Who is your favourite electronics guru?

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

Minor bit of history: It was my talking with Peter about dividing by 2.5 that was the inspiration for that article (not how to *do* it at the logical level, but more about making sure the timing would work out such that it was reliable). I think it was something like a frame grabber (this was back at Electroglas -- the inspection products division, in Corvallis, Oregon, which is now very much dead, possibly related to the fact that we never actually generated one dime of actual *revenue* for Electroglas Corporate down in San Jose) where we had this 106MHz Fibre Channel clock kicking around and were looking for a 40-ish MHz clock to run all the junk in the FPGA.

Yes, keep tagging them "Return to Sender". They stop.

John, they'd give you 100 bucks for that! Ideas for Design is the ideal proof that there is always a $10 solution for a 10 cent problem.

My company actually caters to these same people. But I do notice that their first products are generally self-designed, but as they grow, they contract out more design. These people tend to demonstrate both abilities.

And

Only issue I see with being #1 for "electrical optical converter" is that last month NOT ONE person used those terms for a search (albeit according to Overture, but I would bet it was around that for Google, too). So that top spot may have had ZERO hits. It would be similar to being ranked number one for the random phrase "I eat June bugs in July".

As it happens I've been around when they've done that, and they don't do it well.

When you copy the prior art, you have to understand what the circuit you are copying was intended to do. Electronics has been developing so fast over the past thirty years, that the components are rarely the same from one instantiation of an idea to the next - in the good old days a phase sensitive detector often involved an analog switch, while these days you just didigtise at the right intervals and subtract every second sample (unless you are detecting in-phae and in quadrature, which has gotten a lot easier than it used to be).

He's done a lot more than that.

Not a guru

Only a guru if you think the 555 and the 741 are where it is at.

It's good, but anything Jim has done, someone else seems to have done better - he's a sort of Linus Pauling figure who gets it almost right - the MC4044 and MC4024 were fine, at the time, but someone else came up with the 4046.

Gets my vote

Good

Good - some of my posts have eneded up on his web-site

Hans Camenzind invented the 555, and he is good.

Bob Widlar would have invented the 741, if he had stayed at Fairchild. In fact, he went to National Semiconductor at the wrong moment, and invented the LM307, which is superior to the 741 at a number of points, but it was too late getting to the market. Whoever did the 741 for Fairchild made a horrible mistake in using a lateral PNP in the fully complementary output stage - the LM307 used a complementary Darlington which gave a much better behaved output.

Good

I like them a lot. Can't say that I've ever copied any of them, but every now and then they come up with ideas that I've used, which give me a warm feeling.

It's arguably a lot harder to get the first one "almost right" than to simply stand back, look at the result of someone else's work, and spend a little time tweaking it to improve it. A significant fraction of the "improvers" probably wouldn't have been able to come up with the original work -- from scratch -- in the first place.

I bet you could copy, say, 10-year old IFD articles and sell them back to them as new. They obviously don't look at them much, if at all, before they toss them in the mag. It's just junk "editorial content" required to get the cheap magazine postal rate.

John

Well, our e/o stuff is selling far better than I ever expected, and it's all through google and a few free press releases.

Really, google has changed all the conventional-wisdom rules about starting and running a technology company. I wonder if the business schools have caught on to the shift. There's a book I need to read, "The Long Tail" I think it is.

John

MC4024 and MC4044 (bipolar process) were done in the mid '60's before CMOS became a viable process.

However the 4046 uses the very same phase detectors... it's just processed in CMOS. And it's only been with modern 'HC processes that the top frequency caught up with the MC4024.

The MC4024 was also done in MECL III at the *same time*... MC1658... does 150MHz+... very same circuit... the limit in the MC4024 was due to translating from the differential PECL to TTL (and the gold-doped process). There's presently an ECLinPS LITE version in production... not too shabby for a 41 year old design.

And my MC1648 made synthesizers in radio and TV viable by providing the first integrated low-noise VCO.

A number of the ECLinPS LITE stuff seen on

I do the same sort of stuff in custom BiCMOS today at speeds up to

I've been first in a variety of areas and I have the patents to prove it... and I'm still getting them at my advanced age.

Must be terrible to be a Sloman... fading away with no accomplishments to show for a lifetime... by his own accounts, unemployable, and a kept man... his wife should give him the toss and get a real man ;-)

...Jim Thompson

That's absolutely true. I've seen many duplicates over all these years. Of course the truly funny ones are those that don't work and the editors didn't know any better.

In the old days, IFD was good. I even have an IFD file folder dating back to my youth.

...Jim Thompson

Designed with parts that are paleolithic.

Best regards, Spehro Pefhany

One of the engineers at Microdyne was complaining that were might have to redesign one of our low noise synthesizers because they were having trouble getting the phase detector chips the had used for 20 years. I suggested he replace it with the 4046, and was rudely informed that it wasn't fast enough because the original Phase Detector was ECL. I asked him if he had forgot that the reference frequency was 100 KHz? Boy, did I get dirty looks! I was also banned from the engineering department for a month. ;-)

Why? She might prefer having a pet that can feed and walk itself, and she just doesn't have the heart to have him put down.

I suppose that would be funny, but I had to "put down" a 15 year old dog yesterday :-(

But I look on the positive side... two years ago I thought this dog was a goner, so we, in our 40-some-year style of having at least a pair of Dachshunds, added a new pup.

The old guy perked up and struggled through his arthritis and finally incontinence for another two years.

...Jim Thompson

Here is a challenge for ya, John:

Send in a Design Idea that looks real, but is non-working. Something clever. See if when it is accepted if the SED community can pick it out. Of course, you'd need a pen name.

We could take turns, make it a new sport!

"I eat June bugs in July".

For giggles I bought some books on SE placement a few years back, even got into some of the early algorythms the SE's used. I can own most key phrases I want now, as long as they aren't outrageously pursued by HUGE outfits and "mass traffic" techniques. But you really have to know if anyone is really looking for those keywords in the first place. On your phrase, I couldn't find anyone. Do you have logs from your server? They are very valuable in finding where people are coming from (and often show better ways to pursue those people).

The nice thing is, on a sub $100 budget, you can really pour people into site if you have the time and compete with anyone!

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

Didn't have to worry about that, Jim. The neighbor coming through the driveway at warp speed in January took care of our 12 year old Sheltie for me.

New Sheltie pup in March filled the grieving, but there will always be an empty hole in our lives. She didn't need to go that way.

Bastards.

Jim