Flight Controllers; The Good, The Bad, and The Ugly
- Thread starter Murocflyer
- Start date
I just have to laugh everytime I read that sentence from whatever source, pay a premium price for a board with advanced features but only able to say it flies great in manual mode. Oddly so does my KK 2.0 with the latest firmware and it's about $400 less expensive, seriously...
The best thing to do is rather than just take a list of available flight controllers, first define what it is you're looking for and what features are "must have" vs. would be nice to have. The more features, the more expensive and/or complex it will be to get a system up and running with all of the features functioning as they should.
Once you have a clearly defined list of features and functions then research each of the possible flight controllers for how well those features work on that board. If you just publish a list all you'll get is the respondents favorite and how it's better than all the others, many of which they only know by hearsay which could lead to a not so satisfying experience.
Best way to see how well things work is find the forum where that particular board has the most active postings and do some reading, things that are problem areas will be readily apparent as those will be posted about repeatedly. That can also be an indicator of how easy a particular system is to build and tune, if relative newcomers are having a lot of problems then you may want to pass on that system if ease of setup is one of your criteria.
I have at least one of each of all the boards on your list except for the Quad4 which I've heard of but never researched. Whenever I'm asked which is better the answer is always the same, "it depends". It depends on how much or how little you want to be involved in setup and tuning and what you expect from it. It depends if you want superior advanced functions that your use case will rely on to be 100% functional 100% of the time, some are light years better than others. If you have little to no experience and need a LOT of support that will influence the decision as well and you may be better off having a ready to fly system built for you by a reputable vendor that is geographically close so any problems could reasonably be resolved by perhaps an in person visit.
I suggest you take a big step back and start by defining tthe use case and what is most important then go from there.
Ken
R_Lefebvre
Arducopter Developer
That flight vid was done manually, 3rd person, from quite a distance away as anybody can see. I don't have any FPV or feedback of any kind other than what I can see from 100+m. All the movements it makes are my stick inputs, and I wasn't going for smoothness, I was just flying around, testing it out dynamically. The movements are exactly what I input, so what exactly is the problem? Would you like to see datalogs of input vs. output, so you can see they match up (that is the real measure of a FC's performance)? Keep in mind this is a very responsive 630mm 4 kg octo capable of lifting 7kg additional payload, but which had none on this flight so it's a bit overpowered and it is tuned for maximum responsiveness. It's a bit of a beast. I'm not trying to make excuses for how it flew. It flew *exactly* as it was told to. I was making aggressive manoeuvres to test it out dynamically, this was the first daylight flight I'd had in an open area, everything before that has been hovering in my back yard.
I'd like to know how you can watch a video and conclude that a flight controller is responsible for the apparent stability rather than the pilot. I'd also like to see some video of smoother flying, at distance, unaided, and without a gimbal? Do you have some? Do you have any video of another FC suffer that amount of inflight damage (one half-blade completely gone, and 4-5 more heavily damaged, and a motor tilted over) and not immediately flip over and die. Do you have any video of other VCs doing the bottle test?
Again, I have plenty of video of the thing just sitting there, unmoving. But it's pretty boring so I don't both to upload it. Oh, well here's one boring one, I was trying to see if I could get it to flip on takeoff by intentionally loading up the I-term. There's a bit of flying at the end. Again, all movements are my inputs, I was testing the flying, not trying to shoot Hollywood cinema.
I'm just trying to make this discussion a bit more objective if we are going to be comparing flight controllers. It's a bit ridiculous for anybody to claim that they have judged anything based on watching somebody use it, or based on a given flight video. That goes for any flight controller, not just Arducopter. Because at the end of the day, I think EVERY flight controller available has the whole "stability" thing pretty much licked. 99% chance that if you see one that you think is "unstable" it is the user's tuning which is at fault, not the controller. What really separates the controllers now is ease of use, availability, ground control software, features, robustness to flight failures, and of course price. So why don't we try ranking FC's objectively on these aspects instead?
Where Arducopter suffers is mostly on ease of use. Some users really struggle to get it working properly. It's very powerful, and flexible, it can fly setups that a DJI product can't, but that results in it being complicated to setup well.
Thank you! This is exactly what I'm getting at.
Asking on a forum "which is best?" will only return answers which are basically the respondent's "favourite" system. Of course they think their systems is the best, or else why would they be using it? You'd be better off just conducting a poll and just buying the most popular system if you are going to base your decision on answers like that. Just make sure you conduct the polling in a place where you're likely going to get equal representation of users. ie: don't ask on a DJI forum, which is the most popular controller...
If you only want to manually fly a quad, and you don't want to spend a lot of time tuning, DJI is probably the best. If however, you want to be able to fully scripted GPS missions, that is completely customizable right down to what the yaw controller does in RTL... well your choices are limited. (hint: MR users tend to want the yaw to remain where it last was on RTL so that they can maintain orientation. Helicopter users tend to want the nose to point at home so that the tail is naturally flying behind the helicopter).
Last edited by a moderator:
Benjamin Rowland
Member
Ken, could you briefly give a rundown of the pros and cons of each of the boards you've tried?
thanks,
Ben
Bartman
Welcome to MultiRotorForums.com!!
Ken, seriously, is there ever going to be an end to your personal grudge with Hoverfly? Read all the posts, the PRO flies plain and simple awesome in manual and auto-level plus the resident 2-axis camera stab is beautiful. On top of that it's a high quality item with accessible factory support plus a knowledgeable forums community.
It's not as cheap as some of the other boards mentioned but it's a very capable board that can be used in a variety of applications.
Bart
Well Bart that just hasn't been my experience. I wouldn't call it a personal grudge so much as I've tried every firmware version from the very first to the very lastest on the H/F Pro board I have and I'm just not seeing it, the truth from my end, plain and simple. I've tried it on just about every possible configuration of multirotor as well, and just for the record the comment was more about everyone saying how well it flies in manual. If you want to just fly around in manual mode all the time then why pay a lot of $$$ for features you aren't going to use when there are cheaper alternatives that work just as well, seriously.
Ken
If you search this forum you may find exactly that, I do believe I've posted it here before and if not I have on RCG. I could type it up again but to include all of the systems I currently own or have owned would take a couple hours that I don't have to spare at the moment. Briefly I see it like this, asking which flight controller is best is a bit like asking which car is best, everyone has their opinion and there is a huge variety of reasons and explanations as to why, they ALL have pros and cons and depending on what your expectations are will determine what you list in each column. With the possible exception of the Gaui GU344 I have yet to own a flight controller that didn't at least fulfill the basic functions of being able to take off, fly around, and then land without major malfunction the majority of the time. How well each one does that depends on a lot of factors not the least of which is the quality of the physical components and the firmware that runs all of it, advanced functions are another topic entirely.
The bottom line is exactly what I typed previously, define the use case and what features are important to you, same as would when buying a car. Then do the homework to find out if the systems you are considering are even capable of meeting the defined objectives. If they are, come up with a short list of candidates and then do more research, ultimately you will find a board that can do the job, if it's the right one for you is a question that only you can answer after doing all of the above.
Ken
R_Lefebvre
Arducopter Developer
Bart, I don't even think the Hoverfly is that badly priced. I mean, it's a bit highish, but if as you say, they actually have actual support (ie: you can talk to somebody on the phone who will actually help you) then there's a price for that obviously, that is not built into other controllers. It does start to get up there with the GPS add-on. Hopefully when they come out with the actual waypoint and telemetry software, it won't be another add-on (ie: pay more for software)
I'm curious about the claim for "16 parallel processors" however. What does that mean? Maybe they're running a Parallax Propeller in which case this isn't really a bragging point, but I digress. Ok, I just looked and yes, it is in fact a Propeller chip. 2 in fact, each having an "8 core processor". Each "cog" is 20 MIPS, so they have 320 MIPS total. That beats the pants off an Atmel 2560's 16 MIPS. It's even more than an STM32F4 which runs about 210 MIPS I think. Wow, I wonder what they do with all that power for "just flying around?" Holy crap, and then they have another 2 propellers on the GPS board!
What is the software like?
I hear about the Hoverfly a bit, but I don't know a lot about it.
I'm curious about the claim for "16 parallel processors" however. What does that mean? Maybe they're running a Parallax Propeller in which case this isn't really a bragging point, but I digress. Ok, I just looked and yes, it is in fact a Propeller chip. 2 in fact, each having an "8 core processor". Each "cog" is 20 MIPS, so they have 320 MIPS total. That beats the pants off an Atmel 2560's 16 MIPS. It's even more than an STM32F4 which runs about 210 MIPS I think. Wow, I wonder what they do with all that power for "just flying around?" Holy crap, and then they have another 2 propellers on the GPS board!
What is the software like?
I hear about the Hoverfly a bit, but I don't know a lot about it.
Last edited by a moderator:
Hello,
The list is missing YS-X6! It's an awesome system around price of wookong, just way better in every aspect in my experience. I sold my wookong as it's practically same as naza gps just with no waypoints nor octo support. I have naza now in my 800mm Y6 setup with AV130 and it's fine if it's dead calm but in windy it's just pain and scary to fly. I have not had a single issue with YS-X6, it's rock solid even in high winds, and freaking precise gps hold. Today i flew my cinestar/YS in 16m/s gusts without problems but i couldn't even imagine to fly naza or wookong in those conditions.
Zerouav support is also very helpful via skype if needed. They are releasing little brother YS-X4 in next few days which i'm sure will be amazing system too. YS might be slightly more challenging to setup than dji if it's first multi but nothing painful if done patiently with reading manual. I'm amazed that YS is not more popular at the moment, they have made a great product, software updates have been bug free and well tested before releasing and whole system is mature already after just few months after release. No surprises, no flip of deaths but just great reliable performance. I still don't understand how Hoverfly cannot get decent gps to their system, otherwise i might be using it rather than dji in my Y6 setup. Zero uav just have released their system and it's there, no need for fixing firmware bugs or hardware design failures afterwards.
The list is missing YS-X6! It's an awesome system around price of wookong, just way better in every aspect in my experience. I sold my wookong as it's practically same as naza gps just with no waypoints nor octo support. I have naza now in my 800mm Y6 setup with AV130 and it's fine if it's dead calm but in windy it's just pain and scary to fly. I have not had a single issue with YS-X6, it's rock solid even in high winds, and freaking precise gps hold. Today i flew my cinestar/YS in 16m/s gusts without problems but i couldn't even imagine to fly naza or wookong in those conditions.
Zerouav support is also very helpful via skype if needed. They are releasing little brother YS-X4 in next few days which i'm sure will be amazing system too. YS might be slightly more challenging to setup than dji if it's first multi but nothing painful if done patiently with reading manual. I'm amazed that YS is not more popular at the moment, they have made a great product, software updates have been bug free and well tested before releasing and whole system is mature already after just few months after release. No surprises, no flip of deaths but just great reliable performance. I still don't understand how Hoverfly cannot get decent gps to their system, otherwise i might be using it rather than dji in my Y6 setup. Zero uav just have released their system and it's there, no need for fixing firmware bugs or hardware design failures afterwards.
Last edited by a moderator:
nicwilke
Active Member
+1, I too use ZeroUAV, and being an ex WKM and other FC user, I can say its terrific. The use of datalink (or wifi if you want less range for video type use) is great, I get heaps of realtime telemetry to my PC or android phone or tablet in the GCS app. I can see stats like vibration, GPS sats locked in, what the Rx is doing, amps used (with the power management module added) and also have 16 waypoints functionality for autonomous flight. I dont use this much, only for fun, but could pose useful for single operator AP setups. Set a GPS route, and then have control of camera only. The PTZ lock will be handy, I'm about to test it out (target lock) to encircle a GPS point with camera. The YS-X4 will have most of these features, for under $500.
redridinghood
Member
Don't forget the Autoquad 6. I know the calibration process is a pain but I find the FC outperforms Wkm and YS-X6 when setup properly.
Bartman
Welcome to MultiRotorForums.com!!
Ken,
I'd be happy to have a go with your board to see if it performs similarly to the one I have. If not we can take it to Hoverfly to see what's up with it.
Send it on down, seriously.
Bart
PeterLester
Member
If you start out with a Flamewheel F550 c/w Naza GPS you will have a good starting platform that can later be expanded to take many additional features and can even upgrade the Naza GPS to the WooKong M GPS before you move up to an Octocopter such as the S800......... at which point the sky is the limit. MikroKopter and WooKong M are leading from my perspective.... USA needs to provide Germany and China with some serious competition asap though....... then there will be no more serious communication challenges and super long delivery and service issues.
R_Lefebvre
Arducopter Developer
The S800 is a Hex, no Octa.
I don't imagine you'll see a closed-source system coming out of the USA... well... ever. It's simply too much money, and they could never compete. Arducopter is sort of American, but it's open source. But then MikroKopter is OS as well (well, sort of).... I'm not quite sure what your criteria is.
I don't imagine you'll see a closed-source system coming out of the USA... well... ever. It's simply too much money, and they could never compete. Arducopter is sort of American, but it's open source. But then MikroKopter is OS as well (well, sort of).... I'm not quite sure what your criteria is.
But that is exactly what exists, the only American made flight controller is closed source to the point the only place the firmware can be obtained from is their servers and only when downloading directly to the board, no way to save a copy locally, it doesn't get much more closed than that.
To that point I can see the reasoning, the Chinese will clone anything they can get their hands on, heck they're even cloning each other's stuff now and I think that is hilarious. If it keeps up much longer the clone pricing will get so low they'll be paying us to take them...:highly_amused:
Ken
nicwilke
Active Member
Funny you ask, had a chat to Mr. Rowland about this today.
Having used both, I can say that the WKM is terrific, but with a series of bad issues with firmware and hardware failure, I lost interest and sold it. The YS-X6 feels better with less tuning of gains. Handles all size props well. Descending is flawless, you can see it slow down to avoid prop wash wobble, and it flies steady. I'm not bagging WKM, it's certainly one of the best, but I think YS-X6 has the edge. Is Zero have one thing to step up to, it's the manual. It's a little hard to understand, but enough of us around on these forums to ask.
The GCS on android is really handy, and it drops a data file into a directory to look at later if needed. That is something I never did with the wkm, so can't comment on that.
ZeroTech also need a lesson on GUI, the PC software could do with a little skeuomorphic enhancement.
The YS-X4 is stated as hobby/entry grade apparently, and the YS-X6 is getting firmware upgrade features soon to further enhance functionality. I was told this today by their tech guy when I was asking some calibration questions.
R_Lefebvre
Arducopter Developer
Right, duh, can't believe I forgot that when I just finished talking about it. It's just such a small player that I keep overlooking it.
It's pretty much an example of what the market is like. DJI dominate, it's what most people think about first. And it's not necessarily because it's the best, it's marketing driven.
Where does this rumor come from that Arducopter (well, ArdupilotMega) hardware comes from China? I keep hearing this, but it's not true. They used to be all Made in USA. Now they're mostly made in Tijuana Mexico, though some of it is still done in the US, and design and engineering is still in San Diego.