As far as I know for the new Vulkans layers and dx12 implementation there is a “translation layer” from the old dx implementation to the most updated one. This is the main reason why old games runs faster on Proton than in w7 for the same hw. Even if they were designed for w7 specifically.
Last time I checked this was done during the booting of the game, but i have to admit this was time ago and it could have been changed.
Ummmm sure?
I don’t want to start that extremely old flame war of native VS jit code but…
Proton is not an emulation, it is a translation to native code, and while it has some drawbacks (more memory usage, more time at start up to compile things) it can unlocks a lot of potential when the hw support new capabilities, this is the reason that some dx10 games run faster on Linux…
Nginx is considered battle tested.
Very few products have this level of puic scrutiny and and a good record of being safe.
Once this is said, the majority of problems come from misconfigurations, so triple check the things
OK I understood that the request came from internet and tailscale was to link the reverse proxy and the server.
In this case try ipv6, pretty sure you have ipv6 and you will have a public address.
But for this case you will need a dns in your network so example.com can be resolved and then your proxy will make the right request.
Yes doable.
Yes, it is possible, but you need a domain (example.com) that I guess you want to be resolved from internet and a public face ip.
After that, yep, if the reverse proxy can resolve tailscale names (basically it has tailscale installed in the same machine) and the service is reachable via tailscale, then it is perfect.
In fact in my setup I have a public domain name that is translated into a private domain name in the reverse proxy (exactly what you want with the addition of tailscale)
Did you run the new recommends?
Usually this is enough for regular drivers
It is almost impossible to achieve those numbers, specially the idle, and the load we can only to be close to them when we use the watt/ per-useful-instruction metric.
recentest hw will have decent idle consumption (n95 is around 2w) but the rest of the hw will consume more and usually you attach more powerful/energy hungry devices so expect <10w in idle and perhaps <25w under load.
Older hw could not be so bad under load but they will idle less and will consume more
Strange, I didn’t have any issue, but it is true that I have an old gtx.
First of all, connect both displays to the same board, or you may have strange issues with the vrr and accurate color display.
Second use this Guidefor the drivers, once the repos are added you can use the graphical package manager for selecting the Nvidia drivers.
During the installation, Nvidia drivers needs to be compiled for your specific kernel version, so it always takes a bit of time.
And nothing else to do, Nvidia are always a pain in the ass, specially with his tendency of making the drivers incompatible between minor kernel updates, so if you didn’t select it during install, activate the automatic snapshot for the zfs before and after updates, in case there is something not properly updated you can rollback, continuous working for a few days until all the changes are propagated to all packages and update again.
Start with a nas, the rest will naturally come when you try to access your data for outside, or organize your data, or save more data types to your nas.
Your nas should be the central device and you build the rest around it.
Now, The question is, which nas? I would recommend synology, they are not too performance, a bit expensive and the company is lately doing suspicious moves, but the sw and the hw are rock solid and they are quite good for beginners from almost all angles. Extra point for how many howtos and tutorials are present in internet.
Once you are comfortable with them, you will realize the rest
How, because Overton window says the opposite
Honestly, this seems more than your assumptions and fears than anything else.
Clients are open source and foss, the moment something strange is in, Headscale could adapt.
If they finally goes to full close source, then the community will fork (emby - jellyfinn drama).
Your arguments don’t hold together, for the good of the community
I really don’t see how the enshitification could work when we have the fee version of the central servers Headscale
So yes, it could be a pay wall for some advance featuresike funnels and so on. But the primary use is secured…
You are partially right, I was fast and sloppy and I gave the impression all is jitted and it is not the reality.
The part of the translation is fine. However there are parts that are compiled beforehand (shaders for example and I can recall something about arm or other architectures, not sure now). And this is a crucial point of the extra performance, because some parts can be ported to more updated/efficient implementations, not because there are less services in the background.