Serve - Simplest Way to Snap???

[Raul_SJ]

"It's the turning to high five at the last millisecond that gives you the juice."

"at the last millisecond" is not correct. The purpose of high speed video is to observe positions of objects vs milliseconds. Find a serve video that shows you from the frame of your picture to the frame of impact (about 25 milliseconds later).

There are players that literally believe what you said and in the late 80s I was one of them.

I took a clinic and the Pro's instruction was exactly that: "Like you are going to high five someone at last instant.."
In any case, how would a coaching cue of "at the last millisecond" be harmful when players really cannot perceive differences between 1 millisecond and 25 milliseconds?

This illustrates that the racket head motion from "edge on to the ball at the Big L Position" to impact rotates to face the ball in a smooth fashion. And that there is no special motion during the "last millisecond" or last few inches before impact. Always refer to pictures like these or high speed videos to check your ideas. The racket is about "edge on" to the ball at Big L and "on edge" moving from Big L to impact.

Counting from the bottom, which racquet pic most closely corresponds to the "Big L" position?
#3? Or #4?

We can look at slow motion youtube video step frame by frame with ">" key.
But how do we know the frame rate of the video below? YouTube "stats" says 25 FPS but it looks to be far higher frame rate
since I have to click ">" 25 times from below frame to reach contact frame.

 

[Chas Tennis]

I took a clinic and the Pro's instruction was exactly that: "Like you are going to high five someone at last instant.."
In any case, how would a coaching cue of "at the last millisecond" be harmful when players really cannot perceive differences between 1 millisecond and 25 milliseconds?

Counting from the bottom, which racquet pic most closely corresponds to the "Big L" position?
#3? Or #4?

We can look at slow motion youtube video step frame by frame with ">" key.
But how do we know the frame rate of the video below? YouTube "stats" says 25 FPS but it looks to be far higher frame rate
since I have to click ">" 25 times from below frame to reach contact frame.

I'd guess the bottom racket position is closest to the Big L Position. ?

A video has a recording rate, for example, 240 fps, and that determines how much timing information is available. Then that video is processed in the camera into a video file with a common playback rate, often 30 fps. The Youtube video player can be varied after that, YT has several different playback speeds available from 0.25 to 2?, click the 'gear' icon on the YT. "Stats for nerds" usually shows playback information and not the more important recording information. The recording speed may be in the video file in a file called 'metadata'. I don't know how to access that metadata file or if it is still in the YT file.

To get an estimate of the frame rate for the serve, I usually consider that 100 MPH is 1760"/sec. A 100 MPH object in one frame time moves 1760"/sec / 240 fr/sec = 7.3" per frame. The first picture in your post, the serve from overhead, looks as if FYB recorded that serve at around 240 fps or so. I only estimate unknown frame rates and do not consider them to be accurate. Another way to estimate unknown frame rates would be to look at the top of a tennis ball toss - gravity would always give the ball a signature, it would fall X ball diameters in a given time. See the strobe picture #4, the peak of the toss. Needs work....

I have never done that but it should work, for example, on old film recordings. There may be some other trick - you need a motion with a known distance and time recorded in the film.

No one that speaks of a 'snap' or a special motion at the 'last instant' has attempted to point out when that might occur in a high speed video. The upper arm is rotated by ISR and it starts, smoothly accelerates to higher rpms and then impacts the ball. I showed that. The muddled wording 'last instant', 'last second', 'just before' (impact) can't be shown in high speed videos because there is nothing to be seen. What is the reason that no one that believes in some special motion is willing to point it out in videos or pictures. Why does 'Snap' only exist in the word world but not in the video world?

There may be a feeling that feels very fast, even like a Snap. But show the image of the 'snap' or special 'last millisecond' or admit that you are unable to show images. Maybe the readers don't notice that there has never been a video of a snap (on the forum since 2004?).

Morphing muddled and ambiguous wording by rewording what was 'really meant',........the readers here are too sophisticated for that.............

 

[Raul_SJ]

I'd guess the bottom racket position is closest to the Big L Position. ?

A video has a recording rate, for example, 240 fps, and that determines how much timing information is available. Then that video is processed in the camera into a video file with a common playback rate, often 30 fps. The Youtube video player can be varied after that, YT has several different playback speeds available from 0.25 to 2?, click the 'gear' icon on the YT. "Stats for nerds" usually shows playback information and not the more important recording information. The recording speed may be in the video file in a file called 'metadata'. I don't know how to access that metadata file or if it is still in the YT file.

I had watched that above YouTube Nadal serve at normal speed, starting from t=00:09 seconds (roughly Big L position with racquet approaching perfectly on edge).

See the screen shot above.(also note that the YouTube video automatically slows down playback at that 9 second point. Not sure how the Youtube uploader controls that setting).

By stepping thru frame by frame with" >" key, from that 9 second Big L point to contact point, I can click see several racquet position frames before contact.

Given that time to contact from that that Big L position at the is approximately 20 milliseconds,
can we estimate the recording rate of the video by the number of racquet position frames we see from 00:09 to contact?

So if we assume 20 milliseconds from big L to contact and we observe 10 racquet positions stepping thru frame by frame on the video, we can conclude that the Original Video Recording Rate was at least 1000FPS... Although it would be nice if YouTube simply provided that info straight away from the Stats command...

I believe max YouTube playback is 60 FPS.
So basically if we purchase a 1000FPS camera, and record a one second clip, there is no way to step thru all 1000 frames after uploading to YouTube? YouTube will playback that clip at 60 FPS. Grabbing only 60 frames and dropping the 940 other frames.

 

[40L0VE]

Learn the basics from any decent tennis manual. Apply what you learnt and if you do it correctly all these concerns about getting enough wrist snap and wrist pronation will go away as they naturally happen when hitting the shot with correct mechanics.

 

[DrWang]

Here is my eleven year old son serving. Relax your wrist more.

 

[5263]

Here is my eleven year old son serving. Relax your wrist more.

excellent motion

 

[DrWang]

excellent motion

Thx. He has an excellent forehand as well. I wish I could swing like that. Kids make it look so easy.

 

[5263]

Thx. He has an excellent forehand as well. I wish I could swing like that. Kids make it look so easy.

yes, off to a great start here as well...Check you pm when you get a chance.

 

[Trinity110]

I knew some middlesmart would come in and get hung up on "the last millisecond"

It's a figure of speech. Not literally the last millisecond.

You're right, its probably about 25milliseconds... but its useless to know that. Nobody can move fast enough and perceive real time in high enough resolution to actually differentiate between "last millisecond" and "25 milliseconds" when executing a serve.

"Last millisecond" gets the intended effect across, which is that you want to stay on edge as long as possible.

I'm totally a third to last millisecond believer. I am accepting of those who believe in anything between penultimate and 5th to last.

 

[Chas Tennis]

I had watched that above YouTube Nadal serve at normal speed, starting from t=00:09 seconds (roughly Big L position with racquet approaching perfectly on edge).

See the screen shot above.(also note that the YouTube video automatically slows down playback at that 9 second point. Not sure how the Youtube uploader controls that setting).

I believe that is a separate close up video clip from a second camera or a later serve. The background of the ball is different, so it's not the same camera view.

By stepping thru frame by frame with" >" key, from that 9 second Big L point to contact point, I can click see several racquet position frames before contact.

When I try to step through pressing the ">" / "period" key, it does not move on some key presses. ? I don't understand that or frame skipping that I have seen on some YT videos?? It may be that the recording camera (in Europe with 50 Hz lighting flicker) would produce a 25 fps playback video. If that 25 fps video is uploaded to a 30 fps Youtube site then maybe the 25 to 30 fps mis-match processing will cause issues. ? I avoid problems like this because you can never be sure. The Youtube often works well when your camera produces a video at 30 fps and you upload to a YT that stores at 30 fps, I don't spend much time trying to figure it out. It is best to deal with videos of known or stated recording frame rates.

Given that time to contact from that that Big L position at the is approximately 20 milliseconds,
can we estimate the recording rate of the video by the number of racquet position frames we see from 00:09 to contact?

It would be best to get other Nadal serves with stated recording time rates and get an average time for the ISR from start of its rotation to impact for the same type of serve. That could be accurate and reproducible for time of start to impact. The Big L Position is often interpretative. The start of the ISR is also interpretative. Also, frames are caught at random times, so you can always be off by a frame time or two. Use higher frame rates to improve accuracy. My rough estimate for the rotation time of ISR is around 20-25 ms (ATP slice or flat) and that could be a little higher or lower. I believe that players vary the angle of ISR and that angle is difficult to estimate using elbow shadows.

So if we assume 20 milliseconds from big L to contact and we observe 10 racquet positions stepping thru frame by frame on the video, we can conclude that the Original Video Recording Rate was at least 1000FPS... Although it would be nice if YouTube simply provided that info straight away from the Stats command...

No, 20 ms / 10 frames = 2 ms / frame - that is 500 fps recordimg speed. 1000 fps is 1 ms/frame. At 1000 fps, you record 20 frames in 20 ms. Youtube always works only with the playback video frame rate that the camera produces and cameras produce common play back rates, like 25 fps, 30 fps, 50m fps, 60 fps, etc. My high speed Casio camera produces 30 fps and may have a European setting to produce 25 fps. In Europe, I believe that there is a law that cameras sold in Europe must only have 25 fps but not 30 fps - check my facts.

I believe max YouTube playback is 60 FPS.
So basically if we purchase a 1000FPS camera, and record a one second clip, there is no way to step thru all 1000 frames after uploading to YouTube? YouTube will playback that clip at 60 FPS. Grabbing only 60 frames and dropping the 940 other frames.

No. Your camera will record at 1000 fps and produce a playback video at, say, 30 fps (there may be options). You can step through all frames, but sometimes fames are skipped for unknown reasons. Or, for the video above when you press the ">" you may not get any response. When I make side-by-side videos on Kinovea, sometimes frames may be skipped and at other times no problem. I now often put a millisecond time scale on the Kinovea videos and that is huge for observing time and it also tells you if a frame is skipped by YT advance.

[/QUOTE]

When you think about this stuff, think -
1) recording frame rate ( a beautiful world of clarity and understanding)
2) playback frame rate ( a confusing world of computer land with post processing that could do anything, out of your control, sometimes wasting your time...)

 

[Raul_SJ]

No. Your camera will record at 1000 fps and produce a playback video at, say, 30 fps (there may be options). You can step through all frames, but sometimes fames are skipped for unknown reasons. Or, for the video above when you press the ">" you may not get any response. When I make side-by-side videos on Kinovea, sometimes frames may be skipped and at other times no problem. I now often put a millisecond time scale on the Kinovea videos and that is huge for observing time and it also tells you if a frame is skipped by YT advance.

When you think about this stuff, think -
1) recording frame rate ( a beautiful world of clarity and understanding)
2) playback frame rate ( a confusing world of computer land with post processing that could do anything, out of your control, sometimes wasting your time...)

If a 1000 FPS camera records a one second clip, you are saying that the camera produces a one second playback clip.
That clip can be opened by VLC Media Player (or similar) and played for one second at normal playback 30 FPS. If you want to step thru all 1000 frames, you must use VLC frame by frame mode.

If you want to upload to YouTube, that one second clip will have to be edited and expanded to 16.6 seconds.
YouTube will then be able to play that original one second clip for 16.6 seconds at 60 FPS. Wil be able to step thru 1000 frames on YouTube.

Example - you shoot 1 second of footage, at 1000 fps, giving you 1000 frames to play with. You can slow it down, to 60 fps, stretching out the time of the video to 16.6 seconds of footage.​

However, when you upload something to YouTube, it doesn’t care what your base fps is, only what the length of your footage is.​

​

If you upload a 1 second clip, even shot at 1000fps, you get a 1 second YouTube video. If you edit the clip first, to a 16.6 second, 60fps video, then YouTube shows a 16.6 second video.​

​

If a video was shot at 1000 fps and then uploaded to YouTube (max 60 fps), how is that video still a thousand fps? - Quora

www.quora.com

 

[Chas Tennis]

Here is my eleven year old son serving. Relax your wrist more.

You son has a very good motion and needs to be comparing his technique frame-to-frame.

Developmental Issues. There is a little information on the developmental capabilities of young players. There are some discussions of what they can and cannot do at certain ages. I don't know much about that. The book Technique Development for Tennis Stroke Production, Elliott, Reid and Crespo is one reference with some developmental information.

Shoulder Impingment & The Serve. There is also some specific recommendations by Todd Ellenbecker on the angle of the upper arm to the shoulder joint to minimize the risks of shoulder impingement for the high level serve with ISR. His video "Rotator Cuff Injury" is available for viewing on Tennis Resources with a 3 month membership. There are threads on this subject on this forum. ATP servers are nearly all examples of good practice and that angle should be controlled throughout the motion. Study this issue.

You can compare your son's serve to high level serves frame-by-frame and one above the other. Select serves of the same type to compare. Serves from the same camera angle work well. You can also find videos from model posters and post them in the forum next to your son's strokes for comparisons.

To single frame on Youtube, use the period & comma keys. To select a video always use Alt + Left Mouse Click. Go to impact and compare similar positions of the serve. The recording frame rates may be equal.

Your son's head and neck seem back more than usual, compare. Tommy Robredo does that more than most with a stressful looking head motion. ?

Look at how your son gets his racket back to the drop frame-by-frame. Compare his forearm to racket shaft angle. In high level serves, the inertia of the forearm and racket are used with body motions to cause external shoulder rotation (ESR). Thoracic Extension also plays a part. I posted a thread on Thoracic Extension and Flexion on the serve.

Your son's arm appears vertical at impact from this rear camera view. I'd estimate that the ATP players tend to have their arms tilted to the right at impact as seen from behind. Study some ATP servers to get stats on the arm tilt as seen from behind. The side view has the arm tilt forward at impact. Look at the arm and racket tilts from various camera angles.

The video analysis application Kinovea is excellent for comparing two videos side-by-side. You can coordinate both videos on the frames with impact and add a millisecond time scale. If you use your camera closer and in bright sunlight you can use the shadows on the serving elbow to show ISR and compare that.

 

[Chas Tennis]

If a 1000 FPS camera records a one second clip, you are saying that the camera produces a one second playback clip.
That clip can be opened by VLC Media Player (or similar) and played for one second at normal playback 30 FPS. If you want to step thru all 1000 frames, you must use VLC frame by frame mode.

If you want to upload to YouTube, that one second clip will have to be edited and expanded to 16.6 seconds.
YouTube will then be able to play that original one second clip for 16.6 seconds at 60 FPS. Wil be able to step thru 1000 frames on YouTube.

Example - you shoot 1 second of footage, at 1000 fps, giving you 1000 frames to play with. You can slow it down, to 60 fps, stretching out the time of the video to 16.6 seconds of footage.​

However, when you upload something to YouTube, it doesn’t care what your base fps is, only what the length of your footage is.​

​

If you upload a 1 second clip, even shot at 1000fps, you get a 1 second YouTube video. If you edit the clip first, to a 16.6 second, 60fps video, then YouTube shows a 16.6 second video.​

​

If a video was shot at 1000 fps and then uploaded to YouTube (max 60 fps), how is that video still a thousand fps? - Quora

www.quora.com

Youtube has its own purposes for how it saves videos. My camera records in various slow motion speeds and has a "Video Out" option for a video file of 30 fps. ( I can also choose a Video Out for 25 fps (PAL) and Youtube woud do something else with that.) All the recorded frames are in my 30 fps output file. Youtube saves my 30 fps file as a 30 fps YT video file that also plays back normally at 30 fps.

I have not used YT 60 fps and do not know when YT would select to save an uploaded video at 60 fps. ? I guess 60 fps out of a camera would cause YT to save it as 60 fps. ?

YT also only saves videos at certain resolutions and there is a YT list for those specific resolutions. Your camera's output video resolution - if not on the YT list -will be reduced to next lower resolution on the list when saved by YT. YT reduced my 336 line resoultion to 240 lines on their list and it was ugly. I had to find a work-around to prevent that from happening. What a giant waste of time that was.

Maybe we should continue this on Converation or start a new thread?

 

[Raul_SJ]

Youtube has its purposes for how it saves videos. My camera records in various slow motion speeds and has a Video Out video file of 30 fps. ( I can also choose a Video Out for 25 fps (PAL) and Youtube woud do something else with that.) All the recorded frames are there. Youtube save them as a 30 fps YT video file a playback video that normally plays at 30 fps.

I have not used YT 60 fps and do not know when YT would select to save a video at 60 fps. I guess 60 fps out of a camera would cause YT to save it at 60 fps.

YT also only saves videos at certain resolutions and there is a YT list for those resolutions. Your video resolution if not on the list will be reduced to next
lower resolution on the list. YT reduced my 366 line resoultion to 240 lines and it was ugly. I had to find a work-around to prevent that from happening. What a giant waste of time that was.

What is the highest frame capability of your camera? If it is say, 240 FPS, your camera can produce a one second 240 FPS video playback file at 30 FPS.
You can step thru and view all frames on your local device player.
But before uploading to YouTube, you will have to locally convert that one second file to an 8 second file. Only then will you be able to step thru all 240 frames on YouTube at 30 FPS playback.

 

[DrWang]

You son has a very good motion and needs to be comparing his technique frame-to-frame.

Develmental Issues. There is a little information on the developmental capabilities of young players. There are some discussions of what they can and cannot do at certain ages. I don't know much about that. The book Technique Development for Tennis Stroke Production, Elliott, Reid and Crespo is one reference with some developmental information.

Shoulder Impingment & The Serve. There is also some specific recommendations by Todd Ellenbecker on the angle of the upper arm to the shoulder joint to minimize the risks of shoulder impingement for the high level serve with ISR. His video "Rotator Cuff Injury" is available for viewing on Tennis Resources with a 3 month membership. There are threads on this subject on this forum. Study this issue.

You can compare your son's serve to high level serves frame-by-frame and one above the other. Select serves of the same type to compare. Serves from the same camera angle work well. You can also find videos from model posters and post them in the forum next to your son's strokes for comparisons.

To single frame on Youtube, use the period & comma keys. To select a video always use Alt + Left Mouse Click. Go to impact and compare similar as positions of the serve. The recording frame rates may be equal.

Your son's head and neck seem back more than usual, compare. Tommy Robredo does that more than most with a stressful looking head motion. ?

Look at how your son gets his racket back to the drop frame-by-frame. Compare his forearm to racket shaft angle. In high level serves, the inertia of the forearm and racket are used with body motions to cause external shoulder rotation (ESR). Thoracic Extension also plays a part. I posted a thread on Thoracic Extension and Flexion on the serve.

The video analysis application Kinovea is excellent for comparing two videos side by side. You can coordinate both videos on the frames with impact and add a millisecond time scale. If you use your camera closer and in bright sunlight you can use the shadows on the serving arm to show ISR and compare that.

Wow! This is priceless information and I want to personally thank you. I'm going to print this.

 

[SystemicAnomaly]

Wow! This is priceless information and I want to personally thank you. I'm going to print this.

I recall posting an Ellenbecker shoulder / rotator video link 7-8 yrs ago. I guess the specific link I posted is no longer active but it is good to know that the video is at least available -- even tho it is now behind a pay wall at TR.

Here is another one I had posted back then from Jim McLennan:

metacafe.com/watch/1637626/preventing_rotator_cuff_injury/

 

[SystemicAnomaly]

 

[DrWang]

I recall posting an Ellenbecker shoulder / rotator video link 7-8 yrs ago. I guess the specific link I posted is no longer active but it is good to know that the video is at least available -- even tho it is now behind a pay wall at TR.

Here is another one I had posted back then from Jim McLennan:

metacafe.com/watch/1637626/preventing_rotator_cuff_injury/

My son is here with me and we just watched this together. I've never thought about this and yet it makes perfect sense. I am not a tennis coach, but serving is my forte and I'm actually the one who taught my son his serving. This is really good info and I'm going to taking more slow motion videos to monitor his progress. Thank you for sharing this.

 

[SystemicAnomaly]

My son is here with me and we just watched this together. I've never thought about this and yet it makes perfect sense. I am not a tennis coach, but serving is my forte and I'm actually the one who taught my son his serving. This is really good info and I'm going to taking more slow motion videos to monitor his progress. Thank you for sharing this.

Very good. As you can see in the Sampras image above, he has got an extreme shoulder tilt at contact. He has dropped his left shoulder so much, while raising the right shoulder & arm, that his shoulder tilt is nearly vertical.

Admittedly tho, this is a pretty extreme example of the shoulder over shoulder cartwheel action. Many other elite servers have more of a tilted cartwheel action. The primary thing to note in this image is that Pete has almost no signs of impingement at contact. His right elbow (& upper arm) is pretty much directly in line with his shoulder tilt line.

 

[Raul_SJ]

Here is my eleven year old son serving. Relax your wrist more.

Do you know what frame rate was used to create this video? It looks like 500 Frames Per Second?

 

[DrWang]

Do you know what frame rate was used to create this video? It looks like 500 Frames Per Second?

No idea. I used my Samsung s9+ set on slow motion video. There is also a super slow motion setting but that setting only allows like 2 seconds of recording. I know the range is 240-960.

 

[Chas Tennis]

Very good. As you can see in the Sampras image above, he has got an extreme shoulder tilt at contact. He has dropped his left shoulder so much, while raising the right shoulder & arm, that his shoulder tilt is nearly vertical.

Admittedly tho, this is a pretty extreme example of the shoulder over shoulder cartwheel action. Many other elite servers have more of a tilted cartwheel action. The primary thing to note in this image is that Pete has almost no signs of impingement at contact. His right elbow (& upper arm) is pretty much directly in line with his shoulder tilt line.

The details of the upper arm to shoulder have been discussed here several times and one thread on this forum that has some comments by biomechanics researcher David Whiteside. Measuring abduction up from the side of the body has problems.

Shoulder Orientation- Overhead Baseball Pitch & Tennis Serve | ASMI Sports Medicine Forum: 2005-2019

This video by Todd Ellenbecker goes into detail on the shoulder and its orientation in order to minimize the risk of impingement on the tennis serve. http://www.tennisresources.com/index.cfm?area=

asmiforum.proboards.com

Thread discussing shoulder orientation details. David Whiteside post.

Question for Chas Tennis about "High Level Servers"

What do you think about the degree of shoulder abduction in the Wawrinka and Murray serves?

tt.tennis-warehouse.com

There is a range of angles being used. I'd say that Federer is on the high side and above the average. 25+ d above the line between his two shoulders extended? It is also possible that some players compensate for some injuries. Federer had some back issues from about 2013 -2015 I believe. You might want to look at the date Youtubes were posted, news of injuries and consider if players might have been injured.

I would say that paying $32 to see the best available video on any tennis injury issue would always be best use of your money. I'd also say that videos that might reduce the injuries of tennis players should be promoted at no cost by tennis organizations and instructors.

This is my opinion, but the Ellenbecker video is for the high level serving technique that includes significant ISR. That is based on my interpretation of what Ellenbecker says. For other serving techniques the reader should determine if it applies and how.

 

[5263]

yes, I started to show the many pics of this gold standard as well....but....just let it go, lol

 

[5263]

I took a clinic and the Pro's instruction was exactly that: "Like you are going to high five someone at last instant.."
In any case, how would a coaching cue of "at the last millisecond" be harmful ?

The answer is that it is helpful, not harmful. It is also correct since initially from the area that chas likes to call the "big L", you should be trying to stay edge on like with a hatchet, as much as you can while still allowing you to make the next checkpoint. Does the racket start to open up some as you go towards contact? Sure, but you should be looking to minimize that if you are looking to hit big serves. The next checkpoint is that important "crossover moment" which represents the last instant ( moment, millisecond, instance, second) or whatever term floats your boat.

The Crossover moment is that threshold where you go from trying to stay on edge to shifting your intent to going fully with the rotation and pronation prior to contact,... required to elicit the type of contact that you intend for the serve.
The confusion on this likely centers around how far this last 'instance' is before contact. For faster and better servers, it is closer to contact and for more avg servers this will normally happen further from contact or not at all.....often meaning they are not even aware of the crossover issue.

As to seeing this difference on video....well that hardly applies to some things like this. Imagine a normal car is coming by you on video at 50 mph and then stomps the gas to the floor. How much evidence will the video show that he floored the accelerator? Very little, as at that point, the normal car won't change the rate of accel as dramatically as it will add sound, torque, and gas flow, all with minimal initial video effect. Same with a crossbow with D shaped cams instead of the round pulleys, in that even though the video would not capture the increase output so obviously, the accel of the arrow will have the kind of improved performance that is similar to taking a 115mph serve up to 125mph.

 

[Rattler]

Keep your wrist loose and cocked forward

 

[Raul_SJ]

No idea. I used my Samsung s9+ set on slow motion video. There is also a super slow motion setting but that setting only allows like 2 seconds of recording. I know the range is 240-960.

Must be 240 FPS.

Samsung S9+ has two slow motion modes:

Slow motion mode can record at 240 FPS at 1080p resolution.​

"Super" slow motion mode can record at 960 FPS at 720p resolution​

​

 

[DrWang]

Must be 240 FPS.

Samsung S9+ has two slow motion modes:

Slow motion mode can record at 240 FPS at 1080p resolution.​

"Super" slow motion mode can record at 960 FPS at 720p resolution​

​

Hey! Thanks for the info! I have a Nikon z7 with the Trinity lenses. I should play around with it more.

 

[bigservesofthands]

There is no wrist snap in a traditional serve. It is arm extension, internal shoulder rotation and a little bit of wrist extension. The snap occurs at the shoulder which is powerful enough to cause the bicep to wiggle! PoMo, PoPoMo and PoPoPoMo are more advanced than this and result in more than a bicep wiggle. Also notice how much the ball slides on the strings.

 

[Chas Tennis]

There is no wrist snap in a traditional serve. It is arm extension, internal shoulder rotation and a little bit of wrist extension. The snap occurs at the shoulder which is powerful enough to cause the bicep to wiggle! PoMo, PoPoMo and PoPoPoMo are more advanced than this and result in more than a bicep wiggle. Also notice how much the ball slides on the strings.

That is an excellent gif. Do you have the original video? I believe that it might be a kick serve?

The entire motion shown lasts about 30 millisecond or 3 hundreths of a second. It is hard to tell what that feels like.

The joint motions are defined for bone movements. You can see a shadow on a large bone in his elbow (medial epicondyle?) and that accurately shows his ISR with no delay. His biceps motion is delayed relative to the movement of his elbow bone because the muscles flop around due to inertia. His tatoo shows the relative skin motion between his elbow bone and the surface of his skin between his bicep and tricep.

The wrist rotation is the ISR + Pronation. If we could measure in degrees: 1) the wrist rotation from ISR + Pronation at the wrist and 2) the ISR rotation at the elbow, then we could calculate the Pronation. I have had trouble doing that.

For this video, look carefully for a difference between the wrist rotation and ISR. If there is any, it would be Pronation.

If we had control of the server, we could calibrate those rotations in degrees by taking a video with an protactor for calibration in the video.

That is a great video!

 

[SystemicAnomaly]

There is no wrist snap in a traditional serve. It is arm extension, internal shoulder rotation and a little bit of wrist extension. The snap occurs at the shoulder which is powerful enough to cause the bicep to wiggle! PoMo, PoPoMo and PoPoPoMo are more advanced than this and result in more than a bicep wiggle. Also notice how much the ball slides on the strings.

Very good.

Agree with much of this but I don't believe that there is actually much sliding of the ball on the strings for the 4 milliseconds (give or take) of contact. Most of that observed turning of the ball is happening off the stringbed (after contact) -- that is, much of that sliding might be an illusion.

I'm fairly certain that John Yandel has studied hours of a super slow motion HD video and has indicated that the slide or roll of the ball on the strings is extremely slight. (Not his exact words tho). The ball compresses very significantly and is pocketed by the string bed.

Very high speed film might reveal that that ball might turn slightly for the 4 milliseconds it is on the strings. If I did the math correctly, for a ball spinning at 2500 RPM, the ball turns 1/6 of a revolution. But even this much turning is difficult to see since the ball is so distorted while it is on the strings.

Here is a video of a serve recorded at 6000 FPS. Not detecting much, if any, sliding.

Granted, this is a 142 mph serve that might have somewhat less spin than the Murray serve. Nonetheless, we do observe spin on the ball after it has left the strings for this flat-ish serve.

 

[andreh]

Snap, you say? I was once up 40-30 to even out at 5-5. Hit 3 double faults in a row to give away the set. That made me snap. Threw the racket

 

[Raul_SJ]

There is no wrist snap in a traditional serve. It is arm extension, internal shoulder rotation and a little bit of wrist extension.

To clarify. There is wrist flexion on the serve. The wrist moves from an extended position towards flexion.
It does not move much past the neutral position; what is commonly mistaken for "wrist snap".

 

[Chas Tennis]

Tennis Warehouse / Learning Center has some very interesting footage on balls hitting strings, etc. But the best things are hard to find.

http://twu.tennis-warehouse.com/learning_center/spinpatterns.php

Rod Cross has some of the more detailed work in this area. The article discusses the effects of different string patterns on the ball.

 

[bigservesofthands]

That is an excellent gif. Do you have the original video? I believe that it might be a kick serve?

The entire motion shown lasts about 30 millisecond or 3 hundreths of a second. It is hard to tell what that feels like.

The joint motions are defined for bone movements. You can see a shadow on a large bone in his elbow (medial epicondyle?) and that accurately shows his ISR with no delay. His biceps motion is delayed relative to the movement of his elbow bone because the muscles flop around due to inertia. His tatoo shows the relative skin motion between his elbow bone and the surface of his skin between his bicep and tricep.

The wrist rotation is the ISR + Pronation. If we could measure in degrees: 1) the wrist rotation from ISR + Pronation at the wrist and 2) the ISR rotation at the elbow, then we could calculate the Pronation. I have had trouble doing that.

For this video, look carefully for a difference between the wrist rotation and ISR. If there is any, it would be Pronation.

If we had control of the server, we could calibrate those rotations in degrees by taking a video with an protactor for calibration in the video.

That is a great video!

This is a Borna Coric serve recorded by a Phantom Falcon slow motion camera at Delray Beach. Here is the source video

I think for most high end servers with a traditional serving technique and decent shoulder flexibility there is not really much pronation. The entire arm moves as a unit at the shoulder during the serve. I have trained myself to get a little extra forearm load past the maximum ESR shoulder load and then time forearm pronation release a little after the shoulder release. However I do not notice much of this attempt in professional servers. Wrist movements during a traditional professional serve seem to be mostly for fine motor control to get the desired contact.

 

[bigservesofthands]

Tennis Warehouse / Learning Center has some very interesting footage on balls hitting strings, etc. But the best things are hard to find.

http://twu.tennis-warehouse.com/learning_center/spinpatterns.php

Rod Cross has some of the more detailed work in this area. The article discusses the effects of different string patterns on the ball.

A tennis ball contact with the string bed leaves a considerable skid mark in my experience if you were to trace all points of contact of the ball on the string bed. Make interesting arcs depending on the types of serves being attempted (flat, kick, twist, slice ...)

Here is a very simple contact by a robot arm with a very steady racket at an angle. Even here you can see the interesting skid of the ball on the string bed.

This is the source video shot on a Phantom Falcon.

For me the ball travels most of the width of the bed in a interesting arc for speciality shots like top spin lobs. Some of the Po*Mo techniques make use of this effect.

 

[bigservesofthands]

Very good.

Agree with much of this but I don't believe that there is actually much sliding of the ball on the strings for the 4 milliseconds (give or take) of contact. Most of that observed turning of the ball is happening off the stringbed (after contact) -- that is, much of that sliding might be an illusion.

I'm fairly certain that John Yandel has studied hours of a super slow motion HD video and has indicated that the slide or roll of the ball on the strings is extremely slight. (Not his exact words tho). The ball compresses very significantly and is pocketed by the string bed.

Very high speed film might reveal that that ball might turn slightly for the 4 milliseconds it is on the strings. If I did the math correctly, for a ball spinning at 2500 RPM, the ball turns 1/6 of a revolution. But even this much turning is difficult to see since the ball is so distorted while it is on the strings.

Here is a video of a serve recorded at 6000 FPS. Not detecting much, if any, sliding.

Granted, this is a 142 mph serve that might have somewhat less spin than the Murray serve. Nonetheless, we do observe spin on the ball after it has left the strings for this flat-ish serve.

This is the source video from Peter McCraw's study. Shot on an Olympus i-Speed 3. Part of a string bed study who results are proprietary to those companies that commissioned the study. Interesting information though even in the little that was released. We have seen a lot happen with racquet drill patterns, strings and other aspects of equipment since this video was shot.

I

I remember reading then that the intent from the pro hired in this study for this shot was to go all out and try to hit the dead spot flat on. Most tennis serves are eccentric contact.

 

[Chas Tennis]

This is a Borna Coric serve recorded by a Phantom Falcon slow motion camera at Delray Beach. Here is the source video

I think for most high end servers with a traditional serving technique and decent shoulder flexibility there is not really much pronation. The entire arm moves as a unit at the shoulder during the serve. I have trained myself to get a little extra forearm load past the maximum ESR shoulder load and then time forearm pronation release a little after the shoulder release. However I do not notice much of this attempt in professional servers. Wrist movements during a traditional professional serve seem to be mostly for fine motor control to get the desired contact.

Nice video. It shows the ISR by the elbow shadows.
________________________________________________________________________________________________________________________________
As far as timing "forearm pronation release a little after the shoulder release" - if by "release" you mean start - that is interesting. There is one issue with that. If the shoulder can cause enough rotational acceleration with ISR then forearm pronation may not have enough additional torque to cause any more acceleration than ISR. Or you might be causing stretching of your pronation muscles and they could shorten when and if the ISR forces reduce before impact.

Little Acceleration on Big Acceleration Example. Think of a weight lifter with a barbell in a rocket ship. The weight lifter can press 100 lbs normally. But while the rocket is accelerating up then more force is required to lift the barbell off his chest. 100 lbs is adequate to lift the barbell in 1 g. [g = acceleration of earth's gravity] But when the rocket is accelerating up at 3g then 300 lbs will be necessary to get the barbell off his chest. (If you recall my post on the backhand, that is what I think may be happening with the uppermost body turn vs the shoulder joint horizontal abduction.)

If you have a luggage scale you can measure the static strength of the ISR muscles vs the forearm pronator muscles.
_____________________________________________________________________________________________________________________________________

I can't tell if the tennis racket rose enough for a kick serve during impact or the racket head was just getting closer to the camera. ? It's an unfamiliar camera view. The arm appears vertcial from that view and that is usually seen for a kick serve?

If anyone wants to point out 'snap' on the serve, they should mention the time (as 'at 134 seconds plus 12 frame advances shows the start of snap........')

 

[5263]

There is no wrist snap in a traditional serve. It is arm extension, internal shoulder rotation and a little bit of wrist extension. The snap occurs at the shoulder which is powerful enough to cause the bicep to wiggle! PoMo, PoPoMo and PoPoPoMo are more advanced than this and result in more than a bicep wiggle. Also notice how much the ball slides on the strings.

Clearly NOT a kick serve, but
great video of what I've been sharing for years on here related to how the angle at the hand or wrist is used to create more force from the ISR and arm rotation. Also notice how the racket is still leading with the edge to some degree at contact and continues to rotate thru during the contact.

 

[Chas Tennis]

Clearly NOT a kick serve, but
.............................................................

I can't tell. I see the racket tip appear to rise after impact (see frame vs top of camera frame) as a kick serve does. But did it rise or was the tip of the racket just getting closer to the camera and it appeared to rise. I usually can tell looking at the serve from behind.

What are you looking at to identify the serve?

 

[5263]

I can't tell. I see the racket tip appear to rise after impact (see frame vs top of camera frame) as a kick serve does. But did it rise or was the tip of the racket just getting closer to the camera and it appeared to rise. I usually can tell looking at the serve from behind.

What are you looking at to identify the serve?

the spin is only a few degrees above pure sidespin....kick serve should give closer to pure topspin....This looks like a 2nd serve power slice to me although it could be improved in areas....but I do love the focus here of the racket orientation coming to contact.

 

[Chas Tennis]

the spin is only a few degrees above pure sidespin....kick serve should give closer to pure topspin....This looks like a 2nd serve power slice to me although it could be improved in areas....but I do love the focus here of the racket orientation coming to contact.

The kick serve has more side spin than top spin, according to a publication that I have posted. It also has more sidespin than a slice serve because the spin rate of a kicj serve is much higher. Sounds screwy but that is what the publication shows.

(PDF) DIRECTION OF SPIN AXIS AND SPIN RATE OF THE BALL IN TENNIS SERVICE

PDF | On Dec 31, 2007, Shinji Sakurai and others published DIRECTION OF SPIN AXIS AND SPIN RATE OF THE BALL IN TENNIS SERVICE | Find, read and cite all the research you need on ResearchGate

www.researchgate.net

The arrow length is proportional to spin rate. The height of the vertical component of spin, AVz, is proportional to the 'side spin' component, The length of the horizontal component is proportional to the 'top spin' component, AVy. The length in the forward direction, AVx, is the component of gyro spin. The lengths of those lines show total spin, top spin component, side spin component and forward spin component. You can measure the lengths of these lines.


See thread Junior Twist Serve for a lot of information for video observations of the kick serve.

 

[5263]

The kick serve has more side spin than top spin, according to a publication that I have posted. It also has more sidespin than a slice serve because the spin rate of a kicj serve is much higher. Sounds screwy but that is what the publication shows.

(PDF) DIRECTION OF SPIN AXIS AND SPIN RATE OF THE BALL IN TENNIS SERVICE

PDF | On Dec 31, 2007, Shinji Sakurai and others published DIRECTION OF SPIN AXIS AND SPIN RATE OF THE BALL IN TENNIS SERVICE | Find, read and cite all the research you need on ResearchGate

www.researchgate.net

The arrow length is proportional to spin rate. The height of the vertical component of spin, AVz, is proportional to the 'side spin' component, The length of the horizontal component is proportional to the 'top spin' component, AVy. The length in the forward direction, AVx, is the component of gyro spin. The lengths of those lines show total spin, top spin component, side spin component and forward spin component. You can measure the lengths of these lines.


See thread Junior Twist Serve for a lot of information for video observations of the kick serve.

Interesting to see, but not what I've ever seen taught or taught, and doesn't make sense to me. Either way, that spin was closer to the lowest line on your pic that wasn't labeled....or maybe that was the lowest they claimed for the kick serve range... who knows

Honestly, Imo that chart is pretty useless showing flat serves as topspin along with the slice...I guess all with overlapping spin angles and determined more by spin rates than the axis....

 

[Chas Tennis]

Interesting to see, but not what I've ever seen taught or taught, and doesn't make sense to me. Either way, that spin was closer to the lowest line on your pic that wasn't labeled....or maybe that was the lowest they claimed for the kick serve range... who knows

Honestly, Imo that chart is pretty useless showing flat serves as topspin along with the slice...I guess all with overlapping spin angles and determined more by spin rates than the axis....

That display of the spin axes was a revelation to me. But when I considered why I believed what I believed, my old evidence, I changed my way of thinking. There are some researchers that have done a very good job of measuring spins and presenting the results. So much of the other information available is in words. I didn't think about it and believed that spins were made up somehow only of top spin and side spin - as if the 3D world would be described those 2 spins but not need gyro spin. It also appears consistent with high speed videos of the racket head path before, during and after impact.

The spin axes in the drawing are not familiar without some experience with vector analysis. There is one spin axis and to describe its direction you use 3 components. The drawing shows those components. Calling the vertical axis 'side spin' and the horizontal component to the side 'top spin' may confuse the issue for some people. Rod Cross has an article called The Physics of the Kick Serve. It discusses spins and agrees with the publication and ball diagram as far as I see.

 

[5263]

That display of the spin axes was a revelation to me. But when I considered why I believed what I believed, my old evidence, I changed my way of thinking. There are some researchers that have done a very good job of measuring spins and presenting the results. So much of the other information available is in words. I didn't think about it and believed that spins were made up somehow only of top spin and side spin - as if the 3D world would be described those 2 spins but not need gyro spin. It also appears consistent with high speed videos of the racket head path before, during and after impact.

The spin axes in the drawing are not familiar without some vector analysis. There is one spin axis and to describe it you use components. The drawing show those components. Calling the vertical axis the side spin and the horizontal component to the side top spin may confuse the issue for some people. Rod Cross has an article called The Physics of the Kick Serve. It discusses spins and agrees with the publication as far as I see.

well maybe if you can explain that chart better, but Imo it is about varying the plane of the spin (or axis if you prefer) ....I really don't care what people call them as I pretty much just use variations on the top slice with adjustable diagonal plane of serve rotation. I don't teach pure top or pure side spin for any serve, but I teach how to use every part between the 2.

 

[Chas Tennis]

well maybe if you can explain that chart better, but Imo it is about varying the plane of the spin (or axis if you prefer) ....I really don't care what people call them as I pretty much just use variations on the top slice with adjustable diagonal plane of serve rotation. I don't teach pure top or pure side spin for any serve, but I teach how to use every part between the 2.

The need to describe the direction of a line in 3D space was solved centuries ago with vectors and 3 components for those vectors. That is what the ball diagram uses. Describing the direction of a plane was solved by using a line perpendicular to the plane and then describing the direction of that line as you would a vector. Search: vectors

I have learned some things about tennis strokes by no longer thinking of things rotating 'in planes' but by identifying the rotation axes being used in tennis strokes. Mostly I look at 1) the core rotation axis (spine) and 2) the rotation axis of the shoulder joint. I notice when the spine axis is in use and when the shoulder axis is in use. If I view the rotations as occurring in planes, I would not be able to analyze the strokes as well.

 

[5263]

The need to describe the direction of a line in 3D space was solved centuries ago with vectors and 3 components for those vectors. That is what the ball diagram uses. Describing the direction of a plane was solved by using a line perpendicular to the plane and then describing the direction of that line as you would a vector. Search: vectors

I have learned some things about tennis strokes by no longer thinking of things rotating 'in planes' but by identifying the rotation axes being used in tennis strokes. Mostly I look at 1) the core rotation axis (spine) and 2) the rotation axis of the shoulder joint. I notice when the spine axis is in use and when the shoulder axis is in use. If I view the rotations as occurring in planes, I would not be able to analyze the strokes as well.

but I'm not talking about strokes. I'm talking about the simple plane of rotation for the ball....or you can use the spin axis...whichever you prefer, since they are just 90 degrees out....

 

[Chas Tennis]

but I'm not talking about strokes. I'm talking about the simple plane of rotation for the ball....or you can use the spin axis...whichever you prefer, since they are just 90 degrees out....

For strokes, here is an important reason to use the axis instead of the 'plane of rotation'. (This is much easiler to understand while viewing a stroke video.)

For rotations, the speed of a point rotating around an axis depends on the distance between the rotation axis and the point. The racket head speed depends on the radius out from the rotation axis (including the axis extended). For ground strokes, the racket head is located at a radius out from the rotation axis of the core or a rotation axis through the shoulder joint. An axis through the shoulder joint for a one hand shot is located closer to the racket head so the speed from shoulder joint rotation would be less for the same angular rotation rate. You can check this speed effect by rotating your uppermost body (core axis) and then rotating only the shoulder joint.

For the ball, the same principle is very important especially for explaining the kick serve bounce. Consider the ball and its axis of rotation. Now drill a hole through both sides of the ball and push a pencil through the ball. Spin it and lower it to the floor as it would do in a kick serve bounce. The reason that it bounces to the side becomes apparent, the axis is tilted and the felt is moving fast and to the side at the point of first contact. If you try to do this with a 'plane of rotation' concept instead of an axis description, I don's see how it works. Is there a definition of 'the plane of rotation'?

 

[5263]

If you try to do this with a 'plane of rotation' concept instead of an axis description, I don's see how it works. Is there a definition of 'the plane of rotation'?

Ok, that is helpful to me to hear what others can't see since it is obvious to me both ways. I don't have a preference and can see it either ways.....so whatever is better for most people is fine. Now just to see if most people agree with you that they think of spin relative to the axis or the spin direction ....

 

[Chas Tennis]

Here is a definition of my use of the term Spin Vector. This also is the definition used to produce the diagram of the ball and its spin vectors. Google: Spin Vector

Spin Vector Definition & Meaning | YourDictionary

Spin Vector definition: A vector quantity used to represent the spin angular momentum of a particle. Its magnitude represents the magnitude of the spin, while its direction lies along the axis of rotation of the particle, following the right-hand rule.

www.yourdictionary.com

Two things are described:
1) Spin - revolutions per second, AKA 'magnitude'
2) Spin axis direction - specified as components in 3 orthogonal directions such as x, y and z or vertical, horizontal and forward on a tennis court or vertical, along the ball's trajectory, and right-left of the ball's trajectory, etc.

I believe that the directions of the components used in the ball diagram were defined in the publication. I think that they used the court lines and vertical as a coordinate system. ....

 

[Chas Tennis]

Ok, that is helpful to me to hear what others can't see since it is obvious to me both ways. I don't have a preference and can see it either ways.....so whatever is better for most people is fine. Now just to see if most people agree with you that they think of spin relative to the axis or the spin direction ....

What is the "spin direction"?