[Noisebridge-discuss] what could we do with one of these babies..

Joel Jaeggli joelja at bogus.com
Sun Apr 18 03:47:34 UTC 2010



On 04/17/2010 06:09 PM, Jesse Zbikowski wrote:
> On Thu, Apr 15, 2010 at 8:26 AM, Jonathan Foote <jtfoote at ieee.org> wrote:
>> In fact, it's not clear how "open" their hardware platform is (can we
>> get schematics and part sources at cost?)  so "open" may actually be
>> an abuse of the term...
> 
> I don't see any offers of schematics on their website either, and I
> suspect they aren't at liberty to release 100% of their hardware
> design. I believe they justify calling their product "open hardware"
> because the system is built around EtherCAT, which is to my knowledge
> the most open high-performance architecture for connecting components
> in a robot control system. Even EtherCAT is not totally open though,
> since it requires some licensing of the hardware design.
> 
> EtherCAT is a hack on Ethernet which allows you to address 100's of
> devices in microseconds (as opposed to milliseconds). 

Well, the normal lower latency bound for ethernet is the size of the
packet at a given rate and the latency associated with the serdes if
there is one in the path. at 10Gbe rates, a single 1500byte pdu can be
clocked out at a rate of every 1.2us, which goes up to 7us at 9k frame size.

> This means you
> can run control loops in the 10kHz range. The master node can be
> implemented with just commodity Ethernet hardware; however each device
> (slave) requires an FPGA with a core licensed from the EtherCAT
> Technology Group. The master is not totally unencumbered either,
> because you have to accept an agreement not to change the EtherCAT
> protocol in software (which is incompatible with GPL).
> 
> So it seems to me that the best you can do right now with EtherCAT
> would be a totally open hardware platform *except* for this FPGA. The
> alternatives would be to make do with a lower performance bus, or
> design a replacement for EtherCAT (not trivial considering the
> bandwidth and latency requirements).

Or just use 10Gbe, ann ride the price performance power curve down to
it's logical conclusion.

> My impression is that the
> EtherCAT guys are pretty reasonable and only charge a few dollars for
> their design; however they are serious about protecting their work
> against forks.
> 
> http://www.orocos.org/blogs/peter/cheap-and-open-hardware-robotics
> http://www.pc-control.net/pdf/032004/pcc_interview_etg_e.pdf
> _______________________________________________
> Noisebridge-discuss mailing list
> Noisebridge-discuss at lists.noisebridge.net
> https://www.noisebridge.net/mailman/listinfo/noisebridge-discuss
> 



More information about the Noisebridge-discuss mailing list