Custom/Turnkey SAFEnetwork farming hardware


You can put me down for one or two. Fantastic idea to get more people
involved and grow the network. :grinning:


@happybeing I certainly hope that there will be more than one vault system! I like @dirvine’s concept of using it to replace the 21 Bitcoin computer in concept and function. That’s very solid-- imagine everything – emanating truly free internet. I hope that the design proves as scalable as I’ve heard!

@sam_uk Plinth is awesome. It has Freedombox has good support for Sunxi, and is definitley a candidate in my mind.


Now we have the PI3 (ordered a few), Element14 may be a good choice for logistics. I hope to meet with them soon, or as soon a I get my head above water in the network. Custom configuration and logistics handling all makes sense. It lets us all do what we are good at & will spread the load and project a bit further.


I’d like to get some insight into your concept here-- without significant storage, how will these devices generate safecoin?

Is transmsision valuable in this economy you’re building in your head? Compute?

Also… maidsafe really represents a financial coup. Bandwidth does have real value-- as does storage and compute. Probably more than most precious metals in the world, put together are worth. That insight alone may be enough to rock the world.


The beauty is they do not need to have significant storage really. Lots and lots of them with 64/128 Gb will be loads for now (of course larger usb drives or maybe swap hdmi for sata connection is also a consideration).

I am looking to increase the number of vaults considerably on launch and with lots of people being able to plug in and earn then it’s cool. Now to earn, you have farming, of course. I also am playing with the concepts such as sharing bandwidth (pay to relay connections), setting up portals for passers by, allowing registered local clients via udp service_discovery (see crate) to bootstrap of their own device (securely) and a few more items that I need to test first. SDR is down the road as well, but a possibility along with some other mesh tech hardware.

I hope to get a few weeks atr this in April and maybe then can present a neat solution/project for us all to take forward.


I understand.

I understand so completley-- I have been trying to do what you’re describing for >10 years now. If I can ever lend a hand, let me know.


@dirvine break the world, sir, break it so well that somehow, everyone will have a voice. If, collectively, we can pull this off, well, I imagine you know-- it’s the best possible thing for the species. If we don’t pull it off, we will be just that much further from “star trek.”

If it gets as large as I now know you’re hoping, there will end up being payment tiers for everything and specializations. Just for example-- google is looking for less reliable, larger form factor hard disks because Youtube grows at something like 1PB/day (imagine that!)

There’s definitely an advantage to small, though. I suppose the trick, overall, will be being small and large while managing to incentivize (for example) storage on the scale of Youtube. Thanks for the insight :)!


They were Farnell in an earlier life weren’t they?

They can be expensive, but maybe we can haggle the selling price down using RS Electronics as a competitor to the business.

Maybe you can do what the Pi did and promote the work the foundation does with children and education on IoT and like the Pi force them to sell it like RS Electronics had to do with the Pi. Keep it at a certain price or you remove their rights to the license to sell it.



There is no money in hardware. I’m no admonishing you for looking for inexpensive, high-quality hardware, either-- that’s how I make my living from time to time. That said, The margins on a lot of hardware exists solely to subsidize the software development for the platform. Take Rpi, for example-- its advantage over the millions of Chinese linux boxes is that if I just used an RPI, I’d know that the kernel was already taken care of for me, etc.

Point being: I strongly doubt that they’ll budge even a centimeter on price, but I encourage you nonetheless to try :)!


Great idea! Wasn’t there a school teacher in Australia on this forum once upon a time?


You might want to look at our Bitseed unit.

We currently ship with up to 2 Terabyte hard drive capacity, adding 3 Tb if there is interest or in special order. We have had the units operated by customers as full Bitcoin nodes for over a year now. Bitseed V2 has a dual core ARM, 1 Gb RAM, 8 Gb eMMC, 1 Gb Ethernet, 2 USB 2 ports.

It also sells for a lot less than 21, and unlike 21, includes a hard drive or 480 Gb SSD - $169-$299 depending on choice of drive vs $399 for a 21 with no drive included other than a 128 Gb SD card.


Nice, yes I had heard of this but never investigated (sorry). It seems like this would be a way with little effort to get vaults into the hands of many folks quickly. Lets keep this close for sure. I think the thing I am working on being a reference platform (and ARM) will be an easy install here. Additions would be things like SDR mesh etc. which I suspect both projects would really like, so more than happy to collaborate here indeed.

You have got to love open source and what it an bring to communities. Nice interjection and thanks very much.


Like it. Nicely done. But apples/oranges re the 21BC. Lets not forget 21 is also about a community of developers building apps / ecosystem, products and full compliment of slack support. Once the SAFENETWORK is full tilt, and assuming your product meets the highest standards as defined by @dirvine & Team, I encourage you to dedicate some of your web space to marketing the bitseed to MAIDSAFE farmers in the same way you do to folk interested in running Bitcoin nodes. Publishing / promoting information about ROI opportunities (earning safecoin) while using the bitseed can only serve to reach new markets. There are some very competent technical writers/bloggers in this forum that I am certain, when the time is right, would be willing to contribute a thorough introduction to the SAFENETWORK on your weekly blog and how bitseed (or would that be maidseed or safeseed?) products are best integrated.

It could be some time before the SAFENETWORK goes gold but now is the time to start building relationships and introducing your products to the influencers at MAID and on this forum.



Will this “21” alternative PC able to run as it’s own ISP? We need to build meshnet.

cjdns runs perfectly well on Begleblack and this will be a great viable solution to ramp up the maidsafe + meshnet into higher gear.


Thanks, David for your quick reply and interest. Congrats on the upcoming release of the MVP and the appreciation it is receiving in the markets. I was an investor in the ICO and have held since, expecting Maidsafe to do great things. Two of us went to the Maidsafe Meetup in San Francisco a while back and liked what we saw. We demoed the Bitseed unit V1 Bitcoin Node and saw other hardware being demoed as well for Maidsafe.

The whole focus of Bitseed has been to provide an easy to use, turnkey solution for users to operate the hardware and join decentralized networks like Bitcoin and Maidsafe. Adding the personal server functionality for secure file storage, sharing, and messaging is on our roadmap for V3, which will run on a 64 bit CPU with 2 Gb RAM along with the SATA storage.

Interesting that you bring up SDR mesh networking, as this was the original goal of Bitseed. We researched a lot of different SDR based wireless chips including WiFi and TV Whitespace to implement mesh networking for long distance and local higher speed applications.

For mesh networking to work, it needs to have local storage distributed across the nodes along with store and forward packet routing with custodial transfer between nodes, like NASA Delay and Fault Tolerant Networking, which requires a lot more RAM than found in a typical router. The ARM boards with the higher amount of RAM can fill this niche.

While waiting for mesh networking software to develop more fully, we have been focused on developing personal secure storage and messaging, looking at Sandstorm (very nice but only runs on x86, not ARM), Blockstack (still in development), and Freedombox as software stacks.

Maidsafe looks like it will be a perfect fit when it is ready. We have been running CJDNS on test units. The new 64 bit, 2 Gb RAM boards like Pine 64 look very promising once a low cost version is available with SATA.



Looks like I gotta invest into maidseeds.

Edited: Everytime I read turnkey, I read it as turkey.


Thanks GL for your insight on 21 and marketing suggestions. Yes, as you observe, a lot of the attraction of 21 is the API and development ecosystem surrounding it. We have received a shipment of mining chips to experiment with adding it to Bitseed V2 via the Arduino headers inside the unit, though our main focus is not mining. We would welcome partnering with software developers who would like to extend the capabilities of the unit in an open source manner. Working more closely with Maidsafe towards this goal would be ideal.

In marketing the unit for storage based farming, we’re looking at that, but just need to make sure we can present a case where using our unit with internal SATA but at a higher cost per terabyte has an advantage over using a lower cost Raspi based unit with consumer USB external drives. Providing it as a turnkey unit with integrated storage having proven reliability is one way to address this.


Perhaps you can clear up something, early press releases from 21 mentioned that the mining chips would return, i think 25%, of the mining to 21 and the rest to the address given to the chip.

Is this still the case? Have they now hidden that, or do they just reward the address given to the chip only 75% out of the mining pool?

BTW, great sounding system you have there.


And thank you for taking time to respond.

I see the makings of a solid relationship.

Stay tuned.


Super excited.

If I can buy a unit like this one which is plug and play ready as either a farming only rig or something I can use for personal use too and which I can upgrade I guarantee to buy a few if not more of these things in due course!

Question though, would you pre-configure (if thats even possible?) for SAFE and would you remove some of the GPU buy maybe add some ram and tb options?

Also what about cooling? Is that something to consider?


Yes, we pre-configure the Bitcoin units for plug-n-play operation. The Maidsafe version will be as well.

We don’t have much flexibility on the RAM since it is soldered on the board with no expansion slots. The RAM has been the biggest limitation in terms of how much can run on the device, much more so than CPU capacity. We will be going to a 64 bit processor with 2 Gb RAM for V3. The GPU is actually built into the processor so it isn’t separable and doesn’t consume any of the board real estate. Cooling has not been an issue with completely fan-less operation.

We have had some upgrades come in for larger hard drives, since we started with 180 Gb drives, and have migrated through 250 Gb, 320 GB, and now offering 1-3 Tb units.

We see a very promising niche for a turnkey consumer device which runs Maidsafe and provides its resources to the user without him ever having to set up accounts outside of the device or manage account balances. It functions more like a simple router and NAS box with storage backed up by the decentralized cloud as a gateway to the Maidsafe network and all of its features with a simple, easy to use GUI.

The underlying technology of encryption, decentralized storage, and network communication is transparent to the user. The unit has sufficient resources to earn enough Safecoin to pay for what it uses with low enough power consumption to be left on all the time, but doesn’t try to be a highly profitable farming operation. It could be used for higher capacity farming by adding external 6 Tb USB drives but I expect much of this type of activity to take place in data centers.

The unit also acts as a hub for its owner’s other computers, mobile devices, and IoT devices, replacing proprietary cloud based platforms which are security risks to the user.

The level of simplicity I like to see is like Apple Time Machine, which keeps backups with great reliability which the user never has to think about it, and the interface is very intuitive to just go back in the timeline and restore the previous version of a file when desired.