[Hardware] Fully unrolled RC5 on FPGA

[Ben Payne]

Maybe I'm missing something here but I've always envisioned a much "simpler" hardware design.  I've not been envisioning high speed needed between the PC and the FPGA.  So I was thinking USB as the interface.  To keep that even simpler I was thinking of using a USB capable PIC micro controller for the bus interface, and either serial or parallel communication to the FPGA.  Add some serial eeprom and a few connections for JTAG/download (to help debugging the system).  In a production system FPGA firmware could be downloaded via the USB interface.

-> Ben

-----Original Message-----
From: hardware-bounces at lists.distributed.net [mailto:hardware-bounces at lists.distributed.net] On Behalf Of Frédéric Bastien
Sent: Monday, November 13, 2006 10:04 AM
To: Hardware
Subject: Re: [Hardware] Fully unrolled RC5 on FPGA

Why not choose a card already made by a company? This will remove the 
need to manage an inventory and all this stuff. Also, as they do there 
card for more people this us, they should have lower price on many 
different thing.

So the questions is: Can the board actually sold by other company by 
used for our purpose? Is the cost of those board not more then what you 
can have if we do it ourself?

just my tough.

Frédéric Bastien

Steven Nikkel wrote:
> 3.3v PCI might be a mistake, its still pretty rare in a standard 32bit 
> slot.
> 
> On Mon, 13 Nov 2006, John L. Bass wrote:
> 
>> Date: Mon, 13 Nov 2006 10:35:20 -0700
>> From: John L. Bass <jbass at dmsd.com>
>> Reply-To: Hardware <hardware at lists.distributed.net>
>> To: hardware at lists.distributed.net, jbass at dmsd.com, martin at nnytech.net
>> Subject: Re: [Hardware] Fully unrolled RC5 on FPGA
>>
>> Layout for one of these chips for a 3.3V 32bit PCI is a long days work.
>> I tend to use pcb on sf.net, an older version that's pretty fast. My
>> local PCB house is pretty reasonable with 1 wk turn on 6 layer. If we
>> keep this a pretty simple card (small pci form factor) they would be
>> pretty cheap in a short run.
>>
>> Question is, what else would people want on the card ... SRAM? SDRAM?
>>
>> Other than layout time and increased board size, isn't much cost to
>> adding a few simple things. I'd suggest configuring it from a small
>> JTAG PLD tied to a CF socket on the back panel. That way you can
>> easily change it's code by swapping CF cards when it's dead, and when
>> it's working, write the CF from a PCI utility thru the FPGA.
>>
>> Will have to check that XCV4's can actually be connected to a 3.3V PCI
>> bus directly. May have to use an older FPGA as the PCI bus controller.
>>
>> As for PCI express, that's not a hobby project ... some serious engineering
>> and significant IP in cores.
>>
>> John
>>
>> 	Date: Mon, 13 Nov 2006 12:05:23 -0500
>> 	From: Martin Klingensmith <martin at nnytech.net>
>>
>> 	Me too, considering what you were saying in the past.
>> 	I'm going to bet that there would be a dozen people interested in PCI
>> 	boards. Much more if it was versatile for other applications (it would
>> 	be dumb not to).
>> 	I assume the V4 has LVDS for PCI express applications. It might make the
>> 	board cheaper/smaller.
>> 	--
>> 	Martin K
>>
>> 	John L. Bass wrote:
>> 	> I'm currious what the power/heat are.
>> 	>
>> 	> John
>> _______________________________________________
>> Hardware mailing list
>> Hardware at lists.distributed.net
>> http://lists.distributed.net/mailman/listinfo/hardware
>>
> 
> ---
> Steven Nikkel
> steven_nikkel at ertyu.org
> _______________________________________________
> Hardware mailing list
> Hardware at lists.distributed.net
> http://lists.distributed.net/mailman/listinfo/hardware
_______________________________________________
Hardware mailing list
Hardware at lists.distributed.net
http://lists.distributed.net/mailman/listinfo/hardware