Open Source please?

Please open source the Moai firmware. I bought this printer because it was promoted as being customizable and open to experiment. I find the firmware seriously lacking and have ideas and ability to improve it. How can I gain access to the firmware source?

Thanks.

2 Likes

Amen. I understand why @peopoly would not want to have a bunch of firmware variants complicating the debug/support but this is easily mitigated by simply requiring users first revert to official firmware before asking for any debug help.

If the issue is simply time - I imagine Mark is very busy - then delegate. I have no doubt volunteers could be found to manage the repo (I myself, for example).

This question has been asked before when we launched the project back in March 2017 and appeared from time to time. I will loop in one more person because he ask about it several times @Will_Kostelecky

We developed Moai firmware in-house as there is no open source galvo firmware and runs it on a custom 32bit ARM board. It is not related to Arduino project or RepRap project.

There are several reasons why we kept it closed source. The main reason is the survival of our team depends heavily on protecting our intellectual property. We are intimately familiar with the 3D printer supply chain here in Asia and understand what would happen the minute we open source the firmware. There will be companies who are happy to make a $50 on each sale by making Moai clones, use the cheapest possible parts without QC and offer no customer support or development. It would be a race to the bottom and has happened in 3D printing before.

We are always looking for feedback and see how we can improve the firmware while balancing the usability and customization. While open source isn’t an option at the moment, it is never off the table as situation may change in the future.

As Mark knows from my comments here and on FB I am also of the feeling that the firmware (user side, not galvo control) is not nearly the quality of the hardware, the kit design, and the non-UI aspects of the firmware. I suspected that the reason for not doing open-source is as Mark states…the nature of the cloning franchise. My only comment there would be that the Ultimaker business model flies in the face of the same headwinds but survives and seems to thrive. They are, however, the exception and even there the UM3 is, for the moment anyway, a closed machine. Other printer vendors, having been burned by the cloners, have withdrawn completely from open source endeavors.

So, for the immediate future anyway, the Moai firmware is probably not going to be showing up on GitHub. So what to do? Mark, I don’t know how big your team is, or what your backgrounds are, but I suspect that you are talented hardware and electronics engineers, not software engineers. If this rings true then I would strongly recommend that you find a way to bring in a software person.

A way forward could be to use this forum to put together a design for a new version of firmware that would be of the same quality as your electronics and hardware. Use your experience as a disciplined program manager to drive the development of that firmware, but outsource the actual coding to someone that you either contract for short term or to an external agency. I personally would do the former as control and flexibility would be better. Obviously there would be an investment required but in my opinion it will pay off with future sales.

Just my thoughts.
Will

1 Like

Will,

we always appreciate your input as well as many users in the community. The current Moai board is close to the hardware limitation and would need revamped hardware and thus a new firmware to further adding functionalities. We are committed to improve Moai and develop new hardware as we go forward. Look for new developments as we have been working hard the past few months.

Mark, bad news that you are at capacity now but good news that you are thinking about a future upgrade. I think that upgrading the firmware to match the level of professionalism that is exhibited by your hardware design can only serve you well. I am a retired CIO (I ran IT for Black & Decker) and do this whole 3D printing thing for fun…!

Will

I must agree with Will on the polish of the firmware, being a software developer myself, it hurts a bit to see the UI every time I use the Moai.

There is a possible middle ground - you could encapsulate the parts where your expertise comes in - galvo & laser control, etc - behind a public interface, and then open-source the rest - the UI, file system handling, screen driver, etc. That would allow you to take in contributions from people with experience in those areas without giving up your IP. Distributing the closed-source parts as a lib file wouldn’t expose them to reverse-engineering any more than the finished binary blob already does.

2 Likes

I suspect (hope?) that there will soon be some after-market controller boards available which will be able to drive the galvanometer directly. Or, speed permitting, maybe an interface board that plugs into standard stepper driver sockets and converts step/dir to the proper analog voltages. Sounds like a fun weekend project… maybe I’ll try to convince my son to work on it. :slight_smile:
Then we will have more options (USB, WIFI, touch screen, web interface, spare channels for DIY auto-levelling, O2 wiper, etc). From what I can tell the only missing link is the galvo conrol and then we can tweak the open source firmware as needed.

P.S. I know I’m over simplifying - I was exaggerating the simplicity to make a point. If the Peopoly survival plan hinges on the notion that a galvo driver is IP that noone else can make relatively easily, well… I hope there is a plan B.

When I read the comment about “protecting our IP” I literally LOL’d. Am I the only one who sees the irony here? Ever wonder what Formlabs or Smart3d.net (whose name is still stamped on the test ring you ship with the product) thinks about the Moai? Sorry for the harsh words but a little introspection is in order here.

Peopoly is cheap enough and the quality is high enough that it will continue to attract customers. Yes, there will be cheaper knockoffs. You can’t stop that from happening. You win with good customer support and a product that works well.

Fair point. However, IP is more than just about galvo control, getting the firmware to work well with a specific chipset, which galvo driver and there are many know-hows. None of that is open sourced and there is value to IP even if someone else already knows how to do the same because it takes time and resource to even just recreate. And we did more than just recreate. Even if a team knows about galvo control, it will take time to create a full firmware solution. It is not about stopping it from ever happening. It is about creating enough niche for one to generate enough margin to grow as a team. There is a reason why Formlabs never open source or Makerbot closed source after three generations as well as Ultimaker.

2 Likes

Don’t forget you do have more than a few users with significant development experience (on many target environments) who might be contracted at rather low rates to help out with coding, code review, QE, etc. Of course, they would have to sign a contract with the company, including non-disclosure terms (which are taken more seriously outside of Asia).

yes, there are lots very capable users in the Moai community. If we can strike a balance like @kathanon said it would be great. This board is tapped out so it would be something down the road with new board and new firmware if we can find that balance.

sorry but… I don’t agree with the IP protection thing.

Look at Prusa printers. They are open hardware, open software machines and their business thrives! There are dozens knock-offs but they win on innovation and high quality. I think that’s what should be all about, let the cloners copy your work while you work on the next best thing (hardware and software wise). They will always be one step behind and --most importantly-- make you free publicity.

Just my 2c, I’m not an expert so I guess you did your homework, but I feel people is way too scared of open source and don’t see the actual value.

What does EnvisionTEC think about the Form1,2 :smiley:

Prusa i3 has the biggest marketshare of 3d printers in 2017. When you roll in money it’s easy to invest in innovation. If you are a small company clones will hurt you much more.

Btw. in the photo of the logic board in the moai instructions the markings on the cpu are blured. There you can see that peopoly was worried about clones right from the beginning.

sorry but that just proves my point. Prusa didn’t start as multinational with infinite budget, it was one guy with few friends and an idea. They always kept everything open and look where they are now.

Yes, that’s true. The point i was trying to make is that product cloning was a problem for quite a few 3d-printer manufacturers in the past.

I surely don’t want to argument against open source, let’s hope that the Moai firmware will become open source at some point.

maybe the problem is that while cartesian CNC is common knowledge by now and there’s no point in hiding the code, galvo is still quite an obscure art. Anyway, I feel that you should beat the competitors on innovation and product quality not with IP, but that’s just me. I’m just glad @peopoly released an affordable more-than-decent-quality printer, not going to spit on the plate I’m eating on :smiley:

2 Likes

Unfortunately, that is no longer the situation in the 3D printing world. Prusa and others took advantage of the expiration of the Cartesian FDM patents to start the businesses, much as the SLA makers have done with the SLA core-patents. However, in the time since Prusa launched there has been a massive influx of low-end manufacturers into the space, which has destroyed margins. Even Prusa has had to build IP walled-gardens with his Mark 3: the coated removable build surface is a proprietary process developed in-house, as is the print recovery/head location detection system.

While open-source is a nice idea, in a world of large-scale dedicated copyists it is not a viable business model for the start-ups or even mid-sized one-trick ponies.

Finally, it is not the galvo driver that is the issue: it is the interpretation of gcode into galvo motion commands that is the IP. The maths for that conversion are complicated, particularly since the gcode was originally written to drive a milling head. I imagine there was a lot of time spent developing the conversion algorithms. Formlabs did the same thing to develop their stl converter, which is why Preform is proprietary. The hardware is somewhat trivial, but the conversion software is not.

1 Like

What do you mean by tapped out?
Out of storage or RAM?
Fixing bugs and improving UI doesn’t have to increase usage of either. Improving code often involves decreasing code size.

At the limit of being able to keep up with the rest of the hardware speed-wise?
Same thing there, plus that most of the problems I see are in the UI, where adding a few milliseconds wouldn’t matter.

Even if you do not have room for new features, it’s at least as important to create a good user experience. The UI needs to make sense and work properly. You don’t want spend effort navigating the UI when you could devote it to whatever you need the printed part for instead. When the user experience is filled with annoyances, then that will be a large part of what you think about when you think of the printer.

1 Like

U talk bout IP but u still clone lcd printer and make it bigger and declaring as not open build firmware.

Been fed up with my moai. Pm motor issue still problem. Vat isnt tilting. U should focus on moai rather than clone a lcd printer

1 Like