A team at Google has spent a lot of time recently playing table tennis, purportedly only for science. Their goal was to see whether they could construct a robot which would not only play table tennis, but even keep up with practiced human players. In the paper available on ArXiv, they detail what it took to make it happen. The team also set up a site with a simplified explanation and some videos of the robot in action.

In the end, it took twenty motion-capture cameras, a pair of 125 FPS cameras, a 6 DOF robot on two linear rails, a special table tennis paddle, and a very large annotated dataset to train multiple convolutional network networks (CNN) on to analyze the incoming visual data. This visual data was then combined with details like knowledge of the paddle’s position to churn out a value for use in the look-up table that forms the core of the high-level controller (HLC). This look-up table then decides which low-level controller (LLC) is picked to perform a certain action. In order to prevent the CNNs of the LLCs from ‘forgetting’ the training data, a total of 17 different CNNs were used, one per LLC.

The robot was tested with a range of players from a local table tennis club which made clear that while it could easily defeat beginners, intermediate players pose a serious threat. Advanced players completely demolished the table tennis robot. Clearly we do not have to fear our robotic table tennis playing overlords just yet, but the robot did receive praise for being an interesting practice partner.

From the “why didn’t we think of that” department comes [dupontgu’s] pong mouse project. The mouse appears and acts like a normal computer mouse until you click the scroll wheel. When you do, the mouse rapidly moves the cursor on the connected computer to play pong. Obviously, though, the paddles and the ball all look like your cursor, whatever that happens to be. So, how do you tell the score? Well, when a score happens, the cursor shows between the two paddles. In the middle means the game is tied. Otherwise, the player closest to the score indicator is winning. The mouse is an off-the-shelf unit, but inside is a tiny XIAO RP2040 board that can act as both a USB host and a USB device. The RP2040 intercepts the USB traffic and can modify it as it sees fit or just pass it unchanged. Part of the secret sauce is to make the mouse use absolute mode so that it can teleport the cursor between two spots. This is common with, for example, touchscreen drivers. However, it is unusual for a mouse to use this mode. Of course, the cursor is actually only in one place at a time, but your eye thinks it is in multiple places at the same time.

The code itself isn’t very long. A few hundred lines of C++ and, of course, plenty of libraries to handle the USB. We can think of lots of reasons we might want to filter a USB device.

This version of pong doesn’t take a lot of mechanical parts. Or, you could try making one with no parts at all.

As far as video games go, Pong is already about as simple as it gets. But if even two dimensions is a bit more than you’re looking to tackle, [mircemk] shows how you can distill the core gameplay of this iconic title to its absolute minimum using an Arduino and a row of LEDs.

While [mircemk] brings their usual design aesthetic and flash to the project, this one could truly be done as a parts bin build. All you really need is a microcontroller with enough I/O pins (here, an Arduino Nano is used), a couple of buttons, and the aforementioned LEDs. A 16×2 LCD and a buzzer have been added to improve on the user interface a bit, but even that isn’t strictly required.

To play, each user holds their button and gets ready to hit it as soon as the LED closest to them lights up. Again, [mircemk] spruces this build up by offering both integrated buttons on the front panel of the game, as well as a pair of external “controllers” so you don’t have to crowd around the main unit. In this incarnation the score is shown on the LCD, but swapping that out for a pair of seven-segment LEDs could give the whole thing a bit more of a retro flair.

This isn’t the first time [mircemk] has tackled 1D Pong — if you can spring for addressable LEDs, you can pull the whole thing off with significantly less wiring.