An issue with my fly away with gps doing strange things was already opened weeks ago. I think got pretty much ignored, devs too busy debugging each and every card any chinese manufacturer puts out…
I am dev and I never play with chinese card so that is not true … I can also make false assumption
Some are indeed working with hardware as plenty user are using them, got issue because … you know … that chinese card so no support, but user blame ArduPilot for their lose …
We got very limited time (or deal with my wife…), but if you want I can try to teach you how to look for bug cause and how to make correction… most of the time it ins’t hard but need time to invest the true problem root correctly !
@anon67614380 Independent of what specific manufacturers do (that’s up to them), I’ll point you to this: https://github.com/ArduPilot/ardupilot/pull/10020
CANBus development is happening. I understand your frustration, but it’s a bit inaccurate in this case.
And that is not everything is fine (copyright Lego the movie) !
If there are problem, well, somebody will solve it …
As said in other threads, my job is to test stuff and risk machines. I have no knoledge to code or debug. All figures are needed.
If we all debug/write code and nobody flies than we have a problem, so to each its own. Otherwise whenever a dev asks for a log of a problem in flight we could answer, “i can teach you how to build a drone that carries 12KG so that you can fly it yourself and find out what the problem is”.
What i meant with my previous post and try to make a point out of it is that the system will eventually fall back if all the innovations are support for more and more boards. I think there are key features we still lack that are more important but it is just my point of view, so worth nearly nothing.
As said before, supporting too many different board flavours and deal with all the small quirks each has will lead to death.
For any new major function added there will be more and more thing to fix for the little differences in all the hardware supported, this will take ages and slow down to nearly no new features in future (wich is already happening). Just take a look at the latest release notes and you can see for yourself what the 70% of new code is about (70% is a lower bet).
If I may clarify. I think Corrado_Steri original concern was wondering when ProfiCNC will produce a native UAVCAN GPS.
In addition he believes that spending a lot of time porting Ardupilot to a lot of small F4/F7 boards is not the best use of Ardupilot resources, which I agree with.
Did I get it right Corrado?
AHAHAH!!! Absolutely correct
About ProfiCNC just to make things clear, they already produced an UAVCAN gps, the Here2, saying they would have released a firmware update to make it work using UAVCAN protocol.
Per board only one file is needed. Usually all the bugfixes land in that file. So it is a lot less complex to do it then you might think.
But I do agree that it does not come for free.
I have no tech background in programming so can’t get too technical, i base my comments on the release notes i see once new versions are released. It looks like a lot of time is spent in fixing bugs that appears only on some boards.
In my opinion too many boards are supported at the moment. As said, in my opinion, would be much better to support 3-4 boards and spend dev time (wich is precious) on new important features.
I don’t think where the bug fix lands is what makes a difference but the time needed to track it down and write a fix, that is what takes time.
On the Hardware Options page i can count 27 different boards, now that is not just wrong, it is plain crazy, in my modest opinion and can’t really see what good does it do to the system.
I cannot comment on @proficnc support model but I can certainly comment about ArduPilot Community.
My perception of a Beta Tester is a balance of knowledge between Hardware and Software added to mastering art of flying , analysing and finding problems. Being involved deeper into ArduPilot code is something that I consider as an essential tool in order to optimise the prototypes I am working on. I dont pretend to be a programmer but I know enough to get guidance from the team, get bugs fixed and enhancement being applied rapidly.
The support on ArduPilot is quite exceptionnal, I can work on a project on a Sunday Morning and get assistance in realtime from talented mainteners like @OXINARF. I have rarely experienced such support with paid products, that is generally go through a Level 1 chatting queue , that are filtering your request with stupid preframed questions, and sometimes they offer to upgrade to level 2 with additionnal $$$ , and you will never get acces to developpers, except for writing a wish list … ‘‘Dear Santa…’’
Maybe it is a matter of perception , but in my case I prefer getting envolved instead of going for a commercial autopilot and pay big time for every parts of it , including support.
Ok, understand your opinion and respect it, just do not agree.
Is there a reason you keep making new issues here?
I have answered you on another thread with the exact same question.
I have explained that we have hired full time Development team members to our staff to contribute to the open source Ardupilot code.
However, we have supplied a fully functional CAN GPS hardware, with a fully functional CAN boot loader.
The normal approach here is that the community, assuming there is desire, would then take the hardware and make software for this.
The price you pay is for hardware ONLY, we don’t charge for firmware, you get that for free from the awesome community.
Yet you are jumping on here, making posts attacking ProfiCNC, and Zubax, when we are all supplying you with great hardware.
Then you are attacking Ardupilot for supporting a massive variety of options for you to use. What right do you think you have to come to an open source community, without paying them a cent, and attack them for not doing things your way!
Without Zubax, we would not have any UAVCAN at all! And without ProfiCNC, Zubax would not have had his first hardware for CAN ESC.
We are a community! We cooperate, we work together for a common good.
But I’ve done a review of your posts, and to be frank, your not doing yourself or the community any favours.
CAN will be released on the hereV2 when I am satisfied that it is safe. That could be next week or it could be next year.
I won’t be bullied into releasing code early.
Please stop attacking Zubax
Please stop attacking ProfiCNC
Please stop attacking the Ardupilot team
Code will be released when it’s safe
Thank you for that,I hope it will be ready next week but am prepared to wait for a proper working system,and to be honest I really dont no if it will help me but I have A here V2 GPS since release and it is fitted and working great,new features are all ways welcome
Just asking a release date based on YOUR words “uavcan support is on the way” from 5 months ago. Rest of your post i’ll just ignore.
This post was flagged by the community and is temporarily hidden.
This post was flagged by the community and is temporarily hidden.
Gentlemen, please respect the forum rules
Sorry, to close the topic temporarily.
Will reopen in one hour, so that cooler heads will prevail.
Please use civility and keep on topic.