HN user
I’m not entirely convinced, there are many dimensions to progress. For example, DeepSeek has had a few very impressive innovations that all models could benefit from. There’s also the law of diminishing returns, the US labs have plenty of CAPEX already.
Sometimes, constraints, like sanctions, can also be a source if innovation.
How is /e/ less open than Graphene? As far as I understand, they are both pretty open minus firmware that they can't control?
I'm actually curious if there's something I don't know about /e/
I'm also on Emacs and Niri, seems to be a popular combo, and I don't have any performance problem.
At home I'm driving an Ultrawide (3440x1440@75Hz) at scale 1.1.
At work I'm driving two 4k screens at scale 1.2.
I might be less sensitive to latency but it could also be a graphics driver issue or something similar. I'm using Arch with emacs-wayland (pgtk) with a strix halo (all AMD) laptop.
People are paranoid that everything they read was written by a LLM. It's likely that they are even more suspicious because you used one to code.
You should be proud of both the work you did and the article :-).
The sales are a nice thing but for me the biggest benefit of Steam is knowing that I can use my games library on other/future devices.
My games have been working on my desktop from 10 years ago, the SteamDeck, my laptop and likely any future computer I buy that runs Linux.
Why a team... give me a few days ;-)
I'm using a Strix Halo laptop (~3k, 64GiB) and with Gemma 4 and Qwen 3.6, both at 8 bits, I'm seeing very impressive results.
As a work tool, this is reasonably priced. You can save a bit of money by opting for a non-laptop form factor.
It only reversed it for within the US, I learned that when the company I worked for (owner was a US company) closed.
Nothing of that is angering me, I live in the EU and life is pretty nice.
My comment above was meant to provide context and to point out that both sides of the argument can be legitimate.
The comment is neutral, you shouldn't be able to infer what I think about the EU from it. It simply points out facts and says that reducing a point of view to being uninformed is too easy.
My stance on the EU is complex: it does good, it does bad, it affects each member states differently, it changes over time. Those are all pretty neutral statements too, only siths live in extremes ;-).
I think the grandparent references many referendums that rejected EU treaties:
- Maastricht Treaty, 1992, Denmark, - European Constitution, 2005, France and Netherlands, - Lisbon Treaty, 2008, Ireland.
The votes were usually redone and passed. The exception is Norway which refused to join twice.
Clearly not everyone loves the EU but the majorities are very much in favour (and certainly this is the case among the people who actually understand politics, economics, etc.).
This is quite the loaded statement. I've heard from people on either both sides of the argument that understood politics and economics very well (professors, lecturers, etc.)
I had a similar experience to your family member. I have a 2 year old and when she started eating solids, the smell got to me much more than before.
Still not too hard but much harder in my case ;-).
I had many conversations with a former boss about the Azure sales team. They would come in, say they can do it cheaper, simpler and better — he was immediately convinced.
I would do a calculation based on their public price plan and come up with a 5-10x price compared to the bare metal OVH solution that perfectly fit our use case. I would then ask the sales team where I made a mistake in my calculation and hear nothing back.
A few months later, they would come back with the same pitch and the whole process would repeat...
Many of the proposed EU digital solutions require a Google or Apple verified phone. This makes escaping American companies difficult.
I got a HP g1a for about 3k€ with 64gb of ram when it came out
It should be easy with a Q4 (quantization to 4 bits per weight) and a smallish context.
You won't have much RAM left over though :-/.
At Q4, ~20 GiB
I've been using opencode and would be curious to try something else. What would recommend for self hosted llms?
I'm really curious what will happen to this space when Valve releases great open source drivers for the Qualcomm chip they have in the Steam Frame. It might be one of the first, very powerful, GPU accelerated SoC you can buy that has mainline support.
I can image having a very usable ARM linux laptop and tablet as a result of this — maybe even cellphone when the modems get mainlined or used via USB.
This will allow you to distribute your creations to a limited number of devices without going through the full verification requirements.
How can they count the number of devices you install the app on without being the ones to give a permission to install it?
They took nothing back, they are still putting in place the requirement that Google gives permission to install apps on your phone. They are misleading us about it too which is also terrible.
I made a time sync library over local network that had to be more precise than NTP and used i128 to make sure the i64 math I was doing couldn't overflow.
I32 didn't cover enough time span and f64 has edge cases from the nature of floats. This was for Windows (MACC not GCC) so I had to roll out my own i128.
I'm using niri with two screens at work and it's been very nice. I don't open windows on the side as you suggest but I believe that can be done with custom bindings and/or window rules.
This is how I also read it.
it's still the best OS by a mile
I'm not so sure about that, modern Linux is pretty good — I was able to configure it to fit my needs much better than I could a mac. It's also free of dark patterns (looking at Windows).
If you're willing and able to configure Linux, I would say that, for some people, it's much better than a mac.
Wouldn't it be accurate to say that you can no longer install apps on your phone, only Google can?
Those are very good prices, it's 3700€ in Austria
The fan curve also drives me nuts. It seems to be related to the processor's speed so I turn off boost — I don't notice a perf diff for my workloads. The fan noise is much better.
echo 0 > /sys/devices/system/cpu/cpufreq/boost
I have one as well and haven't had this problem. I've been using mostly the HP G4 dock and a Dell dock at the office with an occasional use of the charger. I'm running Linux 6.17.8-arch1-1, it could also be fixed in newer kernels. When I play demanding games, it pulls quite a few watts and all the chargers are keeping up.
It might be your device that is unlucky, maybe contacting support might help.
I have an HP ZBook g1a ultra in Austria. I've had mine for about 4-5 months and I got the 64gb version because I needed a new laptop and the 128gb wasn't yet available. It should now be widely available everywhere.
It's pricey but I have to say it's the best Linux laptop I've used. I'm a Linux user so for me that's as good as it gets :-).
I had the opposite experience going from a OnePlus 8T stock to Lineage OS. Having root means being able to reduce the amount of apps and wake up — no google play service was the key. This was a while ago but I went from 1-2 days of battery life to about 4-5 days. This is with light use, screen on time was equally draining with both setups.
I would assume that an iPhone has similar amounts of unwanted background apps and would also be able to gain battery life instead of losing it if rooted. Obviously if you install spyware, you lose a lot of battery life. Funnily enough, I remember that a few years ago, people were surprised to find that uninstalling facebook increased battery life because it behaved much like spyware.
I'm not the AI expert in the company but one of my colleagues creates image segmentation models for our specific use case. I've been able to run the PyTorch training code on my computer without any issues. These are smaller models that are destined to run on Jetson boards so they're limited compared to larger LLMs.
edit: just to be clear, I can't train anything competitive with even the smallest LLMs.
I have a Strix Halo based HP ZBook G1A and it's been pretty easy getting local models to run on it. Training small LLMs on it has been a bit harder but doable as well. Mind you, I 'only' have 64 GB with mine.
Under Linux, getting LM Studio to work using the Vulkan backend was trivial. Llama.cpp was a bit more involved. ROCm worked surprisingly well with Arch — I would credit the package maintainers. The only hard part was sorting out Python packaging for PyTorch (use local packages with system's ROCm).
I wouldn't say it's perfect but it's definitely not as bad as it used to be. I think the biggest downside is the difference in environment when you use this as a dev machine and then run the models on NVIDIA hardware for prod.