nromiun

11 ms·

181.
by nromiun 1y ago
Exactly. The rich take huge bets to get even richer and if they fail the poor people are the most affected. See all the hedge funds margin called, 2000 recession or 2008.
182.
by nromiun 1y ago
An economic recession is not selective. It will affect all of us.
183.
by nromiun 1y ago
So much money is being pumped into this whole thing that we might get a global economic shock if/when it unravels.
184.
by nromiun 1y ago
I bet not even 5% of all programs are multi-threaded, or even concurrent. Memory safety is a much bigger problem.
185.
by nromiun 1y ago
Not every program needs thread/task cancellation. Somehow people have been convinced that threading is the same as goto and it is obviously the wrong thing to do. goto is goto, you can't take anything you dislike and say it will d
186.
by nromiun 1y ago
They don't "need" to maintain backwards compatibility. Several major languages have broken it by now.
187.
How to handle people dismissing io_uring as insecure? (2024)(github.com)
122 points by nromiun 1y ago|108 comments
188.
by nromiun 1y ago
Slightly off topic but anyone knows when/if Google is going to enable io_uring for Android?
189.
by nromiun 1y ago
Every number on computers is converted to binary internally, so yes this works on decimal numbers too.
190.
by nromiun 1y ago
Very well for my banking apps. With root and developers options enabled on my phone as well. If your banking apps does not work complain to your bank.
191.
by nromiun 1y ago
Famous because of how rare it was.
192.
by nromiun 1y ago
For benchmarking code like this CUDA, HIP and OpenCL are almost the same. You will only see the difference in big codebases, where you launch multiple kernels and move data between them. Otherwise OpenCL is very good as well, with the added
193.
by nromiun 2y ago
What about unified memory? I know these APUs are slower than traditional GPUs but still it seems like the simpler programming model will be worth it. The biggest problem is that most APUs don't even support full unified memory (system
194.
by nromiun 2y ago
Easier said than done. Social media companies have virtually unlimited money to promote their platforms.
195.
by nromiun 2y ago
> For a long time, well after we had a stable release, people kept claiming Asahi Linux and Fedora Asahi Remix in particular were “alpha” and “unstable” and “not suitable for a daily driver” (despite thousands of users, myself included,
196.
by nromiun 2y ago
How many people only turn on GPS for maps? Aside from that all tracking methods are pretty inaccurate anyway.
197.
by nromiun 2y ago
I don't know about this model but people have been running local models in Android phones for years now. You just need a large amount of ram (8-12 GB), ggml and Termux. I tried it once with a tiny model and it worked really well.
198.
by nromiun 2y ago
So basically Nvidia is like Windows and AMD is like Wine. I think trying to emulate CUDA and using forked Nvidia libraries is not the best strategy for AMD. They should have made a clean break and come out with a fresh API, like Apple'
199.
by nromiun 2y ago
I really want to know about their growth rate. Their valuation has already priced in several decades of insane profit, so I want to if they will be able to pull it off in a decade or two.
200.
by nromiun 2y ago
BDFL model seems to have worked pretty well for so many decades. Guido should have just transferred power to someone he trusted. Now look at him, he can't even comment freely.
201.
by nromiun 2y ago
Very sad to see this happening to the Python community. Maybe we need a fork with only technical discussion allowed. This CoC/vote stuff seems to poison every community it touches.
202.
by nromiun 3y ago
Both, it is an inherent drawback of using chiplets.
203.
by nromiun 3y ago
I also see a 7 years old private repo from my account. But I may have made it public initially, I don't really remember it. Anyone knows how to check if a repo was ever made public?
204.
by nromiun 3y ago
AMD desktop chiplet CPUs draw almost 3x power compared to Intel at idle. It makes a huge difference in overall power consumption. Of course they are much more efficient in the top end, but my PC is not running full load most of the time, so
205.
by nromiun 3y ago
There are a lot of them: https://getdeploying.com/reference/data-egress
206.
by nromiun 3y ago
I know AWS has a reputation for being expensive overall. But it is crazy that it has the cheapest VMs. See here: https://cloudoptimizer.io/ But of course other things like egress, serverless services more than make up for i
207.
by nromiun 3y ago
In this case the PC is the proxy, which sometimes does not have a public IP address (behind ISP NAT).
208.
by nromiun 3y ago
If anyone wants to use GitHub from their IPv6 only VMs I would recommend a SSH reverse tunnel. It works like this: https://unix.stackexchange.com/questions/116191/give-server-...
209.
by nromiun 3y ago
It's not really a mystery. Compare Intel Macbooks with M1.
210.
by nromiun 3y ago
> And sadly, Nvidia does not support OpenCL’s FP16 extension, so FP16 throughput couldn’t be tested What a bummer. But is this really true? I know Nvidia does not report the cl_khr_fp_16 extension, but I saw somewhere that you can still

More ›