Skip Navigation

AlmightySnoo 🐢🇮🇱🇺🇦

@ AlmightySnoo @lemmy.world

Posts

379
Comments

1352
Joined

3 yr. ago

Yoko, Shinobu ni, eto... 🤔
עַם יִשְׂרָאֵל חַי Slava Ukraini 🇺🇦 ❤️ 🇮🇱

Moderating

Google Pixel lemmy.world

2y ago

Removed

CNN blocks Firefox with uBo
Jump
AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

TIL!
and it's mobile-friendly and distraction-free too

2y ago

Permanently Deleted

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

My bad, I'll move there then

2y ago

CNN blocks Firefox with uBo

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

With how MS Teams and now CNN have been reported here to be blocking Firefox, you know that Firefox is doing things right. If web giants are ganging up against it, it's all the more reason to switch to it to make a statement and prevent big tech from making privacy violation the norm.

2y ago

I went back in time to kill Hitler but instead we fell in love and our romance averted the war but then I was kidnapped back to the present and Hitler's broken heart made him do WW2 anyway. AMA

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

I thought this was the name of an isekai for a second

2y ago

The quackening

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

They won't be spared 😔

2y ago

a miniature vase made on a potters' wheel, glazed and fired (cat for scale)

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

2y ago

Fedora 40 Looks To Ship AMD ROCm 6 For End-To-End Open-Source GPU Acceleration

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

Hard to tell as it's really dependent on your use. I'm mostly writing my own kernels (so, as if you're doing CUDA basically), and doing "scientific ML" (SciML) stuff that doesn't need anything beyond doing backprop on stuff with matrix multiplications and elementwise nonlinearities and some convolutions, and so far everything works. If you want some specific simple examples from computer vision: ResNet18 and VGG19 work fine.

2y ago

Fedora 40 Looks To Ship AMD ROCm 6 For End-To-End Open-Source GPU Acceleration

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

Works out of the box on my laptop (the export below is to force ROCm to accept my APU since it's not officially supported yet, but the 7900XTX should have official support):

Last year only compiling and running your own kernels with hipcc worked on this same laptop, the AMD devs are really doing god's work here.

2y ago

Pixel Launcher to soon allow users to change the default search engine

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

I just tried Neo Launcher and set it as the default one and split screen seems to work fine.

2y ago

Pixel Launcher to soon allow users to change the default search engine

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

It's lighter as I get to decide what to keep, no loss of functionality as I can still install the Pixel apps from the Play Store if I want to (but with the advantage of added security from GrapheneOS as I can disable their network access among other things), except for Google Pay NFC payments which don't work for now. We have a sticky thread about GrapheneOS if you want to know others' impressions: https://lemmy.world/post/9432155

2y ago

Fedora 40 Looks To Ship AMD ROCm 6 For End-To-End Open-Source GPU Acceleration

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

Yup, it's definitely about the "open-source" part. That's in contrast with Nvidia's ecosystem: CUDA and the drivers are proprietary, and the drivers' EULA prohibit you from using your gaming GPU for datacenter uses.

2y ago

Pixel Launcher to soon allow users to change the default search engine

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

apart from curiosity, the search bar is the main reason I switched to GrapheneOS on my 7a

2y ago

Fedora 40 Looks To Ship AMD ROCm 6 For End-To-End Open-Source GPU Acceleration

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

ROCm is that its very unstable

That's true, but ROCm does get better very quickly. Before last summer it was impossible for me to compile and run HIP code on my laptop, and then after one magic update everything worked. I can't speak for rendering as that's not my field, but I've done plenty of computational code with HIP and the performance was really good.

But my point was more about coding in HIP, not really about using stuff other people made with HIP. If you write your code with HIP in mind from the start, the results are usually good and you get good intuition about the hardware differences (warps for instance are of size 32 on NVidia but can be 32 or 64 on AMD and that makes a difference if your code makes use of warp intrinsics). If however you just use AMD's CUDA-to-HIP porting tool, then yeah chances are things won't work on the first run and you need to refine by hand, starting with all the implicit assumptions you made about how the NVidia hardware works.

2y ago

Fedora 40 Looks To Ship AMD ROCm 6 For End-To-End Open-Source GPU Acceleration

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

HIP is amazing. For everyone saying "nah it can't be the same, CUDA rulez", just try it, it works on NVidia GPUs too (there are basically macros and stuff that remap everything to CUDA API calls) so if you code for HIP you're basically targetting at least two GPU vendors. ROCm is the only framework that allows me to do GPGPU programming in CUDA style on a thin laptop sporting an AMD APU while still enjoying 6 to 8 hours of battery life when I don't do GPU stuff. With CUDA, in terms of mobility, the only choices you get are a beefy and expensive gaming laptop with a pathetic battery life and heating issues, or a light laptop + SSHing into a server with an NVidia GPU.

2y ago

Cloudflare Employee records her final meeting where HR tries to fire her

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

This. I don't think people here realize that HR doesn't really have a say in this, they aren't the ones deciding on the firing and they aren't the ones who can undo it since they aren't the ones providing the team's budget.

HR's job in these situations is to do the dirty part: handle the announcement to each employee and damage control if necessary.

The girl in the video is saying that her manager was "pleased" with her work and she didn't understand why strangers in the HR department are doing the announcement to her: that's the whole point, it's very likely that it's that "nice" manager who threw you under the bus when he had to make a choice on which people he needs to keep after top management told him to downsize his team but he didn't have the guts to tell you that personally.

2y ago

I don't...

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

Wayland reduces bugs

As I have to give a few lectures, I can't say I'm pleased with how screen-sharing or using a projector in the classroom fails almost half of the time and always embarrasses me in front of everyone. I've ended up purging the Wayland stuff and going back to good ol' i3 and I haven't had a display-related issue ever since.

X11 works, it may not be as sexy or modern as Wayland but it's battle-tested and just works and for the vast majority of the people, excluding Wayland's bugs, the differences are not even noticeable.

2y ago

Not even poor Notepad is safe from Microsoft's AI obsession

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

2y ago

Don't be that guy.

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

They're worse than us Arch users (btw)

2y ago

And never show it at public because I don't know how create folder in github

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

It's a lifelong ~~learning~~ nerding process

2y ago

Google Pixel (7a, 8 Pro and Fold) dominates MKBHD's blind camera test

Jump

AlmightySnoo 🐢🇮🇱🇺🇦 @lemmy.world 2y ago

You can install Pixel Camera from the Play Store. While it's proprietary, you can still just cut off network access for it on GrapheneOS, and honestly it's just superior.