negativegate
negativegate on Hacker News
Am I still SOL on AMD (9070 XT) when it comes to this stuff?
Would it be feasible to patch firefox or chromium to add the ability to dump the stream from <video> elements?
All the more reason to keep posting this stuff. People who do this seriously are why we can't have nice things.
nod-ai/SHARK from the original submission is by far the fastest way I've found to run Stable Diffusion on a 5700 XT.
For 50 iterations:
* ONNX on Windows was 4-5 minutes
* ROCm on Arch Linux was ~2.5 minutes
* SHARK on Windows is ~30 seconds
File-scoped namespaces is nice to not have every class already sitting at one level of indentation. I don't see what it has to do with nested folders.
I wish these phones had better US band support. Or that there were comparable phones for specs / price.
If you've also been vaccinated is it possible to distinguish that?
Edit: I suppose in this case they could test prior samples, but if you don't have any, you're out of luck?
Settings > Personalization > Start, turn off Show suggestions occasionally in Start
This is the first I've heard of Termux and now I'm curious what you use it for. Like are you SSHing into other environments?
It looks like it's broken dark mode support.
Huh, I never knew about this.
Go to Settings > Projecting to this PC to enable it. You may need to install the Wireless Display optional feature first.
Unreal Engine (and Unity) can target HTML5 / WebAssembly. I expect the main issue is the download size of the assets.
You're saying you can't visit any https sites without adding this certificate to your device's trusted certificates?
Did you think Typescript would be the native language in Deno?Yes? I thought that was the whole point. Actually I'm confused about the use of "native" here. It supports running both TS and JS files? And the TS files must be compiled to JS before they can be run anyway. So I'm not seeing any choice that was lost.
That’s what we all do now
I mean, some people do, but I don't. Not everyone uses React / Vue / etc, and afaik it's only React that is forgoing stylesheets. And it's not like stylesheets are dead or that styling elements through JS is non-standard. (I do use Vue, with stylesheets)
I'm not following this train of thought.
First of all C# doesn't run in the browser without compiling to WASM which has all sorts of caveats.
Second, I don't see how the fact that one could compile TS to C# would suddenly put everyone "on the .NET stack". Does the same risk exist if we can compile JS to C#? Which would work about just as well (i.e. not). And TS is also open source and produces reasonably readable JS, so it's not like there's even any lock-in.
Wouldn't it make more sense to be worried about compiling C# or other languages to WASM? Which I'm also not worried about.
PixiJS is the only one to reach above 100 FPS on my desktop, though that's inbetween frequent GC pauses. Two.js gets ~70 and Paper.js gets ~48.
There would probably be less GC pauses if the benchmark code wasn't doing things like
[...Array(this.count.value).keys()].forEach(...)
instead of a for loop.
I'm seeing <2ms in Edge Chromium and ~10ms in Firefox on a 144 Hz display. I'm curious how that compares to what other people are seeing.
I've been doing some WebGL work recently and I've noticed that while it reaches ~144 fps using requestAnimationFrame() in Firefox, there's a lot of stuttering. It's very smooth at 144 fps in Edge Chromium, while Edge Legacy is below 80 fps. As far as I can tell it's not CPU bound, and it's definitely not GPU bound. It would be nice if I could get it running smoothly in Firefox but I don't know what to investigate.
Fixed but it looks like it is still hiding behind a toggle on about:config.
It looks like you currently need to go to about:config and enable layout.css.backdrop-filter.enabled. It worked for me without enabling web render. Not that this does much good until it's on by default.
Data-uris don't make network requests.
Cool, I started working on something similar a while back but I didn't get past handling generics. Like if you have a node<T> with two T inputs and a T[] output, then when you attach a number to an input, the other input's type becomes number too and the output becomes number[]. That could need to propagate to connected nodes, too. Also when you disconnect a node, you would want to determine whether T must still be a number.
Is that something you're handling in your project? Do you have any pointers?
The bundle also seems to include dependencies for multiple architectures and PDBs for debugging. Files.dll which would contain most of the application is ~18MB. And since it does not seem to be a managed dll, it has probably been AOT compiled and likely contains the original bytecode in addition to the native machine code.
Interesting, though it does flag the push() as an error if you use "Array<string> | Array<number>" as the argument type.
It almost seems like some sort of co/contravariance issue?
A function that reads from an Array<string | number> should be able to accept either an Array<string> or an Array<number> (or Array<string | number>), but a function that mutates it should only be able to accept Array<string | number>?
While many companies were affected by downtime, only those that were using or had previously used Pardot were affected by the permissions issue.
Hipchat is dead so you don't need to worry about that.
So is this basically the server okaying a replay attack? I'm looking through the technical post but it hasn't clicked yet.
Maybe there will still be a UWP version on Windows 10?