Apple Watch support first arrived back in 2015. That early version belonged to a very different smartwatch era, when third-party watch apps were still finding their feet and most messaging tools on the wrist felt fairly basic. The app was later dropped out of view leaving users without software for years. As far as Wear OS, Telegram removed its support for that operating system in 2021.

This new release closes that gap on both sides.

What the new Telegram watch apps can do

The new watch apps give users direct access to chats from the wrist. That means you can open Telegram on Wear OS or Apple Watch, browse conversations, read longer messages and check shared media without waiting for a phone notification to appear.

Voice messages are part of the package too. That is useful for Telegram because many conversations on the service are built around group chats, media sharing and quick voice notes rather than simple one-line replies.

The apps also support replies from the wrist, either by text or voice. So the feature set is quite comprehensive.

What’s worth noting, there are some differences between platforms at launch. Wear OS users can mute and pin chats, as well as delete messages from the watch. Apple Watch users, on the other hand, get location viewing and sticker support.

Telegram says the missing tools should come to each platform in the next update. For now, the two versions overlap on the main features but differ on some controls.

Why this is better than notifications

Telegram alerts could already appear on smartwatches through the phone’s normal notification system. That is useful for quick glances, but it only reacts to incoming messages.

The new apps change the interaction. Users can open Telegram directly on the watch, go into chats, check older messages, view shared media and listen to voice notes without waiting for a fresh alert.

That makes the watch more useful for sorting small Telegram interactions. You can see whether a message needs a reply, send a short voice or text response and leave the longer stuff for the phone.

It also avoids the biggest limitation of notification-only use. If the alert is dismissed or buried, you are not stuck going back to the phone just to find the conversation again.

For a messaging app built around busy group chats, voice notes and shared media, that is a better fit for the wrist. The watch still stays in its lane, but Telegram now gives it enough to do that the app feels worth keeping there.

Source: Telegram

The post Telegram app returns to Wear OS and Apple Watch appeared first on Gadgets & Wearables.

The change has already annoyed some Apple Watch users, and I get why. Walkie-Talkie was never the biggest Apple Watch feature, but it had a very specific charm that newer, smarter features do not really replace.

Apple has not said whether the removal is permanent. That is worth stressing, because this is still beta software. Features can disappear, return, move around or get reworked before the final release lands later this year.

Still, the fact that Walkie-Talkie has gone missing from two obvious places is hard to ignore. If Apple was simply redesigning the app grid or Control Center, you would expect some trace of it to remain somewhere. Right now, users on the beta appear to be looking at an Apple Watch without Walkie-Talkie at all.

Why this one stings for some users

Walkie-Talkie arrived as one of those odd Apple Watch ideas that made the device feel more playful. You pressed a button, talked into your wrist and the other person heard you almost instantly. It was not exactly essential, but it made the watch feel like more than a tiny iPhone extension.

That is probably why the Reddit reaction is more interesting than the feature removal itself. Some users said they barely touched it because it was unreliable. Others said they used it all the time with partners, children or family members inside the same house.

That split tells you a lot about Walkie-Talkie. It was not a mainstream killer feature. It was a niche habit. But when it clicked, it really clicked.

One person’s forgotten app is another person’s household intercom. That is the awkward bit for Apple. Removing underused features makes sense on paper, but the people who do use them often rely on them in ways that product metrics do not fully capture.

It was never perfect

Walkie-Talkie also had baggage. The feature has long had a reputation for being inconsistent. Users have complained over the years about connection problems, unavailable contacts and messages that do not always go through cleanly.

Apple’s own support page also points to the way the feature depends on FaceTime. If FaceTime is deleted from the paired iPhone or not set up properly, Walkie-Talkie may not appear or work as expected. That probably explains some of the confusion around the feature over the years.

So there is a practical argument for Apple cutting it. A feature that only some people use and that often causes reliability complaints becomes harder to justify, especially as watchOS grows more crowded.

But Apple loses a bit of personality here

The problem is that Apple Watch has become increasingly serious. It tracks health, workouts, sleep, safety, recovery and notifications. All useful. But some of the earlier fun has been squeezed out along the way.

Walkie-Talkie was one of those features that gave the watch a bit of personality. It felt slightly silly, but in a good way. You could imagine kids loving it, couples using it around the house or someone quickly checking in without starting a full phone call.

That is hard to replace with a smarter Siri or a redesigned app grid. Those features may be more advanced, but they do not fill the same little social gap.

Of course, Apple may bring Walkie-Talkie back in a later watchOS 27 beta. The company could also be rebuilding it, moving it into another communication feature or temporarily removing it while it sorts out compatibility issues.

The post Apple may be done with Walkie-Talkie on Apple Watch appeared first on Gadgets & Wearables.

What changed

It has been a while since the Helio Strap last received a firmware update, so this one is worth noting. Version 3.18.0.1 is not trying to turn the device into something new. It looks more like a tidy-up release aimed at making the strap behave better inside the wider Amazfit ecosystem.

The first change is probably the more useful one. Zepp Health says the Activity Detection feature has been optimised so the Helio Strap will no longer create duplicate activity records when your watch is already tracking a workout. That sounds like a niche fix until you actually run into it. Then it becomes exactly the sort of thing that makes post-workout data messy.

The Helio Strap sits in a slightly unusual place. It is not a watch. It is not a screen-first wearable. It works best as a lightweight fitness strap that feeds data into the Zepp Health app and can sit alongside a more capable Amazfit watch. That makes clean workout handling important. Users do not want two versions of the same session fighting for attention inside their history.

This update should help people who use the Helio Strap with an Amazfit watch rather than as a standalone tracker. If the watch already records the run, ride or gym session, the strap should now avoid generating a separate duplicate activity record from its own detection system.

This also suggests Zepp Health is still tuning how the Helio Strap behaves in a multi-device setup. That is where the product needs to be reliable. A strap like this does not win people over with a big display or rich on-device features. It needs to disappear into the background and record the right data at the right time.

Heart rate tracking also gets attention

The second listed change is an optimisation to the heart rate monitoring algorithm. Zepp Health says this should improve heart rate tracking accuracy, though the release notes do not give extra detail on what has changed.

It will be interesting to see whether the update improves higher-intensity exercise tracking. Optical heart rate sensors can struggle when movement, strap tension and sweat all get involved. The Helio Strap has a simpler job than a smartwatch in some ways because it does not have to manage a screen-heavy experience. But it still lives or dies by whether people trust the numbers.

Worth installing

The final note simply says other known issues have been fixed. That is the usual catch-all line in firmware updates, but the two named changes are enough to make version 3.18.0.1 worth installing.

All things considered, this is a maintenance release with one very practical fix. But for anyone using the Amazfit Helio Strap alongside a watch, removing duplicate workout records should make the experience cleaner straight away. As always, different regions may receive the update at different times. We have reports of those in the UK and a few other countries already receiving the new firmware.

In other news, the company is working on the gen 2 device. We recently spotted the FCC regulatory filing for that one and expect a launch in the second half of the year.

The post Amazfit Helio Strap firmware targets workout duplication and heart rate accuracy appeared first on Gadgets & Wearables.

The numbers explain the shift. Zepp Health said its average selling price rose by more than 20% year-on-year, helped by stronger demand for higher-end Amazfit models. That is a big change for a brand that built much of its reputation on undercutting Garmin, Fitbit and Apple Watch.

The price ceiling has moved

Now make no mistake – Amazfit still sells plenty of affordable watches. But the top of the range now looks very different. T-Rex 3 Pro comes in at around $399. The recently launched T-Rex Ultra 2 pushes that further to about $549.

That still keeps Amazfit below the obvious premium rivals. Apple Watch Ultra 3 starts at $799, while Garmin Fenix 8 commonly sits around the $999 level before discounts. So Amazfit is not trying to match Garmin or Apple on price. Which is a good thing. If Amazfit can offer enough of the same outdoor and training appeal for less money, the higher pricing still works.

T-Rex is doing the heavy lifting

The T-Rex line is the obvious place to test this. Outdoor watches already carry higher price expectations because buyers care about battery life, materials, GPS, maps and durability. A higher-priced T-Rex makes more sense than a much more expensive Bip or Active watch.

Despite the price hike, Zepp Health said the more expensive T-Rex models accounted for nearly half of T-Rex family unit sales in March and April. That suggests buyers are not just looking at Amazfit because it is cheap. Some are now choosing the more expensive versions inside the range.

The company also said it has more T-Rex family products lined up for 2026, so this does not look like a single premium experiment. Falcon 2 could be part of that push if it lands later this year, and that would likely move the price ceiling even higher.

The same shift shows up elsewhere. Balance 3 and Balance Ultra lean into HYROX, recovery, readiness and hybrid training. Cheetah 2 Pro and Cheetah 2 Ultra take the pitch into running, with Zepp Coach, training plans, recovery insights and third-party platform links.

Value is still the safety net

Zepp Health is not abandoning cheaper models. Active Max, Active 3 Premium and Bip Max still keep Amazfit tied to the value side of the market. That gives the brand useful cover. It can chase higher margins with T-Rex, Balance and Cheetah, while still keeping entry-level buyers in the ecosystem.

The risk is confusion. Amazfit used to be easy to understand. It meant long battery life, lots of features and lower prices. Now the message is broader. Some models still fit that old formula. Others are trying to compete with proper sports watches.

But that may be exactly where Zepp Health wants to land. Amazfit does not need to become Garmin. It just needs enough buyers to see a $549 Amazfit as a credible alternative to an $799 or $999 watch.

The post Amazfit is no longer just the cheaper smartwatch brand appeared first on Gadgets & Wearables.

Verizon appears to have jumped early

Carrier support pages are not always exciting reading, but they can be useful when they reveal software details before the official marketing machine catches up. In this case, Verizon has published update pages for Pixel Watch 2 and Pixel Watch 3 that specifically mention Wear OS 7.

For Pixel Watch 2, Verizon lists the update as System Update 20. For Pixel Watch 3, it appears as System Update 9. Both pages show the same software version, CP2A.260603.001, along with the June 2026 Android security patch.

The language around the update stays fairly plain. Verizon says the software brings the latest Wear OS 7 update, the latest Android security patch and performance and stability improvements.

This fits the Wear OS 7 timeline

Google already detailed Wear OS 7 earlier this year, so the update itself is not a surprise. The timing is the interesting part. A June 9 carrier listing suggests the Pixel Watch rollout may be at the final preparation stage, or already staged for some users depending on region, model and carrier.

Wear OS 7 has always looked like a cleaner platform update rather than a dramatic redesign for the sake of it. The useful parts sit around battery efficiency, smarter watch interactions, better app behaviour and tighter links between Android phones and Wear OS watches.

One of the bigger background changes is the move to Android 16 under the hood. That should give Google and developers a more modern base to work from, even if users mostly notice the surface-level features first. The best updates to a smartwatch often feel boring at first, then prove themselves through better consistency over time.

Google has also talked about battery gains with Wear OS 7. The company previously suggested that supported devices could see improved efficiency, which would be welcome on Pixel Watch hardware. Battery life has never been the Pixel Watch line’s strongest selling point, so even smaller gains could make the daily experience less annoying.

The sensible caveat is that carrier pages do not always mean every user gets the update instantly. Pixel Watch updates can roll out in stages, so one owner may see it before another. Region, carrier and model can all affect timing.

Still, the signs now look solid. Verizon has published the details, the software version is consistent across the Pixel Watch 2 and Pixel Watch 3 pages and the date lines up with a likely June software release window. Pixel Watch owners should probably start checking the system update screen over the next few days.

via Droidlife

The post Wear OS 7 appears ready for Pixel Watch 2 and Pixel Watch 3 appeared first on Gadgets & Wearables.

That is not how most people test a fitness tracker. It is also exactly why I found the result useful. If a device can produce decent data in a messy real-world setup, that tells us something about how flexible the hardware might be.

This follows my earlier Fitbit Air vs Garmin 5K test, where the screenless tracker came very close when the run was started properly from the phone. For this follow-up, I wanted to make life harder for it. I wore it on my ankle and did not start anything manually.

The distance result was solid

My Garmin Forerunner recorded the run at 5.01 km. Fitbit Air logged 5.16 km, which puts it around 150 metres higher over the run. That is not perfect, but for an automatically detected run from the ankle, I would call it a surprisingly decent result.

The Garmin used GPS. Fitbit Air, on the other hand, had to rely on its own motion data and whatever context it could pull together in the background. The important part is that it did not drift wildly. It recognised the session as a run and landed close enough to be useful for casual tracking.

That makes the result more interesting than a simple wrist-to-wrist test. Ankle wear changes the movement pattern completely. The device has to deal with sharper leg motion, higher impact and a position it probably was not designed around as its main use case.

Google only talks about wearing the tracker on the wrist. In practice, though, it can also be worn on the upper arm or even the ankle. I have worn it like this 24/7 for the past week and the only real difference I have noticed is shorter sleep tracking. My guess is that I move my legs more than my arms during sleep, so the tracker may be reading some of that movement as wake time.

The average heart rate needs context

But what about heart rate during the 5K. Well, at first glance, the heart rate comparison looks weaker. The Garmin reported an average heart rate of 144 bpm, while Fitbit Air showed 132 bpm. Taken alone, that looks like a fairly large gap.

But the chart tells a better story. Fitbit Air appears to have started tracking too early and ended too late. Those low heart rate sections at the beginning and end pulled down the average. During the actual running portion, the heart rate trace looked much closer to the Garmin result.

The max heart rate makes the point even clearer. The Garmin recorded 161 bpm, while Fitbit Air reached 160 bpm. That is only 1 bpm apart, which is a strong result for a tracker worn on the ankle during a run.

This was probably the most interesting part of the test for me. Optical heart rate from the ankle sounds like something that should struggle. In this run, it looked surprisingly stable once the effort started properly.

Auto-tracking is the weak link

The main issue was not that Fitbit Air failed to detect the run. It did detect it. The issue was that it wrapped too much time around the actual workout.

The Garmin recorded 28 minutes and 33 seconds for the 5K. Fitbit Air showed 35 minutes and 48 seconds. That extra time changes the whole summary. It makes the average pace look much slower and it drags down the average heart rate.

Fitbit Air*

Order now

This is where Google Health needs better editing tools. There is currently no option to trim the run afterwards. I could view the activity, but I could not cut off the early and late sections that should not have counted as part of the workout.

That limitation is annoying because the raw tracking looked better than the final numbers imply. Strip away the extra time and Fitbit Air would probably get much closer on average heart rate. The problem is that the app does not let me clean up the session.

A useful result with a clear caveat

This test does not prove that Fitbit Air is a replacement for a running watch. It also does not prove that ankle-based heart rate will hold up in every workout. Intervals, hills and faster changes in effort could expose weaknesses that a steady 5K does not show.

Still, I came away encouraged. Fitbit Air got the distance close, detected the run automatically and matched the Garmin’s peak heart rate almost exactly. For a screenless tracker worn on the ankle and left to work passively, that is a better outcome than I expected.

The lesson for me is that Fitbit Air’s hardware may be more capable than the app summary suggests. The sensor performance looks promising, but the automatic tracking window needs more control. A simple trim option in Google Health would make a big difference here because it would let me remove the dead time and keep the useful part of the workout.

For now, this makes Fitbit Air look like an interesting passive fitness tracker rather than a precision running tool. It can capture the run and the ankle heart rate result looks genuinely useful. But if Google wants this kind of device to be trusted for workouts, post-run editing needs to catch up with the hardware.

The post Fitbit Air on the ankle delivers a surprisingly solid 5K run result appeared first on Gadgets & Wearables.

That is not a huge delay, but it does change slightly the shape of the launch. The revived Pebble project now looks like it is choosing a cleaner production run over rushing the first watches out the door.

Pebble Round 2 is close but not in mass production yet

Migicovsky posted the update after reviewing the latest pre-production samples in China.

As can be seen, the watch is currently in beta testing, with US users invited to apply. That suggests the hardware is far enough along for real-world use but still not at the point where Pebble wants to press the production button.

For anyone waiting on a preorder, the current plan is for shipping to start next months Pebble expects to finish shipping all preordered Round 2 units in September, so this will not be a single big launch wave where everyone gets a tracking number at once.

The delay comes down to a case detail

The reason for the hold-up is not software, battery life or a major hardware redesign. Migicovsky says the latest sample had a cosmetic issue near the lugs, caused by a CNC milling step close to where the strap connects to the watch case.

That may sound minor, but on a watch like the Round 2, the case finish does a lot of the heavy lifting. This is not a thick outdoor watch where a tiny mark disappears into the overall chunkiness. It is a small round smartwatch with a simple design, so visible case details stand out quickly.

Pebble is now modifying the metal-injection-moulding tooling to avoid that CNC cut. In plain English, the company wants to fix the production process rather than accept a mark that would probably annoy buyers the moment they noticed it.

Leather straps are coming too

There was also a smaller accessory update. Black and brown leather straps will be available as add-ons for Pebble Round 2.

The watch already has a clean circular case and the sample shown in the update looks more like a simple everyday watch than a gadget trying too hard. Leather straps should push it further in that direction.

This still looks like a careful relaunch

A small manufacturing issue has appeared, the tooling needs tweaking and shipping shifts into July. That is not dramatic, but it is exactly the kind of thing that happens when a small hardware project gets close to production.

Pebble could have framed this more aggressively and promised everything was fine. Instead, Migicovsky showed the problem, explained the fix and gave a clearer shipping window. For a preorder product, that sort of visibility helps more than pretending the schedule never moved.

The post Pebble Round 2 delay pushes shipping into July appeared first on Gadgets & Wearables.

The company opened its keynote from Apple Park in Cupertino. The developer conference continues through June 12, but the keynote was the main public event where Apple set out its next round of software updates across iPhone, iPad, Mac, Apple Watch, Apple TV and Vision Pro.

For Apple Watch users, the focus was watchOS 27. This was not a hardware launch, as Apple normally saves new Watch models for later in the year. But the software still gives a clear steer on where the wrist platform is heading, especially around Siri, Apple Intelligence, health tracking and everyday usability.

Siri AI comes to the wrist

This year it is all about AI. So it should come as no surprise that the main Apple Watch change in watchOS 27 is a custom version of Siri AI built for smaller screens.

Apple says users will be able to ask questions and take actions from the wrist. That makes Siri AI more useful on Apple Watch than a simple voice search tool, assuming it can handle real tasks reliably. The watch has always needed a better assistant because typing, tapping and scrolling on a small display only gets you so far.

The AI conversation also syncs across Apple devices through iCloud. That means a Siri AI interaction does not have to stay trapped on the watch. A user could start from the wrist and continue elsewhere, which fits how Apple devices already work together.

Apple has also placed Siri AI inside a redesigned app grid in watchOS 27. That is a small interface detail, but it shows Apple wants the feature to feel like part of the normal watch experience rather than a separate AI layer bolted on top.

Unfortunately, Siri AI will have a limited launch. The European Union and China are excluded at first, so Apple Watch users there may get watchOS 27 before the new AI assistant features arrive.

Apple Intelligence leans on private cloud computing

Apple Intelligence features on watchOS 27 will use private cloud computing rather than relying on full on-device processing. That makes sense for the Apple Watch. The device has limited space, battery and thermal room compared with an iPhone or Mac, so heavier AI processing needs a different route.

The privacy claim is the important bit here. Apple says the data used for these features cannot be accessed by Apple or third parties. That gives the company a cleaner argument than simply saying it has added AI to the watch.

For wearables, privacy is not just a marketing line. The Apple Watch handles health, location, activity and personal routine data. If Apple wants users to ask more from Siri AI on the wrist, it needs to explain clearly where that data goes and what happens to it.

Smaller changes may be the ones users notice

watchOS 27 is not only about Siri AI. Apple has also added a batch of smaller Apple Watch improvements that should show up in daily use. These include faster music playback, improved Wi-Fi connectivity, quicker app extension launch times and battery optimisation suggestions.

Fitness app step count also now syncs with Health app steps. That should reduce some of the mismatch between Apple’s own activity and health data, which always felt oddly messy for such a polished ecosystem.

Find My also gets cleaned up. Apple has combined Find Devices, Find People and Find Items into a single Apple Watch app with a map-centric view. That should make the feature easier to use from the wrist, especially when trying to locate a person, device or item quickly.

Wallet and Smart Stack get attention too. Apple says Wallet can now create custom passes for cards that use a QR code or barcode, such as a library card, then make them available in Wallet or Smart Stack. Transit cards and IDs can also appear in Smart Stack, alongside new Smart Stack suggestions.

Call Context is another addition. When calling a business, Apple Watch can surface relevant information, such as a confirmation code from Mail when calling an airline. Apple says the feature is coming in English.

Workout Buddy gets more useful

Workout Buddy also gets attention in watchOS 27. Apple says it now uses fitness history to deliver more specific insights, including progress around pace, distance and workout duration.

The useful bit is that Workout Buddy can now work without the iPhone nearby. That makes it more relevant for runners and gym users who train with only the watch. Apple is also adding Spanish support, which broadens the feature beyond its initial language setup.

Cycle tracking adds perimenopause support

Cycle Tracking in the Health app can now provide notifications when logged cycle patterns are suggestive of perimenopause. Users can also track related symptoms and access educational resources.

Apple is not saying the watch diagnoses perimenopause or menopause. It is looking at logged cycle patterns and flagging possible deviations for users aged 40 and above.

This moves Apple’s women’s health tools further into longer-term health context. It is still not a medical diagnosis, but it gives users another reason to keep cycle data inside the Health app rather than using a separate tracker.

Compatibility

Apple’s watchOS 27 page lists support for Apple Watch SE 3, Apple Watch Series 10, Apple Watch Series 11, Apple Watch Ultra 2 and Apple Watch Ultra 3. It seems Series 9 will also support the new operating system, despite it being left off the list initially.

The update also needs a paired iPhone running iOS 27. Apple says that means iPhone 11 or later, or iPhone SE 2nd generation or later. So compatibility depends on both the watch and the phone.

That makes this a more limited update than expected. Owners of older Apple Watch models should check Apple’s final compatibility list before assuming they will get watchOS 27.

The post watchOS 27 brings Siri AI and new health tracking to Apple Watch appeared first on Gadgets & Wearables.

The table tennis range is not built around one device. Pongbot currently has three main table tennis robots in the lineup: Nova S Pro, Omni S Pro and Halo S Pro. They all do the same basic job, firing balls with adjustable speed, spin and placement, but they sit at very different levels.

My review of the Pongbot Pace S Pro was about tennis, but the bigger point was the same. A good training robot only works if it gets out of the way and lets you build repeatable sessions without turning practice into a settings menu.

That is where Pongbot’s table tennis machines make sense on paper. These are not humanoid opponents or robots that play rallies back at you. They are smart serving robots built for repetition, placement, spin work and drill structure.

Nova S Pro is the entry point

Nova S Pro is the smallest and cheapest model in the range. It sits on the table, weighs 4kg and holds around 150 balls, so it is clearly the one designed for home users or players who want something they can move around easily.

The headline specs are still strong for the price. It can fire balls from 2 to 15 metres per second, with a frequency of 30 to 90 balls per minute. Spin goes up to 60 revolutions per second and Pongbot says it supports 9 spin variations. There are also more than 264 built-in drills, plus app and remote control.

Nova S Pro*

Order now

The compromise is adjustment. Nova S Pro has manual side and vertical adjustment, so it does not give you the same automated placement flexibility as the more expensive models. That probably will not bother a casual or intermediate player. It may bother someone trying to build more complex footwork routines.

This is the model that looks easiest to recommend for someone who wants structured solo practice without going deep into club-level equipment. It is still a proper training robot, just stripped back in some places.

Omni S Pro steps things up

Omni S Pro is the clamp-on model and it moves the range into more serious territory. It weighs 9kg, clips onto the table and holds around 250 balls. The ball speed remains 2 to 15 metres per second, while the firing frequency stays at 30 to 90 balls per minute.

The big jump is spin and control. Omni S Pro supports up to 100 revolutions per second and 360-degree spin adjustment. It also adds automatic continuous left and right oscillation, plus automatic vertical and side adjustment. Control comes through the app or Pongbot’s E-Pad touchscreen.

Omni S Pro*

Order now

That changes the use case. With Nova S Pro, you are getting a compact robot that can handle useful drills. With Omni S Pro, you are getting something that can build more realistic patterns across the table. Forehand, backhand, middle ball, short ball, deep ball, spin change. That is where table tennis training starts to become more serious.

It also includes more built-in drills than Nova S Pro. Pongbot’s own listing shows over 522 drills, which gives players a large preset library before they start creating their own routines. For clubs, coaches or players who train often, that extra control is probably where the money goes.

Halo S Pro is the full setup

Halo S Pro is the floor-standing model. It weighs 11kg, holds around 250 balls and uses the same broad performance range as Omni S Pro, with 2 to 15 metres per second speed, 30 to 90 balls per minute and up to 100 revolutions per second of spin.

The difference is the format. Because Halo S Pro stands on the floor, it can be positioned in different locations around the table. That gives it more flexibility than a fixed clip-on robot, especially for players who want to simulate different angles or feed patterns from more natural positions.

Halo S Pro*

Order now

It also gets the largest drill library, with more than 702 built-in drills according to the current Pongbot listing. That makes it the most complete option in the range, but also the one that makes the least sense for someone who just wants a casual hit in the garage.

Halo S Pro looks more like club kit. It is the one for players who already know they will use a robot regularly and want the widest spread of training patterns. For everyone else, Omni S Pro may already cover the useful middle ground.

Pongbot table tennis robot comparison

The table makes the range fairly easy to understand. Nova S Pro is about portability and price. Omni S Pro is about better automation and more serious drill building. Halo S Pro is the most flexible setup, mostly because of its standing design and larger drill library.

You can also see the typical pricing. But it’s worth pointing out, the company frequently runs sales so its worth checking their site from time to time.

The AI label needs some perspective

Pongbot uses a lot of smart training language around these products. That makes sense, especially after using the Pace S Pro, but it is worth separating the useful training features from the marketing gloss.

With the Pace S Pro, the smarter side of the product is more obvious. It can track the player, adjust drills and use positioning data to make tennis practice feel less static than a traditional ball machine. That is a very different setup from the table tennis robots, which are not watching your stroke or reading your movement in the same way.

The table tennis range is better understood as a set of app-controlled serving robots with structured drills, spin variation and programmable placement. That still lines up with the same Pongbot idea I saw in the tennis machine: solo practice should feel more deliberate, less random and easier to repeat.

That does not make the table tennis models less interesting. It just makes the buying decision cleaner. With Pace S Pro, part of the appeal is the way the machine reacts to the player. With Nova S Pro, Omni S Pro and Halo S Pro, the appeal is different. You are paying for repeatable ball delivery, quick drill changes and the ability to build awkward patterns without needing a practice partner.

For table tennis, that is already a lot. You do not need every device to pretend it is a coach. Sometimes the better product is the one that feeds the exact same awkward backspin ball 80 times until your brain finally gives in.

Which one makes the most sense

Nova S Pro is the obvious starting point for most home users. It is portable, much cheaper and still gives you speed, spin and drill programming. For players who want to improve footwork, returns and consistency without spending over $1,000, it is the easiest one to understand.

Omni S Pro is probably the sweet spot for serious players. It adds automatic adjustment, higher spin capability, more drills and a larger ball capacity, while keeping the unit mounted neatly on the table. It looks like the one to consider if the robot will become part of a regular weekly training setup.

Halo S Pro is the big one. It makes sense if placement flexibility is important, or if the robot will sit in a dedicated training space rather than being packed away after each session. It is overkill for casual use, but that is not really the point of it.

What I like about Pongbot’s table tennis range is that the differences are not hard to explain. The company has not buried the buyer in tiny model changes. It has a portable model, a clamp-on model and a floor-standing model. Pick the setup first, then decide how much automation and drill depth you actually need.

After reviewing the Pace S Pro, this range feels like a natural extension of the same idea. Pongbot is clearly interested in solo training that feels more structured than old-school ball machines. The table tennis robots are less ambitious than the tennis machine in terms of tracking and AI, but they may be easier to justify for everyday practice.

*We are a review site that receives a small commission from sales of certain items, but the price is the same for you. Purchasing items by clicking on links in this article allows us to run this website. We are independently owned and all opinions expressed here are our own. See our affiliate disclosure page for more details.

The post Pongbot S Nova, Omni and Halo show where solo table tennis training is going appeared first on Gadgets & Wearables.

Bottom line

Balance 3 looks like the easiest recommendation for most people. It brings the main 2026 upgrades, including the brighter 3,000-nit display, more storage, newer processor, physical flashlight, extra buttons and Zepp OS 6 out of the box, without the Ultra price.

Balance Ultra is for a narrower buyer. It gets those same newer platform upgrades, then adds the bigger battery, more premium build and most physical controls. It is the best hardware package here, but not the best value. The awkward bit here is that the device shares most of its core platform with Balance 3. The main separation is case, materials and battery, so the higher price really needs those things to matter to the buyer.

Balance 2 still has a clear role. It is smaller, lighter and loaded with features, including 10 ATM water resistance, 45 metre diving support, dual-band GNSS, golf maps, HYROX modes and strong battery life. It just needs the discount to be strong enough.

For most buyers, the answer is likely Balance 3. Go Ultra if battery and build quality sit at the top of your list. Go Balance 2 if the price drops far enough, or if the smaller case is more important to you than the 2026 hardware upgrades.

Balance 2 is the smaller and lighter watch

One thing Balance 2 has in its favour is wearability. It is the smallest and lightest watch in this comparison, measuring 47.4 mm diameter and weighing 43g without the strap. That makes it easier to live with day and night.

Balance 3 and Balance Ultra are much larger. Balance 3 has a 51.4 mm diamter, while Balance Ultra is slightly larger again at 51.8 mm. They have more of a big sports-watch feel, which will suit some users, but not everyone wants that much case on the wrist. For those with slim wrists, these watches are not realistic options.

The Balance 3 version also matters. The stainless steel model is the heavier one at 62g and uses a plastic bottom shell, while the titanium version drops to 55g and gets closer to Ultra on materials. Balance Ultra weighs 57g and uses more grade 5 titanium across the bezel, frame, buttons and back cover.

That gives Ultra the more premium build, but not necessarily the easiest fit. Balance 2 may be older, but its smaller case and lighter body remain real advantages if comfort is high on the list.

The button setup also changes the feel of the watches. Balance 2 has two buttons, Balance 3 has four and Balance Ultra has five. That gives the newer models more direct control during workouts, map use and wet conditions, which helps explain their more serious sports-watch positioning.

Materials also move up as you go newer and more expensive. Balance 3 comes in stainless steel and titanium versions, while Balance Ultra leans harder into grade 5 titanium across the bezel, frame, buttons and back cover. That gives Ultra the more premium build, but it does not automatically make it the easier watch to wear.

The button setup is also worth mentioning. Balance 2 has two buttons, Balance 3 has four and Balance Ultra has five. That gives the newer models more direct control during workouts, map use and wet conditions.

Screen brightness is the first obvious upgrade

One of the easier upgrades to explain is outdoor visibility. Balance 2 tops out at 2,000 nits, while Balance 3 and Balance Ultra reach 3,000 nits. The size and sharpness stay familiar, as all three use a 1.5-inch AMOLED display with 480 x 480 resolution and sapphire glass.

That means the newer watches are not giving you a larger canvas. They are giving you a brighter one. For indoor use, that will not change much. For outdoor training, maps, hiking, golf or sunny runs, it should make a bit of a difference.

The newer chip and extra storage matter for maps

Maps and navigation are where the internal hardware starts to matter more. Balance 3 and Balance Ultra move to the newer ZPS3044s chip and 64GB of storage, while Balance 2 uses the older ZPS3044 and 32GB.

That sounds like background hardware housekeeping, but it fits where these watches are going. Zepp Health is putting more weight on maps, routes, navigation and structured training screens. A watch that leans into those features needs enough storage and enough speed to avoid feeling awkward.

This is also one of the reasons Balance 3 looks like the smart upgrade. You get the faster chip and doubled storage without stepping up to Ultra. If map performance really has improved, that may be one of the more noticeable differences between Balance 2 and the 2026 models.

The LED flashlight on new models

The physical LED flashlight could easily sound like filler, but it is not a useless extra. It gives Balance 3 and Balance Ultra a simple practical tool that Balance 2 does not have. Anyone who has used a watch flashlight at night knows it quickly becomes one of those small conveniences you miss when it is gone.

Battery life is where Balance Ultra pulls ahead

Battery life is the clearest reason to look at Ultra. Balance 2 and Balance 3 both carry a 21-day typical-use claim, but Balance 3 improves the GPS numbers. It moves from 33 to 41 hours in accurate GPS mode and from 67 to 84 hours in power-saving GPS mode.

That makes Balance 3 a stronger outdoor watch than Balance 2, even though the general battery claim stays the same. It is a useful improvement for longer rides, hikes and events, but it does not completely change the character of the watch.

Balance Ultra is different. Its larger battery stretches typical use to 30 days and pushes accurate GPS endurance to 50 hours. That is the strongest practical reason to pay more. If you travel often, do long outdoor sessions or simply hate charging devices, Ultra has a clear job.

Hybrid training is the direction of travel

Zepp Health is clearly pushing the Balance line further into hybrid training. Balance 3 and Balance Ultra arrive with HybridCharge, LifeLoad, Weekly Focus, Training Balance and deeper HYROX tools built into the launch story, which gives the newer watches a more specific training angle than Balance 2 had at launch.

HybridCharge needs careful wording, though. Zepp Health is rolling it out more widely across Amazfit watches through the Zepp Health app, so this is not a simple case of Balance 3 and Ultra getting it while Balance 2 misses out. The newer watches mainly have the advantage of arriving with this training framework already front and centre.

HYROX is the stronger difference. Balance 2 already has HYROX modes, but Balance 3 and Balance Ultra go further with a training library, race strategies, virtual pacing and post-race analysis. Zepp Health also has the global HYROX licensing for Amazfit, which gives these watches a native HYROX angle rather than treating it as a generic hybrid workout.

The race strategy tools can work around event layouts, with targets for run sections and stations, then show whether you are gaining or losing time against the plan. That makes the newer HYROX setup feel more like race preparation than basic workout tracking, which explains why Zepp Health is leaning so heavily into this area.

Software is an advantage for the newer models, but not a settled one

Balance 3 and Balance Ultra ship with Zepp OS 6. Balance 2 currently runs Zepp OS 5. That gives the newer watches the cleaner software story today.

The caveat is important. Balance 2 may still get Zepp OS 6, but Zepp Health has not confirmed that yet. So this is a current advantage for Balance 3 and Ultra, not something that should be presented as a permanent split.

Hardware and specification comparison

The post Amazfit Balance 3, Balance Ultra vs Balance 2 compared appeared first on Gadgets & Wearables.