Is open source hardware next revolution?

[gopi_sonsivri]

Hi guys,

I have been observing a lot of open source hardware boards like Arduino, Rasberry Pi
and Beagle-board coming to market for enthusiasts with open designs for PCB and free tools.

This reminds me of linux revolution .

I was wondering  Can open hardware reduce cost of electronics and products in coming days
and improve innovation?

Thanks

[Gallymimu]

I think it's going to increase the number of hobbyists and the number of people who THINK they can design hardware.

Ultimately it's a bit different from software since there is always a cost for components, boards and assembly.  Since there is a physical turn around time on hardware it's never going to be quite a simple or easy to play and try things as it is with software.  That said rapid prototyping gets us closer every day.

I think open hardware will act similarly to open software in that it will undercut pricing for designers which can be good and bad.  If people donate time to support projects it's fine.  If no one can make money it might leave designs and hardware approaches completely unsupported since no one can make a dime on it.

Posted on: May 29, 2013, 05:03:58 17:03 - Automerged
BTW I don't think raspberry pi is really open source hardware.  I think only the pi foundation can build it due to license with broadcom.  Please correct me if I am wrong.

[solutions]

I wouldn't discount hardware design becoming more mainstream if stuff like this catches on in schools:

http://www.kickstarter.com/projects/lightup/lightup-learn-by-making

Nifty - magnets are used to "solder" the nodes together and a smartphone photo is used for connectivity verification, troubleshooting.and current flow visualization.

[Gallymimu]

I wouldn't discount hardware design becoming more mainstream if stuff like this catches on in schools:

http://www.kickstarter.com/projects/lightup/lightup-learn-by-making

Nifty - magnets are used to "solder" the nodes together and a smartphone photo is used for connectivity verification, troubleshooting.and current flow visualization.

I've seen those, very cool idea.  If I had kids I buy these right up.  Heck I want them anyway!  I do love that "average joe" can now get into hardware hacking.

I guess I see it like the accessibility of DSLRs.  Pro photographers often complain that it undermines their business (well personally I think it's because they are crappy photographers if they have nothing more to offer than me with a T2i and a crappy stock lens).

Does it really lower the barrier for good hardware designs or just introduce a bunch of crappy designers.  It certainly should put price pressure on companies that sell simple stuff like temperature monitors and relay modules for $500 a piece!

[solutions]

Some colleges already "introduce a bunch of crappy designers". There's definitely a gaussian distribution of design greatness, with the mean centered at "getting by".

Nobody said anything about "good designs" - just "designs", and the occasional electrocution because everybody, no matter the activity, gets to that stage of comfort where caution is thrown to the winds, or the magnitude of ignorance is blinded by the arrogance of having just enough skill to keep repeating bad habits until bit.

[optikon]

 If no one can make money it might leave designs and hardware approaches completely unsupported since no one can make a dime on it.

Posted on: May 29, 2013, 05:03:58 17:03 - Automerged

Ding ding ding! you got it. If money cant be made then nobody is going to spend the time. Rasperry pi and its time wasting brothers & sisters are toys to get people interested in the field of electronic design. Great for hobby & tinkerers. It is born there and dies there. If you want to take it to professional levels and foster innovation and new sustainable technologies, I'm sorry but it takes a lot of cash. Yes there is the occasional backyard genius or a garage shop miracle that changes the world but those don't count since there is no telling when the next one will be along.

[alien]

Dear All,
IMHO all these above mentioned boards are designed around some SOCs ..so where is the real hardware learning stuff ? Majority of these boards have some SOCs whose datasheets are even shared with the user community..to get those you will first have to commit large volume purchase then only they will forward you NDAs etc to sign...
Though i am not an electronics engineer academically but i learn t the stuff my own...my first step in this Microcontrollers/microprocessors domain was to start with 80xx/8031 type parts...there was some real stuff in that... how to interface external memory/memory mapped I/Os and other stuff...that all really helped in understanding what actually goes inside a microprocessor and its external peripherals...
so my point is...majority of people using those above devices are either already used to Linux type environments or are using only to make a mediacenter or some server type service out of these boards...the actual motive seems to be not getting forward as it should have been.

[solutions]

The motive is to quickly and inexpensively achieve the means to an end - the end is a system implementation and not having to deal with even understanding the hardware.

Stuff as simple as blinking a LED or turning on a solenoid is now in the grasp of the completely ignorant.

No, they can't design the hardware, but a) they no longer have to and b) as was said, it's so commoditized that few if any that do understand hardware are motivated to build it for them.

[mgparrish]

Hi guys,

I have been observing a lot of open source hardware boards like Arduino, Rasberry Pi
and Beagle-board coming to market for enthusiasts with open designs for PCB and free tools.

This reminds me of linux revolution .

I was wondering  Can open hardware reduce cost of electronics and products in coming days
and improve innovation?

Thanks

In terms of actual products that are used commercially or in industrial applications ... no.  Really comes down to the semiconductor companies innovating in terms of real hardware costs for any kind of cost down. Getting more functions into die is where real cost down comes from.  I do find companies like TI encouraging some open hardware and software, but it's because you are buying their silicon. Open hardware and open software are really just tool alternatives. Think of the "razors and razor blade" economic concept. Make the razor cheap and sell a lot of razors. 

[Jeckson]

Is that Raspberry open their PCB like Beagle ?

[gopi_sonsivri]

^^ yeas they give out gerber files ..i believe gerber is pcb standard format

[Gallymimu]

Raspberry Pi is NOT open.  There is an exclusive manf agreement AND broadcom does not make the SoC available to the public for purchase.  (unless something has changed in the last few months)

Posted on: September 18, 2013, 05:45:22 05:45 - Automerged

^^ yeas they give out gerber files ..i believe gerber is pcb standard format

I'd be interested to know where you saw gerber files for a Pi!

[myheadhurts]

Bunnie has an open source laptop that he's currently developing.

Code:

http://www.bunniestudios.com/blog/

For those of you who have never heard of Bunnie....He's the guy who developed the Chumby

The blog entries relating to his trips to China to oversee development and production of the Chumby are fascinating.

Rob

[Vineyards]

Nice reading. As it is told there, what makes an advancement possible is another that lies under it. With the advent of desktop 3D printing tons of new applications have come into existence. If one day, we have a chip printer at our disposal we will be able to produce customized devices. We can think about zillions of possible applications should such a thing becomes true.

These days they are talking about printable circuits using gallium and indium (conductive metals which melt at room temperatures.) This will help with our practical needs when fully realized. Bendable screens are becoming a reality. I remember reading about displays that could be folded or bent easily and that was perhaps 15-20 years ago. I think what we see is the tip of the iceberg at the moment. There is a lot more to come.

[solutions]

3D printing, like drones, are over rated.

You already have your customized IC - it's called an FPGA

And a circuit that drips on the floor at room temperature? Unless it's for a British motorcycle manufacturer, no thanks.

[hate]

Unfortunately FPGA'a are far from being custom ICs with their practically zero analog processing capabilities. It's not even possible to implement a transmission gate let alone implementing anything analog without the help of external chips. They are very suitable to implement parallel hardware (only digital) though like a neural network and it's good to have something like that in hand even though they are only digital.

[CocaCola]

No, they can't design the hardware, but a) they no longer have to and b) as was said, it's so commoditized that few if any that do understand hardware are motivated to build it for them.

Exactly, having browsed many of the 'Arduino' like forums I can say most can't and don't design anything and if they do about half the time it's wrong and when you try to explain to them what they did wrong and how to properly design said circuit most reply back "You are wrong, it works just fine. I have had it running for x number of day and it works perfect, no need to change anything the Ebay seller said this is the way it's done..."

These toys are plug and play hacks that give newbies a platform to experiment on and get something working with little or no hardware knowledge...  These people will but an full sized Arduino and 4 big plug in modules for something that could be done with an 8 pin micro and about $1 worth of other components, and then complain that it doesn't fit in their HO gauge train...

Due note that is a generalization, some do progress and some do learn hardware, but in my experience most don't...

[Vineyards]

And a circuit that drips on the floor at room temperature? Unless it's for a British motorcycle manufacturer, no thanks.

No. Actually, they use an alloy based on these metals, this alloy is solid at the room temperatures but easily melts at a certain above room temperature. They are not meant for durable, military class circuits. Instead, you can only print your prototypes with that technology.

Posted on: January 20, 2014, 10:01:42 10:01 - Automerged

These toys are plug and play hacks that give newbies a platform to experiment on and get something working with little or no hardware knowledge...  

Well, we must give ourselves some credit for the things that we know don't we? Many an old-timer electronics guru has found it difficult to step a foot in the microprocessor domain. Earlier the same was more or less true for analog/digital transition. There will always be those who consider electronics as a pastime hobby and given their overwhelming number, there will always be a huge market addressing their needs. This is normal. They are actually the lucky guys. They do their simple projects with love and dedication and take great pride at coming up with something that at least performs a function (regardless of how). We are earning our lives thanks to electronics and we are expected to know a few things better than the rest already.

[solutions]

Unfortunately FPGA'a are far from being custom ICs with their practically zero analog processing capabilities. It's not even possible to implement a transmission gate let alone implementing anything analog without the help of external chips. They are very suitable to implement parallel hardware (only digital) though like a neural network and it's good to have something like that in hand even though they are only digital.

He said chip printer, and of course, you pick a corner case to fight over- analog.

In any case, if you're going to do decent analog, it is not going to be on a printer. A digital process completely sucks at analog,not to mention substrate noise problems. It's also not going to be within the realm of ability of Joe Shmoe artist or teen hacker, which is the context of this thread.

Meanwhile, PSOC is programmable analog. Microsemi has/had mixed signal programmables. There is also a programmable analog chip from Anadigm.

FPGAs encompass PLDs and can create some pretty powerful capabilities in minimal gate counts, they even have processors on board now. Sure you can buy huge ones for parallel processing, but you can also implement a custom chip for less money than you ever could with a foundry if your volumes are low. There are some now that even claim to be viable on cellphones.

So, you do have the ability to do a custom chip now and digital and analog are tough for some of the best designers (I've worked with) on the planet - what's next for you to nit over...1000V I/O?

Posted on: January 20, 2014, 10:39:41 10:39 - Automerged

No. Actually, they use an alloy based on these metals, this alloy is solid at the room temperatures but easily melts at a certain above room temperature. They are not meant for durable, military class circuits. Instead, you can only print your prototypes with that technology.

Your gallium is going to be wonderful around the aluminum wire bonds inside chips and those aluminum electrolytic caps...

For those that don't know about gallium and aluminum, a fun video here: http://www.youtube.com/watch?v=UN3bHBTSaHQ

It's egghead stuff - they narrow the scenario to get a paper published and move on. If they didn't resort to handwaving, they'd be making real products or curing cancer
 

[hate]

Solutions: What I'm saying is, FPGAs are FAR from custom ICs as in ASIC. I don't know which part of this do you oppose to when they lack the ability to implement a single, simple opamp.

[CocaCola]

Many an old-timer electronics guru has found it difficult to step a foot in the microprocessor domain.

If a so called 'old timer guru' can't read a datasheet, follow the example circuit in said datasheet and manage a basic IDE and programmer to get a micro up and running his 'guru' status is quite misplaced...  Boards like the Arduino are not designed for 'gurus' they are designed for entry level newbies, students and part time garage hobbyist not 'gurus'...

Micros are not exactly new, yeah they are dirt cheap and more and more prevalent today but any 'guru' should have been getting his feet wet decades ago when PICs and AVRs started to roll out or even before that with 8051s...  Lets face the facts, entry level micro development boards like for example the BASIC Stamp have been around for over 20 years now, a 'guru' that is 20+ years behind the tech is hardly a 'guru'!

[Vineyards]

CocaCola, a guru is someone who is good at a certain thing and not necessarily at all the variations of it. If you are caught with a radical change at a senior age; you are still a guru of the thing that you know but you may be too late to excel in another field. This is how I perceive it. Here is an example, I am don't remember the name but I guess the editor of the Q&A section at the Nuts&Volts is someone whom people consult with regarding analog electronics but he can barely write a simple program in mikro Basic.

[CocaCola]

If you are caught with a radical change at a senior age;

As I explained it's not an all the sudden radical change, it's been building for many decades...  If the guy is decades behind in his proclaimed 'guru' field, I consider him a 'past guru' and instead is now he is a newbie once again, just like I suggest these type of boards are for...

Here is an example, I am don't remember the name but I guess the editor of the Q&A section at the Nuts&Volts is someone whom people consult with regarding analog electronics but he can barely write a simple program in mikro Basic.

How does a board like the Arduino help the guy in your example when he would still have to write the program?  Forgive me but I though we were talking hardware not software as the thread title implies?  I'm guessing that the guy from Nuts & Volts was able to read the datasheet and the included example schematic and wire/solder up a micro, no?  Seriously what kind of basic hardware design does it take to get a simplistic modern micro up and running (like the Arduino board) that you think an old timer 'guru' couldn't figure out with the datasheet in front of him?

If you want to talk code examples and libraries that is an entirely different story...  But, be it open or closed source there is a plethora of that to be found for just about any compiler/IDE...

[Vineyards]

OK different perspectives.

[optikon]

Solutions: What I'm saying is, FPGAs are FAR from custom ICs as in ASIC. I don't know which part of this do you oppose to when they lack the ability to implement a single, simple opamp.

I am aware of some companies who use FPGA's to prototype ASICS, so they can be a "low cost" springboard. As far as opamps built into digital ASIC, the longer they can keep the two apart, the happier I will be. I concur with solutions that the digital processes suck for analog. You can buy today "analog" FPGA's and guess what? Those implementations of analog are very low performing.