[sdiy] FPGA multiplier

[Kevin Downs]

> -----Original Message-----
> From: owner-synth-diy at dropmix.xs4all.nl 
> [mailto:owner-synth-diy at dropmix.xs4all.nl] On Behalf Of James Patchell
> Sent: Sunday, 7 December 2003 4:22 AM
> To: jbv; synth-diy at dropmix.xs4all.nl
> Subject: Re: [sdiy] FPGA multiplier
> 
> 

> Right now, I would say that the Xilinx Spartan 3 chips are 
> you best bang for the buck.  These chips are very inexpensive.
> I would be willing to bet that we are talking about $20 US for 
> a fairly modest chip....(Spartan 3 chips start at 50,000 gates, 
> if I remember correctly).
> 

Yup, the Spartan 3's are very cool.... My company is doing a design on
one at the moment, and all those multipliers are wonderful! The 3's also
have a much more friendly ratio of RAM to logic cells than the 2E's

Unfortunately, they're a bugger to get hold of at the moment, and still
fairly pricey.  Only the initial, engineering sample, 50J parts seem to
be in wide circulation (at about US$25 one-off) - these are 50,000 gate
parts without BlockRAMs or DLLs :-(

Production of all parts in the Spartan range should be ramping up over
the next 6 months, then the prices will fall considerably. In the
meantime, the 2E's are very, very much cheaper, and you could do a
16x16->32 multiplier even on the smallest (about US$7).

As a side-note, don't forget that most FPGA's need to be fed with
configuration data on each power-up. This is normally done using a
serial eeprom; which can add a surprising amount to the cost! If you are
doing something that uses a micro-processor/controller, you can always
feed the config data from that and save yourself some parts-cost.

regards,
Kevin