Has anyone spec’ed out how much it would be to build the @Autodesk #Ember printer themselves? Now that Autodesk has open sourced the electronics / firmware, I would think that all the parts are there…someone could even upload the 6-layer PCB to get a quote…just wondering if anyone has yet?
http://learn.ember.autodesk.com/blog/ember-open-source-electronics-and-firmware
They open sourced the printer because they couldn’t get around copyrights for some parts.
Great! All the better for everyone!
You might be able to make one but it looks like they have a lot of molded plastic parts that might be difficult to create on a 3D printer.
They forget the most important part of what I call Open Source, the steps.
Sure anyone can work on something for 10 man years (or 20 people working for 6 months) then release it. That’s why Open Source always have some sort of revision control and keep track of it.
Could you imagine the Linux kernel if they kept the development up to the releases private?
@Andrew_Hodel They weren’t planning to have it be open source. They only released the hardware files as open source because they couldn’t sell it commercially. That’s why there are no steps, the plans changed.
@Adam_Steinmark what genius figured that out that couldn’t figure it out before?
Are you saying they didn’t keep internal revision control? They just have a SMB share or what?
Shouldn’t be too hard to use this data to push open source designs forward a bit. The open source firmware could help as well.
@Andrew_Hodel They were trying to get around the copyrights but couldn’t. That’s why development was only 6 months. Paperwork takes time. They probably didn’t want to release the files until they were sure they couldn’t sell it commercially.
@Adam_Steinmark I am wondering why they can’t also give the revisions, I’m sure they have them on disk and that would help people to understand the steps taken to get to that final product or final firmware code.
For example, if you are writing code to control positioning of steppers or PWM for DC motors (haven’t looked at the printer) then you write the pulses for that first… then you write the gcode interpreter (if you are using gcode etc).
It doesn’t just all come in a dream and someone sits down and writes it for 6 months straight, there are steps.
@Andrew_Hodel Oh you’re wondering why they don’t release it now. Maybe nobody has asked. Maybe you should.
@Andrew_Hodel but in industries for small products/projects sometimes their is no revisions.
If you are lucky the files have an autors name and a explanation.
@Sebastian_Schmidt git makes that quite easy. It makes things easier for the developer too. When you need some code block you can just find it in the revision history.
Any person or organisation can do that and should for their own benefit.
Sure a lot of minds are either “stick with what works” or “gung ho for the latest and supposedly greatest” but both miss the middle that sticks with what works until something honestly better comes along.
Marketing and striving to be filthy rich at any means doesn’t help and neither does doing nothing new.
I don’t think using git is asking too much. I would agree that an industry pushing a Dev to make something he doesn’t care about is going to lead to exactly what you are talking about though. That’s a fault of specialization and multiple identities is not a solution as so many devs are doing nowadays as a way to escape the mundane.
The honest truth is they want to see a previous experience and skill list of exactly what they want so here is the “industry”.
Either people scared of change or people willing to lie to make change. It makes more sense to talk and be open in my opinion.
As far as I read: It should take roughly the same (6k$) to build it in DIY.
Saw the Ember in person. Other than the tiny print envelop, the print quality is amazing.
The Spark program was always going to be open source. Since day 1 they said “we will be releasing the plans and documentation for this printer”. They released the resins 6 months ago now they are doing the printer. They are just going a different direction with the development than what a lot of FOSS projects do with immediately releasing their idea, they are releasing it after they get a polish on it.
The printer was released as a Fusion360 project in June I think. Basically what I’m saying is, this was always part of the plan, not a fall back.
Spark was never meant to be a DIY platform. That has been made clear from the beginning. The platform is and always has been providing manufacturers with a platform. End users are free to develop for the platform if they wish.
What the firmware release does is allow those that are developing machines to get an idea of how to integrate your specific firmware to the platform. A Faye pub/sub is opened and data is exchanged with JSON. The platform needs to communicate directly with your firmware, or at least what it thinks is your firmware. It can’t happen with the current Marlin but I’ve been kicking around ways to integrate Octoprint as the middleware with the platform. Marlin would still control the internal machine functions but Octoprint would act as the public side of the host, just as it does now.
@Adam_Steinmark The statement about them wanting to “get around” copyrights is absolutely false. The stack is large and complex and they have the attributions and copyrights available and will provide you source for any of the parts of the stack used if you ask. That’s all permissible under the license.
Here is the Github https://github.com/spark3dp