Analog may be coming back

[Grant Richter]

I reread the paper and that IS what he said.
I also noticed the date on the paper is 1994.

The only way I see that could make sense
is the current draw from high speed CMOS switching.
That is CMOS current draw goes up with clock speed.

I think, since the circuits are monolithic and vacuum sealed,
you could use a much higher internal impedance that
would be practical on a PCB.

I still can't get over the fact that for EPROMs,
you stuff some electrons in a insulated well and they stay there.
Pure voodoo.

----------
> From: Don Tillman <don at till.com>
> To: grichter at execpc.com
> Cc: synth-diy at mailhost.bpa.nl
> Subject: Re: Analog may be coming back
> Date: Saturday, July 31, 1999 1:49 PM
> 
>    From: "Grant Richter" <grichter at execpc.com>
>    Date: Fri, 30 Jul 1999 21:37:19 -0500
> 
>    Here's an interesting article on the cost benefits of analog VLSI
>    http://cnmat.cnmat.berkeley.edu/~adrian/ICMC1994.html
> 
> Is he really claiming that a bank of 100 OTA-based analog filters can
> be placed on a chip, including 200 capacitors, with a power
> comsumption that is "easily 1/100th that of" the DSP approach?  
> 
> Wha?
> 
>   -- Don