A nice place to discuss rumors, happenings, innovations, and challenges in the technology sphere. We also welcome discussions on the intersections of technology and society. If it’s technological news or discussion of technology, it probably belongs here.
Remember the overriding ethos on Beehaw: Be(e) Nice. Each user you encounter here is a person, and should be treated with kindness (even if they’re wrong, or use a Linux distro you don’t like). Personal attacks will not be tolerated.
Subcommunities on Beehaw:
This community's icon was made by Aaron Schneider, under the CC-BY-NC-SA 4.0 license.
The equipment required to "make" a motherboard is orders of magnitude more expensive than anything you could afford.
There's a reason why it's all custom designed and there's only a handful of board manufacturers in the entire world, most laptop companies don't even have their own fabrication for these pieces, they just do the design and final assembly.
Yah, that’s kind of what I figured from looking around.
You can order a board from the likes of PCBWay... you can even ask them to populate it and solder the components for you... but just designing a board with the complexity of a laptop, will require a lot of expertise, a lot of work... and most likely require several iterations where you'll end up with a bunch of trash boards with some flaw in them, and you'll have to pay for it all.
It's still a tool you could use, but I'd recommend sticking to something like simple adapter boards at most.
See, I knew about PCB way and was wondering if their products would even really work. Or if the sockets for the chips would be available for small batch stuff.
Even if you were able to make your own PCB and somehow solder everything onto it, one of the things that makes complex boards like motherboards so tough to make is signal path lengths. Ever see how some of the traces on motherboards are squiggly and take up more space than the straight ones? That isn’t just for fun - all of the traces have to be incredibly specific lengths for a whole number of reasons, including signal timing and interference with other traces.
"Sockets for the chips" are the least of the problems, most chips don't use sockets at all, they instead get soldered to a board through BGAs. Even multi-chip boards like CPUs, usually get soldered through BGAs. Then, what you need to think about is what kind of boards you want to connect and how.
If you wanted to make something low spec, you might look at the Raspberry Pi Compute Module, that only needs a sort of "breakout board", where you could place a bunch of M.2 connectors for daughter boards... but chances are you will be really hard pressed to make a whole laptop that's cheaper than a much higher spec brand one.
You could order from the same supply chains that Brands do but the up front costs and lot size requirements would be too much for anyone other than a major brand. Plus it would have to be worth their time and not upset their other customers. Big barrier to entry. I do not doubt it could be done, but at a very high unit cost.
If you don't need power then why not buy old laptops? Plenty of companies around that sell corporate laptops 5+ years old that of course still work fine. Also gives bonus points for reuse and thus reducing ewaste instead of adding to it.
Maybe get one of those Mini-PCs from Amazon and see if you can fit its guts into your laptop case.
There also are pre-made laptop projects that take SBCs like the Raspberry.
It's not cheap by any means, but MNT Research has done a lot of work for you. You can buy their Reform and Pocket Reform laptops, but also all of their designs are open source, so you can start there and tweak it to your own design if you really want.
Maybe you can find a second hand pi-top laptop, and tinker with your own SBC, and go on from there making your own ?
I will be interested in what others say too.
Use to be that there were quite a few white box laptop suppliers but not so many these days. Even with the white box suppliers lot of the components were custom. This is one reason my wife and I still have desktop computers. The laptop and mobile market kind of sucks for standardization.
I can kind of get why the standardization sucks. Everything is so packed together in most designs, having standardized elements might limit a companies options for building something compact.
I do wonder how much extra bulk a bunch of standardized elements would actually add though. Like would the average person really mind the laptop body being half cm thicker?
The "average person" wants a laptop that's: cheap, unbreakable, lightweight, thin, high resolution, small, fast, and with a ton of battery. Suffice to say they're contradictory requirements.
The most "standardized" laptops, as in easier to find spare parts for, are from brands popular with businesses, like Lenovo or Dell ones.
I think it is about product differentiation and the cutting edge. Also Apple has historically defined the trends even for the non-Apple markets. Apple reaches for the next high end thing, then it filters to others after they prove it out. Similarly on the PC side one has to sell an idea to the brand's first. So design is heavily brand driven and hence custom. I use to work in the industry. You would not believe the pressure to drive out a mm of thickness, an ounce of weight, a watt of power.
So do customers care. I do not myself. More normal people, well they will buy what companies like Best Buy choose to sell and push though the size, weight, features, and design do matter presumably.
It’s a common issue in a lot of industries, where internal priorities and concerns end up having more influence on a companies decision making than actual consumer wants and needs. Why the idea of “the free market will optimize for the best products” kind of falls apart in reality so often.
What concerns you about Framework's chipsets?
If you need cheap, and "built for a group", an existing SBC and custom-printed enclosures are likely the best option.
I was interested in using older cheaper CPUs but they only have relatively modern stuff, which makes sense, but isn’t what I was looking for, and I’m looking at the SBC stuff now because that seems like the best option.
Oh I see what you meant than. Older and cheaper would definitely fit your application better and all of Framework's stuff is expensive. Repairable and upgradeable comes at a cost!
Something I always found interesting was a cyberdeck which is kind of like a mishmash of components. Nowadays you can use sbcs for the hardware but I've always been given the families ewaste and doing some research you can reuse a lot of it. I have an old celeron laptop that's too slow for anything useful. It's just the bare board now ziptied to the lid of a shoebox and I use it for low power learning server. But found out the display connector it uses for the screen, which is gone, is very common and found a 9 inch screen that would work with it. With a 3d printer you can build cases for everything.
Yah, that’s where I kind of started from in thinking about this, but then I started looking in to getting off the shelf components and ran in to the reality that short of cramming a desktop motherboard in to a laptop shaped box, there was just no way to make it work. The cooling alone was a no go, so it was a matter of using SBCs or ripping something out of an existing laptop, which kind of defeats the point unless something else is broken in it.
You can buy just the motherboard of a Framework laptop and build something around that. There are even cases for them available to print.
Interesting idea. The netbooks that ASUS introduced and Google Chromebooks are good examples if minimal hardware. ASUS had to stop promoting and apologize to Microsoft for competing with them but Google of cousre did not because it could go it on its own and has deep legal pockets.
Other thing is to think in terms of minimum usable product. A barrier there is that it needs to be enough hardware to drive a web browser and a screen that is or is near HD. This tends to mean a lot of memory and computing power.