Date   

Re: UDRX Processor Board

bhhoyer@...
 

Hi Michael,

Well said. We are keenly aware of these issues.


Re: 44 addresses / JNOS 56K / Gateway Security ?

"John D. Hays" <john@...>
 



Please be aware, there is no graphics display on the UDRX-440 -- so graphics capability is not something on our checklist.


--


John D. Hays
K7VE

PO Box 1223, Edmonds, WA 98020-1223 
  


Re: 44 addresses / JNOS 56K / Gateway Security ?

dsp_stap@...
 

---In UniversalDigitalRadio@..., <dwhall256@...> wrote :
> The basis of your rant against the RPi is out of date.
> Please read (especially the last paragraph of the Broadcom section):
> http://en.wikipedia.org/wiki/Free_and_open-source_graphics_device_driver#Broadcom

Oh how I wish you were correct.

I read the entire article.  I also followed up on some of the footnotes.

Footnote 48 is this ArsTechnica article:
http://ArsTechnica.com/information-technology/2014/02/raspberry-pi-marks-2nd-birthday-with-plan-for-open-source-graphics-driver/

which reveals that Eben Upton is a Broadcom employee as well as being the CEO of Raspberry Pi.  It also reveals that Upton wrote Broadcom's announcement that they were providing more documentation on the BCM21553 cellphone chip which the first article lauds.

Footnote 49 points you to this article:
http://www.RaspberryPi.org/a-birthday-present-from-broadcom/

in which Eben Upton is a little bit more honest than he was when he authored Broadcom's press release.  He admitted that there was still a porting effort needed for the source code release of the graphics stack, because it was for the BCM21553 cellphone chip and not the BCM2835 which the Raspberry Pi uses.  He also admitted that "there are still significant parts of the multimedia hardware on BCM2835 which are only accessible via the blob".

Broadcom certainly has the full source code for the graphics stack on the BM2835 which the RPi uses.  Why don't they release it?  And why are there "still significant parts of the multimedia hardware on BCM2835 which are only accessible via the blob"?

So very sadly, the basis of my rant is not at all out of date.  Broadcom continues to be hostile to open source software developers.

My issue is documentation on interfaces -- not necessarily documentation on the hardware.  The Raspberry Pi is not open hardware, unlike the Beagle Bone and Arduino.  That's OK with me.  But when I buy a device, I expect to know fully how to use it.  That requires that interfaces be completely documented.  They aren't.  I also expect that if the device uses software for its essential operation, that software be made available and fully documented, so that when (not if) bugs are found in the future, they can be fixed, and my hardware does not remain permanently broken.

Broadcom, and anything built with Broadcom hardware, completely and totally fails this test.  They are indeed hostile to open source software developers.

They do have a good economic reason to do so.  When fielded Broadcom products are found to be defective, and broken, and vulnerable to attacks over a network, because they are not fixable by the end user, they must be thrown out and replaced.  This leads to the increased sales of newer Broadcom products.  It is good for Broadcom's bottom line as long as consumers keep on putting up with it.  But it is evil.

So I do not believe you are correct.  I do not trust Broadcom, based upon what they have done in the past.  I do not believe that the zebra has changed his stripes.  And since Ebon Upton and Broadcom have called this a "birthday present", I also note that the Achaeans offered a beautiful gift of art to the city of Troy: a very large carved wooden horse.

73,
Ken N8KH


 


Re: UDRX Processor Board Was: 44 addresses / JNOS 56K / Gateway Security ?

Robert Copelan <rcopelan@...>
 

Hmmm... Ken,  Have you homebrewed all of your radios using non-encumbered components?   Do you have a vehicle that has no proprietary components or software?  Do you have a smart phone that is totally open source with no closed source apps?  Are all of the appliances in your kitchen able to be copied without patent issues?    If the answer to any of those questions is yes then we all would like to know what you use so that we can support open source in those areas.     The PI is about as open as I've seen for processor boards in ANY price range.

The life of the project depends on a working processor board that keeps the costs in line and is in line with the goals of the project.   Yes, those who don't like the RPi may use that as a reason not to buy the product but  there will be many more that have no issue and will be happy to see the product being delivered.  
With all due respect,
Robert
WB4DHC
 

On Sat, Mar 28, 2015 at 2:36 AM, dsp_stap@... [UniversalDigitalRadio] <UniversalDigitalRadio@...> wrote:
 

---In UniversalDigitalRadio@..., wrote :

> NW Digital Radio is looking at alternative
> “off the shelf” computer boards. The Raspberry
> Pi 2 is a candidate, but we are still in the
> investigative stage -- no selection has been
> made.

Choose wisely.  The life of your project depends upon it.  There are thousands who will not go near it if it is not truly free (as in speech).  We value liberty.

Without outside free software development, your project goes nowhere.

I'm willing to pay for freedom.  I will not pay one cent for bondage.

73,
Ken N8KH




UDRX Processor Board

"Michael E Fox \(N6MEF\)" <n6mef@...>
 

Some important considerations for a reliable device:

 

1)  main board temperature (and humidity) tolerance

2)  storage temperature (and humidity) tolerance

3)  storage reliability

 

RE 1 & 2:

Anything that might be deployed in a remote location could be subject to a broad range of temperatures and humidity levels.  For example, one of our mountain-top facilities suffered an A/C failure last year.  Due to all of the radio and computer gear in the building, the temperature reached 115 *F and humidity dropped to 2%.  The A/C failure occurred on a Friday afternoon and the site owner was not able to get an A/C company to go up there until Monday morning.  In the meantime, the backup A/C also failed.  So all of the equipment baked in a sauna for the entire weekend.  The radios we use (mobile Kenwoods and Alincos) continued to work just fine.

 

Obviously, the heat pipe & heat sink on the chassis plays a key role.  But since the UDRX is also a linux machine with storage, then the storage must also be able to survive such temperatures, not just the processor connected to the heat pipe / heat sink.

 

RE 3: 

The board needs to support at least commercial grade SSD if it’s going to be reliable.  SSD storage is generally categorized as consumer, commercial or industrial.   The differences are generally related to the quality of the write leveling algorithm and the temperature tolerance. 

 

A consumer-grade SD card in a Raspberry Pi is fine for indoor tinker-toy projects.  But those who operate servers (such as BBSs, mail servers, etc., with lots of files being created and deleted, lots of logging) know they will fail after significant write activity and/or temperature changes such as described above.  Consider a commercial grade mSATA or M.2 SATA rated at 50+ *C. (BTW, there are also important configuration issues which impact the write volume, such as unpartitioned space for garbage collection, use of the TRIM command, etc.)

 

Remember that a failed, non-RAID storage device can take a system out of service for many hours or even days.  If you are immediately alerted, and read to go, then it’s at least several hours:  time to retrieve spare parts, load up the car, drive to the site (>1hr in many cases), unload, set up, replace the drive, restore from backup, etc.  But if you’re at work, out of town, on vacation, etc, it may need to wait for several days or longer.  So, a cheap little SD card from Amazon can render your solution worthless.  That’s why the devices we deploy as servers (JNOS BBSs, mail gateways, etc.) use dual commercial-grade SSDs configured with software RAID-1 (mirroring).  The extra $50-75 or so is nothing compared to being out of service for hours, days, or longer.

 

Michael

N6MEF


Re: UDRX Processor Board Was: 44 addresses / JNOS 56K / Gateway Security ?

"Timothy J. Salo" <salo@...>
 

I suspect that UDR faces pretty much the same challenges in selecting a
processor solution as any other project. First there are the more
fundamental considerations, such as whether a potential solution
meets the needs of the project (e.g., peripheral support, memory,
processor power, software availability, etc. etc.).

Beyond that, there are more prosaic considerations, such as:

o For how long is a processor board expected to be available? I
suspect that the UDR will be available for longer than most
inexpensive processor boards. As a result, the UDR team will
probably need to manage this, perhaps by integrating new processor
boards as availability changes, or by purchasing an inventory of
processor boards. More expensive processor boards may be backed
with a promise that they will be available for at least a certain
period of time or that replacement boards will be drop-in
replacements. Less-expensive boards usually don't come with any
of these sorts of promises.

o I don't know what the UDR specs are, but it may be hard to find
an inexpensive processor board that is designed to operate in harsher
conditions, such as over a wider-than-typical temperature range.

Considerations such as these two may drive an vendor towards building
its own processor board. Of course this merely sifts the problem to
the long-term availability of certain chips, rather than the long-term
availability of certain boards. And, it imposes a cost.

What I would really like to see is a processor board and software
distribution built by and for radio amateurs. Hopefully, this board
could be leveraged across multiple amateur radio projects. This board
ought to be designed to be available for a number of years, and future
boards ought to be drop-in replacements for previous boards (as
processors and memory chips advance). This amateur radio board is
likely to be different than other solutions in that it ought to support
industrial temperature ranges (so people can run the board outside), it
ought to support a bunch of interfaces useful to radio amateurs, and it
doesn't need much display power. I would also like to see a variant of
this board that could be used in satellites and balloons.

It would also be really nice to see a Linux distribution that is
tailored for this board. Among other things, this distribution
ought to support really, really unattended remote operation (think
satellites, where systems may have a dead-man timer and may boot
alternative images, so that you never, ever lose control of the
satellite because of a bad software image).

Of course, I recognize that the impediments to an amateur radio
processor board and software distribution are substantial.

For the record, when I envision an amateur radio processor board,
no, I don't think a PIC will be adequate...

-tjs


Re: UDRX Processor Board Was: 44 addresses / JNOS 56K / Gateway Security ?

"Timothy J. Salo" <salo@...>
 

Choose wisely. The life of your project depends upon it. There are
thousands who will not go near it if it is not truly free (as in
speech). We value liberty.
Ideologue alert.

"The life of your project depends upon it."

So much drama.

"Thousands"???

-tjs


Re: 44 addresses / JNOS 56K / Gateway Security ?

cory@...
 

Too late... Ken's rant has already told us who he is ;)

"When somebody shows you who he is, believe him the first time."


Re: 44 addresses / JNOS 56K / Gateway Security ?

Dean Gibson AE7Q <yahu.stuff@...>
 


On 2015-03-27 23:29, dsp_stap@... [UniversalDigitalRadio] wrote:
... No self-respecting free (as in free speech) software lover will go near it now.  ...

Since you are clearly insulting a great many people (including me, and I've released most of my software under the GNU GPL), I wonder if we are now free to insult you?


Re: 44 addresses / JNOS 56K / Gateway Security ?

Dean Hall <dwhall256@...>
 

Ken,

The basis of your rant against the RPi is out of date.
Please read (especially the last paragraph of the Broadcom section):
http://en.wikipedia.org/wiki/Free_and_open-source_graphics_device_driver#Broadcom

!!Dean

On Mar 28, 2015, at 1:29 AM, dsp_stap@... [UniversalDigitalRadio] wrote:

---In UniversalDigitalRadio@..., <steve.n8gnj@...> wrote :

NWDR is going to spin the design one more time
and :-( and make the Raspberry Pi 2 the computer board.
My disappointment is without limits.

Now the UDRX has a Raspberry Pi problem: proprietary interfaces and no documentation. As far as I am concerned, Bryan has killed his project. No self-respecting free (as in free speech) software lover will go near it now. It is closed and proprietary, and not documented. It is dead.


Re: UDRX Processor Board Was: 44 addresses / JNOS 56K / Gateway Security ?

"del_swt@..." <del_swt@...>
 

I have a pi here, I'm very happy with it, I can run all my open source code on it without problems, I don't understand why you hate it so. in my view, a pi as the core would increase it's attractiveness

Sent from my Sony Xperia™ smartphone


Re: UDRX Processor Board Was: 44 addresses / JNOS 56K / Gateway Security ?

dsp_stap@...
 

---In UniversalDigitalRadio@..., <john@...> wrote :
> NW Digital Radio is looking at alternative
> “off the shelf” computer boards. The Raspberry
> Pi 2 is a candidate, but we are still in the
> investigative stage -- no selection has been
> made.

Choose wisely.  The life of your project depends upon it.  There are thousands who will not go near it if it is not truly free (as in speech).  We value liberty.

Without outside free software development, your project goes nowhere.

I'm willing to pay for freedom.  I will not pay one cent for bondage.

73,
Ken N8KH



Re: 44 addresses / JNOS 56K / Gateway Security ?

dsp_stap@...
 

---In UniversalDigitalRadio@..., <steve.n8gnj@...> wrote :
> NWDR is going to spin the design one more time
> and :-( and make the Raspberry Pi 2 the computer board.

My disappointment is without limits.

Now the UDRX has a Raspberry Pi problem: proprietary interfaces and no documentation.  As far as I am concerned, Bryan has killed his project. No self-respecting free (as in free speech) software lover will go near it now.  It is closed and proprietary, and not documented.  It is dead.

Too bad.  It showed so much promise.

It didn't have to be that way.  Bryan could have chosen other CPUs.  He could have chosen something like the Beagle boards.  He could have chosen something like the Tegra. But he doomed his project by choosing something closed and proprietary and not documented ... and therefore evil.

The URDX is dead.

> The potential is VAST, folks.

Potential is not actual.  And the potenial is killed by using closed, proprietary, and non-documented interfaces.

I will not buy one. I will not spend even a moment developing software for it.  I value liberty.

> There's other projects out there, and eventually one will
> pop loose with a good-enough SDR for VHF/UHF, and
> we're off.

Maybe.  But I'm skeptical.  I'm from Missouri.  Show me.

73,
Ken N8KH

PS  I regret the time I've spent investigating the UDRX.  I wish I could have that time back.

PPS Unfortunately, based upon who is in charge, I must have the same opinion of the ARETF.

PPPS  When somebody shows you who he is, believe him the first time.



Re: UDRX Processor Board Was: 44 addresses / JNOS 56K / Gateway Security ?

myyahoo@...
 

Hi, John:

I can't say that I saw the perfect combination of tech and price, but I'll look through the stuff that I brought home and see if any of the options come close...

- Richard


Re: UDRX Processor Board Was: 44 addresses / JNOS 56K / Gateway Security ?

"John D. Hays" <john@...>
 

Yes -- it's in our list. It does have some mechanical issues, but is "in the mix" ....

On Fri, Mar 27, 2015 at 3:19 PM, Donald Jacob wb5eku@... [UniversalDigitalRadio] <UniversalDigitalRadio@...> wrote:
 

Have you looked at the Beagle bone Black?
Don WB5EKU

John D. Hays
K7VE

PO Box 1223, Edmonds, WA 98020-1223 
  


Re: UDRX Processor Board Was: 44 addresses / JNOS 56K / Gateway Security ?

Donald Jacob <wb5eku@...>
 

Have you looked at the Beagle bone Black?
Don WB5EKU

On Mar 27, 2015 3:03 PM, "'John D. Hays' john@... [UniversalDigitalRadio]" <UniversalDigitalRadio@...> wrote:
 

Hi Richard,

To maintain the price point for the UDRX, we need something in the $50 and below range (including CPU, Memory, Network, and USB interfaces).

The interface to the radio requires a SPI bus, I2S bus, and various GPIO.

Our DSP code is fixed point, so highly efficient and doesn't strain a CPU.  However, having multiple cores, plenty of RAM, etc. is important for applications.

We have been watching various boards but a lot of otherwise good candidates are lacking one or more of the needed interfaces.

We have no video processing needs in the radio.

The Raspberry Pi processors put Ethernet on the USB bus but for normal operations should be sufficient, but it is one of the questions we have about the board.

Did you see any boards that match the listed items that we should look at?

On Fri, Mar 27, 2015 at 2:50 PM, myyahoo@... [UniversalDigitalRadio] <UniversalDigitalRadio@...> wrote:
 

I just attended the Embedded Linux 2015 conference, and there are a lot of choices (and a varying price range).

e.g., at $99 Intel had an Atom-based board with full-speed GigE, SATA, USB3, HDMI, and other interconnects. It was running at about 4 watts while doing HD video decoding.

I have a Raspberry PI 2, and while it has more processing power than the original (four, faster cores instead of one slower core), it is still constrained for I/O. If we don't need the I/O, the $35 price tag is very attractive.

I'm putting GNUradio onto it to see how well it does.

- Richard


--


John D. Hays
K7VE

PO Box 1223, Edmonds, WA 98020-1223 
  


Re: UDRX Processor Board Was: 44 addresses / JNOS 56K / Gateway Security ?

"John D. Hays" <john@...>
 

Hi Richard,

To maintain the price point for the UDRX, we need something in the $50 and below range (including CPU, Memory, Network, and USB interfaces).

The interface to the radio requires a SPI bus, I2S bus, and various GPIO.

Our DSP code is fixed point, so highly efficient and doesn't strain a CPU.  However, having multiple cores, plenty of RAM, etc. is important for applications.

We have been watching various boards but a lot of otherwise good candidates are lacking one or more of the needed interfaces.

We have no video processing needs in the radio.

The Raspberry Pi processors put Ethernet on the USB bus but for normal operations should be sufficient, but it is one of the questions we have about the board.

Did you see any boards that match the listed items that we should look at?

On Fri, Mar 27, 2015 at 2:50 PM, myyahoo@... [UniversalDigitalRadio] <UniversalDigitalRadio@...> wrote:
 

I just attended the Embedded Linux 2015 conference, and there are a lot of choices (and a varying price range).

e.g., at $99 Intel had an Atom-based board with full-speed GigE, SATA, USB3, HDMI, and other interconnects. It was running at about 4 watts while doing HD video decoding.

I have a Raspberry PI 2, and while it has more processing power than the original (four, faster cores instead of one slower core), it is still constrained for I/O. If we don't need the I/O, the $35 price tag is very attractive.

I'm putting GNUradio onto it to see how well it does.

- Richard


--


John D. Hays
K7VE

PO Box 1223, Edmonds, WA 98020-1223 
  


Re: UDRX Processor Board Was: 44 addresses / JNOS 56K / Gateway Security ?

myyahoo@...
 

I just attended the Embedded Linux 2015 conference, and there are a lot of choices (and a varying price range).

e.g., at $99 Intel had an Atom-based board with full-speed GigE, SATA, USB3, HDMI, and other interconnects. It was running at about 4 watts while doing HD video decoding.

I have a Raspberry PI 2, and while it has more processing power than the original (four, faster cores instead of one slower core), it is still constrained for I/O. If we don't need the I/O, the $35 price tag is very attractive.

I'm putting GNUradio onto it to see how well it does.

- Richard


Re: 44 addresses / JNOS 56K / Gateway Security ?

Barry Volpe <ag6wk4bv@...>
 

Steve,

Very nice write up.

I read "the groups" (so my comments are for what I see in general) but rarely engage because I can take what I want and throw the rest out as I please in that way.

I have to say that it didn't matter really what you said here (although definitely interesting and I understand the discussion of not reinventing something and following a broad standard that is already in place) and you rolled it out well in an enjoyable manner.

The "groups" would bode well if we all followed this in some way, shape or form.

73,

Barry AG6WK



On Thu, Mar 26, 2015 at 8:38 PM, Steve Stroh N8GNJ steve.n8gnj@... [UniversalDigitalRadio] <UniversalDigitalRadio@...> wrote:
 

Wow... ignore this list for a month, and some interesting discussions bloom!

This might focus the discussion a bit more. At the MicroHAMS Digital
Conference last weekend in Redmond, WA, Bryan Hoyer of NWDR (UDRX)
confessed ruefully that technology had passed THEM by. The UDRX has
two halves - an ARM-based computer board, and the RF board. The
amazing thing is that the off-the-shelf-for-$35 (really... just gotta
look hard) Raspberry Pi 2, with its quad core processor and 1 GB RAM,
was now cheaper AND more capable than the computer board in the UDRX.
So, since the UDRX still hasn't shipped :-(... NWDR is going to spin
the design one more time and :-( and make the Raspberry Pi 2 the
computer board. When I asked if I had it right... that the RF board
was just going to glom on top of the Raspberry Pi 2, Bryan stated
"There will be NO glomming" :-) Which I took to mean that they're
going to do a careful design. But... now they're going to have to redo
their hardware... AND software... again... perhaps to be compatible
with more standard Linux distros common on the Raspberry Pi. But,
we're going to get a more capable... and fun... UDRX. Eventually :-(

What does the UDRX change?

For starters, watch this video - https://youtu.be/SmyVEwjhG_k... or at
least the last ten minutes of it. David Rowe discusses the concept of
a TDMA repeater. Think digipeater (as in single frequency) only the
T/R switching is done at 5 mS intervals. Yep - a REAL TIME (close
enough for humans) single frequency repeater. He's going to do it with
a minimal SDR and good software, but there's nothing in David's
concept that the UDRX can't already do.

Remember all the hype about 2.4 GHz Wi-Fi hobbyist mesh networks -
every experimenter was a node in the mesh? And more recently the ham
mesh based on the little blue routers? Good idea, but lousy
technology. Hams had pretty good "mesh networking" 20 years earlier -
remember Net/ROM? Good range - the network covered entire regions (and
small countries in Europe), but it was low capacity (1200 baud, Z80's,
1 MHz, 32 MB RAM - nuff said). So fast forward 20... 30 years... and
now we have the technology to marry the longer-range radios with
decent compute power and better RF technology. Let 100,000 real-time
digipeaters bloom at every ham's house... and apartment balcony.
MANUAL spectrum coordination to minimize interference... that's
laughable now. Let the MACHINES do it. We have ample spectrum when we
don't have to worry about fossilized spectrum coordinations for
reserved spectrum for ghost repeaters and affording duplexers that
mandate fixed frequencies.

We're going to have SO much fun playing with new RF technology - in
3-5 years, we're going to just blow past what's possible on commercial
/ public safety two-way networks. One guy, or a small team, with a
great idea posts his alpha software and one weekend later hundreds of
people are testing it for him. A few have 64 GB USB flash drives
plugged into their UDRX for diagnostic logging that can be uploaded to
the dev team to figure out where the holes in THAT project are. Don't
like THAT project - there's dozens of others begging for participants.

The potential is VAST, folks.

We WILL get makers, hackers, etc. flooding in just to play with this
stuff. There's going to be tech tests at Maker Spaces and Maker Fairs.
There will be endless variants of SDRs for VHF/UHF as soon as it's
abundantly clear to vendors (think Chinese radios, folks) that yes,
there are legacy users of the Amateur Radio bands, but the old fart
incumbents react in tens of minutes when the machines can adapt to
changing conditions in milliseconds. So the old farts and their legacy
modes will be properly classified as interference to the interesting
things, and routed around.

I've seen enough to believe that NWDR can make UDRX real, but they
just haven't had the courage to pull the trigger on releasing anything
less than a perfect design upon us, the great unwashed. I have some
sympathy for that - my day job is frustrating, maddening, tiring tech
support for a technical product. But at some point, to unleash the
potential, we gotta have hardware in our hands to play, and for the
code jockeys to hack on.

Good news is that this is no longer a one-horse race, with us
depending on NWDR, or nothing. There's other projects out there, and
eventually one will pop loose with a good-enough SDR for VHF/UHF, and
we're off. I vividly remember the Xerox 820 board? What a POS that
thing was... but it was enough of a platform to start a revolution in
hundreds... thousands (?) of BBS'. All we need is a good-enough SDR
for VHF/UHF and we'll really start having some fun.

Steve N8GNJ

On Fri, Feb 27, 2015 at 7:52 AM, 'Tyrell Jentink, KD7KUJ'
tyrell@... [UniversalDigitalRadio]
<UniversalDigitalRadio@...> wrote:
>
>
>
> A few years ago, there was a crash at the National Championship Air Races in Reno, NV, injured dozens, killed a few. I wasn't there... I was 10 miles away at the time; Yet, the cell network was down, and I couldn't make calls or send texts. My amateur radio saved the day, not only getting me directions to where I needed to go, but also kept me updated faster, and with more accurate information than the news was providing.
>
> I was the president of the student amateur radio club at the University of Nevada. That story excited kids at conventions, but it didn't even attract new hams from the Red Cross club (Although, we did get 2 people from the Doomsday Prepper club... But I think they were interested before I talked to them). The IEEE kids looked at us like we were some relic of the past, and they looked at our shack like it was a museum. I heard a resounding "I don't wanna take ANOTHER test, just to get to play with technology that my cell phone laughs at, just in the off chance that we have another city wide emergency like the plane crash."
>
> We shouldn't have to emulate the commercial world... But EmComm isn't attracting the kids. The promise of talking to Australia got me to get my license when I was young... But I still haven't done it, because 15 years later, I still can't afford an HF radio. There are a lot of half-baked promises built into this hobby... Things that sound like they should attract new people, but when faced with the reality of time and money prioritization, just don't pan out with the majority of young people.
>
> We need creativity. We need something that has a coolness factor that can't be beat by the commercial world. In the '80s and '90s, packet seemed to meet that description... But we haven't done anything revolutionary since the iPhone.



UDRX Processor Board Was: 44 addresses / JNOS 56K / Gateway Security ?

"John D. Hays" <john@...>
 

Clarification on comments referenced by N8GNJ

NW Digital Radio is looking at alternative “off the shelf” computer boards. The Raspberry Pi 2 is a candidate, but we are still in the investigative stage -- no selection has been made. We have certain requirements for interfacing the radio to such a computer board.

We believe that using such an “off the shelf” board could expedite delivery of the UDRX-440 as the board that was designed a couple of years ago uses a less available SOC processor, which now has extended lead times.

We also believe this will allow the UDRX to evolve with newer processors as they become available, providing increased speed and functionality over time.

The software effort to move from one Debian based ARM processor to another is minimal as we largely use drivers already in the operating system and any additional drivers that we might develop will mostly be a recompile. All application software should also either run or simply require a new compile.


We also agree that hearty and widely used computer platforms will enhance the delivery of even more solutions based on the UDRX architecture.