Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Chrome ships WebGPU

developer.chrome.com

Post reply on HN
Print view View original

331–340 of 405 posts

Re: Chrome ships WebGPU

#331

Post by tgsovlerkhgsel »

beebeepka wrote: post #273
tgsovlerkhgsel wrote: post #248

Earlier quoted context omitted.

I think browsers should distinguish more aggressively between "web application", "web site", and "user hostile web site". Many APIs should be gated behind being a web application. This itself could be a permission dialog already, with a big warning that this enables tracking and "no reputable web site will ask for it unless it is clear why this permission is needed - in doubt, choose no". Collect opt-in telemetry. We…

Do you have something specific in mind with your opening paragraph? Because defining what is a web site and what's an app, strikes me as particularly impractical idea. You correctly point out that yes, there are a number of powerful APIs that should be behind permissions. But there are a number of permissions already, so we need to start bundling them and also figure out how to present all this to the regular user. F…

News sites are a particular category that I expect to spam people with permission prompts, as they did when notifications became a thing. Without the deterrent of possibly landing in the naughty box, they'd all do it. With it, I still expect some of them to try until they land in the box.

Re: Chrome ships WebGPU

#332

Post by bitL »

Is there a PyTorch port running on WebGPU somewhere? So that I could add local processing for ML pipelines into a webapp, bypassing cloud.

Re: Chrome ships WebGPU

#333

Post by johntb86 »

yamtaddle wrote: post #275
codedokode wrote: post #168

Earlier quoted context omitted.

Everything that can be used for fingerprinting should be behind a permission. Almost all sites I use (like Google, Hacker News or Youtube) need none of those technologies.

Main thing that ought to be behind a permission is letting Javascript initiate connections or modify anything that might be sent in a request. Should be possible, but ought to require asking first. If the data can't be exfiltrated, who cares if they can fingerprint? Letting JS communicate with servers without the user's explicit consent was the original sin of web dev, that ruined everything. Turned it from a user-co…

If javascript can modify the set of URLs the page can access (e.g. put an image tag on the page or tweak what images need to be downloaded using CSS) then it can signal information to the server. Without those basic capabilities, what's the point of using javascript?

Re: Chrome ships WebGPU

#335

Post by tormeh »

kllrnohj wrote: post #240
tormeh wrote: post #36

Earlier quoted context omitted.

Not really. I don't see how any of the vendors could break it without being extremely overt about it. And it has no direct competition. Nothing else runs on any OS and any GPU, except WebGL, which is abandoned.

Vulkan & GLES are both the obvious competition here, I don't know why you seem to be ignoring them?

Vulkan doesn't run on Apple platforms. And Apple refuses to upgrade their OpenGL drivers. So no, they don't count. Then there are the consoles, which also have their own APIs. Expecting all vendors to implement a common API is a fool's errand, and we should stop trying. The vendors don't want this, and we can't force them. Any common GPU API must be an abstraction over vendor-specific ones.

Re: Chrome ships WebGPU

#336

Post by tormeh »

kllrnohj wrote: post #242
tormeh wrote: post #88

Anyone know the reason Google created Dawn instead of going with Wgpu? Attachment to C++, or NIH syndrome?

What does Mozilla use Spidermonkey instead of V8? Why did Apple create B3 JIT instead of just using Turbofan? Competing implementations are a cornerstone of standards. Indeed in many domains it's a requirement for a spec to have multiple compliant implementations to be considered complete at all.

The older I get the more I disagree with this POV. All other things equal, a single open source implementation is superior to several ones. Several implementations lead to duplication of effort, both for those developing them, and more importantly for those developing for those implementations. Software has to be tested separately for each implementation, often with vendor-specific hacks.

Re: Chrome ships WebGPU

#337

Post by geysersam »

pjc50 wrote: post #108
geysersam wrote: post #98

Earlier quoted context omitted.

Why would I want bit-identical output? Genuinely curious. I see there's some increase in confidence perhaps, although the result can still be deterministically wrong...

It's very hard to do tests of the form assert(result == expected) if they're not identical every time. And it can waste a horrendous amount of time if something is non-bit-identical only on a customer machine and not when you try to reproduce it ...

It's not that hard. You'll just have to decide what level of accuracy you want to have.

Asserting == with floating point numbers is basically a kind of rounding anyway.

Re: Chrome ships WebGPU

#338

Post by victor96 »

paulryanrogers wrote: post #246
victor96 wrote: post #245

This is super exciting for us as we develop interactive browser UIs. I hope future versions of WebGPU will be backwards compatible to all those with Chrome 113. We need an iOS like pace of updating for browsers!

> We need an iOS like pace of updating for browsers! How do you mean? Browsers often update every 6 weeks. IOS releases with new behavior are annual.

The difficulty is the number of people still running old versions, as soon as new iOS versions come out a super high percentage of people switch. And with browsers you have to account for the update cycles of every browser on the market.

Re: Chrome ships WebGPU

#339

Post by BulgarianIdiot »

sebzim4500 wrote: post #45
BulgarianIdiot wrote: post #3

Now I want to see someone figure out a way to performantly distribute a trillion parameter model on a million smartphones looking at a web page.

The bandwidth would end up costing way more than the compute would.

Depends how you do it. LLM are huge, but their input and output is miniscule bits of text. If you find a way to put "narrow paths" in the hidden layers, basically to subdivide a model into smaller interconnected models, then the bandwidth will be similarly massively reduced.

This is not without precedent, look up how your brain hemispheres and the regions within are connected.

Re: Chrome ships WebGPU

#340

Post by codedokode »

mastax wrote: post #252
codedokode wrote: post #168

Earlier quoted context omitted.

Everything that can be used for fingerprinting should be behind a permission. Almost all sites I use (like Google, Hacker News or Youtube) need none of those technologies.

So CSS should be behind a permission?

CSS should not leak fingerprinting information. After all this is just a set of rules to lay out blocks on the page.
Post reply on HN