AN: GuruGram #72 is newly ready for free download

Don, even you've been pushing "magic sinewaves" for something like a decade now, no? And the math behind it is at least two hundred years old, with commercial applications probably dating back to about the '70s?

Probably earlier. One of my colleagues at Plessey Pacific in 1969-70 had previously worked on using PWM to synthesisie three phase sine waves from a (big) DC source, basically a long distance link between power grids.

This was before TTL was cheap, to the point that transistor-transistor-logic was cheaper when made with discrete transistors, but when you are talking generating station power levels, this doesn't matter much.

schrieb im Newsbeitrag news: snipped-for-privacy@80g2000cwy.googlegroups.com...

Hi Bill -

Where can I read uproad of Don about Magic Sinewaves?

regards - Henry

--

The underlying math is a VERY RECENT discovery that utterly and completely blows away conventional PWM.

More at http:/

If you really think this is the case you should submit a paper to your favorite IEEE journal about it, Don. We know you're good at writing! ;-)

Heck, these days they'd probably accept it regardless of whether or not there's plenty of prior art. :-)

You are thus suggesting that I pay a lot of money to be published in a journal of miniscule circulation whose subscription costs are so high that my own library cannot afford it?

After 35+ books and 1900+ articles in print, I think not.

It is probably still the cheapest way of finding out about the prior art. Referees for academic journals don't get paid, and most of them do a pretty good job.

Your "underlying math" is presumably a better way of calculating where the edges should go in the PWM waveform - people have been doing this with sufficient accuracy (if cruder math) for quite some time now.

Some of these computational tricks get discovered several times before the field pays any attentiion. The much-hyped Fast Fourier Transform turned out to have been invented several times before Cooley and Tukey got their little band-waggon rolling.

Danielson-Lanczos invented their version in the days of mechanical calculators, and IIRR someone else invented a version of the algorithm about ten years before Coley and Tukey for sorting out multi-element antennas.

It is probably still the cheapest way of finding out about the prior art. Referees for academic journals don't get paid, and most of them do a pretty good job.

Your "underlying math" is presumably a better way of calculating where the edges should go in the PWM waveform - people have been doing this with sufficient accuracy (if cruder math) for quite some time now.

Some of these computational tricks get discovered several times before the field pays any attentiion. The much-hyped Fast Fourier Transform turned out to have been invented several times before Cooley and Tukey got their little band-waggon rolling.

Danielson-Lanczos invented their version in the days of mechanical calculators, and IIRR someone else invented a version of the algorithm about ten years before Coley and Tukey for sorting out multi-element antennas.

Don -

I already noticed your site (back one year or so). But I found it difficult to read and there was no reference to others. Something without reference is suspictious or really new or sophisticated.

(OK, I already know your answer at least in part).

- Henry

--

Scholarly journals no longer serve a useful purpose and their very survival is extremely unlikely unless they dramatically change their services and policies.

The IEEE, of course, is by far the worst offender and the most in risk.

An individual has a choice: They can instantly publish on the web to a worldwide audience. Or they can pay lots of money and waste lots of time having a scholarly journal publish for them. With a miniscule and fast diminishing aucience and a subscription price so high that even their own library cannot afford a subscription.

If the journals are to survive at all, they MUST provide instant and free no hassle downloads of all papers older than three years, and MUST have a ridiculously faster method of getting into print.

See

Even with the Internet, there's probably still somewhat better long-term "persistance" with IEEE journals (or similar) than most other publications. Besides, academics love keeping names around, so 100 years from now people would be talking about "Lancaster Sine Waves" -- maybe I can appeal to your vainity? :-)

If you did up a local college professor you can probably get them to foot the fees if they get to stick their name somewhere on your paper. Normally this might also get a bit of "name recognition," but in your case I would hope that even most academics would be familiar with you.

I am playing devil's advocate a little here; there are plenty of pros and cons of the IEEE and the publishing empire they head up.

---Joel

All of the scientific journals are cons.

And I am a college professor, sort of.

In a previous job I went to a seminar on publishing journals, with speakers from a few famous publishing houses. The basic idea is to keep raising the price to make money and hope subscribers don't cancel. When too many cancel start a new journal, but keep the old one alive. Now you are selling 2 subscriptions. And so on.

The contents of the journals was not really a subject in the seminar.

Thomas

The IEEE isn't committed to making a profit. In Europe, Elsevier is seen as far worse.

Without the benefit of the refereeing system, which protects the world from a great many misleading publications - believe me, I've done enough referring to know that Sturgeons Law (90% of everything is rubbish) is pretty accurate.

I've never paid a cent to publish in any scholarly journal. Sometimes I've been invited to pay page fees, but since I've never published grant-supported work, I've always declined. If you want to identify the journals involved search for "A.W.Sloman" on Google Scholar.

high that even

Have a look a the Public Library of Science for an example of how sucfh a system might work

They still publish refereed articles, so one can't publish half-baked thoughts before the sub-concious has done a bit of error-checking, but it does get the greedy publishers out of the loop.

Not exactly detailed - and your comments about music on page 2 miss the very real problem with the "decoupled" electriconic piano, where the "two switch" piano key does an absolutely dire job of capturing the timing and amplitude information provided by the pianists fingers, to the point where I can play better on my piano teacher's acoustic German Steinway - whch I get to play once a week - than I can on my electronic Roland 330, which I play every day.

-- Bill Sloman, Nijmegen (but in Sydney at the moment).