Gemini vs. Ray-Ban Meta glasses: Which one guides me better in a Dutch supermarket?

Earlier this month, Gemini picked up a new ability called Guided Vision, which lets it look through your phone's camera and answer questions about whatever it's looking at.

I got curious, so I walked out to my balcony, aimed the camera at the view, and asked Gemini Live to guess where I was.

It named the exact neighborhood by recognizing a single building in the distance, which is impressive. The only thing is, my Ray-Ban Meta glasses have been able to do that for a while now.

It felt like time to put both to the test and see what Gemini's Guided Vision can do that the Meta glasses can't.

Related

I talked to Google Gemini Live for an hour; the lag was never the problem

Gemini Live fixed every awkward pause in AI voice chat. After an hour of talking, I learned the pauses were never the real problem

Posts By  Nimrod Aldea

Why a Dutch supermarket makes the perfect test

Thousands of products and not a word I can read

I recently moved to Amsterdam, and while I'm in love with the city, I have one major problem: I don't speak Dutch.

That's not as big a problem as I originally thought it would be, since almost everyone here speaks fluent English, but every sign, every advertisement, and every information board is still in Dutch.

That's why I chose to hold this duel in one of the hardest places to navigate as a non-Dutch speaker: the supermarket.

Everything is written in Dutch, thousands of products compete for your attention, big signs announce sales you don't want to miss, the packaging is covered in tiny print, and there are people all around you, silently judging. Or at least that's how it feels.

Talking to your glasses in public is weirder than it looks

And pointing your phone at a shelf isn't much better

As I walked into the supermarket, ready to start my very professional testing, the first thing I noticed was how self-conscious I felt.

Standing next to a stranger and suddenly saying "Hey Meta, help me find the sriracha sauce" out loud was something I couldn't bring myself to do, at least not until the woman next to me picked up what she needed and walked away.

When I finally asked, Meta didn't even take a picture. It just answered, "You can find sriracha sauce in the hot sauce section, and you can identify it easily by its signature green cap."

That wasn't what I was after, since I wanted Meta to actually look for it, but I can't deny that the second it said "signature green cap," my eyes jumped straight to the bottle on the shelf.

It's not like pulling out my phone, pointing it at the shelves, and asking Gemini for help felt any less awkward. Sure, people probably assumed I was on a video call, but the act still feels weird and unnatural.

Related

I didn't expect to love this Gemini feature, but now I use it every day

I no longer rely on Circle to Search

Posts 2
By  Marvin Velasco

Gemini's live camera found what the glasses couldn't

One photo against a slow pan down the aisle

Putting the awkwardness aside, how did the two do on the actual test? To keep things fair, I gave both the same request: help me find the coffee filters. The coffee section at Albert Heijn is huge, and I genuinely couldn't find them.

The Ray-Ban Meta glasses took a single picture and told me they couldn't find them here. And that was it.

Gemini came out of the gate swinging and showed just how much has changed in the last few years.

After I pointed my phone's camera at the coffee shelf and asked for help, Gemini replied, "I can't see it in front of me; perhaps you can pan around, and I will let you know when I spot it."

I took a step back and started slowly panning from left to right when Gemini suddenly exclaimed, "Found it! Go back to your left, and it's on the top shelf." It was.

This was a clear point for Gemini. The Meta glasses can also help locate a product based on what their camera sees, but they're limited to still pictures.

To scan the entire aisle, I'd have to go section by section asking, "Do you see it here?" and by that point, I'm basically scanning the aisle myself.

Being able to sweep the whole aisle with my phone and have Gemini stop me when it spots the product is genuinely useful.

It also made another customer stop in their tracks and wait for me to finish. So, instead of me waiting for someone to leave so I could talk to my glasses, this time someone had to wait for me to stop talking to my phone. Just as awkward.

The Meta glasses win when you're holding something

Fine print, chalkboards, and no phone in sight

As I kept shopping, the difference between Gemini and the Meta glasses kept growing.

I expected both to do the same thing, but I quickly realized each was better for a different reason, and the Meta glasses are more practical for translating or pulling information out of big chunks of text.

For example, I picked up a carton of orange juice that had its nutritional values printed in seven languages, in the tiniest font imaginable.

I could ask Meta for what I wanted to know, which was the amount of carbohydrates in the juice. I didn't have to look for it. The glasses just gave me the answer.

The Meta glasses also shined in the cheese section, in front of a giant blackboard where the types of cheese were written in chalk, in cursive, in Dutch.

When I asked what the sign said, Meta didn't just translate it. It explained that it was a list of cheeses and their prices, then asked if I wanted the price of a particular one.

That turned out to be the most helpful part of the trip: holding something in my hand and asking a question about it, or asking what a sign says. It also feels much less awkward than pointing my phone at it.

Related

Gemini 2.0: Five major changes you need to know

Gemini 2.0 brings new AI capabilities to Google's ecosystem

Posts 3
By  Nathaniel Wilkins

Gemini is the better guide, but I'd bring the glasses

Both Gemini and the Meta glasses performed perfectly. They could both find products on the shelf, read any kind of text, translate, and answer general questions.

While Gemini might have the edge in responsiveness and overall abilities, I'd rather take my Ray-Ban Metas on my next shopping trip.

I'm sure that if I were traveling somewhere new, Gemini would be the smarter choice, since being able to scan my surroundings and get real-time feedback would be useful.

But in the supermarket, I don't really need to be guided. I can usually find any product if I look for it.

My main takeaway from this shopping trip is that the answer to what I'm looking for is sitting right there. I just can't reach it yet.

My entire interaction with Gemini went through my phone, but I heard it talking back to me through, ironically, the Meta glasses' speakers.

What I really need is Gemini's capabilities running through the glasses. If Gemini could look through the Meta glasses' camera, I'd have the best of both worlds.

Maybe smart glasses need a camera and speakers, and you bring whichever AI you want.