ITT people who don’t understand that generative ML models for imagery take up TB of active memory and TFLOPs of compute to process.
That’s wrong. You can do it on your home PC with stable diffusion.
And a lot of those require models that are multiple Gigabytes in size that then need to be loaded into memory and are processed on a high end video card that would generate enough heat to ruin your phones battery if they could somehow shrink it to fit inside a phone. This just isn’t feasible on phones yet. Is it technically possible today? Yes, absolutely. Are the tradeoffs worth it? Not for the average person.
deleted by creator
“He’s off by multiple orders of magnitude, and he doesn’t even mention the resource that GenAI models require in large amounts (GPU), but he’s not wrong”
deleted by creator
You can for example run some upscaling models on your phone just fine (I mentioned the SuperImage app in the photography tips megathread). Yes the most powerful and memory-hungry models will need more RAM than what your phone can offer but it’s a bit misleading if Google doesn’t say that those are being run on the cloud.
So much for the brilliant AI-specialized Tensor processor
It’s basically just a mediocre processor that offloads interesting things to the mothership.
Yep. What a joke. Goes to show you that google could Make these pixel features available to all android devices if they wanted to.
This really doesn’t surprise me.
Yeah, obviously. The storage and compute required to actually run these AI generative models is absolutely massive, how would that fit in a phone?
They aught to just slap a phone in a 3090
Fuuuuck that.
Why ?
Because it’s a privacy nightmare.
Using Google products has always been a “privacy nightmare” - it’s not like this is some mega open source phone or anything it’s literally Google’s flagship. Is this really surprising? Playing with fire gets you burned.
Googles phones are the easiest to de-google surprisingly
deleted by creator
I’m not surprised at all, I’ve been trying to not use googles services for years
And if you do, they won’t upload your stuff to the cloud anymore.
Wait - is that praise or criticism?
both! some people want Google’s features and don’t care about privacy.
In what way ? Everything you do on phone Google has access to and therefore any govt body as well.
Privacy on a phone is a myth unless you use a modded phone
Even ignoring all the privacy issues with that, it’s kinda shit to unnecessarily lose phone features when you’ve got no signal
That’s how phones have always worked. As long as they are not automatically ingesting data from people without permission. I don’t see an issue with this. At least in this instance we have some choice in what we send or don’t. It’s no more privacy nightmare than mid-Journey doll-e or any of the others.
Isn’t that kinda the dream. We have devices that remote the os. So we get a super powerful device that keeps getting updated and upgraded. We just need a receiver ?
Isn’t that what we want. Can reduce down the bulk on devices. Just a slab with battery, screen and modem and soc that can power the remote application ?
Sometimes that’s what people dream about. On the other hand that hybrid cloud model is giving up the last remnants of computing autonomy and control over devices we own.
What I would like is something that gives me the framework to host my own server-side computations at home.
That would be, if Google wasn’t constantly killing things that didn’t do good enough. Especially given how expensive generative AI can be to run remotely. Just look at what happened with Stadia
Also, it just feels disappointing. Ever since chatGPT, they’ve been pouring near infinite budget into stuff like this by hiring the top talent, and working them to the latest hours of the night. And the best that they could come up with for the pixel 8 is feeding it data from the cloud.
And I can’t even really believe the whole “consumer hardware isn’t powerful enough” thing given that there’s quite a few ARM processors, Apple especially, That’s been able to build consumer hardware capable of performing generative AI (I’ve personally been able to run stable, diffusion and whisper on my M1 MacBook). Maybe not at the scale or quality of the cloud, but still capable of doing so regardless.
deleted by creator
I mean it sucks for offline situations or for actions that need very good latency.
But the phone’s battery would be happier if the processing is not done locally.
For some things I prefer to do the stuff locally, for other things on the cloud.
Some cloud Vs local examples I’m thinking about:
-
For example (example not related to the Pixel) if I’m generating an image with stable diffusion at home I prefer to use my RX 6800 on my private local Linux rather than a cloud computer with a subscription. But if I had to do the same on a mobile phone with tiny processing power and battery capacity I’d prefer to do it on the cloud.
-
(Another non AI example): for gaming, I prefer to run things natively. At least until it is possible to stream a game without added latency. Obviously I don’t play games on a phone.
-
Another example (Google photos): I would prefer to connect Google photos to a local NAS server at home, and “own the storage on premise” and then pay a lower fee to Google photos for the extra services it brings. Browsing, face and objects recognition, etc. But this is not an option. With this I would be able invest on my own storage, even if I had dozens of gigabytes in 4K60FPS videos.
-
Obviously, I mean, Google did so well with Stadia.
/s
Stadia was great. People were just attacking it. Bad marketing from the get go. Have you tried it ? Did you just follow what everyone else says or actually try things for yourself
No, how could I try it? it doesn’t exist anymore. Poor tiny little startup Google, bullied by the big bad user mob who paid for the service they sold and actually expected to receive what they had bought.
Removed by mod
I’m not from the USA, how could I possibly have tried it? It died before it could’ve even become a product. Even if they would’ve gone to rural areas or the developing world it would’ve failed because there’s not sturdy enough infrastructure to support the service. And as the experience proved, not even South California has enough infrastructure to support the service without massive lag. I remember reading that even people with Google Fiber had issues. So it was a failed service for a strong technical reason, not just the awful marketing and even worse service offering. Paying a subscription and buying games was the most retarded move Google made at the moment, but they make so many stupid moves so frequently that it’s hard to tell if it was the worst.
Removed by mod
I tried it. It worked reasonably well when wired and pretty terribly wireless, which isn’t that surprising. The free starter kits they gave out came with a Chromecast Ultra for a reason: it had Ethernet on the power brick.
On the other hand, I hated it ideologically from the start. You bought games that were only available on their service and could only play with an active Internet connection. That’s also to be expected. Not ideal, but not a big deal in itself so long as you know what you’re getting into. What isn’t okay is that you could lose access to the service permanently and there was no guarantee you would be compensated in any way whatsoever. At most they gave vague “don’t worry about it wink wink” answers when asked in Reddit AMAs. Fortunately they did refund everyone when they shut it down, but they were under no legal obligation to do so. People who bought in were lucky. I wish I could say it was purely out of goodwill, but it’s more likely that a bean counter at Google decided it was cheaper to refund everyone than to shoulder the bad press from not doing so.
Thank goodness there were also no notable exclusives on the service. The idea of games just completely disappearing from history just because Google got bored with a product like they always do is nauseating. It’s bad enough that they bought out Typhoon Studios only to pretty much immediately shut them down. Again, thank goodness they were able to rise from the ashes and reform the studio under a new name.
Interesting. I only had access to it wireless. I used a OnePlus one VPN to North America and HDMI to a screen. It wasn’t flawless but I didn’t expect it to be.
Oh I agree but also I don’t. Ubisoft games are still the same. They can drop access at any point and they likely wouldn’t refund. Unfortunately unless we build an open source internet that is backward compatible I think we are fucked.
Will Xbox and PS5 games work without a patch. Could they yank a game and stop it being playable? Can steam epic gog ?
Yeah there was a game that was exclusive? Was that the one you mentioned?
Fuck Google but I loved the idea and now use GeForce. Similar concept I suppose
Good idea, while they’re at it, they could tweak their prices accordingly.
This just strengthens the argument to install privacy/security first operating systems like CalyxOS and GrapheneOS. I don’t was a phone that’s more a service subscription than it is hardware. I have the pixel 8 and didn’t get the pro due to the offloading to Google servers for some “features”.
Just waiting for GrapheneOS to be released for the 8… Until then, I’m sitting uncomfortable knowing my phone is uploading telemetry to Google servers…
I’ve bitten the bullet, replaced my Redmi 13 to a Pixel 7 + GrapheneOS - because the MIUI spyware and google spyware are just too much… I still don’t get why , as an owner of a phone , I’m not the owner of the phone.
deleted by creator
I can’t put grapheneOS on my p7 since at&t sells phones with locked bootloaders. I didn’t know this when I bought it or I wouldn’t have bought it.
Oh yeah, only unlocked phones can boot custom os, carrier locked are not. Unfortunate you had to find out the hard way. I could’ve sworn it was on graphene’s website or something.
Well, yes, however, installing GrapheneOS may mean not having access to “features” which were part of the marketing for the phone… Which begs the question around what are you paying for? To me, it’s like you are paying for a product and also for services. Although the p8 and p8p both have 7 years of support, I wonder if that extends to features and services that are cloud based?
deleted by creator
Yea, Google features are very anti privacy, it’s funny how it’s still one of the most secure phones on the market. I got the pixel 8 for the fact its going to be supported with GrapheneOS, and get 7 years of security updates.
Currently using GrapheneOS on my 8 Pro, and the experience has been extremely smooth so far.
Yea I was tempted for a second to get it, but there was a comment that it “should” be possible to upgrade to the released version when available, I don’t want to have to mess around in the event that it’s not. So I’m just going to wait a week or so.
deleted by creator
People didn’t think this was the case already?
When some company praises the ground braking AI capability of a new SoC they have been built, you might get the idea that it’s doing these tasks on said SoC.
Why would you think otherwise?
A list of what this phone does offline and what not would be great.
And?
deleted by creator
So no more pixel phones for me.
How did you think it worked before this?
deleted by creator
deleted by creator
I think you have no clue what the g3 is capable of. Magic eraser for example can be used entirely on device. THAT’S what even Google admits.
So now tell me why it needs to upload the pictures…just for computing? 🤡
Well, the Magic Eraser is just an algorithmic fill. It’s not an actual generative ai. So it doesn’t require so much compute power. If Pixel 8 had the new generative Magic Eraser on device, it would be either really bad or take minutes to process.
It is generative ai via ml. What are you talking about… If it wasn’t then it proves even more my point. Why uploading it for computing?!
It’s obvious that you don’t understand this topic at all. So it would be better to end this discussion. But I will try to explain. Old Magic Eraser: definitely not generative ai, because it doesn’t “generate” anything - just completes the missing part with the other parts of the image. So it’s not that resource heavy and can be processed on device. It doesn’t require complex diffusion generative models. New Magic Eraser: uses actual generative ai, because it generates whole new objects that replace the part you want to erase. And that requires a lot of processing, because it needs large image generation models that currently can’t be run on the device. Although, Pixel 8 could run some small image generation models, they would be far worse than the large models that Google runs on their servers. In the future, the technology will get better and more lightweight. So in a few years, it will be possible to run a good image gen models on device. But that’s not the current focus of the industry.
Are you not capable to read the press statement from Google itself? GOOGLE SAYS THEY COMPUTE MAGIC ERASER ENTIRLY ON G3.
Now they also upload the photos AFTER the process.
I think we both misunderstood each other. I looked up the press release and they say that there were some generative improvements to Magic Eraser that run on device. But that’s not in the “Magic Editor” UI I meant, when I was talking about the new features of Pixel 8. You were mad at Google because of the upload to the cloud. That happens ONLY if you use the Magic Editor that uses large models in their datacenters for best results. But if you use the normal Magic Eraser it will work on device and hence offline. Although it does use some generative AI now - I 100% guarantee you that it will be worse than the Magic Editor, which uses image models that are too powerful for the Tensor G3. So you were angry for nothing.
Pixels have never been “stock android” instead they have been close to stock but with extra AI features over it (e.g Magic Eraser, Now Playing, etc)
Surprisingly they’re the easiest phones to degoogle as well
