fulafel

18 ms·

421.
by fulafel 11mo ago
That might exclude a lot of your user base. For example a big chunk of Android users, or Linux workstation users in enterprise settings who are on older LTS distributions.
422.
by fulafel 11mo ago
For Python: PyTorch has Vulkan support according to https://docs.pytorch.org/executorch/stable/backends/vulkan/v... - wonder how performance is there.
423.
by fulafel 11mo ago
It's also totally US centric, whereas the blog post is written to sound general. (They do cop to it in the caveats section)
424.
by fulafel 11mo ago
Vulkan Compute is catching up with HIP (or whatever the compatibility stuff is called now), which seems like a welcome break from CUDA - in this benchmark it beats CUDA in some benchmarks on AMD: https://www.phoronix.com/rev
425.
by fulafel 11mo ago
The context in the GP comment was generally getting v6 connectivity for containers. "Internal" is a context dependent term that you introduced. But to give a use case for that, for example you might want to have (maybe at a future
426.
by fulafel 11mo ago
For a lot of use cases a major advantage of IPv6 is to get away from ambiguous rfc1918 addressing. You can then just put an allow rule between arbitrary v6 addresses anywhere on the internet when you need connectivity without any other hack
427.
by fulafel 11mo ago
> my state will need at least 5GW of power to literally keep the lights on. I think this abstraction is missing the elasticity of demand that can by unlocked by end-to-end dynamic pricing. Probably if the production was cut in half for s
428.
by fulafel 11mo ago
Intel seems vulnerable to Trump tariffs, would seem more likely than TSMC getting into trouble.
429.
by fulafel 11mo ago
Vulkan compute is also getting some good press as a local llm platform (at least on the linux side), will be interesting to see which crosses the line to "can ship production quality apps on this" first.
430.
by fulafel 11mo ago
You might think that a dGPU is always faster but the limited memory capacity bites you there (unless you go to datacenter dGPUs that cost tens of thousnds). Look at eg https://www.ywian.com/blog/amd-ryzen-ai-max-plus-39
431.
by fulafel 11mo ago
I guess it's the physical HDMI port that's needed, as Minis and the Pro laptops have working monitor HDMI monitor support?
432.
by fulafel 11mo ago
Does consumer hardware (non-MI) need proprietary kernel drivers for running rocm + pytorch?
433.
by fulafel 11mo ago
It'll be interesting down the road to see how this affects the rate of security bugs that are not memory safety related.
434.
by fulafel 1y ago
How does it compare with SST?
435.
by fulafel 1y ago
Well, the brits eventually decided against it after a thorough investigation found it to be a waste of money - "a January 2018 report by the National Audit Office found that the UK had incurred many billions of pounds in extra costs fo
436.
by fulafel 1y ago
You don't need an external machine. Since games are set up to allow twitch etc streaming, it's easy for apps on the same machine to get access to the video.
437.
by fulafel 1y ago
His recent stuff is pretty relevant to tech. Eg https://www.yanisvaroufakis.eu/category/books/technofeudalis...
438.
by fulafel 1y ago
Don't know about britain, but generally a popular neoliberal thing with PPPs is to copy the private sector and "turn capex into opex". If you lease a road or school for 20 years, it doesn't show up as debt in your books,
439.
by fulafel 1y ago
I had an accidental "not" in my comment (and edit window closed). Yep, that's indeed the message.
440.
by fulafel 1y ago
It's not indeed more at risk than previously thought - https://www.politico.eu/article/gulf-stream-could-collapse-l...
441.
by fulafel 1y ago
Germany allowed it in 2021 or so. The biggest variation is above 6 GHz: most of the world allows 5.9-6.4 but reserves 6.5-7 GHz for cellular or haven't decided yet if it'll be for wifi or cellular. There's a nice map on http
442.
by fulafel 1y ago
It would be good if we had a term that didn't confuse linking with translation. In English compiling means joining together many parts, after all.
443.
by fulafel 1y ago
Not the case - lots of games including AAA ones have these things on the Steam Deck.
444.
by fulafel 1y ago
I still suspect JS. It's much harder to shoot yourself in the foot with Python. Even if you use JSON: >>> json.loads('{"nr": 348555896224571969}') {'nr': 348555896224571969} >>> t
445.
by fulafel 1y ago
What's the Android situation there? Last I heard Google didn't license Android there and they were using Chinese app stores with forked AOSP Android. Which would seem to put the sideloading decision in the hands of the forked OS.
446.
by fulafel 1y ago
The Steam Deck AMD chip is rumored to be a design for the Magic Leap 2, not for Valve.
447.
by fulafel 1y ago
Smells like floating point. Python prompt: >>> int(float('348555896224571969')) 348555896224571968 It just exceeds the mantissa bits of doubles: >>> math.log2(34855589622457196) 54.952239550875795
448.
by fulafel 1y ago
Thanks. They seem to be pretty proactive indeed if you look at the feed: https://lore.kernel.org/linux-cve-announce/
449.
by fulafel 1y ago
There are other CVE numbering authorities you can report a vulnerability to and apply for a CVE, or appeal, but this does possibly have a chilling effect if the vendor's CNA refuses valid vulns. (Like with MS in https://news
450.
by fulafel 1y ago
They can if someone manages to develop an exploit. Let's not confuse vulnerabilities and exploits.

More ›