Re: [Discuss-gnuradio] Welcome and brief update

[Mark Smith]

On Fri, Jan 10, 2003 at 10:47:10AM -0800, Daniel Rogers wrote:
> I was thinking about this as well.  My idea was that I could use a 
> single chip ADC and connect it to an existing hardware link (such as 
> PCI, or Firewire-2, or Ethernet).  There are tons of single chip ADC 
> solutions out there.  One I stumbled upon is this one from Texas 
> Instruments:
> ADS800, 12-bit, 40 MSPS ADC SE/Diff inputs.
> http://focus.ti.com/docs/prod/productfolder.jhtml?genericPartNumber=ADS800
> ...

Oy.  40MSps?  Is that a bit overkill for a starting point?  Since the
chip is easy to interact with (I'm taking your word for it; I've not
looked at the spec sheet), maybe we should go ahead with that chip, but
not expect to push a full 480Mbps for the first rev, if it'll make
things easier.

I agree that it would be nice to be able to finally get to that, but if
it means getting something out quicker and cheaper, I'd be all for limiting
to only (say) 400kHz sample rate for FM broadcast, or 12MHz for NTSC.

Now.  Having said that, if we find a simple solution that will get us
the full 40MHz sample rate, by all means, go for it.  I'm just suggesting
that it shouldn't be a requirement for the first rev of the board.

> But the idea of a single input, single output, firewire powered, 
> external, high sample rate ADC is very sexy, and, it would seem, fairly 
> close to reality.  Not to mention a WHOLE lot cheaper that some of the 
> other commericial ADC's out there.

I'm in total agreement there.  Having the ADC seperate from the PC (as 
opposed to having it on a PCI card) would help a lot with interferance 
as well.

> I am curious what others who may have experience in things like this think.

Although I can't claim gobs of experience, I think it's a great idea.

-Mark
-- 
Mark Smith - finger address@hidden for GPG v1.0.1 public key

pgpOoVdyiCIIX.pgp
Description: PGP signature