TTL computing

From: Richard Erlacher <edick_at_idcomm.com>
Date: Mon Apr 15 21:37:26 2002

There's a pin on those old-timers that enables the EPROM during download, then
disables the EPROM when the download is complete, so you then recover the use
of the pins. If you had a problem with the EPROM connections, as I said, the
XILINX folks provided an app-note that illustrated a circuit which took the
data from the EPROM and serialized it, thereby minimizing the connections to
the FPGA. It wasn't necessary, except on the smaller devices, since most of
their pins would be used in the EPROM hookup.

With the smaller devices, it wasn't unusual to share an EPROM between a local
microcontroller and the FPGA, and have the microcontroller feed the FPGA, as
you say, pretending to be one of the serial EEPROMs.

Dick

----- Original Message -----
From: "Tony Duell" <ard_at_p850ug1.demon.co.uk>
To: <classiccmp_at_classiccmp.org>
Sent: Monday, April 15, 2002 5:56 PM
Subject: Re: TTL computing


> >
> > Won't these parts work with ordinary EPROMS? That's what we used back
when I
> > was routinely using XILINX parts. It's possible to build a PAL (though
the
>
> For the Xilinx RAM-based FPGAs (3000 and 4000 series) that I used some 5
> years ago, the actual method of getting the configuration data into the
> device was documented in the data sheet. The meaning of that data wasn't,
> of course (meaning you had to use their design tools :-(), but you could
> load the device from any logic circuit you wanted.
>
> IIRC, at least with the 3000 series you could connect up a normal
> parallel EPROM (like a 27256 or something). It took a fair number of I/O
> pins to do this (which is a good reason for not doing it!), but it was
> possible. AFAIK you didn't need any other parts, just the FPGA and the
EPROM.
>
> Or you could use a microcontroller to read out a normal EPROM (or E2PROM,
> or...) and turn it into the right type of serial bitstream. And pretend
> to be one of the special serial PROMs. Again, the timing for the
> bitstream was documented in the datasheet, even if the meaning of the
> bits wasn't.
>
> -tony
>
>
Received on Mon Apr 15 2002 - 21:37:26 BST

This archive was generated by hypermail 2.3.0 : Fri Oct 10 2014 - 23:34:31 BST