Which controller will eventually replace RAMBo?

A place for anything else electronics related
User avatar
U.S. Water Rockets
Printmaster!
Posts: 157
Joined: Mon Apr 06, 2015 7:19 pm
Location: Galway, NY USA
Contact:

Which controller will eventually replace RAMBo?

Postby U.S. Water Rockets » Sat Mar 26, 2016 12:27 pm

When I bought my Rostock Max V2 about 2 years ago next month, I assumed that the RAMBo that it came with would soon be replaced by one of the more powerful boards that were coming to market at that time. The Smoothie, or Duet, or Replicape, etc. all seemed to be much more capable boards and especially in the world of Deltas, the extra processing power would translate into better and faster prints. 2 years ago, the RAMBo, with it's loved/hated Arduino single threaded Wiring based programming IDE seemed like a de-facto standard that everyone was using in spite of it's weaknesses, and it was going to be replaced very soon with something vastly improved.

Fast forward 2 years, and it seems even new printers come with RAMBo and similar variants. Even some very costly machines with advanced web based interfaces still use the RAMBo type boards in them, and add their advanced features by kludging a RasPi or Beaglebone in and hanging the RAMBo off the USB port. This seems like a really inefficient design.

I see many smart folks on this forum that have written directions on up to upgrade their RMAX to newer controller boards, like Smoothie or Duet, but I still don't get a good feeling about the direction of the industry and people really getting behind a RAMBo replacement. Most of the replacement boards and tutorials seem to involve a lot of tweaking and hacking to get them installed and running, and the firmware for these boards really seems to be in a huge state of flux.

I was wondering if anyone else was out there that wants to upgrade to a newer board, but because things are so jumbled that they are worried about wasting a lot of time and money and ending up with sub-optimal results. My RMAX works great like it is now, and I'd like to have Ethernet and a better LCD, but it's not broke now, so I am inclined to not want to "fix" it, and make it print badly in the process, or create something that needs constant attention to keep running.

What I was hoping was that SeeMeCNC would have moved the RMAX to a newer board, and I would buy whatever upgrade parts they made for older models, and then I would follow their progression to a new board, and I would have some confidence that there were people on the forums here with the exact same setup, so there would be commonality and a good resource for assistance in the event of issues.

As it stands right now, it seems everyone who's upgraded is doing a one-off custom build and no two printers are alike. Having a unique printer with custom firmware is appealing to some, but I have a lot of other projects going on and I don't have the bandwidth to make and maintain a custom machine, so I'd like something more advanced that was somewhat standardized with what everyone else used.

Does anyone else have the seem trepidations about upgrading that I have? What direction do you guys see the industry (especially deltas) moving to? I'd be curious how everyone else felt about new controllers in general, and if there was a consensus about who the new frontrunner or preferred controller might be.

User avatar
Nylocke
ULTIMATE 3D JEDDI
Posts: 1421
Joined: Sun Jun 23, 2013 1:43 pm
Location: Iowa

Re: Which controller will eventually replace RAMBo?

Postby Nylocke » Sat Mar 26, 2016 12:47 pm

I think its a bit too early to tell, theres a lot of people working on a lot of different things, all pulling in different directions. Last year I was betting my money on BBB based controllers because of the real time co-processors, but it seems MachineKit development is a bit fragmented and confusing from what JFettig has told me.

Theres the Smoothieboard V2 that is supposedly coming out sometime this year with a bunch of beefed up specs that should be plenty fancy, and 626Pilot I think will be working to develop his calibration improvements for that.

The Duet is pretty decent and I think (?) I heard something regarding an upgrade coming for that as well. MHackney is really happy with the firmware for the calibration, and I think its a little bit better than smoothieware since it actually supports G2-G3 arc commands as arcs instead of line segments.

I was hearing some rumors about SeeMe and Ultimachine working together on the "next version of the RAMBo" some time or another in the last couple years, but I haven't seen much about that lately.

I think the Duet or Smoothieboard are probably the go-to boards right now from the information I've heard. It might be good to wait on the V2 smoothie board if you're going that route

User avatar
mhackney
ULTIMATE 3D JEDDI
Posts: 5398
Joined: Mon Mar 26, 2012 4:15 pm
Location: MA, USA
Contact:

Re: Which controller will eventually replace RAMBo?

Postby mhackney » Sat Mar 26, 2016 12:59 pm

Like all things, there will be no one winner. My bet is on the Smoothie and Duet camps and their respective firmware. I run all my machines on Duet now except one Smoothie hold out simply so I can keep my foot in that camp too. But Duet is so much easier and better at this point.

The Duet firmware is much much more than just great auto calibration. It is the ONLY firmware out there that calculates each POINT of movement, not line segments like the others. Does this translate to better prints? It depends on a lot of factors but I'm getting the best prints I've ever got with it. The RepRapFirmware also has a wonderful integrated web interface. The Duet has great ethernet support and with the new web interface and firmware there is no need to use USB to print OR to upgrade the firmware itself. No RPis, no Octoprint, nothing else required and it all works beautifully. There also is no compiling firmware, you download and upload the binary from the web interface, you update/edit config from the web interface, you upload and run gcode files from the web interface. And there is great support for macros - which are gcode files that can do all sorts of useful things. And if that wasn't enough, ALL configuration in the config files is legitimate gcode, not name-value pairs, not EEPROM parameters, not compiled-in parameters, simple gcode.

There is a new Duet in development that should hit the streets later this year, but frankly, the Duet .8.5 is pretty darned good. No memory limitations, fast processor, onboard support for 2 extruders and hot ends and with the expansion board you can add another 4 extruders/hot ends. Duet and RRF is almost 100% focused on 3D printing and not trying to be a jack of all trades CNC controller. In my opinion, this is what's needed as there are enough variants of 3D printers to deal with.

Sublime Layers - my blog on Musings and Experiments in 3D Printing Technology and Art

Start Here:
A Strategy for Successful (and Great) Prints

Strategies for Resolving Print Artifacts

The Eclectic Angler

User avatar
bot
Printmaster!
Posts: 993
Joined: Thu Sep 25, 2014 12:18 am
Location: Vancouver
Contact:

Re: Which controller will eventually replace RAMBo?

Postby bot » Sat Mar 26, 2016 3:10 pm

Duet, for sure right now. Smoothie 2 or the new Duet in a year or two. We'll see. It's like the old days of gaming PC. There is such quick advancements, that one day TNT2 is good, the next day Geforce256, and now the rest is history.
*not actually a robot

User avatar
Eaglezsoar
ULTIMATE 3D JEDDI
Posts: 7185
Joined: Sun Apr 01, 2012 5:26 pm

Re: Which controller will eventually replace RAMBo?

Postby Eaglezsoar » Sat Mar 26, 2016 3:47 pm

I agree that the Duet is a great board and the display is fine also but the price of the display is high in comparison to the rest of the RepRap world.
I would like to be able to connect the Duet to the same display that is being used on the Rostock Max now and I am sure that there is a way to do it
but I have seen no schematics for such a thing yet.

I realize that the quality displays like the Duo and the Vicki demand higher prices, but not all of us need the premium displays. Touch screens that are
physically small leave a lot to be desired unless you have small fingers and like to use them as a slow pecker.
“ Do Not Regret Growing Older. It is a Privilege Denied to Many. ”

User avatar
mhackney
ULTIMATE 3D JEDDI
Posts: 5398
Joined: Mon Mar 26, 2012 4:15 pm
Location: MA, USA
Contact:

Re: Which controller will eventually replace RAMBo?

Postby mhackney » Sat Mar 26, 2016 6:05 pm

Eagle, the display for Duet is much more than a simple display, it's more like a controller like the MatterHacker controller. And, you really don't need it with the great web interface. That said, $90 wish for a display really isn't that bad especially for what you get - color, touch screen, no stealing memory or processor from the controller. And the 7" version is just spectacular.

This said, I wouldn't be surprised to see a less expensive, albeit less capable, display sometime this year.

Sublime Layers - my blog on Musings and Experiments in 3D Printing Technology and Art

Start Here:
A Strategy for Successful (and Great) Prints

Strategies for Resolving Print Artifacts

The Eclectic Angler

User avatar
Xenocrates
ULTIMATE 3D JEDDI
Posts: 1351
Joined: Wed Sep 23, 2015 2:55 pm

Re: Which controller will eventually replace RAMBo?

Postby Xenocrates » Sat Mar 26, 2016 6:24 pm

I think the biggest issue between them is standards fragmentation (Ala https://xkcd.com/927/. All of them have different feature sets they offer, and all of them are working on more with different priorities, but with the difference in architecture and code-base, as well as essentially firmware mono-cultures, the Rambo (And all the other arduino boards like the Rumba/RAMPS) is here to stay for a while, since it's so well entrenched and has so many hardware/firmware/UI options. If you want a given feature, or a modification, someone has likely developed it (Heck, from what I understand PT100's are essentially Arduino only for now, and SMCNC+Ultimachine have delta calibration working with it using a the best probing solution so far). They are small, commonly available (although not that cheap, the Rambo seems to be pricier than the BBB+Replicape 149 Vs 180). They are easy to hack at, and fairly robust, and were what seems to be a standard controller for printer manufacturers, meaning most people are familiar and comfortable with them. But I would love to see them either be supplanted by a better controller, or upgraded (While keeping most of the firmware compatibility)
Machines:
Rostock Max V2, 760W corsair modular PSU, PT100 enabled E3D V6 and volcano, self adjusting carriages, Raymond style enclosure
Automation Technology 60W laser cutter/engraver

Sic Transit Gloria Mundi
01-10011-11111100001

Eric
Printmaster!
Posts: 725
Joined: Sat Aug 18, 2012 4:09 am
Location: Chula Vista, CA

Re: Which controller will eventually replace RAMBo?

Postby Eric » Sat Mar 26, 2016 9:24 pm

Homebrew CNC controllers are stillll like the early days of the PC industry, back when Microsoft and Apple camps still had different hardware (today, the only real difference between a PC and a Mac is the operating system). My advice in those days was find the software and support that you need, then buy the OS and hardware it runs on. That's why so many professional video and audio engineers were/are in the Mac camp...that's where the best audio/video editing software was when they started.

Anyway, it's still that way for us. Find the software that works like you want, and buy the hardware it needs if the price is acceptable. And make sure the support is there...the best and greatest is useless to you if there's nobody to tell you how to use it when it gets confusing.

The arduino boards retain an edge because they're so well supported, both in the variety of software options and the availability of technical help. The fact that they tend to be the cheapest option helps as well. For bundlers like SeeMe, the price and support issues factor into whether they stick with what they have or move on to newer tech. Innovation isn't always worth the price of change if it isn't a big enough leap.

User avatar
Jimustanguitar
ULTIMATE 3D JEDDI
Posts: 2591
Joined: Sun Mar 31, 2013 1:35 am
Location: Notre Dame area
Contact:

Re: Which controller will eventually replace RAMBo?

Postby Jimustanguitar » Sun Mar 27, 2016 2:47 am

We'd be fools to think that Johnny and the crew at Ultimachine weren't working on some kind of super rambo. I've heard some rumors... I'm sure they'll have a Due or ARM based contender soon.

User avatar
U.S. Water Rockets
Printmaster!
Posts: 157
Joined: Mon Apr 06, 2015 7:19 pm
Location: Galway, NY USA
Contact:

Re: Which controller will eventually replace RAMBo?

Postby U.S. Water Rockets » Mon Mar 28, 2016 10:29 am

Wow, thanks for all of the thoughtful responses!

I suppose a big issue for me is that there are so many options and combinations of options that the choices become dizzying. What type of stepper drivers, how many extruders, what endstop inputs, what CPU, etc. It is rather like buying a car and instead of having several models to choose from, you have to pick engine, wheels, seats, glass, transmission, suspension geometry, doors, hood, trunk, gauges, gas tank, etc. from a list of choices which you know are not all compatible with each other. Fear of buying an engine that won't fit in the frame without extensive cutting and welding makes it hard to commit.

I suppose ultimately, I want a controller that is powerful enough to compute the kinematics with a high frequency and is capable of computing actual arcs rather than subdividing into line segments. Even these requirements are not easy to pin down because the software and hardware need to work in unison to do this. You can have capable hardware and load software on it that meets neither of these criteria.

Another thing that would be great is safety improvements, like redundant temperature sensors, hopefully using RTDs, as they are more reliable and accurate. Software that senses failures in the hot end and cancels the print would also be a huge improvement. I work with sensors quite often, and these things are not terribly difficult. I guess I'd even like to get involved adding these kinds of things to help the community, but that takes me back to the initial problem of not really knowing what board most people will adopt and fear of wasting effort on a board that is a dead end makes it a tough choice.

Anyhow, thanks for all the thoughtful replies.

Polygonhell
ULTIMATE 3D JEDDI
Posts: 2430
Joined: Mon Mar 26, 2012 1:44 pm
Location: Redmond WA

Re: Which controller will eventually replace RAMBo?

Postby Polygonhell » Mon Mar 28, 2016 1:43 pm

The reason we haven't seen a wholesale move away from arduino based solutions is that the bulk of the reprap market is insanely price sensitive, the fact you can pick up a basic arduino based board for <$60, means that most people do. And that means that most firmware development is done on and for those boards.
Boards like RAMBO are the highend versions of those cheap boards, that leverage that same software.

Now more recently we've seen various ARM based solutions, with new control software, Smoothie, Duet, BB etc.
But like all open software/hardware, it's going to evolve in the direction the authors are most interested in. Unlike commercial software, there aren't any real market forces driving development, their most important customers (the developers themselves) are already locked in.
The various GRBL derivatives Repetier/Marlin/whatever, exist because their developers didn't like what was already there.

As to G2 arcs, I know of no slicer that generates them, and they are only useful for actual round holes anyway.

User avatar
jmjcoke@gmail.com
Printmaster!
Posts: 39
Joined: Mon Sep 15, 2014 1:58 pm
Location: Belvidere IL

Re: Which controller will eventually replace RAMBo?

Postby jmjcoke@gmail.com » Thu Feb 16, 2017 7:57 pm

We'd be fools to think that Johnny and the crew at Ultimachine weren't working on some kind of super rambo. I've heard some rumors... I'm sure they'll have a Due or ARM based contender soon.


Yes they have, but until Seeme adds firmware for this, Marlin is your only choice.
Attachments
05348462-226A-4D95-9C1F-257D0C8E0D4A.JPG
Max 2
xyz DaVinci 1.0 with custom FW for Repetier-Host
Future plans
NextEngine scanner
Droplite or other SLA

dc42
Printmaster!
Posts: 267
Joined: Mon Mar 07, 2016 10:17 am

Re: Which controller will eventually replace RAMBo?

Postby dc42 » Fri Feb 17, 2017 6:08 pm

Going back two posts, I just read this:

Now more recently we've seen various ARM based solutions, with new control software, Smoothie, Duet, BB etc.
But like all open software/hardware, it's going to evolve in the direction the authors are most interested in. Unlike commercial software, there aren't any real market forces driving development, their most important customers (the developers themselves) are already locked in


That's not true for RepRapFirmware any more. The Duet and Duet WiFi have been hugely successful with OEMs, so we are very sensitive to their needs. Mostly these are to make things easy and simple to use when everything is working well, and to provide good diagnostics when it isn't.


Return to “OTHER Electronics”

Who is online

Users browsing this forum: No registered users and 1 guest