SIYI A8 for gimbal/camera control testing

Topic:Siyi A8 for flight code testing and development

Proposal type: Hardware [x]

Description:

@peterbarker very often does weekend testing of various new features. It would be very helpful if he could get a SIYI A8 in order to help test the ongoing gimbal and camera improvements that I’m working on (a recent example PR is here).

In a more general sense aerial photography (including mapping) is one of the core use cases of AP so it makes sense for developers to equip their vehicles with a camera and gimbal. The SIYI A8 is a good choice because it is relatively inexpensive, small enough to fit on nearly any vehicle and has all the basic controls we need for testing including 3-axis gimbal (including nearly 360 deg yaw) and a camera with features including take-picture, record-video, zoom, etc.

This will also likely help double check the QGC enhancements (see related already approved funding request)

Planned amount $$ (USD):$350

Estimated time for completion: 1 month

5 Likes

We’re getting a lot of proposals for hardware for testing and drivers and such and in most of those cases the vendor should be supplying the hardware. The funding team would REALLY like to start enforcing a rule where there’s some sort of effort/attempt to get the hardware sponsored before it comes to the funding committee as a proposal for Ardupilot.org to pay out for hardware.

We know it’s important that the developers have the proper hardware and we don’t want to inhubit that. It would be better if there is some engagement from the vendors. If not, then that’s what the org is for.

Here’s an example of a propose from Henry. We asked if he could get it for free/discounted… and they gave him %50 off. https://discuss.ardupilot.org/t/approved-funding-for-walksnail-avatar-hd-goggle-set/98675.

2 Likes

I really should have asked for the discount/free up front since this was NOT a partner…I have been told in the past that we should not ask our partners for hardware for support since they already pay for that…okay, if its a new development…but maybe I miss-understood what I was told previously

Whoever told you that was wrong to do so: when hardware is required to add support for a vendor product the preference is always that the vendor provide the hardware. Partner or not. It makes no sense for a partner to donate $1000 and us hand it straight back to buy their product in order to test support.

2 Likes

@MagicRuB @james_pattison

OK, so if we need a particular vendor’s hardware, before they become a Partner AP will pay for it… but once they become a partner then AP won’t pay for it and we definitely need to get the Partner to supply it for free or we do without. That’s how it works?

I understand the need to be fiscally responsible but by being too strict you’re delaying devs getting the hardware they need and/or they end up buying it themselves.

1 Like

Re-read my previous.

Even 50% off feels disapointing, should be free imho.
They are getting development work for free, and ultimately free or greatly reduced support, after all. Not to mention increased visibility/marketing.

Also not very fair for competitors who are also partners who end up paying for it, even though it ends up being a very small sum.

I just want to express my disappointment that the funding committee is unwilling to fund a $300 camera for a core developer. I should not have to go back again and again asking a Partner for the hardware we need just because they happen to have it. This leads to us wearing out our welcome.

Maybe some of the confusion is that the funding committee thinks that I’m asking for a camera so that the driver, specific to this camera can be improved. That’s not the only reason. The reason is that many more general purpose features need a camera (any camera) but the Siyi A8 is the best value option (e.g cheapest and easiest).

1 Like

Once again an example of the funding committee hamstringing development just like in a previous proposal to investigate firmware issues. They are constantly looking at the “partners” while forgetting the others that have donated to this project. Seems odd that this is being second guessed especially when there is a developer willing and able to spend the time to make this work right. One of the biggest issues I have seen is lack of gimbal/camera integration and this will further this goal for a very reasonable price! now if this was for say a product that is thousands of dollars I would expect them to help foot the bill, but given the small price and ROI I think this should be funded. I just saw you guys fund a set of goggle so come on! @rmackay9 please get ahold of me on the side and I will gladly send you the funds. So silly funding committee!! I put my money where my mouth is so please get ahold of me.

@rmackay9 is intentionally misrepresenting the feedback I’ve given him. The funding committee strongly support this line of development, but want Randy to actually plan things properly so we can provide Peter what he needs. That feedback was provided months ago and again earlier today.

1 Like

I highly doubt he is mincing words but offer still stands as you have still not approved this proposal! you just approved a set of goggles that although will help the community I am not sure will have as big of benefit as gimbal/camera enhancements. I believe the cost for the camera is pretty trivial given the amount of benefit it will have. Gimbal/camera integration has been quite a lacking point in ardu. Often times I see folks asking for a feature to be funded. The response is always the same find someone to do it and ask them. Well you have a developer asking right now for a pretty low cost even zero if @rmackay9 will take my offer!

Direct cut & paste:
‘’’
“1. The A8 is the best choice for PeterB because it’ll fit on the vehicles he flies and has all the basic features (almost 360deg yaw, take pics, record video, zoom, focus controls)”
But he can’t use it because those aircraft have DJI FPV fitted, which isn’t compatible, and testing with analog would be very backwards looking. As per months ago, let’s plan what he needs.
“2. PeterB will add value to AP by regularly flying with a camera/gimbal far beyond the $300 cost. It’ll give him experience and understanding of how our users incl Partners use them and he’ll surely find and fix issues he runs into. He’s already contributed to PRs related to cameras/gimbals and he’s the backup mentor on the camera/gimbal GSoC project this summer.”
So let’s scope out what is actually best to get him, not what’s cheapest. Value isn’t just dollars.
“3. Cameras and gimbals are an important feature to AP. In general we need more devs using them.”
Absolutely!

I’m a broken record. This isn’t about $300. It’s about a poorly scoped proposal and lack of a forward test plan.
‘’’

I would also add that would be great to have @peterbarker input on this. I have no doubt he is well qualified for the task but will he do it or have time to do it ?

The whole point about having a funding committee is to clear some points about how we are using the project funds, so I think that if some points need to be explain we should respect and just discuss about them, maybe a 5min call will do it.

2 Likes

@james_pattison I look forward to when this says approved. Seems silly you are hindering development and furthering the community to prove a point to one person.

I’ll say this once, because I think it’s worth saying, but then return to keeping my own counsel.

You don’t get to gainsay the decisions of the committee because this is literally how democracy works. Representatives are elected and then make decisions which they believe are in the best interests of the project based on the mandate they have been given. That mandate is reviewed once a year and if decisions are generally felt to be in error then new committee members can be elected. The process simply does not work if every single decision can be reversed by someone complaining. At that point you have a rubber-stamping process and no need for a committee - maybe that’s the direction the project will decide to go in the future but it’s not the direction right now. Perhaps one of the problems is that it is so rare for a proposal to be turned down - maybe everyone just assumes it is a rubber-stamping process, when it is not.

On this particular proposal I am ambivalent, but I will defend to the hilt James’ mandate to ask for reasonable modifications based on his view of what is in the project’s best interests.

2 Likes

From what I have read in this chain, it looks like the Committee is simply asking that the proposers reach out to the vendor to see if they can acquire the hardware/software at a reduced cost based. I don’t think the Committee is requiring that a discount is actually achieved - just that the proposer at least asks for a discount.

Doesn’t seem like too much of an ask and I wouldn’t see why this would delay the process by a lot (hopefully we aren’t trying to acquire from vendors that don’t check email/phone/etc.).

For those who are against this - why?

Thomas from IRLock has kindly sent PeterB a Siyi camera. Thanks very much Thomas!

3 Likes

@rmackay9 please work with @peterbarker to put together the list of the things he needs to be able to use it properly. I had a good chat with him on Sunday, and there’s a few things that he doesn’t have/needs to upgrade (video link, rc rx, second camera compatible with video link if you want him to test camera switching, perhaps more).

1 Like