New FP2 Motherboard

From my side of view the cooperation from the company with the community needs more roots.
What @Leo_TheCrafter (and some other very motivated poeple here) did so far is an enormous effort and deserves all honour.
But there is no communication between the company and the community an I think this is huge gab, as the efforts just remain as a drawing or a prototype an finally end up in the trash can.
Together we all could achieve so much more if both partys just talk together.
Don´t get me worng, I do not complain, I just like to have more efficiency.
My idea:
The community proved they have some very nice skills. Let´s do something the company needs. Thy could come up with a project board, where they offer some projects to keep their workforce clean for other, more important projects.
Of course that can´t be made in a public, but the skilled community members could register themselves as a creator, programmer, designer, solderer, whatever. So they can apply for a project an get involved, also maybe as a team of several.
With this organization a new bottom module (or maybe even a new core module) should be able to create in just some weeks, even the prototyping should not be a problem. And finally the company could let it be manufactured at the fair certified companies they uses for the other parts.

6 Likes

As an owner of both, the FP2 and a GPD Pocket1 I really love the idea of combining those two devices. Especially the x86 CPU would make a lot of things easier - like being able to use mainline kernels and a desktop OS. I generally perceive the android-SW environment as kind of toxic.

Eventhough I wasn’t reading the whole thread I still would like to share some of my thoughts:
) the CPU seems a bit overpowered for purely passive operation; wouldn’t a slower CPU be a better choice so it doesn’t throttle after a few seconds? As there’s not enough space for a passive cooling concept this must be quite critical IMO.
) I remember the Pocket 1 to have had some serious issues with some sleep modes in Windows and Linux; I think it was s3 or hibernate. There were rumors it’s due to a limitation of the cherry trail architecture. Are those issues fixed already? I guess these must be pretty important on a mobile (I don’t use them on my laptop so I can’t really say if it works already).
) 8GB of RAM is really nice; as I’m doing a lot of CAD/modelling/slicing on my Pocket I’m very glad of it. I wonder what applications on a phone would require that amount of RAM. Ok - I could use it in convergence mode but then the purely passive CPU will soon throttle to uselessness. But maybe some kind of fan-driven docking station could help here.

However - if you really manage to produce some of these PCBs I’d be glad to buy one for approx. up to 500€

3 Likes

The CPUs are availeble in different clock speeds, and the RAM is too. Even for the eMMC there are one’s availeble for 32GB, but do you really want that, if you have to spend a few hundred € for a new motherboard. But the lower spec CPU is a good option.

1 Like

For phones you don’t want S3, you need a state in which wireless connections are maintained, and apps are able to wake up the CPU at scheduled times to check if there are any notifications.

S3 is too deep for this, in S3 the CPU only turns on when the user sends a wake up event that is recognised by the BIOS (for instance, power button).

3 Likes

Hello,
@Leo_TheCrafter are you making process?

Stefan

No,
even though I have a design, prototyping it would be too expensive, but
I have some hope with diy PCB

3 Likes

I hope the best, what can and what cannot be achieved by DIY PCB?

Any feedback from the Fairphone team yet?

1 Like

Would it be an option for you to take part on the #efct19 ? That would offer the possibility to talk to the Fairphone team and discuss your ideas face to face.

6 Likes

This looks like an opportunity for crowd sourcing to me. Did you consider a campaign at Kickstarter or Indigogo, yet? Of course, there is a risk to promise too much too soon. But how about using the exact state of the project. That is, with no relevant development to be done during or after the campaign.

The campaign would not offer a completely assembled working device but a kit. The kit would include the PCB, a solder paste stencil, components and full documentation including instructions required to assemble and do some basic testing. The crowd sourcing would not need to collect thousands of orders. It might even be wise to limit the volume to maybe 50 sets.
The goal would be to bring down the cost per piece compared to a single prototype. At the same time the low number would avoid to get overwhelmed by the demands of mass production.

I for one would immediately jump to accept such an offer. (Yes, I have the means to assemble BGA components. Could even offer to help out others on a case by case base.)

The prototypes would be expensive compared to a Samsung with the same specs. Also, production of the PCB would probably be standard, so not explicitly more fair than others. But I’d see this as a necessary step towards a fair and open phone. My emphasis would be on “open” as in “open source hardware”. That is, all necessary design files publicly available under some kind of free and open license and done with free and open software.

4 Likes

Well, I guess starting a fundraising project on kickstarter or the lilke would need some legal planning, as you are (hopefully) actually selling some product. As far as I know, you will be a business and have to regard all the legal liabilities (consumer protection, product liability etc.).
It takes - at least - a business plan and some calculations, that take all the costs for repair- and warranty cases into account, when you fix a price for your product.

1 Like

Please don’t underestimate how much work running a good crowdfunding campaign is. I’ve been involved with 2 and it’s a full time job if you want to succeed. And that’s not even taking into account the need for a good video and text.

4 Likes

I would much rather prefer producing the PCB together with someone that knows what they are doing. Maybe someone on the forum knows someone that knows someone . . .

There’s a few iterations on that product (GPD Pocket 2, for example, which I have). I have trouble with warranty with GPD (clear artifact spot of display assembly), so I cannot recommend the device. The device also becomes very hot under minimal load. Reapplying thermal paste [1] or applying thermal pads [2] can help.

[1] Applying new thermal paste on a GPD Pocket 2 for better thermals

[2] https://www.reddit.com/r/GPDPocket/comments/ayynmr/thermal_mod/

That being said, how will you handle the heat issues on this Atom SOC?

1 Like

I’m not sure. The thing is the atom processor was the only SOC I could get my hands on a datasheet. If anyone got a suggestion for an other SOC with a availabe datasheet I would love to hear

2 Likes

Yes, your choice seems to make sense because you want to run Microsoft Windows on it, it has to be AMD64 (x86-64) because Windows doesn’t run on say ARM yet. AFAIK everyone uses either ARM or Atom (in that order). I don’t think AMD has anything in this segment yet, though their Ryzen are going to be seen in “netbooks” more often.

The Librem 5 appears to be using the i.MX8 https://en.wikipedia.org/wiki/I.MX#i.MX_8_series but that is an ARM-based SOC.

The new Gemini PDA called Cosmo Communicator (which I also backed) is going to be using MTK so its RIP right after release.

2 Likes

I’m OK with using an ARM SoC, so the i.MX8 might be a good choice, but I have to some research, but I was able to get the datasheet.
Thanks for this idea.

2 Likes

I think this CPU is usefull, I just have one problem: Where can I manufacture it ?

It would be nice to know someone that can prototype these PCBs or even someone that has already designed HDI PCBs.
Maybe someone out there knows someone.

Not necessarily afaik. I backed several campaigns on Kickstarter & Indiegogo. But my return rate was way better from Indiegogo than from Kickstarter as Kickstarter not really assure delivery or refund. So there were some campaigns where I lost my money with many other backers.

(Excuse me don’t have much time right now to write this well…)

https://www.nxp.com/support/sample-buy:SAMPLE-BUY

https://www.nxp.com/support/sample-buy/buy-direct/buy-direct-and-samples-faq:WTOBUY_FAQ

https://www.nxp.com/support/sample-buy/distributor-network:DISTRIBUTORS

https://www.nxp.com/products/processors-and-microcontrollers/arm-based-processors-and-mcus/i.mx-applications-processors/i.mx-8-processors:IMX8-SERIES

1 Like

Maybe I did not make my point clear or I do misunderstand you right now:
Projects on Kickstarter, Indiegogo and the like do not guarantee that you will receive the product. The campaign may fail and you loose your money.
But if the product is delivered, then the producer is liable and has to provide legal warranty etc.
Well, that’s obviously no legally reliable statement, just my understanding.

2 Likes