Originally Posted by
ad_bfl
I just got a Galil motion controller based project off the ground, and I have been using Mach3 and EMC both for a couple of years.
All I can say after having used this "industrial" strength Galil motion controller I will never go back to General purpose desktop CPU motion controller/environment, even a really good one with a RTOS like EMC, or a Mach3/smoothstepper combo.
Clean pulse trains, quick response to a E Stops, homing, syncing multiple servos, gearing, I/O, make the cost worthwhile.
One thing I have not read in this thread is issues with RTOS based solutions in general. Dig a bit in that direction, try writing some "real time" code on your RTOS of choice and suddenly a stand alone motion system based a CPU and a BIG FPGA make a lot of sense.
And before everyone starts talking about how expensive those solutions can become, look at the overall cost and the performance/results. IMHO it is like the difference between buying good tooling vs low end stuff, no comparison.
Absolutely on the mark. I know a large collection of folks who have taken this path, and my own personal experience matches.
Whether you prefer Mach3 or EMC2, adding hardware between the machine and the PC control will make either a much better solution. That's the gist of my two part series on motion controllers:
(Part 1 of the series: Motion Control Boards Take Mach3 From Hobby Class to Industrial Grade, Part 1 « CNCCookbook)
The future of low end CNC is going to be about these hardware solutions moreso than Mach3 vs EMC2. In fact, the more likely you think it is that you'll drive these machines from non-PC's like tablets, the more of a requirement it will be to backstop the weaker machine with hardware.
Some on the thread seem to regard hardware assist as a hack, but in fact they have a long and distinguished history as being how the better controllers in the industrial world work. The hobbyists didn't invent them, they're only just now getting access to them.
We're already seeing tons of hardware development, more than ever before. We're seeing folks build simple g-code interpreters on Arduinos. The college kids have this bit firmly between their teeth and they'll make some innovative things happen. Hardware has just gotten too cheap not to take advantage of it. Even Gecko are talking about going to a higher level interface and putting more intelligence in the drives.
As for some of the other points being made:
- ger21, you may be willing to conceded EMC is a much better control, but I'm not. I use both Linux and Windows, and I have more machines running Linux, so that isn't my issue. I see far too many arguments made on religion that don't successfully translate to a real benefit. These two are both good controls with different strengths and weaknesses and there are real benefits for EMC as well as Mach. So choose carefully and don't get caught up in the religion.
- Agree 100% that the free argument is a red herring. My controller PC's are all used and older machines. They already have Windows. If anything, it's a pain to strip it off and reconfigure. There's tons more older machines available for the taking. My mill is run by a $100 laptop that doesn't even have a parallel port. Works great with Mach3 because of the Smoothstepper.
As was pointed out, the OS cost was a fraction of the cost of the overall machine. What wasn't pointed out, is the time to learn a new OS if you don't already know it is much more expensive than the cost of Windows and Mach3 combined. You're kidding yourself if you think a casual user is going to pick up Linux in an hour or two.
As I said before, if I had a machine with idiosyncracies I'd choose EMC and write the custom code to deal with it. If not, I'll choose Mach3 precisely because it'll cost me less time to get it to work well. What are idiosyncracies?
Something other than the usual 3-4 axis mill or lathe. Something that Mach3 just couldn't handle. There isn't much out there along those lines, but that's not to say such things never come up.
Others will have their own criteria for why they choose. That's really the point. Don't let somebody else make the decision for you. Gather all the info. Decide which problems you need to solve and which advantages you'll need to be successful. Then make your choice.
If you don't feel like you're ready for that, best pick the one that has the largest availability of knowledgeable folks to help. That's a big reason why Mach3 has such a following.
Cheers,
BW
Try G-Wizard Machinist's Calculator for free:
http://www.cnccookbook.com/CCGWizard.html