Out of interest, what are the main concerns you have? You've got a tried and tested design, have experience writing content and creating videos, you have an existing good reputation and followers, etc. You just need to get the pricing right, and that primarily just means being realistic.
I do not want to make a profit from selling the boards (we do not sell boards, we have enough income from our other activities), but because there is a lot of people who like the board, I would like to give them opportunity to buy it.
Have you considered talking with Chinese vendors such as Seeeds and let them handle the manufacturing sales and fulfilment? If you don't care about profit, just about making them available, than cloning should not be an issue.
What I am worried about is, that a lot of people do not see the real cost hidden behind manufacturing and selling the boards.
+1. So far i am left with figuring out on my own what is it that would make me buy a Rex board.
I wonder if you are able to explain to the potential funders why this is a better board for learning than existing ARM boards like the Raspberry Pi, BeagleBone Black, etc.
Guys, please help, what the project would need to have, so you would pay $300 USD to play with it?
Anyway, you are one of the world leading expert in Altium Designer, and frankly speaking, your design document along worth more than the price you wanted.
Maybe you could follow Arduino examples .... And since you recognize the problem of difficulty, you may have a chance at this. Good luck
- I think US$300 would significantly limit the sales.
- Seeed Studio are pretty good to deal with, they've built and sold one of my designs.
- Alternatively have you considered teaming up with someone in Europe that also does open hardware - like Olimex or MikroElectronika?
- Maybe a bundle at ~ with both the course and the Rex?
- BTW, that's the most beautiful schematic I've seen in my life. What a piece of utter excellence!
If you want to cut the cost to lower than $150, try doing these:
1. Change part numbers of connectors to non branded ones, and buy them directly from Alibaba. Good quality connectors make no sense, except for the USB ones.
2. Use cheaper sensors. There are lots of new products that are not mature, so you need more time on developing their drivers, but the silicons are much cheaper.
3. Remove the SHT21, I do not see any reason why an ARM board should have it.
4. Use cheaper RAM. The SoC supports only up to 533MHz RAM, so do not waste money on DDR3 1600.
5. You can cut the cost of the temp sensor, use the integrated ones instead (some MEMS have internal temp sensors, your MCU has, and probably your SoC also has one).
6. Change tantalum caps to aluminum ones. This is very important as cable inductance, when applied on a very low ESR cap, can cause over voltage.
7. Replace low current DC/DC converters with LDO regulators.
8. Replace your Ethernet PHY to cheaper ones from Realtek or Marvell. They do not offer documents to small players, but illegal copies of their datasheet are easily available online.
9. Do not buy parts from DigiKey. Mouser and Newark are considerably cheaper than DigiKey for most of the times. On the SoC itself you can save up to 6 bucks at 1kpcs.
10. Get a cheap PCB assembly fab. Some Chinese companies charge you ~$500~$1000 setup fee, then they charge you very less per board.
11. Seeed or iTead are not the only Chinese manufacturers. Try to find services from factories directly. You may need a Chinese translator. Get a college kid and pay him for $5/hr.
Applying all of them can you get ~$100 cost for a board at 1kpcs. Including logistics and tax and software maintenance cost, I think you can sell them for less than $150 each.
- In other words, the intention of the buyers are actually to rip off your design and integrate them to their own products.
- Then you are targeting at high level pros. Therefore, price it higher, and I do not think you can sell even 1k of them even at $150.
- Does Seeed pay everything (including components) by themselves and they just send you agreed profit or you have to pay them something before they start manufacturing?
If you have a look at the other companies, they have teams of people for PR and Marketing. And, it's only 2 of us. OpenRex is not even our main project.
I wanted to work on something what we would enjoy designing and what would be dedicated only for people to play with. Something what I could handle to my son, he could start playing with a microcontroller, LEDs, buttons, motors, sensors, ... and when he is ready or need something more powerful he would move up to the CPU and Linux. No installation.
There are plenty of boards and plenty of software, but it is actually not very easy to learn program.
When I was learning (on ATARI), it was easy. I had one book and I wrote the instructions and that was it. Now, you have to install a huge software first, configure it and when you create a new project it already has tons of code inserted. I have to sit down with my son and explain him what things mean, it's hard for people who are starting ... to actually start.
That's may plan about OpenRex - to teach how to design quite complex hardware, to teach how to write firmware for microcontrollers and to teach how to do Bootloader & Linux stuff for ARM boards. All in one place.
Unfortunately, I am not a marketing guy (I am an engineer) and we do not have team of people creating nice videos and blogging / tweaking / facebooking, so I may not be able to explain people why OpenRex is better than the other boards
Guys, please help, what the project would need to have, so you would pay $300 USD to play with it?
And what I hate most are the software guys who don't want to share their know how. It's so frustrating. Most of the programming stuff is very simple if you know how to do it - but no one wants to tell these simple things, they keep it for themselves. If you ask a software guy, he will tell something like: "It's simple, just write there this ... google it ... follow the manual ...", but they never tell you the exact steps and you have to spent hours and days trying to find the one command line which will do it.
But realise that all the other board makers' intentions were precisely that. Seeing how they obviously haven't done so well even with relatively large staffing dedicated to it I'd wonder how 2 people for whom it's not the main project could manage to create an ecosystem of libraries, docs, courses and examples that would shake the world. The amount of work to create this, then more importantly (where it usually crumbles) maintain it is tremendous.
Would love it and for sure wish you luck, but without much hope
Then how are you going to make courses/tutorials that are appealing to the current generations? It's (for good or bad) essential nowadays to create things in a "modern" format and with matching promotion.
I'm starting to think you should pretty much forget about the board beyond being the example for the existing "hardware design course" product and not lose your time/money on producing it/offering it for sale like most others, but rather focus on what you actually want to do i.e. the software and documentation work and use another existing and cheaper board as a support like the UDOO neo that shares much of the philosophy (and hardware) and that you couldn't compete with at $50. Maybe you can sell your "soft" material for $250 to use on an UDOO board, instead of letting it be an "aside" for a $300 board that will cost you $200 to produce.