Regardless of budget, I have found the following setup has afforded me all the comfort upsides of mobility and console gaming, with none of the performance downsides.

1. Build a standard desktop gaming pc to your budget, setting aside ~$150, give or take.

2. Make sure it’s wired into your network and not using wifi. Setup Steam on it as usual.

3a. (Console experience) Buy a Google TV with Chromecast, or whatever it’s called now. Install Steam Link app on it and connect it to your gaming pc. Get a Bluetooth compatible Xbox controller, connect it to the chromecast. Enjoy a console experience with your gaming pc. If you have the chromecast on a wired ethernet lime you’ll have maybe 1ms of input lag, very playable.

3b. (Laptop experience), buy a dirt cheap laptop, install steam on it, use Steam Streaming fu ctionaloty to stream from gaming pc to laptop. If you plug the laptop into ethernet you should have sub 1ms input lag.

This let’s you get all the horsepower of a gaming pc, at gaming pc hardware prices, but the portability of a laptop and/or couch gaming comfort of a console.

And since it’s all centralized to your 1 “server” machine, of you make changes in setup A (ie change am in game setting or etc), it’ll persist even if you swap over.

IE if I change my settings or preferences on the console, I’ll persist that over on my laptop and won’t have to change it again.

Furthermore no network save game synching needed, no waiting for a game to download a second time, no need to update the fane multiple times, etc.

It’s all centralized to your own core machine and everything else is just a thin client.

PS: this works with the Steam Deck too, you can stream from gaming pc to steam deck and use it as a thin client 👍

Dunno why ppl are down voting you, this is 100% the way.

Architecture as code is amazing, being able to completely wipe your server, re-install fresh, and turn it on and it goes right back to how it was is awesome.

GitOps version controlled architecture is easy to maintain, easy to rollback, and easy to modify.

I use k8s for my entire homelab, it has some initial learning curve but once you “get it” and have working configs on github, it becomes so trivial to add more stuff to it, scale it up, etc.

The one thing that sucks is this doesn’t cover gear stored in the extra mog… whatever it was called storage you could pay extra for.

And you can’t pay just for that inventory, you have to pay for your account before you can pay to enable the storage.

And finally, if you had important shit stored in those bonus inventories, you can’t access it til you pay for it

Result: players that prior paid for and used the extra inventories basically can’t leverage the free login, they can login but can’t access their stuff unless they opt out of the free login and fully pay for the month + inventories…

RIP

They genuinely looked identical to me.

Either way, gogs dies what I need it to, git server for backing up my code and super basic git web Hooks to trigger my build server.

Couldn’t ask for anything more.

Gogs and Gitea are very similiar, Gitea is a fork of Gogs with a bit more features as I understand it.

However when I tried to get Gitea working personally a year and a half ago, it had some rough issues with redirect looping onto itself infinitely, could never get it to work.

On the other hand Gogs didn’t have this issue, and was much more painless to stand up, so it’s what I use now.

You have to be able to convey business value to get approval on anything corporate deems “extra”

At the end of the day, the project manager is going to have to be able to “prove” that color blind settings will translate to $$$ to the people above them, and not only that, but reliably more $$$ than it will cost to implement.

Which means first you need to know how much money it actually is likely to make, and we have actually very little data on what % of gamers that enjoy (genre) are colorblind.

So you’re already off to a pretty dang rough start.

Usually you only actually get these features when the CEO themself has buy in, like, “Oh yeah my cousin is colorblind and told me how much games suck about it, so make sure we include that feature”

Thats pretty much the only way you’ll be seeing that sort of inclusivity, when you have direct buy in to the movement of inclusivity coming from the very top at a company culture level.

Closer to a week or two, speaking as an actual software dev.

You have to first include the investigation into “how do we do it? What our are best options?” which is a day or two

Then the couple meetings as you go over your findings and get the sign off and approval that you can go ahead with it.

Then a couple days to implement it, write some tests for the code.

Another day for all the documentation to be added to Confluence, detailing all the above.

Another day or two for the code review process back and forth.

Another day or two for the QA testers to validate things are working.

There’s many many steps involved in going from “Idea” to “Implemented, reviewed, and tested”, and the human element in the back and forth stretches it out as you wait for people to take their lunch breaks, join the zoom meeting, the usual “your mic is muted mate” “oh jeez sorry” back and forth, etc etc…

I was able to connect to the DB with Cloudbeaver, but it straight up wasn’t providing the diagram tab in the way the picture said it outta. The example pic even specifically is using a postgres DB as its example!

I pretty much had the exact same view, but no diagram tab. Unfortunately the wiki article doesn’t go into much detail, it just says:

“(if the tab is not presented then the object does not support the diagram presentation)”

With no information provided further listing off what is, and is not, supported for diagram presentation.

Lack of documentation it seems, which is unfortunate. It seemed like it has potential but I spent a good 20 minutes fiddling with it, trying different configurations and settings, nothing made it start working and it seems like (as is the case on a few of these tools) the ERD tooling is often a bit of an afterthought and poorly supported.

Many of the tools are sql first, ERD… third? fourth? forgotten and lacking most features :(

I have no idea how to respond to this. I wrote up a few attempts but it just felt weird.

Why do you think I need to explain what ERD stands for, in the context of my question?

Trying it out, the wiki says it has an ERD editor, but its documentation is kind of lacking.

It’s example image here: https://github.com/dbeaver/cloudbeaver/wiki/Entity-Diagrams

Shows it interacting with a postgres database, but when I try the same I am not getting a Diagram tab. Its also proving to be pretty awkward to try and work with.

So far best I have found is Azimutt, which is pretty close to what I want but its interface is lacking atm, and I couldnt get it to successfully connect to my postgres database in the end (kept giving NOT FOUND errors even though I tested inside the docker image to validate the connection and it could indeed TCP the postgres database’s port)

Let me know what specific questions you have that my post did not cover and I can try and help out.

I feel like I specified exactly what I am looking for and even gave some examples of existing (non self hosted) tools, so not sure what else to add.

Are you familiar with what an ERD is, with respect to database management?

Have you used tools like Luna/Moon, Navicat, Visual Paradigm, etc?

Im looking for such a tool, but self hosted via docker in the form of a Web App.

I have a K3OS cluster built out of a bunch of raspberry pis, it works well.

The big reason I like kubernetes is that once it is up and running with git ops style management, adding another service becomes trivial.

I just copy paste one if my e is ting services, tweak the names/namespaces, and then change the specific for the pods to match what their docker configuration needs, ie what folders need mounting and any other secrets or configs.

I then just commit the changes to github and apply them to the cluster.

The process of being able to roll back changes via git is awesome

If you set a secondary DNS server, your clients will just use it to get ads, negating the effect of your pihole.

Unless, of course, the second DNS server is a second pi-hole ;p

When the raspberry pi (running pi hole) is down for any reason

You can have 2 DNS servers provided over DHCP assignment, if this is really an issue for you (ideally it shouldnt) you should be running 2x PiHoles for failover protection, that way you can take 1 offline and all devices will auto swap to the second one for DNS.

This will also horizontally scale your DNS querying, as devices will just 50/50 flip flop between the two while both are online and it will sort of auto-load balance.

Largely speaking thats your best bet though, just 2x PiHoles for failover protection.

For DHCP just use one of the pi-holes DHCP, or your routers, or whatever, they all largely perform the same but I personally find the pi-hole’s DHCP settings to be easiest to work with (you can even super fast modify them by SSHing in and using nano/vim to text edit the config file, so instead of using a UI you can just copy-paste all the entries in quickly)

If you need to take the DHCP pi-hole offline, you can just flip DHCP on for the second one, however as long as new devices arent connecting you don’t actually need the DHCP server online once everyone has an IP assigned until their registration time is up (24 hours by default) so as long as you get the pi-hole back online quickly, you’re network typically wont even have a problem.