[hpsdr] Xilinx Code and Oleg Skydan work

[Alex]

Bob,
Thanks for the new link to a working website for Oleg. I have an entire copy of the website down for preservation.
I had been able to pull about half of the other version of his website out of Google Cache. 
The only problem is that Oleg's copy of his QEX article is missing pages 31 and 32.

We should post links to ALL the "prior art" and "prior shoulders we are standing on", and give proper credit to everyone.

I agree that we should create an open source version of the "Parabolic Taylor Series Corrected DDS" code and place it on opencores.org.

An interesting bit in Oleg's article is on page 33 where he discusses constructing a frequency synthesizer 
with a multiplication factor of 1 using a combination of US patent #5,847,615 and his design from the QEX article.
He also states that with a low-noise reference oscillator, a highly linear PFD and wide loop bandwidth, one could get 
"very good" VCO noise suppression. Then end result being a "super synthesizer".

Alex, N3NP

----- Original Message ----
From: Bob McGwier <n4hy at idaccr.org>
To: Alex <harvilchuck at yahoo.com>
Cc: hpsdr at hpsdr.org
Sent: Wednesday, June 14, 2006 3:14:28 PM
Subject: Re: [hpsdr] Xilinx Code  and Oleg Skydan work

At the risk of finally proving to myself that I am only talking to 
myself,  let me repeat with more emphasis and less math,  what I said in 
my last about the Taylor Series DDS.  At its heart, it is a trivial 
piece of nothing work done by trig identities .  My daughter learned 
these in the seventh grade.   The hype stuff they have added is the 
control fixtures to allow you to make changes and settings on the fly.  
This will ONLY work with their boards.  The stuff is not useful to us.

We should do our own verilog design for this and stick it in opencores 
and advertise it on the hpsdr web site with about as much hype as they 
did.  We can add a second section of correction to the process and call 
it  Parabolic Taylor Series Corrected DDS and put out 2^12 * 2^10 * 2^10 
table lookup based corrections and do a 64 bit accumulator and show the 
noise floor at -190 dBm.  It will mean as little as this hype does.


I have been a fan of Oleg for a long time (the first note that comes up 
in google has my name in it).  His work is brilliant and done on a shoe 
string.

For some reason,  I still had the URL to a pdf of his synthesizer.  That 
is STILL ACTIVE.  Download it now before that too disappears.

http://skydan.in.ua/T03DSP/Download/SchSynth.pdf

There is no copyright or other notice on the schematic.  If we place it 
someplace (hpsdr or hamsdr), I insist that we give Oleg full credit in 
the blurb.  I would prefer modifying it with adobe pro or similar tool 
to have Oleg's name and callsign on it.


Bob
N4HY

Alex wrote:
> ***** High Performance Software Defined Radio Discussion List *****
>
>
>   
DDS stuff snipped.
> Alex, N3NP
>
>
>
>   

-- 
Robert W. McGwier, Ph.D.
805 Bunn Drive
Princeton, NJ 08540
(609)-924-4600
(sig required by employer)






 1150315030.0