Last summer mention was made of a DSP / FPGA book by Ray Andraka hitting the (online) shelves this fall:
Has this come to pass?
Aside from Meyer-Baese's "Digital Signal Processing with Field Programmable Gate Arrays" book, which I found somewhat difficult to read, are there any other FPGA-specific DSP books out there?
Thanks! Stephen
No, unfortunately, I am still working on it, and my publisher (elsevier) is sitting on my rather firmly to get it done. Only so many hours per day. They have set a new deadline for me to have it to them by August so that they can get it out in the fall.
I noticed this very old post about the "book". Is it to be published soon?
On 2013-01-25, snipped-for-privacy@GMail.com sent: |---------------------------------------------------------------------------| |"I noticed this very old post about the "book". Is it to be published soon?| | | | | |On Wednesday, February 22, 2006 9:28:42 AM UTC-7, Ray Andraka wrote: | |> Stephen Craven wrote: | |> > Last summer mention was made of a DSP / FPGA book by Ray Andraka | |> > hitting the (online) shelves this fall: | |> >
I was not aware of this book but I am aware of Ray Andraka and he is very competent. You could ask him yourself:
---|
on?|
| | | | | | | | | | | | | | | |r) |
er |
st |
|---|
I am afraid, many 2005 techniques, like, for example, cordix, distributed arithmetics, utilization of SRL16 for delay lines, are not relevant today, when even low end FPGAs have plenty of hard multipliers and embedded memory blocks.
I disagree in some cases, although multipliers are bountiful, relatively speaking. Many applications have no problem using up all the multipliers and still unsatisfied. Being able to calculate an arctan with a pipelined cordic is still relevant.
Regards, Chris
Yeah, and not all FPGA projects use gargantuan devices that dim the city when powered up. I am right by a nuclear power plant and I can hear the turbines groan when some of you guys run your designs.
I'm currently working on a powerless design. It will be so low power it scavenges power from stray electrons, quanta and fields, maybe even thermal gradients. I could power it from the glow of some of the high end FPGA designs, or even the convection air currents!
Yes, I think CORDIC is still useful, although I plan to do it with a table lookup... I can live with six bits out.
Rick
I agree as well, I imagine there are many low-power low resource designs. A good book on DSP in FPGAs will cover both ends of the spectrum. How to easily use up all the multiplier resources as well as all the neat little tricks to save resources/power.
What I don't like in books like this is device specific information. Seems pointless to provide specific FPGA architectures (changes too often). Just need the generic view of the architecture not the specific slice organization, or optimization on device specific components, again it should talk in device generalities.
Regards, Chris
If I had a nickel for every time I've heard throughout my career about this or that technology no longer being relevant...
Technology is like fashion: whatever is old will be new again someday, with a new spin and a new relevance. Don't throw it away; just keep it in the b ack of your closet, and you will be able to use it again. And for those tha t missed it the first time around, the second-hand stores are always full o f these still-useful articles from bygone times.
I remember my college digital design coursework included implementing boole an logic functions with multiplexers and decoders. Then PALs came along and changed that to sum-of-products. Then FPGAs came along and changed it back (pre-HDL). Then HDL came along and changed it again.
The Cordic algorithms were not new when FPGAs came along. They were dusted off from the ancient spells of the priests of the order of multiplierless m icroprocessors and "pieces of eight". And those priests were probably taugh t their craft by the wizards of relays and vacuum tubes.
Andy
that technology no longer being relevant...
new spin and a new relevance. Don't throw it away; just keep it in the back of your closet, and you will be able to use it again. And for those that missed it the first time around, the second-hand stores are always full of these still-useful articles from bygone times.
logic functions with multiplexers and decoders. Then PALs came along and changed that to sum-of-products. Then FPGAs came along and changed it back (pre-HDL). Then HDL came along and changed it again.
from the ancient spells of the priests of the order of multiplierless microprocessors and "pieces of eight". And those priests were probably taught their craft by the wizards of relays and vacuum tubes.
Yes indeed; CORDIC was old when I used it in 1976 on 6800s.
The earliest papers I have date from 1962: J E Meggit, Pseudo division and pseudo multiplication processes, IBM Journal April 1962 1959: Jack E Volder, The CORDIC trigonometric computing technique, IRE Trans Electron Comput ec-8:330-334
Neither reference anything from the time when "computer" was a job title.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.