OOAK Table Tennis Forum


A truly International Table Tennis Community for both Defensive and Offensive styles!
OOAK Forum Links About OOAK Table Tennis Forum OOAK Forum Memory
It is currently 29 Mar 2024, 19:43


Don't want to see any advertising? Become a member and login, and you'll never see an ad again!



All times are UTC + 9:30 hours




Post new topic Reply to topic  [ 28 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: Robot suggestion please
PostPosted: 17 Sep 2014, 15:46 
Offline
New Member

Joined: 17 Sep 2014, 15:37
Posts: 3
Has thanked: 0 time
Been thanked: 0 time
I am looking to purchase a TT robot. Any suggestion between Butterfly Amicus pro and Robo Pong 2050? Pros and Cons.

Thank you,


Top
 Profile  
 


PostPosted: 18 Sep 2014, 07:21 
Offline
Super User

Joined: 11 Mar 2013, 21:12
Posts: 849
Has thanked: 5 times
Been thanked: 40 times
oat wrote:
I am looking to purchase a TT robot. Any suggestion between Butterfly Amicus pro and Robo Pong 2050? Pros and Cons.

Thank you,


If you're looking for value, Y&T are IMO the best choice if you can get one w/o unreasonable markup. They're probably the most popular brand in china. Their 989 or similar line has two wheels (spin independent from speed) and great build quality w/ metal parts for <<$1k usd. I did a significant amount of research when buying mine and got the A9 model which is the table-mount 989e and one of the better choices for home use due to compact size and ease of use. Unfortunately sometimes importers charger >$1k.


Top
 Profile  
 
PostPosted: 18 Sep 2014, 07:45 
Offline
LP Collector
LP Collector
User avatar

Joined: 01 Aug 2012, 06:57
Posts: 2289
Location: Hampshire, UK
Has thanked: 303 times
Been thanked: 333 times
Blade: Yaska Sweden Classic
FH: 802 OX
BH: DHS C8 OX
We purchased the Newgy RoboPong 2050. It was reasonably priced, and works well. I like the integrated ball recycling feature, and it comes with a number of useful pre-programmed drills, as well as the ability to program your own.

I've found it reliable so far. The only real drawback is the lack of an independent speed/spin capability, as it only has one wheel.

_________________
Yasaka Sweden Classic | 802 OX | C8 OX
Check out my blog - LordCope's Latest Learning Log - 10+ years of accumulate mistakes!


Top
 Profile  
 
PostPosted: 18 Sep 2014, 08:05 
Offline
Super User

Joined: 11 Mar 2013, 21:12
Posts: 849
Has thanked: 5 times
Been thanked: 40 times
Other than the one-wheel limitation, I found the Newgy's felt kinda cheap. A lot of give to the parts which make the club level y&t's feel like a mercedes in comparison.

The difference also extends to output quality. I haven't used the 2050 but for example the lower models use pretty simple mechanical oscillation, whereas the y&t has stepper motor with angle sensor so all head movements are exceedingly smooth.


Top
 Profile  
 
PostPosted: 18 Sep 2014, 10:29 
Offline
Senior member

Joined: 20 Dec 2012, 07:57
Posts: 192
Location: Tennessee, USA
Has thanked: 0 time
Been thanked: 30 times
oat wrote:
I am looking to purchase a TT robot. Any suggestion between Butterfly Amicus pro and Robo Pong 2050? Pros and Cons.

I'll take a shot at this. Please keep in mind that I work for Newgy, so of course I am biased. Also I have not used the Amicus Pro yet, but have used the Amicus Basic.

# of wheels--Newgy uses 1 rubber wheel. 1-wheeled robots apply spin similar to a human (1 contact point with ball)--easy to identify what spin is being applied to the ball and motor speed directly translates into amount of spin. This permits the user to determine spin type and amount by looking at robot head and listening to the pitch of the motor. This is visually like watching a player to see what surface of the ball is contacted with the paddle and listening for the contact sound to determine spin amount. The Amicus has 3 foam wheels, and there are no indicators for the user to determine which wheel is spinning, and if more than one wheel is spinning (and there has to be more than one wheel spinning for any spin other than pure underspin), there's also no way to use motor pitch as a gauge to amount of spin.

The 1-wheel design is more limited in its spin/speed range as any increase in motor speed results in a corresponding increase in spin. Therefore it cannot deliver a no-spin shot or high spin/slow speed shots like a slow loop or a short, heavy sidespin/topspin serve that bounces twice on the player's side.

The 3-wheel design does allow for it to deliver no-spin shots. And depending upon whether or not it allows any of its wheels to spin backwards, it may also allow for a much larger spectrum in speed/spin. I've seen posts here that say at least some of the motors do indeed spin backwards and others that say none of them do. If the controls allow for backward spinning, then the full speed/spin spectrum should be available, including slow loops and short topspin/sidespin serves. But if all motors only spin forward then it too will not be able to able to deliver slow loops or short, heavy topspin/sidespin serves.

It will still have an advantage of offering no-spin and probably lighter spin shots, but also at the expense of adding more speed to the ball, so a slow no-spin shot (like a soft anti block) is probably not possible because all 3 wheels would have to be spinning at the exact same rate, and those 3 wheels are all putting speed on the ball, representing more speed than would typically be considered slow. Perhaps a Amicus Pro user can speak up about this so we can all get clarity on this point.

I've never found the 1-wheel design to be all that limiting as I can practice against a wide range of underspin and topspin shots and I'm skillful enough to be able to adjust my racket angle/stroke angle accordingly. A find it easy to adjust to no-spin balls in actual matches as it is simply a matter of adjusting the angle in between what I would for topspin or underspin. And at least for most of my real matches, I don't encounter dead balls all that often except on serves, so it's not something I would practice against all that often anyways. I can understand that if dead balls are a problem area for a player, they would be attracted to a robot capable of producing no-spin shots, but that personally is not the case for me.

It would be nice to start a drill with an underspin serve and then switch to several topspin rallying shots, but again, I don't find that particularly hard to do in a real match as I can normally read the spin and adjust my angles accordingly. Where I usually go wrong in a match is in reading the spin on the ball and therefore I select the wrong stroke to handle that spin. And all robots, regardless of their design, are not well suited for helping the user read disguised spins--that's something you really need to learn by playing a lot of different players, not on a robot.

The 3-wheel design presents 3 times the number of motors and spinning wheels to maintain and troubleshoot if something goes wrong. With previous iterations of the Amicus, the foam wheels weren't very durable. I do think they've made some improvements to them, and I guess the jury is still out as to their durability simply because there are so few units out there so far and there hasn't been enough time to properly judge their durability. The 2050 has now been available for almost 5 years, is widely distributed, and has a very good durability reputation. Parts are readily obtainable and service centers are knowledgable and well-stocked.

I personally don't care for the way the Amicus determines side-to-side location and up/down trajectory. A metal deflector plate with thin edges that moves up and down and side to side sits directly in front of the ball ejection tube. Balls come out of the tube, are grabbed by the 3 wheels, and then thrown into the deflector plate, which determines where the ball is thrown on the table and how high. Furthermore, the plate edges are hard to see and therefore I find it difficult to pick up where the ball will go and how high it's going to be.

By contrast, the entire head of Robo-Pong moves side to side and up and down, so that movement is easy to see. It can still change directions faster than I can react to it, so it's not like I can easily get into position if I have the frequency set at a match like pace. It would be nice to be able to change the up/down trajectory during a drill like the Amicus can, but I don't find that to be a deal-breaker.

I think the Newgy digital interface is far easier to use than the multiple buttons, dials, lights, and sliders that the Amicus uses.

The Robo-Pong control box (CB) is similar to the typical pre-smartphone cell phone interface with buttons for up & down, left & right, select/main menu, on/off, start/stop, and special function. Functions are labeled in plain English (or you can switch to one of 5 other languages) in a relatively large, easy to see font.

Amicus pro has 30 buttons, 1 dial, 6 indicator bars, 20 indicator lights, and a 2-digit LED display on a control panel that is roughly 3 times the size of the Newgy CB. It's a lot of information being presented all at once to the user and many of the labels are pretty small and use somewhat esoteric 2 or 3 letter abbreviations (RDM, rdm, Pr). Newgy also uses some abbreviations but we have a bit more space to explain what the abbreviation means (like OSC CALIB) and they're used mainly for calibration functions that are typically set and then not used again for a long time, not functions that are used often during play.

For Drills, Newgy has these separated from our manual controls into a separate menu. Once you select DRILLS, then you merely have to select a drill number, verify that the head angle and Spin are set as described on the CB, then press the Start button. If the drill is too fast or too slow, you can use the WAIT ADJUST function to speed up or slow down the pace to suit your tastes. Same for the ball speed, you can use SPEED ADJUST to increase/decrease ball speed for the drill on an individual basis. You can also preview the ball locations during the drill by pressing the Special Function button and being presented with a table diagram and a series of dots representing the landing spots in that drill. Or simply consult the Drill Diagram pages in the Owner's Manual for the same information.

The CB comes pre-programmed with 64 drills and covers a wide range of drills for all levels of players (beginner, intermediate, advanced, expert) and many different skills to practice (FH/BH transition, front/back footwork, side-to-side footwork, serve return, random placements, etc.). And I think our random controls are better than the Amicus. There are drills where a random number of shots are delivered to one location before switching to another location and throwing a single ball, or throwing one ball to a set location and throwing the next randomly to one of two locations. I don't think Amicus can do either one of these types of drills.

And when it comes to customizing drills, hook your CB up to a PC and Newgy supplies you with free software to write your drills as an easily understandable sequence of steps like SPEED 15>POSITION 10-18 (robot randomly selects position from 10 to 18)>THROW>WAIT 1 (second)>SPEED 12,15 (robot randomly selects speed 12 or speed 15) >POSITION 3>THROW>WAIT 1. And the number of drills you have access to is limited only by the memory on your computer (drill files are tiny). You can send drill files to friends and they can store them on their PC and run them directly from that PC hooked up to their own CB or transfer the drill to their own Robo-Pong CB so the PC isn't needed to run that drill. You can also give that drill file a unique name to help you remember exactly what that drill does (especially handy when you have dozens of drill files). Newgy also makes available a blank Drill Diagram page for you to diagram your custom drills to reference when you trying to figure out which drill to choose when you're ready to play.

With Amicus you have to program each step by manipulating the array of buttons, sliders, and knobs for each ball, then save that sequence of balls as a number. You can save up to 99 such drills, but you only have a number to identify the drill--no descriptive name, no way of transferring those settings to another Amicus other than writing down each step and then manually programming the other Amicus with the exact same settings. And as far as I know, Butterfly does not supply any pre-programmed drills (what do you do if you've never done drills before and have no idea where to start?). They also don't supply a blank drill diagram for you to fill out to help you remember what drill each number is associated with and what happens in that drill.

And of course there is the price. Here in the states, the 2050 is $699 and the Amicus is $2000. You have to decide on the value proposition each offers, but IMO the 2050 is far and away the better value (more bang for the buck). And I just think you'll find it a much easier robot to use and adjust to your liking. The Amicus does seem like a pretty good robot, and it does a few things the Newgy can't, but I don't see those things as being essential for the vast majority of the things that I want to practice.

There's other things, but I need to stop this missive and go practice some on my Robo-Pong 2050!

Larry


Top
 Profile  
 
PostPosted: 18 Sep 2014, 12:47 
Offline
Super User

Joined: 11 Mar 2013, 21:12
Posts: 849
Has thanked: 5 times
Been thanked: 40 times
It's a disservice to potential buyers to gloss over the difference in build quality for just the feature points. The home machines (Newgy/Supermaster) and club level ones (Y&T/Oukei/Btfly) aren't in the same class/category at all, and those shopping for one aren't going to be happy with the other if they were able to compare them side-by-side.

The second wheel is probably the most important feature for a robot behind re-cycling balls. The problem with one wheel is that spin scales with speed so it's impossible to practice against large categories of shots (slow spinny, or fast hits); anything beyond a slow speed is far more loaded than <2k players face regularly. Frankly it's disappointing that Newgy never bothered to upgrade their machines mechanically over the years when all the competition have.

All the other design diffs are pretty minor. For example Oukei uses foamy wheels and a deflector vs. the direct rubber wheels for Y&T but either work fine.

IMO the best value for robots in the US if buying retail under 1k is probably the AW32 at paddle palace, which is a Y&T A8 with simplified controls, or the Oukei 2700 (Oukei is slightly lesser build, but has a nice knob interface). The simple way to put this is they weigh ~twice as much as the home machines, and it's not because those companies can't figure out how to make them lighter. For a cheaper option, the new Y&T B3 looks the one to get.


Top
 Profile  
 
PostPosted: 18 Sep 2014, 14:00 
Offline
Senior member

Joined: 20 Dec 2012, 07:57
Posts: 192
Location: Tennessee, USA
Has thanked: 0 time
Been thanked: 30 times
agenthex wrote:
The second wheel is probably the most important feature for a robot behind re-cycling balls. The problem with one wheel is that spin scales with speed so it's impossible to practice against large categories of shots (slow spinny, or fast hits)

Partly untrue--2 wheel robots don't necessarily equate with the ability to produce slow, spinny shots. To get slow and spinny, you must be able to reverse the wheels. All Y&T, Oukei, SuperMaster, Killerspin, Paddle Palace robots have wheels that can only spin forward. That means, like the 1-wheel robots, that each wheel spinning by itself produces both speed and spin. When the opposing wheel also spins, it reduces spin and increases speed . So the vast majority of 2-wheel robots cannot make slow, spinny shots like a slow loop, heavy underspin or topspin short serves, or high, heavy underspin floaty chops. So most 2-wheel robots are no better at producing such shots than 1-wheel robots.

For slow, spinny shots, the wheels have to spin backward. Only when one of the wheels spins backward and the other forward, can you then achieve true separation of speed and spin. A wheel that spins backwards reduces speed and increases spin .

agenthex wrote:
anything beyond a slow speed is far more loaded than <2k players face regularly.

I disagree. I play with players rated from 1600 to 2100. Several of the 1700-1800 players have loops as strong as, if not stronger than the Newgy robot can deliver. And when I loop at the 1600 player's long pips, it comes back more loaded with backspin than anything I've practiced against on our robots. I've also seen many sub 2000 players juice up their serves to ridiculous levels. When I play on the Y&T, Killerspin, Paddle Palace, Oukei, and even the Amicus Basic, I always feel that the spin amount is underwhelming compared to the Newgy. Just two different points of view, I guess.

Larry


Last edited by larrythoman on 19 Sep 2014, 02:16, edited 1 time in total.

Top
 Profile  
 
PostPosted: 18 Sep 2014, 14:45 
Offline
Super User

Joined: 11 Mar 2013, 21:12
Posts: 849
Has thanked: 5 times
Been thanked: 40 times
I'm not sure how involved you are in the design of these machines, but with one wheel there's a fixed ratio of speed to spin as dictated by the friction cooefficient of the ball's path through the machine. If this is too biased towards speed, there's inadequate spin for faster shots, and so on.

With two wheels the variables are controlled independently so a bias towards spin by default can be negated by the additional wheel. It's also a fact that club level machines have more powerful motors, however excessive that may be given the higher settings are rarely used even by pretty good players.

We can take a guess whether all manufacturers of higher end robots can't figure out that one wheel is adequate and in their incompetence need two to produce what are apparently worse results, or whether the one company which only makes single-wheelers needs to make their tradeoffs seem an advantage.


Quote:
Several of the 1700-1800 players have loops as strong as, if not stronger than the Newgy robot can deliver.


I haven't used a Newgy in a while, but I did borrow a Supermaster T288 (Newgy equivalent w/ second wheel) for a bit before I got the Y&T. I've certainly played 2k level players and their loopkills are at best similar to the 7 or so setting out of 10; these are shots not meant to be returned. If anything, there wasn't enough fine grain control between 3-5. Perhaps the Newgy uses slower settings?


Top
 Profile  
 
PostPosted: 19 Sep 2014, 00:23 
Offline
Senior member

Joined: 20 Dec 2012, 07:57
Posts: 192
Location: Tennessee, USA
Has thanked: 0 time
Been thanked: 30 times
agenthex wrote:
I'm not sure how involved you are in the design of these machines

I am director of R&D for Newgy. I'm intimately familiar with every aspect of Newgy's (and many other robots) design. Since we're asking, what are your qualifications for analyzing TT robot design?

agenthex wrote:
with one wheel there's a fixed ratio of speed to spin as dictated by the friction cooefficient of the ball's path through the machine. If this is too biased towards speed, there's inadequate spin for faster shots, and so on.

I'm sorry, this makes no sense at all to me. Can you explain further?

agenthex wrote:
With two wheels the variables are controlled independently so a bias towards spin by default can be negated by the additional wheel.

Exactly how are the variables controlled independently? Please explain how the additional wheel can overcome the other wheel's bias towards spin.

agenthex wrote:
all manufacturers of higher end robots can't figure out that one wheel is adequate and in their incompetence need two to produce what are apparently worse results, or whether the one company which only makes single-wheelers needs to make their tradeoffs seem an advantage.

You said this, not me. I've played with almost every robot out there. I'm always analyzing our competitors' designs. We all have borrowed design ideas from each other. And all I can tell you is that as a mid-level player (2100) and someone with 40+ years of training and coaching experience, I find the Newgy 2050 to be the robot that is easiest to use and get it to quickly do what I want it to do so I spend more time training and less time fiddling with controls. Sure part of this is simple familiarity with our design, but there's no denying that virtually every time I play on one of these two wheel robots that spin the wheel only forward that I feel underwhelmed by their capabilities and gladly return to my familiar 2050 to train the way I like to train.

Sure, 1-wheel robots have their limitations, but so do 2-wheel robots, especially if they only spin their wheels forward. In essence, forward spinning 2-wheeled robots can only reduce the amount of spin on the ball and add speed as compared to 1-wheeled robots. The only other advantage of that design, as best I can tell, and which only a few of those type robots can do, is to give topspin and then underspin on alternate shots.

For me, the amount of speed and spin coming from the Newgy seems more familiar to me than the speed/spin coming from the forward spinning 2-wheeled robots when comparing it to the speed/spin I get from the players I compete and train with. I fully understand that if you have to compete with a lot of dead ballers, you would be attracted to a forward spinning two-wheeled robot as that better matches the type of balls you typically receive in actual matches. But if that is not the case, or you want to learn how to deal with the type of speed and spin typical at the mid to upper levels of our sport, I maintain that a 1-wheeled Robo-Pong 2050 is a better match than the vast majority of forward spinning 2-wheel robots available today.

And to get back on topic, I would even make that claim comparing the 2050 to the Amicus Basic, for which I had that same underwhelming feeling when assessing its speed/spin capabilities. The Amicus Pro sounds to me like a much more capable robot compared to the Basic. I'd sure like to get positive confirmation as to whether or not it spins any of its wheels backward. Then that would be a robot that could offer true speed/spin separation. If it doesn't, then it too will not be able to produce shots in the high spin/low speed part of the spin/speed spectrum, like a slow loop or short heavy topspin serve.

Thanks.

Larry


Top
 Profile  
 
PostPosted: 19 Sep 2014, 00:59 
Offline
Super User

Joined: 05 Nov 2012, 02:58
Posts: 596
Has thanked: 51 times
Been thanked: 73 times
A single-wheel is also missing high speed with low or moderate spin, right? Need that to simulate play against pips or simply to provide the variety that occurs in play against a human.

I couldn't accept a linkage between spin and speed because I would become used to it; no human opponent plays like that.


Top
 Profile  
 
PostPosted: 19 Sep 2014, 02:01 
Offline
Senior member

Joined: 20 Dec 2012, 07:57
Posts: 192
Location: Tennessee, USA
Has thanked: 0 time
Been thanked: 30 times
Zhaoyang wrote:
A single-wheel is also missing high speed with low or moderate spin, right? Need that to simulate play against pips or simply to provide the variety that occurs in play against a human.

Yes, that is correct. With 1-wheel robots, high speed = high spin. And yes, there are plenty of shots in a real match that can have higher speeds with low to moderate spin, as there are other shots in the entire speed/spin spectrum. Forward-spinning 2-wheel robots do excel at producing higher speed/lower spin shots. And if dealing with that type of shot is important to you, than a forward spinning 2 wheel robot is an ideal match for you. But in my neck of the woods, those type shots are the exception rather than the rule (and there seem to be fewer and fewer flat hitters every year).

When I train on my 2050, I train the skill of adjusting body position, stroke angle and paddle angle in response to changes in speed and spin on the ball by using a large SPEED RANDOM setting (Drill #8 does this exact thing). While still restricted to low speed/low spin, moderate speed/moderate spin, and high speed/high spin, this forces me move my body position forward/backward (and side to side) while on the fly adjusting my paddle and stroke angles. While this practice indeed misses a high speed/low spin ball (or for that matter a high spin/low speed ball), it incorporates enough changes that I find it relatively easy to handle the missing parts of that spectrum in an actual match as long as I can read the speed/spin correctly in the first place. Where I feel the most pressure in a match is facing a good two-winged looper using high speed/high spin that puts tremendous time pressure on me and the Newgy 2050 does an excellent job at giving me lots of practice against that type of ball.

Thanks.

Larry


Top
 Profile  
 
PostPosted: 19 Sep 2014, 03:52 
Offline
Super User
User avatar

Joined: 05 Feb 2012, 19:52
Posts: 731
Has thanked: 4 times
Been thanked: 21 times
Blade: TT Piet Homemade 6 ply
FH: Spinlord Waran Short Pips
BH: Joola Badman Reloaded 0X
I have 3 robots here.

I happen to have a Praktismate and a Newgy 1050 and an old 38 mm Newgy 2000. I use the nets from the Practismate to collect the balls and either recycle or reload into it, or feed coffee cans with 100 balls to reload the Newgy 1050. I use the 100 balls and that allows me to count my misses to get a percentage of good shots. Its no big deal to reload.

Anyway, if you want all topspin or all backspin and don't need 1800 looper level spin, the Newgy works pretty well, but if you want more than that, you need the multiple wheels and ability to reverse to get oodles of speed, spin, or less of either.

Whats good about the Newgy is being able to easily program balls on the display under normal, or on the computer when you want to get more complex, and the ability to take the Newgy predefined programs and copy and modify them or create your own and save them. Its limited to about 25 balls in per sequence, if I remember correctly, but that should be plenty.

None of the others had a computer hookup to be able to easily program the balls. The Praktismate throws excellent balls, but its all analog dials to tell it what to do, and it can only throw 1 ball over and over and oscillate it across the table.

So most of the time, to practice I use the Newgy to learn new shots if I can. But if I want to practice a nospin ball, or a spinny loop, or a heavy chop, I use the Practismate because the balls can be made more realistic.

I don't think Newgy has kept up with the competition lately at the higher levels. My suggestion to Newgy would be as follows:

1. Come up with a 3 wheel, reversing motor head design, or a 2 wheel reversing motor head design with a 3rd stepper motor to rotate for sidespin, that could bolt on in place of the current 1 wheel design that uses the same or similar reliable components

2. Add a sidespin + or - value to the programming language, and allow for - spin, that being backspin

3. If possibel, add a decimal point for speed, spin and location values to the programming language for finer control and more but smaller variations in random values

4. Have the programming language interface take the raw settings requested and translate them to speeds for the motors for each shot

IMO, then the Newgy would be what I'd really like to have on my table, and would be just as good or better than any of these fancy ones that lack the computer controls, and it could be done pretty reasonable.

Hire me to test it, convert the existing programs, and demo it/user test it at the 5 clubs I play at regularly.

_________________
--- Push/Chop or Attack ---
USATT 1684 left, Lost it all by playing sick and hurt.
No legs left, no more backing up, just smash it!
Blade: TTPiet OFF-/ALL custom
FH: Spinlord Waran Short Pips
BH: Joola Badman Reloaded 0X


Top
 Profile  
 
PostPosted: 19 Sep 2014, 05:08 
Offline
Senior member

Joined: 20 Dec 2012, 07:57
Posts: 192
Location: Tennessee, USA
Has thanked: 0 time
Been thanked: 30 times
Old-Man-Southpaw wrote:
My suggestion to Newgy would be as follows:

1. Come up with a 3 wheel, reversing motor head design, or a 2 wheel reversing motor head design with a 3rd stepper motor to rotate for sidespin, that could bolt on in place of the current 1 wheel design that uses the same or similar reliable components

2. Add a sidespin + or - value to the programming language, and allow for - spin, that being backspin

3. If possibel, add a decimal point for speed, spin and location values to the programming language for finer control and more but smaller variations in random values

4. Have the programming language interface take the raw settings requested and translate them to speeds for the motors for each shot

IMO, then the Newgy would be what I'd really like to have on my table, and would be just as good or better than any of these fancy ones that lack the computer controls, and it could be done pretty reasonable.


Thank you for those suggestions. I'll keep these in mind as we work to develop new and improved models.

Larry


Top
 Profile  
 
PostPosted: 19 Sep 2014, 07:12 
Offline
Super User
User avatar

Joined: 05 Feb 2012, 19:52
Posts: 731
Has thanked: 4 times
Been thanked: 21 times
Blade: TT Piet Homemade 6 ply
FH: Spinlord Waran Short Pips
BH: Joola Badman Reloaded 0X
PS: I also played an evening with a Paddle Palace S4W two head, four wheel robot, but it wasn't mine to "tweak". It had the advantage of being able to do topspin, underspin or nospin on each ball, but I was completely unimpressed with the programming, and I find easy to get my Newgy to send specific sequences of varying balls to specific varying locations by comparison.

BTW, I do a couple hundred balls each afternoon on the Newgy before I go to the club to try to tweak in the shots I need to work on. When you don't have 24/7 access to a good practice partner, it does help, especially learning new shots.

_________________
--- Push/Chop or Attack ---
USATT 1684 left, Lost it all by playing sick and hurt.
No legs left, no more backing up, just smash it!
Blade: TTPiet OFF-/ALL custom
FH: Spinlord Waran Short Pips
BH: Joola Badman Reloaded 0X


Top
 Profile  
 
PostPosted: 19 Sep 2014, 09:09 
Offline
Super User

Joined: 11 Mar 2013, 21:12
Posts: 849
Has thanked: 5 times
Been thanked: 40 times
larrythoman wrote:
agenthex wrote:
I'm not sure how involved you are in the design of these machines

I am director of R&D for Newgy. I'm intimately familiar with every aspect of Newgy's (and many other robots) design. Since we're asking, what are your qualifications for analyzing TT robot design?

I happen to grasp physical mechanics, which seems key to how TT robots work.

Quote:
agenthex wrote:
with one wheel there's a fixed ratio of speed to spin as dictated by the friction cooefficient of the ball's path through the machine. If this is too biased towards speed, there's inadequate spin for faster shots, and so on.

I'm sorry, this makes no sense at all to me. Can you explain further?

Which part is confusing, what a friction coefficient is or what a ratio means? Or that varying amounts of friction on the ball can change spin more than speed or vice versa?

Quote:
agenthex wrote:
With two wheels the variables are controlled independently so a bias towards spin by default can be negated by the additional wheel.

Exactly how are the variables controlled independently? Please explain how the additional wheel can overcome the other wheel's bias towards spin.

Eg. given a normalized speed/spin ratio of 1/1, the other wheel can reduce the latter number compared to a fixed ratio (eg 0.5/0.5) without. This means a robot can be designed to to throw 1/1.5 and thus provide relative independent control of the two var. Why am I explaining this to someone who supposedly designs these?

Quote:
agenthex wrote:
all manufacturers of higher end robots can't figure out that one wheel is adequate and in their incompetence need two to produce what are apparently worse results, or whether the one company which only makes single-wheelers needs to make their tradeoffs seem an advantage.

You said this, not me. I've played with almost every robot out there. I'm always analyzing our competitors' designs. We all have borrowed design ideas from each other. And all I can tell you is that as a mid-level player (2100) and someone with 40+ years of training and coaching experience, I find the Newgy 2050 to be the robot that is easiest to use and get it to quickly do what I want it to do so I spend more time training and less time fiddling with controls. Sure part of this is simple familiarity with our design, but there's no denying that virtually every time I play on one of these two wheel robots that spin the wheel only forward that I feel underwhelmed by their capabilities and gladly return to my familiar 2050 to train the way I like to train.

Sure, 1-wheel robots have their limitations, but so do 2-wheel robots, especially if they only spin their wheels forward. In essence, forward spinning 2-wheeled robots can only reduce the amount of spin on the ball and add speed as compared to 1-wheeled robots. The only other advantage of that design, as best I can tell, and which only a few of those type robots can do, is to give topspin and then underspin on alternate shots.

The aggro top-top loop rally (one of the few shots that a 1wheel bot can accurately replicate) is only one part of the game, and frankly one of the simplest aspects to learn.

If anything the harder shots for breaking into higher levels is opening against low or backspin which 1wheelers tend to overdo.
Quote:
For me, the amount of speed and spin coming from the Newgy seems more familiar to me than the speed/spin coming from the forward spinning 2-wheeled robots when comparing it to the speed/spin I get from the players I compete and train with. I fully understand that if you have to compete with a lot of dead ballers, you would be attracted to a forward spinning two-wheeled robot as that better matches the type of balls you typically receive in actual matches. But if that is not the case, or you want to learn how to deal with the type of speed and spin typical at the mid to upper levels of our sport, I maintain that a 1-wheeled Robo-Pong 2050 is a better match than the vast majority of forward spinning 2-wheel robots available today.

And to get back on topic, I would even make that claim comparing the 2050 to the Amicus Basic, for which I had that same underwhelming feeling when assessing its speed/spin capabilities. The Amicus Pro sounds to me like a much more capable robot compared to the Basic. I'd sure like to get positive confirmation as to whether or not it spins any of its wheels backward. Then that would be a robot that could offer true speed/spin separation. If it doesn't, then it too will not be able to produce shots in the high spin/low speed part of the spin/speed spectrum, like a slow loop or short heavy topspin serve.

Thanks.

Larry


Ok, so nothing is as great as your Newgy because no other designer understands just how much better of a compromise their machines would be without adding another wheel. Everyone who's ever used all these robots and (often) chose something else sure are dumb.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 28 posts ]  Go to page 1, 2  Next




All times are UTC + 9:30 hours


Who is online

Users browsing this forum: No registered users and 11 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Copyright 2018 OOAK Table Tennis Forum. The information on this site cannot be reused without written permission.

Powered by phpBB® Forum Software © phpBB Group