytime I have told you that this is pnatchwey, you have always said this cannot be him.  Believe me now?

fatt, 

fatt wrote: did I sound that serious?