Rabbit, Netburner, or ?

I have an application that will write UP TO
2k bytes of data per day (often, less) for
seven days a week into a battery-backed RAM
chip. The data each day would start at a 2k
boundary and unused bytes from the day before
would be zeroed.  I want to present the data
via HTML as form data to a web browser as long
as the data is non-zero. When a zero is
encountered, the web processor should skip to
the next 2k boundary and start sending form data
again.  The web processor needs to access the RAM
chip only when sending the form data to the web
browser;  the rest of the time, the application
processor will have priority over the address
lines.

Is there a network-enabled processor that can
conditionally send HTML code by addressing data
in an external RAM chip (off-chip address and
data buses) while monitoring a DISENABLE port
pin for permission to address the chip?

I am not prepared to write an OS or TCP/IP stack
application from scratch.  I like the RABBIT's
ease of programming, but I need a 14 bit external
address bus plus extra handshake pins.

Can I add anything to make my question clearer?

Any suggestions would be appreciated.

TIA

Rick

This would be very easy to do on our Netburner MOD5270.
The dynamic HTML is easy see the dynamic html sections in our programmers guide.
http://www.netburner.com/downloads/nndk/ProgManR2p7.pdf

Our low cost MOD5270 dev kit is only $99.00 and we usually ship the same day for any orders
received by noon.

You are comparing a 8 bit Z-80 (rabbit) with a non standard compiler.
To a 32 bit 150Mhz processor with standard C/C++ compiler. NetBurner.
We have had a number of developers switch from Rabbit to Netburner and
we have gotten rave reviews, from them, on the  environment tools and performace. 


Paul
CTO NetBurner.

<pbreed@netburner.com> wrote in message 
news:htqua39tve9ei21djuq4aroudgueap8b8n@4ax.com...
> This would be very easy to do on our Netburner MOD5270.
> The dynamic HTML is easy see the dynamic html sections in our programmers 
> guide.
> http://www.netburner.com/downloads/nndk/ProgManR2p7.pdf
>
> Our low cost MOD5270 dev kit is only $99.00 and we usually ship the same 
> day for any orders
> received by noon.
>
> You are comparing a 8 bit Z-80 (rabbit) with a non standard compiler.
> To a 32 bit 150Mhz processor with standard C/C++ compiler. NetBurner.
> We have had a number of developers switch from Rabbit to Netburner and
> we have gotten rave reviews, from them, on the  environment tools and 
> performace.
>

It's been a few years but the last time I saw the compiler for the Rabbit, 
it looked pretty C like to me!

In article <46b7ab56$0$15222$fa0fcedb@news.zen.co.uk>, Fred says...
> 
> <pbreed@netburner.com> wrote in message 
> news:htqua39tve9ei21djuq4aroudgueap8b8n@4ax.com...
> > This would be very easy to do on our Netburner MOD5270.
> > The dynamic HTML is easy see the dynamic html sections in our programmers 
> > guide.
> > http://www.netburner.com/downloads/nndk/ProgManR2p7.pdf
> >
> > Our low cost MOD5270 dev kit is only $99.00 and we usually ship the same 
> > day for any orders
> > received by noon.
> >
> > You are comparing a 8 bit Z-80 (rabbit) with a non standard compiler.
> > To a 32 bit 150Mhz processor with standard C/C++ compiler. NetBurner.
> > We have had a number of developers switch from Rabbit to Netburner and
> > we have gotten rave reviews, from them, on the  environment tools and 
> > performace.
> >
> 
> It's been a few years but the last time I saw the compiler for the Rabbit, 
> it looked pretty C like to me!

You didn't look closely enough.  My blood pressure kept rising as I was 
using it.  I found it was close enough to be very irritating.  It seemed 
every time I tried something it refused to work since the compiler 
didn't support it.

Mayhap they've become standard in the meantime but the last time I 
checked they seemed pretty resistant to the idea.  Seemed to think their 
approach was superior.

Robert

-- 
Posted via a free Usenet account from http://www.teranews.com

On Jul 31, 2:07 pm, "kelvin_cool_ohm"
<kelvin_cool_...@noemail.invalid> wrote:
> I have an application that will write UP TO
> 2k bytes of data per day (often, less) for
> seven days a week into a battery-backed RAM
> chip. The data each day would start at a 2k
> boundary and unused bytes from the day before
> would be zeroed.  I want to present the data
> via HTML as form data to a web browser as long
> as the data is non-zero. When a zero is
> encountered, the web processor should skip to
> the next 2k boundary and start sending form data
> again.  The web processor needs to access the RAM
> chip only when sending the form data to the web
> browser;  the rest of the time, the application
> processor will have priority over the address
> lines.
>
> Is there a network-enabled processor that can
> conditionally send HTML code by addressing data
> in an external RAM chip (off-chip address and
> data buses) while monitoring a DISENABLE port
> pin for permission to address the chip?
>
> I am not prepared to write an OS or TCP/IP stack
> application from scratch.  I like the RABBIT's
> ease of programming, but I need a 14 bit external
> address bus plus extra handshake pins.
>
> Can I add anything to make my question clearer?
>
> Any suggestions would be appreciated.
>
> TIA
>
> Rick

Rick, can you do something to reduce your data size? like compression
or  abstraction so the web browser with simple script would fetch the
abstract data to make it informative.

I would also vote for rabbit, they are good enough when it comes to
documentation and sample code, but quite lazy in interactive support.


ali

"Ali" <abdulrazaq@gmail.com> wrote in message 
news:1186461243.176039.23910@w3g2000hsg.googlegroups.com...
> On Jul 31, 2:07 pm, "kelvin_cool_ohm"
> <kelvin_cool_...@noemail.invalid> wrote:
>> I have an application that will write UP TO
>> 2k bytes of data per day (often, less) for
>> seven days a week into a battery-backed RAM
>> chip. The data each day would start at a 2k
>> boundary and unused bytes from the day before
>> would be zeroed.  I want to present the data
>> via HTML as form data to a web browser as long
>> as the data is non-zero. When a zero is
>> encountered, the web processor should skip to
>> the next 2k boundary and start sending form data
>> again.  The web processor needs to access the RAM
>> chip only when sending the form data to the web
>> browser;  the rest of the time, the application
>> processor will have priority over the address
>> lines.
>>
>> Is there a network-enabled processor that can
>> conditionally send HTML code by addressing data
>> in an external RAM chip (off-chip address and
>> data buses) while monitoring a DISENABLE port
>> pin for permission to address the chip?
>>
>> I am not prepared to write an OS or TCP/IP stack
>> application from scratch.  I like the RABBIT's
>> ease of programming, but I need a 14 bit external
>> address bus plus extra handshake pins.
>>
>> Can I add anything to make my question clearer?
>>
>> Any suggestions would be appreciated.
>>
>> TIA
>>
>> Rick
>
> Rick, can you do something to reduce your data size? like compression
> or  abstraction so the web browser with simple script would fetch the
> abstract data to make it informative.
>
> I would also vote for rabbit, they are good enough when it comes to
> documentation and sample code, but quite lazy in interactive support.
>
>
> ali
>

Unless the compiler has improved radically, it also produces about the most 
bloated code I've ever seen.  Having said that, it is a useful processor 
with an affordable compiler.

"Robert Adsett" <sub2@aeolusdevelopment.com> wrote in message 
news:MPG.2121b0e87806eeea98978e@free.teranews.com...
> In article <46b7ab56$0$15222$fa0fcedb@news.zen.co.uk>, Fred says...
>>
>> It's been a few years but the last time I saw the compiler for the 
>> Rabbit,
>> it looked pretty C like to me!
>
> You didn't look closely enough.  My blood pressure kept rising as I was
> using it.  I found it was close enough to be very irritating.  It seemed
> every time I tried something it refused to work since the compiler
> didn't support it.
>
> Mayhap they've become standard in the meantime but the last time I
> checked they seemed pretty resistant to the idea.  Seemed to think their
> approach was superior.
>

I've looked again and would still say it approximates pretty well to C code. 
My main quibble were with arrays of strings and there were a few other minor 
points.  Otherwise it's OK like many other processors on the market.

In article <46b8c4fd$0$31731$db0fefd9@news.zen.co.uk>, Fred says...
> 
> "Robert Adsett" <sub2@aeolusdevelopment.com> wrote in message 
> news:MPG.2121b0e87806eeea98978e@free.teranews.com...
> > In article <46b7ab56$0$15222$fa0fcedb@news.zen.co.uk>, Fred says...
> >>
> >> It's been a few years but the last time I saw the compiler for the 
> >> Rabbit,
> >> it looked pretty C like to me!
> >
> > You didn't look closely enough.  My blood pressure kept rising as I was
> > using it.  I found it was close enough to be very irritating.  It seemed
> > every time I tried something it refused to work since the compiler
> > didn't support it.
> >
> > Mayhap they've become standard in the meantime but the last time I
> > checked they seemed pretty resistant to the idea.  Seemed to think their
> > approach was superior.
> >
> 
> I've looked again and would still say it approximates pretty well to C code. 
> My main quibble were with arrays of strings and there were a few other minor 
> points.  Otherwise it's OK like many other processors on the market.

When I last attempted to use it it wouldn't compile something as basic 
as

static int noItems;

static void ClearItems( void)
{
    noItems = 0;
}

That's so important to proper data hiding that any C compiler that 
doesn't support that I find severely crippled.  

That it didn't support compile time initialization was also a big 
annoyance (especially for tool generated code).  I don't see any 
justification for that on a processor with this much memory and a 
compiler that has reached this age.

The libraries are nice but t'aint C.

Robert

-- 
Posted via a free Usenet account from http://www.teranews.com

>
> "Ali" <abdulrazaq@gmail.com> wrote in message 
> news:1186461243.176039.23910@w3g2000hsg.googlegroups.com...
>> On Jul 31, 2:07 pm, "kelvin_cool_ohm"
>> <kelvin_cool_...@noemail.invalid> wrote:
>>> I have an application that will write UP TO
>>> 2k bytes of data per day (often, less) for
>>> seven days a week into a battery-backed RAM
>>> chip. The data each day would start at a 2k
>>> boundary and unused bytes from the day before
>>> would be zeroed.  I want to present the data
>>> via HTML as form data to a web browser as long
>>> as the data is non-zero. When a zero is
>>> encountered, the web processor should skip to
>>> the next 2k boundary and start sending form data
>>> again.  The web processor needs to access the RAM
>>> chip only when sending the form data to the web
>>> browser;  the rest of the time, the application
>>> processor will have priority over the address
>>> lines.
>>>
>>> Is there a network-enabled processor that can
>>> conditionally send HTML code by addressing data
>>> in an external RAM chip (off-chip address and
>>> data buses) while monitoring a DISENABLE port
>>> pin for permission to address the chip?
>>>
>>> I am not prepared to write an OS or TCP/IP stack
>>> application from scratch.  I like the RABBIT's
>>> ease of programming, but I need a 14 bit external
>>> address bus plus extra handshake pins.
>>>
>>> Can I add anything to make my question clearer?
>>>
>>> Any suggestions would be appreciated.
>>>
>>> TIA
>>>
>>> Rick
>>


Whats wrong with a serial dataflash with battery backup?
The AT45DB642D has enough storage to last 4096 days @ 2kB /day
without ever overwriting the memory (> 10 years)
The part has two SRAM buffers of 1 kB +, so you can write
to the buffer during the day, and commit the write to flash
at the end of each day.

Then you can use an AT91SAM7X128 to do the job.
www.freertos.org for OS/IP stack

-- 
Best Regards,
Ulf Samuelsson
This is intended to be my personal opinion which may,
or may not be shared by my employer Atmel Nordic AB