PlayStation VR2’s first hand tracking game is also available on Quest, offering a comparison between the two headsets.
Waltz of the Wizard added hand tracking on PlayStation VR2 last week, the first title to do so, just under five years after it became one of the first to add the feature on the original Oculus Quest too. Now, its developer Aldin Dynamics has shared a direct hand tracking comparison video it made by wearing a PlayStation VR2, Quest 3, and wearable camera at the same time.
0:00
Video from Waltz of the Wizard’s developer Aldin Dynamics.
It’s difficult to holistically assess the quality of hand tracking from a video clip alone. What Aldin’s video does clearly show is that Quest’s hand tracking, by default, doesn’t handle fast hand motions as well, and has more “overshoot” than PS VR2’s, with the hand visibly bouncing upon the cessation of fast motion.
But I say “by default” because there’s more to this story, and I went hands-on with hand tracking on both headsets to review its quality.
Henry Stockdale
As revealed by Sony’s SIGGRAPH Asia 2024 booth, PlayStation VR2’s hand tracking runs at 60Hz. In contrast, by default, Quest’s hand tracking runs at 30Hz, and this is what Aldin is using in the Quest build of Waltz of the Wizard. This likely explains the “overshoot” visible in Aldin’s comparison video. With twice the time between camera frames, Quest has to rely more heavily on prediction.
Quest’s hand tracking can, however, run at 60Hz (or 50Hz in some countries) too, a feature called Fast Motion Mode, which developers can enable for their apps. Testing Fast Motion Mode in Meta’s Move Fast demo and XRWorkout, I found that it solves the overshoot issue and handles fast motion just as well as, and perhaps even slightly better than, PlayStation VR2.
So why doesn’t Waltz of the Wizard use Quest hand tracking’s Fast Motion Mode? The answer is that it comes with tradeoffs.
Firstly, on Quest 2, Fast Motion Mode reduces the maximum GPU clockspeed, harming rendering performance, and on Quest Pro it cannot be used at the same time as eye tracking, which Waltz of the Wizard leverages for gameplay.
Neither of these limitations apply to Quest 3 and Quest 3S, but there is another that applies to all Quest headsets using Fast Motion Mode: increased jitter and slightly decreased accuracy. And it’s this reason that Aldin says it doesn’t use Fast Motion Mode. According to the developer, this reduced accuracy “prevents gesture recognition from functioning properly”, referring to the game’s many highly specific hand tracking gestures used for gameplay. Thus, Waltz of the Wizard sticks with 30Hz.
0:00
Footage from XRWorkout’s developer showing Quest 3 hand tracking in Fast Motion Mode.
However, trying PlayStation VR2’s hand tracking in Waltz of the Wizard myself I found that, in my home at least, it actually has more jitter and inferior accuracy to even Quest’s Fast Motion Mode, which I tested in Meta’s Move Fast demo and XRWorkout. XRWorkout’s developer also defended the accuracy of Fast Motion Mode, sharing the clip seen above.
This disparity in results underscores how different environmental lighting conditions can significantly affect the quality of hand tracking. This is why Quest 3S having infrared illuminators for hand tracking is so significant – in theory, it nullifies the detriments of low light conditions.
Anyways, with all this aside, let’s get to my personal comparison of hand tracking in Waltz of the Wizard between PlayStation VR2 and Quest 3 (in its default 30Hz mode), in my home.
Latency
How soon after you move your hand do you see movement represented on your virtual hand? That’s latency – the delay between real and virtual movement.
In theory, Waltz of the Wizard on PlayStation VR2 should entirely have the edge here, as the 60Hz tracking should mean there’s half as long between camera samples.
In practice, for hand motion this is certainly true, as PS VR2 feels like it has an ever so slightly less delay between moving your hand and seeing your virtual hand move. But interestingly, Quest 3 feels like it has less latency for finger motion, such as curling your fingers. These differences feel very minor though, and some of it could be down to differences in the refresh rate of the app itself, as well as graphics rendering latency differences.
WINNER: Draw
Accuracy
Accuracy refers to how closely your virtual hand and finger movements match your hand in real life, regardless of the delay.
With PlayStation VR2, I find the system fairly often gets the direction of my finger curls wrong, showing them curling outwards instead of inwards for example. I also find that when touching one finger to another in reality, there’s a gap between my virtual fingers.
0:00
Here’s Sony’s hand tracking misinterpreting my finger curl direction on my right hand.
On Quest 3, in the same lighting conditions the system almost never gets my finger poses wrong. Further, when touching my fingers together, my virtual fingers touch too. Quest’s hand tracking, in its default mode at least, is simply a more accurate representation.
WINNER: Meta Quest 3
Jitter
When you keep your hands still, do your virtual hands stay still too? If not, that’s called jitter.
0:00
PlayStation VR2
The difference between the two headsets here is very subtle, but Quest 3 has a slight edge in stability, with PlayStation VR2 having slightly more noticeable jitter.
0:00
Quest 3
WINNER: Meta Quest 3
Fast Motion
Nowhere is the difference between PlayStation VR2’s 60Hz hand tracking and Quest 3’s default 30Hz tracking more evident in Waltz of the Wizard than with fast hand motions.
0:00
Aldin’s own video demonstrates fast motion well.
Moving my hands quickly on PlayStation VR2 sees the virtual hands almost always keep up, with tracking loss being rare and reestablishing quickly. On Quest 3 on the other hand, owing to the 30Hz tracking, moving my hands at even a medium speed, anything faster than slowly, results in constant hand tracking loss, where your hand momentarily disappears throughout the movement.
WINNER: PlayStation VR2
Hand Overlap
Everything I’ve described so far has been in the context of your hands being separated, easily tracked separately. But what happens if you bring your hands together, overlapping one with the other and attempting hand-to-hand interactions?
0:00
Quest 3 handles hand-to-hand interactions well.
Quest 3 handles this significantly better than PlayStation VR2. Occluding one hand with another results in far less disruption, and finger tracking accuracy is more often maintained when fingers from each hand interact. Hand-to-hand interactions are thus far more practical on Meta’s headset than Sony’s.
0:00
PlayStation VR2 does not, for the most part.
WINNER: Meta Quest 3
Launching UX
Quest’s Horizon OS supports hand tracking throughout the system, meaning you can simply don the headset and use your hands to launch Waltz of the Wizard, even setting up the boundary if you need, without ever touching your controllers.
In contrast, PlayStation VR2’s system interface itself does not support hand tracking. You’ll still need to charge and pick up your controllers to launch Waltz of the Wizard. Once in the game, you’ll need to toggle passthrough to place your controllers down, then untoggle it to continue with hand tracking.
WINNER: Meta Quest 3
Conclusion: So Which Is Better?
In Waltz of the Wizard, PlayStation VR2’s hand tracking offers similar latency to Quest 3’s, yet handles fast motions significantly better thanks to being 60Hz by default.
However, even at 30Hz, Quest 3’s hand tracking delivers less jitter, superior accuracy, and better handling of hand overlap, as well as a far less frictionful UX to getting into hand tracking in the first place.
David Heaney
Of course, Meta’s hand tracking is currently on version 2.3, after seeing seen more than five years of software updates. Sony’s PS VR2 hand tracking was made available less than three months ago, so we’ll keep a close eye on whether the company improves it with firmware updates throughout the rest of the year and into 2026.
