Non-rechargeable batteries is a terrible idea from an ecological point of view.
Most people won’t replace an internal battery. If the battery dies, they buy a new controller.
I’m not sure whether this is actually better for the environment than even if they were using non-rechargeables.
You’re already able to. I wonder whether they’re using Waydroid.
I mistakingly wrote 6600, it’s similar to a lower clocked RX 7600. Though it’s performance is not that much different than the 6600. It’s a okay GPU. It’s a entry level GPU, i.e. the most inexpensive to get while still getting good value.
deleted by creator
In my experience Bluetooth is really bad because of its inconsistent latency. Even if the latency doesn’t appear to be much higher than with cable/2.4GHz dongle, it feels much worse (similar to WiFi Vs Ethernet).
It’s likely these issues are worse because my Bluetooth dongle isn’t great, but I avoid Bluetooth like the plague if I want to have a good experience.
Edit: Though I have to say that taking a break from something after trying for a while without improvement can actually result in rapid improvement. A part of it is mindset (calm, without expectations) but I do believe the break also helps the brain to learn/reinforce.
They are not worth it. First party controllers are expensive for what they offer, given there’s better gamepads for less money.
Just buy an 8bitdo Ultimate 2C or another one of their controllers with the 2.4GHz dongle. The wireless dongle has less latency than wired xbox/PS5 controllers, not to mention Bluetooth.
YeetMouse includes substantial improvements like switching to fixed point math and having a GUI.
There’s some really bad mouse accelerations built into desktop OS. For example Windows’ mouse accel curve is an abomination and I really don’t know how they came up with it. Maybe they threw a dice. [1] Comparatively a curve like [2] is much more suitable for gaming, as it essentially allows for precise tracking, small flicks with constant sensitivity, while still allowing for rapid 180 degree turns. Especially quake-like arena shooter benefit from this as you can play long range and close range with the same settings (Quake is where this kind of mouse accel was first implemented).
[1] 
[2] 
With the right curve accel is not necessarily less consistent.
I’ve used linear acceleration, a limit of 2.0 and an for me suitable offset with the same consistency as without. This is because the acceleration is kicking in only after an acceleration threshold (offset) has been passed. So unless I really yank my arm, there’s no difference between having accel on and off.
The result is that normal aiming (tracking, small flicks) benefits of exactly the same muscle memory with and without accel on.
You can think of it as having two sensitivities: one for tracking, and one for quick 180° turns. Because of the linear increase in sensitivity between the two sensitivities (one caused by the offset, the other one set by the limit), it feels quite natural and not like “traditional” mouse accel.
The following 1 minute video explains this kind of custom mouse acceleration quite well.
It’s packagekit which is slow. I’ve used Gnome Software on Fedora Atomic and it is quite fast (since a few big optimizations about two years ago) because it only has support for flatpak enabled.
Niri already had support for X11 through xwayland-satellite, although $DISPLAY and xwayland-satellite had to be started manually.
That said, being enabled by default is great and the other changes are awesome too.
Idk whether there’s delay, but this also applies to minor version changes of Nvidia drivers. They must match exactly.
The benchmarks go against the narrative that Windows and Linux are pretty much equal in performance. I’ve read regularly that Linux is “often” faster than Windows for gaming, especially from more recent Linux users.
5 years ago, 15% performance difference were the expected performance loss through DXVK and wine/proton, so these benchmarks would’ve been the expected result.
waypipe is a proxy for Wayland clients. It forwards Wayland messages and serializes changes to shared memory buffers over a single socket. This makes application forwarding similar to ssh -X feasible.
Dev needs to eat.
Arch requires reading the manual to install it, so installing it successfully is an accomplishment.
It’s rolling release with a large repo which fits perfectly for regularly used systems which require up-to-date drivers. In that sense it’s quite unique as e.g. OpenSUSE Tumbleweed has less packages.
It has basically any desktop available without any preference or customisations by default.
They have a great short name and solid logo.
Arch is community-based and is quite pragmatic when it comes to packaging. E.g. they don’t remove proprietary codecs like e.g. Fedora.
Ubuntu is made by a company and Canonical wants to shape their OS and user experience as they think is best. This makes them develop things like snap to work for them (as it’s their project) instead of using e.g. flatpak (which is only an alternative for a subset of snaps features). This corporate mindset clashes with the terminally online Linux desktop community.
Also, they seem to focus more on their enterprise server experience, as that is where their income stream comes from.
But like always, people with strong opinions are those voicing them loudly. Most Linux users don’t care and use what works best for them. For that crowd Ubuntu is a good default without any major downsides.
Edit: A major advantage of Ubuntu are their extended security updates not found on any other distro (others simply do not patch them). Those are locked behind a subscription for companies and a free account for a few devices for personal use.
I thought the verses might be relevant in some way to the project but they seem to be pretty generic.
Weird choice to put them at the top of their refreshingly comprehensive readme. I believe I’ve seen projects with a quote at the top before, although they’ve been shorter.
Some books have quotes before each chapter and they can be inspiring and/or interesting.
They explained quite well why those comparisons aren’t valid either way. They are freezing their setup from updates so performance claims are mostly invalidated almost immediately etc.
They going to lengths trying to remove variables and compare apple to apple, which is not possible when comparing different platforms.