You wanna lock something down then do what they are doing, keep the design a secret. Then only allow a select few access to this type of information and this will force most tinkers to buy commercialized interfaces. Which seem to be very expensive compared to the Pi. This way they can make some profit from that side of the business, then aren't making anything on the Pi. Hey they have to make a living too! I've ran into similar issues with other companies where they wouldn't release critical information about accessing data in their chips (I2C protocol). This was an even bigger problem as you can imagine. Instead the wanted me to purchase their interface board and give me their code. I of course decided to drop that chip and go somewhere else. After that I finished the project I had been working on using other ICs. Later, in my (spare time) I decided to hack the chip. That was a lot of fun and eventually I figured it out on my own and in doing so mapped the entire chip from how it addressed down to the # of registers it had. Same thing can be done here. Might burn up a few boards in the process but that's how you learn to design. So although it would be grate if they release the schematic I can certainly understand why they don't. This is just another one of those challenges that help us grow. Be grateful for what they've done to date as this Pi is just incredible. It's one of the most innovative ideas since the invention of the computer. With that in mind, having a schematic might just take out all the fun in building new interfaces and deprive us from experience that could otherwise lead us to even greater achievements. After all most of the fun is in the struggle to understand, so having a schematic is just a minor inconvenience that can be over come through careful testing. For me, the GPIO pins are the most fun as all the other stuff has been worked out. Have a great day and have FUN with your Pi.
1) The only interface you can "commercialize" on the Pi (GPIO) is documented. So anybody can, and many already have "commercialized" it. There are
several Pi "look a likes" that use different SoCs but include a Pi compatible (or nearly compatible in the case of the ODROID C1 and C2) GPIO header.
And ALL of them that I know of publish their schematics.
2) The internal interfaces of the Broadcom SoC (for I2C, SPI, Serial, DMA, interrupts etc.) and the registers they use are fully documented so anybody can
write software that uses them. I think the USB interface which is not Broadcom IP is the only exception to this.
3) Even the GPU in the SoC is fully documented which is something that no other ARM SoC has done. This allows the GPU's processing power to be
used for purposes (innovative or otherwise) other than graphics. Because of it's complexity it is also a part of the SoC that you could never (OK
within your lifetime even with help) fully hack.
The only thing that is not documented on the B+, 2, and 3 is the schematic ! With the main component of the Pi so well documented it makes no sense
to not publish the schematics and they have never published anything that even vaguely approaches a reasonable explanation for this. About the only
thing left is that this is to prevent a company from directly cloning a Pi using the same Broadcom SoCs. ODROID tried to do this but when push came to
shove they could not get Broadcom to sell them the chips.
If you really want to understand a schematic is the one of the two most fundamental parts of that. And it is essential if you trying to determine why that
"innovative idea" (hardware, software, or both) is not working. It also helps to greatly reduce the number of boards sacrificed in the "innovative" process.
The Raspberry Pi Foundation (AKA Broadcom ?) made an incredibly stupid decision with no advantages to them and a great disadvantage to anybody
that "really wants to to understand" the Raspberry Pi in all its variations. This is underscored by the fact that they have never make any attempt to justify
1) The only interface you can "commercialize" on the Pi (GPIO) is documented. So anybody can, and many already have "commercialized" it. There are
several Pi "look a likes" that use different SoCs but include a Pi compatible (or nearly compatible in the case of the ODROID C1 and C2) GPIO header.
And ALL of them that I know of publish their schematics.
2) The internal interfaces of the Broadcom SoC (for I2C, SPI, Serial, DMA, interrupts etc.) and the registers they use are fully documented so anybody can
write software that uses them. I think the USB interface which is not Broadcom IP is the only exception to this.
3) Even the GPU in the SoC is fully documented which is something that no other ARM SoC has done. This allows the GPU's processing power to be
used for purposes (innovative or otherwise) other than graphics. Because of it's complexity it is also a part of the SoC that you could never (OK
within your lifetime even with help) fully hack.
The only thing that is not documented on the B+, 2, and 3 is the schematic ! With the main component of the Pi so well documented it makes no sense
to not publish the schematics and they have never published anything that even vaguely approaches a reasonable explanation for this. About the only
thing left is that this is to prevent a company from directly cloning a Pi using the same Broadcom SoCs. ODROID tried to do this but when push came to
shove they could not get Broadcom to sell them the chips.
If you really want to understand a schematic is the one of the two most fundamental parts of that. And it is essential if you trying to determine why that
"innovative idea" (hardware, software, or both) is not working. It also helps to greatly reduce the number of boards sacrificed in the "innovative" process.
The Raspberry Pi Foundation (AKA Broadcom ?) made an incredibly stupid decision with no advantages to them and a great disadvantage to anybody
that "really wants to to understand" the Raspberry Pi in all its variations. This is underscored by the fact that they have never make any attempt to justify
