What are the advantages and disadvantages or each?


The PX4 looks like the very old days when APM had many boards and no case. Just looking at the advertisement I have no idea why I would want to trade in my APM 2.5+ for a PX4?




Views: 51765

Reply to This

Replies to This Discussion

According to the main OP guy...

You are reading way too much in to this Kevo, PixHawk have a new product planned and are basing it on the OpenPilot code base, the intention I think is so we can work together one one code base rather than fragmenting things.

We have a few members of PixHawk here and one even has a CC, they are good guys. No one is accusing anyone of anything but there is a lot of curiosity as to what is being planned and if we can work together, certainly no one is saying anybody has done anything wrong, with a project like OpenPilot, keeping things in the open is what we are about.

For me personally I would like to see OpenPilot kept as one big happy family and welcome new people and we all work together. The problem with working in secret is it causes these issues.

Not long after that I heard that Pixhawk gave OP the cold shoulder, signed a deal with 3DR, and took down all their schematics.

Most servos operate based off a PWM signal, I'm not suggesting there's any real point in trying to change that.  If you want to support all the receivers out there you'll have to deal with PPM.  OTOH it would be easy enough to get rid of that in favor of an all digital solution like the 3DR radios.

Using an analog signal like PPM for inter-processor communication just seems plain foolish to me.  Given the limitations and problems you guys have had I'd think you'd be eager to use a proper digital method.

There's no hardware support for PPM and you guys have a whole chip already dedicated to PPM and PWM. With as tight as resources are on the APM you would think you'd be trying to use the built in hardware peripherals wherever possible.  I2C, serial, and SPI were designed for inter-chip communication.  Bit-banging an analog signal between chips just seems wasteful.

I haven't looked into the history of PX4, but If that is the case, they would not be the first to leave OpenPilot and I am not all that surprised. There seem to be A LOT of internal struggle in that project, with lots of negativity and disgruntled developers leaving to fork their own projects. Reading the OpenPilot forums is one of my guilty pleasures that I sometimes do for the entertainment factor.

And let me be clear that this is my personal view from reading theirs and other forum discussions, and not related to 3DR or DIYd in any shape, way or form. So no miss use of quotes please..

I'm really surprised how much calculated deception goes on in the autopilot game.  OP has always seemed pretty positive to me, but they sure have had their problems.

I guess that's what happens when you can't get the hardware out.

The balance between commercial interests and open source has always been tricky. But that's why you have licenses. If you don't agree with a project license, don't contribute. It's really that simple. But sadly some people seem more interested in making semi-political standpoints then writing actual working code. I can't help but think how ironic it is to see the close mindedness of some "open" source defenders.

And the idea that sharing code, functions and ideas between different open source projects, somehow is a bad thing (especially with autopilots) is just ridiculous. And shows no respect for the true spirit of open source.


You're spot on.   As for right now the APM is better supported and in my opinion is the better option for keeping things flying without bugs.  In the future as the PX4 is developed, I think we'll see cool uses (like 360* sonar) etc on the PX4.

John, I do the same thing, occasionally perusing their board.  I had the same impression as you.

I just read through that thread on OP: "Hey Pixhawk guys", and unfortunately, it seems the same as so many other things I've read coming from OP.  That is, a lot of baseless accusations, with no supporting evidence.  It's like an echo chamber.

It's exactly the same thing that happened during my first interaction with DAnkers.  He made statements that Arducopter was stealing/copying their code.  I asked for him to provide a single line of code which he could show demonstrates that fact.  His response was basically "I'm too busy to do that."

Well, if you're going to make an accusation, you better be ready to back it up with evidence.

Now the fact that Gary Mortimer started it... ?

But I love the part where one guy says basically "oh yeah, it's true, just search on google".  The response is "I did, there's no evidence of it".  There's no response to that.  Where is the evidence? That thread proves nothing about the accusation, other than as I say, it can be like an echo chamber.

I'm not interested in the politicking. 

Drone.. I'm not entering a pissing match with you, other then to say that you clearly missed what I meant by "true spirit of open source. Hint, it has nothing to do with licenses, and more to do with philosophy.

For example. I have put a lot of time into the ArduPPM code. But I could not care less if for example OpenCopter decided to start using it. As long as I get to read the code to see how it evolves in the hands of others, I'm happy. And if they did some "bad" changes I would most likely try to help then out.

Kevin, the take-down notice had nothing to do with the fact the hardware was cloned.  The take-downs were because they violated a copyright.  The APM name, and in some cases, the entire silk-screening to include the 3DR logo.  These were not merely a case of somebody else making boards that were the same, they were literally *cloned*.

The inevitable result of which, is potential customers thinking they were buying authentic hardware.  And thus when hardware problems occurred, they would call 3DR for support.  That is the part that is unfair, and surely you can agree with that.

AFAIK, 3DR has taken no action against RCTimer after they changed the name to ArduFlyer.

And on a personal level, can you really fault Chris for having hard feelings about the fact that somebody would simply make a carbon-copy of the APM, without putting any effort into improving it?  I mean, mano-a-mano here.  I think you'd be a little annoyed too.

I mean, there is LOTS of room for somebody to make improvements to the APM, and I'd love if somebody did.  But they didn't.  It's rather disappointing to me too, and I don't have a dog in this fight.

I've been thinking about buying an ArduFlyer just based on price.  But I'd rather have the option of buying a competitive products which have different features and functions instead.  How about having the IMU seperate like we've all be crying for?  How about an remote mounted servo output rail?  How about a daughterboard to take a decased Rx?  How about a GPS unit with a built-in Mag?  How about a power supply that better suits my needs?  There's lots of room for improvement, but all we get is a cloned board with a made in China pricetag. 

And to John's point... again you bring up the issue of licenses.  John was talking about "the spirit" of OS.  The OP project has it's own issues with it's OS license.  As evidenced here, they release their code under an open license, and then roam around the internet accusing all the other OS AP's of "stealing" their code?  How does one "steal" OS code?  I understand attribution would be in order.  I'd be totally on-board with you if OP ever provided evidence of this "stealing".  But all I've seen are the accusations.

Even today, if you google "Pixhawk Open Pilot" you really only get 3 valid results, all of which are threads where OP people are talking about Pixhawk stealing the OP code, where the only evidence given is "just google it".

One of those valid results is now this thread.  The second is the one on the OP forum.  And the third is another thread on this forum, where Jake brought up this exact same point.

@Drone: Design a hardware ppm encoder? It's an ATMega chips with an external 16mhz crytal and some load capacitors. And the rest is selecting and tracing out pin connectors for servo inputs. There is no design to speak of at all, just putting the chip on a board. When we "redesigned" the pin outs for the 32U2 version to make it work as a Arduino USB converter and PPM encoder at the same time, it was literary a 5 minute job.

OK, to be honest maybe I used a couple of hours to make sure we selected the right pins, and Jordi a couple more to trace out the PCB board..

I'm not dissing PCB layout. it's an art form to get just right with no electrical problems, especially at higher frequencies. But the PPM encoder is as simple a PCB as it gets.

Yes, in future the PX4 or equivalent would be the natural evolution. That said, I think a lot more benefit will come from improving what we have, in stead of adding more features. Take the tuning process of a multirotor Arducopter vs naza: Multiple rates each with a P I and D for different modes for Arducopter, vs. 4 sliders for naza. I'm a Arduplane guy personally but I was shocked when a friend of mine switched to naza after months of labouring the PIDs on an Arducopter for months. Within 20 minutes on the naza it was flying great.

If someone could build a tuning tool such as that, the existing (brilliant!) features of Arducopter would be leveraged.

Another important thing to keep in mind is that even if we where using lots of OP code in our project (we aren't using any to my knowledge), it would not be stealing. The notion that we or others are stealing code/ideas is pointless. As you said yourself, as long as both source code projects are released under the same GPL license, it is technically impossible to steal any code. If they don't want others to reuse their code, don't release it under a GPL license.

Reply to Discussion


© 2020   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service