Still the best option for Orange Pi 5!

[guynich]

I just updated an Orange Pi 5 server OS - it was running an older joshua-riek version with 22.04 and 5.10 kernel.

My goals:

• Ubuntu server version 24.04
• Linux 6.x vendor kernel
• rkmpp hardware acceleration

I test installed these OS server versions

• Armbian (25.5, May 2025)
• Orange Pi "official" (v1.2.2, July 2025) - Ubuntu 22.04 server only
• joshua-riek (v2.3.2, September 2024)

I selected joshua-riek (v2.3.2). The others were good but lacked rkmpp hardware acceleration. Bravo!

EDITED: clarified Orange Pi official server version I tested was 22.04 (24.04 server not available from Orange Pi v1.2.2) so didn't meet my goals

[BEEFY-JOE]

Unfortunate that this project seems dead since Joshua announced his leave.
Which is a bummer since I am also here same device and same goals.

[briantopping]

From a couple months ago, #1231 (comment).

If we put ourselves in his shoes, this is a relatively thankless job. The board vendors aren't bad people, but motivating them to pay for something that already works is very. difficult. It's also difficult to get them to pay for something that doesn't work at all or something that supports their competitors.

How do we make this work for everyone? Joshua is our man... he knows these boards like nobody else. But until we can convince the board sellers to help pay his expenses, nothing changes!

If we can get the project working again, then when certain boards start don't work with new releases, we encourage the board vendors to get involved. New board? Get involved. Etc.

Ultimately what has to happen is everything is broken down into Debian packages and integrated back to the package repositories. Joshua started this project because nobody else was doing it. Presumably he did it the way he did is because building everything as packages is a lot harder than building a board support package tree with correct dependency resolution. My guess is it started off as something very small and he didn't stop. It's exhilarating to create something that a lot of people use, but this script setup isn't sustainable long term. I've personally been in the situation where what I did was better than anything else and still wasn't good enough to take it to the next level. It hurts.

What I proposed here is that we start slowly converting the scripts to packages, working from the bottom to the top. For instance, let's convert all the patches to packages. Let's not change his top level scripts to start with, rather lets first prove the process works with moving some parts to Debian packages and build some momentum. The rest will take care of itself.

[frozen-cdn]

Original model Orange Pi 5 checking in. Was curious to see if there's anything else out there besides the single working Armbian Vendor Gnome image. Nothing else works properly, or has any acceleration in Chromium.

I guess I found out about this too late and it's been abandoned.

What a shame!

[briantopping]

What a shame

Where is the shame? Who owns this shame?

Have you tried to fix it? What questions do you have about contributing? Do you think it will ever get fixed if all people do is “shame” the people who have contributed so far?

[frozen-cdn]

I don't think you understand English, that's not what it means whatsoever. Where did you even come up with that?

[guynich]

Original model Orange Pi 5 checking in. Was curious to see if there's anything else out there besides the single working Armbian Vendor Gnome image. Nothing else works properly, or has any acceleration in Chromium.

Curious: there are many Armbian OS builds for Orange Pi 5 - please could you share which is the one you referred to?

[guynich]

Just to clarify—my original post was meant as a celebration of the great work done here. I appreciate the efforts that went into it and simply wanted to acknowledge that.

[briantopping]

Thanks! I am just another user, but I think we can do better than shaming the project, its state, or anyone involved with it. What it needs is resources and momentum - enough to make up for the loss of the most important person on the project. How can we get there?

[guynich]

Thanks - I like your questions and motivational suggestions