Mailing list for discussions and development of KORUZA, a system for wireless optical networks (in English).

List archive Help

Re: [koruza] KORUZA grows: World Wide Experiment & gen 5 alpha release candidate

Chronological Thread 
  • From: Caleb James DeLisle < >
  • To:
  • Subject: Re: [koruza] KORUZA grows: World Wide Experiment & gen 5 alpha release candidate
  • Date: Thu, 28 Aug 2014 03:19:27 +0200

Highfive nice work!

Now lets see those cad files and the BOM, I don't own a 3d printer but
I still want to study the design.

Regarding the Marvell PHY you wrote about in the other mail thread,
writing new protocols for Ethernet PHY doesn't look like rocket science,
just a matter of getting an FPGA and teaching it to talk RGMII to the PHY.
Whether that's worth it... Probably not for gigabit, IMO any serious FPGA
development should be in the 10Gb range at this point.

All said, a lot of ethernet cards will provide (and even send) invalid
frames if you ask them nicely. Have you actually verified that the TPLink
devices you're using drop invalid frames? It looks like doing an ifconfig
on the eth device should tell you...
It seems that in the RX section, "errors" is CRC errors and "frame" is
misaligned frames, non-zero in either case would suggest no.


On 08/27/2014 10:43 PM, Musti wrote:
> Hi everyone,
> I am pleased to show the product of recent publication silence,
> generation 5 KORUZA prototype named: World Wide Experiment & Release
> candidate prototype
> World Wide KORUZA is a coordinated and organized experiment to test the
> system for 12 months. I am inviting primarily organizations to
> participate and sponsor a deployment at their location. This will enable
> me to construct a set of devices and while doing that prepare all the
> documentation, sources and other thing required for open source release.
> Please see more information and two videos here:
> You can see real time data from two generation 4 systems currently deployed:
> The release KORUZA alpha version - generation 5 prototype is planned for
> this fall. While this will be rather experimental, experienced hackers
> should probably success in building a working system.
> Generation 5 is now designed/coded in OpenSCAD, as this is one of the
> rare approaches that enables collaboration and tracking of changes
> without expensive CAD software.
> With enough interest I will consider providing sets of parts to build
> devices, sourcing them directly from manufacturers and packing it up.
> This should be much more economic then buying individual parts locally
> (can get ridiculously expensive) or spending a fortune on shipping low
> volume of individual units.
> Starting October I will continue to work on KORUZA almost full time,
> doing my Masters thesis on improving the system and implementing
> auto-tracking-
> Feel free to send questions, ideas or anything really to this mailing list!
> Kind regards,
> Musti

Archive powered by MHonArc 2.6.18.

Top of page