Because then I have two seperate boards to power, house and protect. Combining those into one would make for a smaller design footprint, in our case. With this particular product, portability and compactness is ideal, since it is intended to be used on job sites to mark steel banding or tags. Often times, these guys are working out of a very dirty trailer that's already crammed with equipment, so space is at a premium for them. The machine does nothing special, so size tends to be where the competative edge is in that admittedly niche market.

▲

mschuster91 3 days ago | parent | next [-]

> Because then I have two seperate boards to power, house and protect.

Well at the moment you have to power, house and protect an entire Windows PC...

Assuming a Raspberry Pi 5 is powerful enough to run the control software, I'd go designing a carrier board for a Compute Module.

▲

0xEF 3 days ago | parent [-]

> Well at the moment you have to power, house and protect an entire Windows PC...

In the case of this specific product, the customer supplies their own PC and we supply the software for it. No matter how you cut the mustard, it's not an ideal setup in my eyes. It's rare to see them us an existing PC that serves more than one purpose on these machines. What I usually see them do is buy a cheap Windows laptop and have that running next to the machine. So, as I mentioned, if I can eliminate the need for a separate PC at all, that's a step forward.

I'm currently exploring the Pi 5 idea, too. I've only been with the company for a few years, but this design was made about 15 years ago? The speed of getting them to change anything, especially when what they've been doing has been working, is glacial, at best, even though they're a small company with the potential to be pretty agile. I have a Pi 5 in-house right now with Windows 10 on it (courtesy of https://github.com/Botspot/bvm) as a sort of proof-of-concept that we would not even have to port our software to work on a different platform, but it's in the corner of my workroom collecting dust due to their lack of interest. Some days, I'm not sure why I try.

I see another commenter pointed out some Seeed boards that was not aware of, so I guess it's not that the options never existed, but more so the higher-ups aren't all that interested in changing something for the sake of future-proofing or being more efficient and cost-effective. This is, unfortunately, an extremely common problem in industrial automation. That and, as much as I try to keep up, I'm also admittedly not fully cognizant of all the SBC/microcontroller options out there, so I appreciate discussions like this.

	▲	HeyLaughingBoy 3 days ago \| parent [-]
		It can be hard to make a business case for something like that, so I'm somewhat sympathetic to their position. I've run into the same problem: the cost of changing can often be greater than the amount you, or the customer, will save by doing so. As far as future proofing goes, my experience is that no one cares until they can't buy parts. Then panic ensues. At my last job, we kept running into this problem with one customer. Every time they placed an order for the hardware we built for them, it kept getting more and more expensive because we had to search for obsolete parts and charge them a premium. But the amount that our price increased was dwarfed by the amount of money that they made selling their machine, so they literally didn't care. And as long as they happily paid us, we didn't really care that much either.

▲

raphman 3 days ago | parent | prev [-]

FWIW, Seeed Studios offers a range of Intel SBCs with an embedded microcontroller that is explicitly meant to be used like an Arduino board [1].

[1] https://www.seeedstudio.com/ODYSSEY-X86-v2-board-p-5075.html

	▲	0xEF 3 days ago \| parent [-]
		I was not aware of those, thanks. Certainly up for consideration but I can't imagine that price point being attractive to the decision-makers in my company.