Rock PI4c

[britus]

Armbianmonitor:

http://ix.io/2xKH

Hi,
I’m using this panel:
https://osoyoo.com/2019/11/12/instruction-for-raspberry-pi-7-dsi-touch-screen

 

I've installed image:

Armbian_20.08.1_Rockpi-4c_bionic_current_5.8.6_minimal.img

 

The system does not work with the Raspberry PI Touchscreen. The OSOYOO panel works fine if connected to Raspberry PI 3B+.

Any solutions?

 

 

[Igor]

3 hours ago, britus said:

Any solutions?

Armbian forum for RK3399 does not provide / cover solutions on requests - only limited support for onboard hw. 3rd party hardware is not covered. 

General support terms:

https://github.com/armbian/build#support 

If you really want this to happen, you will probably need to hire an expert.

Moving to Features requests.

[britus]

Dear Igor, thank you for your explanation. But I think it doesn't help me in any way. I was already in contact with Tom von Radxa. I wish that the cooperation, especially for the mainline kernel, would be on a higher and more professional level. For me it is incomprehensible that the DSI port is not taken into account in the mainline kernel. According to Tom, there is support for your project. I am having a bit of a problem with both worlds (Armbian Mainline Kernel and the outdated Radxa Kernel). I also think that with my 40 years of experience in software engineering, I do not necessarily need a dedicated developer ... My ambition is too great to solve the problem here. However, I am not very familiar with the current state of the entire kernel and the DTS feature is new to me. I 'know' the kernel from another time - at least a decade before today ...
Anyway, thank you for the answer and I wish you a healthy and good time
Regards Bjoern

[NicoD]

2 hours ago, britus said:

Armbian Mainline Kernel and the outdated Radxa Kernel

Armbian Legacy kernel is not from Radxa. It is the same kernel for al RK3399 devices.
If it works for your goals then why use mainline? Use what works, don't try to make it work for your use.

 

2 hours ago, britus said:

I also think that with my 40 years of experience in software engineering

Then you should know nothing is really that simple to do.
Linux support for SBCs is a work of many people with many years of experience. Everyone has his own skills that combined created what you use. You can't do that on your own. Linus(or Igor) didn't do it all by himself.

[Igor]

14 hours ago, britus said:

But I think it doesn't help me in any way.

 

I am sorry, but helping has limits - we are just a few and project doesn't cover costs while there are thousands of you seeking for our attention, help, serious development. Both costs time and emotional enerrgy. I would easily lost 500 EUR of time per day just by talking to you - demand is very intense - and I really hate doing that. Since I can't afford to waste that much for exactly nothing, I have to forward you to predefined texts. Which you obviously didn't understand.

 

14 hours ago, britus said:

For me it is incomprehensible that the DSI port is not taken into account in the mainline kernel.

That small problem - porting all features from legacy kernel to mainline but without maintaining them - costs million(s) or several hundred of thasands at best. Luckily most of those millions were already poured into the code, otherwise you would not have support already on such a great level. Usually also not all features are ported (by pro and community) and you are on your own. Asking us to sponsor that for you is not just funny, but also outrageus.

 

14 hours ago, britus said:

According to Tom, there is support for your project.

We are talking, yes, but until today we only have friendly relationship, which means we haven't been able to hire and train developers that will support project maintanace and perhaps - if we were extremly lucky - know something about camera subsystems, that will be able to solve your "small" problem.

 

Its a bit harder to pay for the time we lost dealing with the code and you. On average hourly rate we currenty lost around 3000-5000 EUR per day and only a few % of that is covered by industry, where Radxa thinks to join. Around 0.2% is covered by general public /donations. More and more % of work comes to the areas that no-one is willing to do. Voolonteers don't do the dirty job - just what's fun and interesting.

 

When this is covered in 100% we can talk about professional levels. Until then, its a project we do whatever we like to do and you are also welcome to join.

 

14 hours ago, britus said:

I also think that with my 40 years of experience in software engineering, I do not necessarily need a dedicated developer ...

I would assume you are retired or near? Most of us are also highly experienced with decades of work in the area, but mainly we have full time jobs and work here on a side ...  my job is the same, but its not near as demanding and tensed as this. There I rest from this projects demands. Which is also a main reason to have it. To have a safe playground.

Do you really think we need more work? And why we would move scarse and precious resources to your case, skip 1000 other people that asked for some help before you ... and leave the project in ruins. We are too small to do both.

There are several companies that do driver / kernel development for hard cash. Not sure if you can easily find one that has time and is cheap enough, but that is not our problem. Our budget for investing more into - your business or afternoon project - is exactly zero. Why is that so hard to understand???

 

Without adding a single line of new code, we would need few full time developers to releave overstressed staff if nothing else. The greatest problem is actually not developing, but maintaining and implementing that code - into a live and moving mainline kernel code - since you expect top quality also weeks/years after making a POC.