Skip to main content
  1. Home
  2. Phones
  3. Mobile
  4. Reviews

Hands on: Trainerbot

The world's first smart ping pong robot wants to turn you into a champion

Add as a preferred source on Google
Trainerbot
Ted Kritsonis/Digital Trends

“Why you can trust Digital Trends – We have a 20-year history of testing, reviewing, and rating products, services and apps to help you make a sound buying decision. Find out more about how we test and score products.“

Ever felt like you could use a highly skilled opponent to improve your ping pong game? Trainerbot is “the world’s first smart ping pong robot” that claims to simulate specific shots and games to help players at any level practice and get better.

Recommended Videos

Similar to how a batting cage hurls baseballs out like a pitcher, the Trainerbot sends ping pong balls at you like some sort of robotic Forest Gump. The only difference is that it’s entirely programmable through an accompanying app, letting players pick and choose almost every aspect of the type of game they want. With a Kickstarter crowdfunding campaign now underway, Digital Trends picked up a paddle and got a harsh lesson in how calculated this bot truly is.

Bucket of balls

The advanced prototype we got to try out is essentially going to be the final design, save for some operational hiccups we noticed along the way that will require fixing. It’s a bucket-style design that holds up to 30 balls, with five motors inside that spit them out at varying speeds based on intervals programmed through the app. Of the two versions that will be available, the Basic one sits on the table, while the Pro comes with a stand that clamps onto the end of the table. The latter is the one we got to play with.

Fairly lightweight, and small enough to fit into a backpack or duffle bag, the Trainerbot is definitely portable enough to take to a friend’s home or even to a rec center or place of business with a ping pong table. Both versions are universal that way, as any table should be compatible, and the setup only takes about a minute, assuming a power outlet is nearby.

Trainerbot
Ted Kritsonis/Digital Trends
Ted Kritsonis/Digital Trends

Designed and co-founded by brothers Alexander and Harrison Chen, each of which left Taiwan and studied at different universities in California and Toronto, Trainerbot was born out of frustration. The two inventors came up with the idea because they weren’t able to play each other. Irritated by a lack of suitable and consistent opponents to keep improving, they figured a robot could step in instead.

“If you had a shot that kept beating me all the time, I could program that shot and spin into the app, secretly practice at home, and then counter it the next time we meet,” says Alexander Chen. “Catching a serve is one thing, but if there are only three ways to do so, you can learn how to handle all of them. Or you could set up a serve to practice your own returns.”

App-based training

What’s most interesting about the Trainerbot is how the app customizes play by controlling the five motors inside. Dividing the player’s side of the table into nine sections, it’s possible to program a sequence where the ball will land in the same section each time, in a series of different sections, or even randomly. Topspin, backspin, sidespin, and five other spin types can also be chosen. And then there’s the interval at which the balls can be spit out, going as fast as 0.4 seconds between each one.

It’s a bucket-style design that holds up to 30 balls, with five motors inside that spit them out.

Beyond that is technique. Returning serve, overhand smashes, quick forehand-to-backhand rallies, and much more are within reach. The Chen brothers say the app is also meant to be a repository for members to design their own custom modes, where they could adjust shot frequency (time between each ball), angle, and table section. It’s an interesting way to try replaying a scenario that may have happened in an actual real-life match between pros.

Not to mention an arch-nemesis opponent. Trick shots used by others can be replicated — or at least you can try — through the app to practice how to deal with them. We figure these are the kind of features aspiring and advanced players will really take to, since ping pong is a such a game of inches when played at a high level.

“The cool part in using the robot is that it can be personalized to your level of play,” said Alexander Chen. “Some of the modes in the app are based on YouTube. We’d watch a video on there, see where and how quickly the ball lands, and add it so that we could practice how to better play that shot or sequence.”

Of course, the Trainerbot never told us how well or poorly we hit the ball. For that, we had to use our own eyes. So, in that respect, the unit was simply an automated server that shot the balls over to be hit in any which way. The spins used, however, were impressive. A sidespin was very much as advertised, fooling us completely on the first run. The rapid fire sessions, which the Chens called “crazy mode”, were also pretty overwhelming.

The app doesn’t seem to track anything because there’s no way for it to assess technique or performance in an empirical way. That’s why the Trainerbot functions more like a smart batting cage than app-based practice trackers from the likes of Zepp Labs and 94Fifty that measure the mechanics of swinging a golf club, bat, racquet or shooting a basketball.

The network is supposed to do that instead. By comparing notes with other players, in what they are calling “the first social and connected training for ping pong”, human feedback and assistance stands in for analytics.

Nuts and bolts

The unit wasn’t without a couple of hiccups, which the Chens assured us will be fixed in the final version going to backers. Trainerbot sometimes stuttered between shot intervals, and there was a noticeable delay before spitting out the first ball. The robot’s jerky motion when reorienting itself also wasn’t all that pretty, but these are all going to be remedied, the two co-founders said.

During sequences where the Trainerbot had to move from one side to another, or up or down to change the direction or trajectory of a shot, it was fairly quick, but could be faster. For example, really short intervals between balls are harder to do when the robot has to switch from one side to another. Making up for that are drills that spit out balls in rapid succession, based on the set interval time, to practice reaction time. While placement was less important in that particular case, at least for us, being quick and making good contact with the ball was the main purpose.

Trainerbot
Ted Kritsonis/Digital Trends
Ted Kritsonis/Digital Trends

The app will be free for iOS and Android, and interestingly, both will be updated simultaneously because of the unified code they’re using. The user interface may get a slight facelift between now and the first batch of units that go to backers.

Conclusion

At the end of the day, the Trainerbot is definitely an enthusiast product, but we felt that it had a high enough “fun factor” to be attractive to ping pong players at any level. If you have a table at home, this is a nice way to get more action out of it when no human beings are around to play against. Assuming the co-founders address the issues we noted, it’s a robot we can see making some players better.

The Kickstarter campaign is live now, starting at $329 for a Basic model, and $379 for a Pro model, with shipping in March 2017.

Highs

  • Lightweight and portable design
  • Shot and spin selection
  • App support is superb
  • Bucket fits 30 balls at once
  • Easy setup

Lows

  • Not quite fast enough yet
  • Initial delay in spitting first ball out
  • Expensive
Ted Kritsonis
A tech journalism vet, Ted covers has written for a number of publications in Canada and the U.S. Ted loves hockey, history…
Ask YouTube expands to mobile, web, and TVs across the US
YouTube’s AI search is now available across all major screens.
Ask YouTube on TV and desktop

Searching for something online by simply describing what you want is becoming pretty normal now. Google Search can do it. ChatGPT can do it. And YouTube has been slowly turning its own search box into something that works in a similar fashion.

YouTube has now announced a wider rollout for Ask YouTube, its conversational AI search tool. The feature is available to users in the U.S. across the web, mobile devices, and TVs. Mobile is the newest addition here, after Ask YouTube previously expanded to the web and smart TVs. Instead of putting a few keywords into the search bar, you can ask YouTube a complete question, describe what you are looking for, and then refine the results using follow-up questions.

Read more
Google cuts prices on refurbished Pixel 9 phones and throws in a free AI subscription to sweeten the deal
Save on select refurbished Pixel phones and get a six-month AI subscription with 2TB of storage.
A person holding the Google Pixel 9, showing the camera.

Google has added the Pixel 9, Pixel 9 Pro, and Pixel 9 Pro XL to its Certified Refurbished store (via 9to5Google). The lineup launched back in 2024 starting at $799, but you can now grab a refurbished unit for as little as $499.

How much can you save on a refurbished Pixel 9?

Read more
OnePlus 16 lands with some insane perks. Alas, you can’t get it in the US
The OnePlus 16 pushes battery, display, and performance specs to new extremes, but it is staying out of the US.
OnePlus 16 all colors

OnePlus has launched the OnePlus 16 in China, and there are some wild numbers on the spec sheet. A 9,000mAh battery. A display that can hit 185Hz. A Snapdragon chip crossing the 5GHz mark. Even the main camera has jumped to 200MP. Sadly, none of it matters much if you are in the US. The OnePlus 16 is not officially available here, so American buyers will have to admire all this hardware from afar.

The battery situation is frankly ridiculous

Read more