[hpsdr] Sasquatch II

[Steven Wilson]

Hmmm - let me give a  data point for this discussion. 

I've actually routed an ARM7 in a Vertex II a few years ago.  It ran at
around 25Mhz. The cost of the parts was around $2K.  So I'm not sure
that is a cost effective route and it kinda falls short in "High
Perfomance" area due to the low clock rate you can expect from a general
purpose processor implementation in an FPGA. 

Now I'll be the first to admit that the ARM7TDMI isn't really FPGA
friendly. It's got some long and hairy paths in it which subsequent ARMs
have pipelined. However, I think the clock rate is representative of
what a GP design with an MMU (something you really need/want to run
Linux) will give you.

73 de Steve KA6S

Philip Covington wrote:
> ***** High Performance Software Defined Radio Discussion List *****
>
> On Thu, Dec 11, 2008 at 12:16 PM, Ramakrishnan Muthukrishnan
> <vu3rdd at gmail.com> wrote:
>   
>> ***** High Performance Software Defined Radio Discussion List *****
>>
>> On Thu, Dec 11, 2008 at 7:53 PM, Lyle Johnson <kk7p at wavecable.com> wrote:
>>     
>>> Sasquatch was originally going to be based on the TMS320C6726 DSP, which is
>>> in a TQFP package.  The new DSPs include the TMS320C6745 which is in a 176
>>> pin TQFP and should be able to be accommodated on a 4- or 6-layer PC board.
>>>  This is the same DSP as contained in the OMAPL137.
>>>       
>> How about a C67xx only on sasquatch with a host processor implemented
>> in Ozy FPGA?
>>
>> --
>>  Ramakrishnan
>>     
>
> How about doing the DSP and processor in the biggest QFP FPGA we can find?
>
> Phil N8VB
> _______________________________________________
> HPSDR Discussion List
> To post msg: hpsdr at hpsdr.org
> Subscription help: http://lists.hpsdr.org/listinfo.cgi/hpsdr-hpsdr.org
> HPSDR web page: http://hpsdr.org
> Archives: http://lists.hpsdr.org/pipermail/hpsdr-hpsdr.org/
>   


 1229026335.0