What is Open Source Hardware?

ptorrone writes "In their piece 'What is open source hardware?', MAKE magazine divides up electronic hardware into layers, each of which has different document types and licensing concerns: Hardware (mechanical) diagrams, schematics & circuit diagrams, layout diagrams, core/firmware, software/API — each layer has an example provided and links to many of the open source hardware projects currently being worked on."

In a nutshell... OpenSPARC from Sun Microsystems

[CoreTech]

OpenSPARC [opensparc.net] is available from Sun Microsystems. The SPARC architecture is still highly relevant. Open source hardware projects like this are worth noting.

Free Telephony Project

[mo]

One very interesting example of open-source hardware is the Free Telephony Project [rowetel.com].
David Rowe, the author has almost single-handedly designed an embedded computer using a blackfin processor combined with FXO/FXS (PSTN lines) chips to produce an extremely low-cost PBX running uclinux and asterisk. Recent posts indicate he's also close to producing a T1 interface as well. The amazing thing about this project is how open it all is. The cirucuit design, and layout for all of the boards are open. Also, he's committed to using only open-source software to do the design (and contributed a number of enhancements back to these projects, such as pcb [sourceforge.net]). Not to mention also developing the uclinux based distribution, astfin, as well as a number of custom modifications to asterisk itself to use some of the Blackfin's special DSP capabilities.

Another site (mostly RTL level)

[Anonymous Coward]

http://www.opencores.org/ [opencores.org] has quite a few hardware designs (mostly RTL) ranging from cryptographic engines to complete processors. They were also instrumental in developing Wishbone, a completely open SOC bus architecture, akin to something like ARM's AMBA. IIRC you can also buy a pack of cd's which contain open-source or free-of-charge EDA software.

I'm not sure if anybody's said it explicitly, but a hardware equivalent to SourceForge would be a great asset to the community, where people can share RTL, schematics, PCB and chip layouts, and so forth.

Re:hmm...: answered

[zen611]

The open source rapid prototyper: http://www.fabathome.org/wiki/index.php?title=Main _Page [fabathome.org]

A good example

[serviscope_minor]

Ronja [twibright.com]

The schematics for electronics and mechanical design are available, including in enditable source form.

There's a foundation all about Open Hardware!

[Theovon]

Check out:

http://www.openhardwarefoundation.org/ [openhardwa...dation.org]
http://www.opengraphics.org/ [opengraphics.org]

A lot of people are really taking this idea of open hardware designs very seriously, especially in graphics, where we have a really hard time getting docs out of GPU vendors to write Free Software drivers. One of the commenters on this article said something about how he and his colleagues who know how to do this stuff have no interest in doing it outside of work. This isn't true for everyone. The founder and leader of the Open Graphics Project is an experienced graphics chip designer.

Re:Open source medical equipment

[fcc3]

Welcome to the OpenEEG project http://openeeg.sourceforge.net/ [sourceforge.net]

Many people are interested in what is called neurofeedback or EEG biofeedback training, a generic mental training method which makes the trainee consciously aware of the general activity in the brain. This method shows great potential for improving many mental capabilities and exploring consciousness. Other people want to do experiments with brain-computer interfaces or just want to have a look at their brain at work.

Unfortunately, commercial EEG devices are generally too expensive to become a hobbyist tool or toy.

The OpenEEG project is about making plans and software for do-it-yourself EEG devices available for free (as in GPL). It is aimed toward amateurs who would like to experiment with EEG. However, if you are a pro in any of the fields of electronics, neurofeedback, software development etc., you are of course welcome to join the mailing-list and share your wisdom.

Right now, this site is mostly about the hardware; schematics, part lists, building instructions etc. However, a few members have developed some useful software which is hosted on their own websites. You can find these through the software pages.

Open hardware is NOT useless!

[chris_mulhearn]

To those of you saying "I have the skills, but why bother?" or "Why would some company bother?", I can give a halfway decent example of why "closed" hardware can kind of stink. Of course, its also an example of why "closed" hardware can kind of be nice, for the manufacturer, especially if they are also sell content for that hardware.

I ported uClinux to the Sony PSP (check it out at http://df38.dot5hosting.com/~remember/chris/ [dot5hosting.com]. It wasn't on slashdot because they had more important stories to run, like what operating systems the iPhone DOESN'T run. [slashdot.org] But I'm not bitter ;)

Anyways, this project was a major pain in the ass, and at the moment its kind of stalled. The problem is that the hardware design is closed, and most of the components are all consolidated onto one giant Sony ASIC so inspecting the motherboard is no help.

So... It took a million years just to get the memory map right, and understand the exception/interrupt plumbing of this customized not-quite-MIPS-R4400 cpu core, and understand how to talk to the video, serialport, cache controller, etc.

Now, I happen to know at least 10 people who would buy a PSP if it had a really well supported linux. Allowing people to target a popular platform, rather than a proprietary one, would allow really neat applications that made use of this things built-in Wifi/Audio/etc. In short, if it was an open platform, it would allow for better interoperability, more diverse applications, blah blah blah. And if its Truly Truly "open", people could even make new PSP's that maintained compatibility with current PSP software, but optimized it for other tasks, etc. Whatever, you name it.

Of course, Sony is a content company. The memory stick interface, from which you can launch applications, makes one think "Maybe I can write some software to dump my friend's UMD game to memory stick, and then launch that game from the memory stick, so I can copy it." But thats REMARKABLY difficult on the PSP because they will only run memory stick code that is cryptographically signed by Sony (unless buffer overflow exploits are used, etc. which exist and are what make uClinux/PSP possible) and due to the closed nature of the platform, I don't think anyone even knows how to get a signed copy of a UMD onto a memory stick.

So it works pretty well for Sony. And it stinks for me.

Re:Today is that glorious future

[josath]

Here are two places I've seen that have very good prices for low-quantity custom PCB's:

http://pcbnet.com/ [pcbnet.com]
http://www.goldphoenixpcb.biz/special_price.php [goldphoenixpcb.biz]

Re:Another site (mostly RTL level)

[Anonymous Coward]

Exactly; in fact, most of the design on OpenCores are most often used on FPGAs for that reason (fabrication is expensive). For a solution to that, try http://www.mosis.com/ [mosis.com]. They combine many low-volume projecs onto MPWs (multi-project wafers) and lower the cost to the point where college VLSI design students can afford to fabricate a few of their class project. Granted, I'm not aware of a way to move from this very low-volume stage to something like production at a foundry like TMSC or UMC, there is a huge price jump I believe, but I'm not sure that's a very big obstacle. If your design is useful enough to justify creating an ASIC in the first place (as opposed to an FPGA or microcontroller+specialized hardware), it's useful enough to fabricate a bunch of them, and if you're not dead-set on the latest and greatest process node, a mask set's cost is manageable as long as your production run is sufficiently large.

Another thing I would like to see, personally, is something like MPWs at TSMC. Somebody correct me if I'm wrong, but I'm not aware of a way to amortize the cost of a mask set over many small projects. This, of course, would tie the production numbers of all related projects today, but if you're only making a few hundred wafers, perhaps this doesnt matter.