From p. 7 of the datasheet of Agilent's new 81150 series pulse gens.:
formatting link
"You can change the timing parameters (delay, frequency, transition time, width, duty cycle) without dropouts or glitches. This patented, industry-leading feature means..."
Here's a video:
formatting link
So this means all you waveform generator designers who ever made a generator which didn't do what I show below, must pay Agilent for their patent on not having glitch bugs or risk getting sued.
You see, when I bought a new Agilent 33522A function generator about 1.5 years ago, I expected I could dial in new pulse widths like on my much older Tek AFG3022 without glitches like this:
formatting link
formatting link
formatting link
formatting link
formatting link
In the above, the 33522A was told via the front panel knob to increment the 10s place of the pulse width from 100ns to 110ns, while driving an LED with a bank of 4 parallel TC4422A MOSFET drivers with supply voltage of about 15-18V. This circuit can put up to 15A into typical 1mm*1mm LEDs, which don't complain as long as the pulse width is about
Dude, inventions solve problems. The glitch bugs are problems that the Agilent invention solves. You are free to solve the glitch bug in whatever alternative way occurs to you, that doesn't infringe on the Agilent (or any other) patent.
That is sad. When I worked at HP [pre-Agilent days] we were told to always design for the unspecced, yet EXPECTED, performance. During reviews, discussed.
What you described is no different [at least to me] than having a power supply that every time you change the setting, it decides to go to MAXIMUM voltage for a short time before acquiring the dialed in voltage level! What's the difference? Really sad.
I've grown to expect IC's sometimes to do weird things when below their input voltage range - including destory themselves, which is equally assinine.
Wow, what a ding to a reputation ...unless all units get recalled, then I'd attribute it to 'exceesivley rushing to market place, but owning up to their errors', and then they could at least maintain a slightly tarnished image.'
Well designed chips can be held at any voltage less than abs max and not blow up.
This "fix" that HP created sounds to me like the difference between FSK and CPFSK. Of course you do everything smoothly. The customer should expect nothing less.
Then they didn't invent it, so their patent is invalid. Start lining up your evidence. When did you start selling your product with this invention? Do you have any documentation? Has Agilent sent you a letter giving you a heads-up about their patent?
s
tic?
is the improvement new, useful, and nonobvious? Then it probably can be patented.
The HP33120 function generator has the same problem. Also changing the amplitude makes a lot of glitches, relays clicking. Amazingly, it also has no function to turn the output on/off!
Developing code is an interactive process with th computah. You clearly, and please do not see this as an insult, but just as an observation, have a very limited programming experience.
After I wrote my reply it occurred to me I forgot to mention a very important aspect,
GUI development.
Now you can tell me you design a GUI based program reading late at night in bed, I tell you that building a user friendly GUI is an interactive process. The way *I* do it is to write large parts of the GUI first, so I have the call backs so I can test the various functions. Making a good layout, that sometimes involves many forms with many buttons, browsers, indicators, 'lights', meters', is purely interactive, : resize this, move this, change that a bit, color choices, all the inter-activity (buttons change color when pressed etc), all the safety locks, choices in pull down menus, interactive graphs, all step by step development, and yes that is programming, and yes it is very different from writing a few hundred lines for your embedded processor in asm. And again all the libraries. you have no clue(tm) it seems how to write a big program that runs on a modern computah with a modern OS, and the tools you need. You look down on programmers because you are no programmer, just a tinker who writes
- and lets take your DDS driver (because it is no more than that) as example -, some very basic close to hardware extremely user unfriendly limited piece of aspiring driver code and sell it for more money than it is worth. OK, stick that up!
Back to saying nice things now, like all the board modifications, or do you use wires? Why not wire wrap the hole f*chin thing.
PS I was reading in nytimes.com I think it was (some do not want me to read that paper because it is made by liberals?) that San Francisco is changing into a home base for them 'programmers' you so despise (2500 man Twitter will move there IIRC), rents going up, middle class going out of business, but YOU WILL HAVE ALL THE PROGRAMMERS YOU NEED applying for a job, maybe they will bust down your door and maybe you should buy a gun like Jimmy did to defend yourself against all them programmers who's software will replace your hardware[1]? Whatever ... LOL
[1] Not sure that is possible, but who knows, with HDLs, but from your simulation of reality POV something that could worry you perhaps.
On a sunny day (Fri, 08 Jun 2012 17:43:32 -0700) it happened "Mr.CRC" wrote in :
This hole patenting thing has gone way out of control. I think some are beginning to see how it actually hurts industry. There is Apple, Samsung, Nokia, Google, HTC, Motorola, etc trying to use patents to stop the sale of the competition's products. Apple even used the argument 'look alike' successfully to stop Samsung.
If we did away with the patent system everything would be better and cheaper. But your politicians are lawyers...
Three RTOSs, a couple of compilers, maybe 100 real-world (as in commercial, sold to real science and aerospace customers) embedded apps, maybe 1000 or so PC apps, mostly engineeering things. I actually designed one CPU, two boards of ssi/msi TTL logic, for shipboard bell and data logging; hundreds were built and sold.
A very small fraction of my deployed embedded programs had bugs, and only one could be made to crash, and that was in a test mode. I treat software as an engineering design: do it carefully, document it thoroughly, check it hard, test it hard. It takes less time than deploying 20 versions of sloppy junk, rev 1.3.24b sort of nonsense.
In todays technological world, the worst thing that gets done is programming. There's no doubt about that.
aspect,
This is an electronic design group, and modt of our stuff uses hard embedded processors. Some benchtop instruments use GUIs... the slow, buggy ones.
We live in the dark ages of programming. Things have to change.
--
John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com
Donald Knuth said programs should be human-readable firstly and machine-readable secondly, but very few programmers read books like The Art of Computer Programming. Engineers read multiple texts that have a similar approach to teaching how something should be done. The kind of books most programmers do read are similar in style to a part datasheet. They describe the functions of a new language or compiler.
Any interesting cases?
--
Reply in group, but if emailing add one more
zero, and remove the last word.
LT Spice is coded, I hear, by one very talented guy. It's excellent because he's good. That's an exception. There's lots of commercial bloatware that is slow and buggy. People are being killed by bad code in xray machines and Airbus planes. The bigger the project, the worse the code.
Google software bug death and get stuff like
formatting link
Again, what do you mean by that?
--
John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com
The patent holder argues infringement while the defendant argues invalidity. If the patent holder offers a low cost license I don't see a problem.
Didn't Apple patent computer icons? The trash can, etc. I agree that enforcing such things will stifle competition.
per.
Eliminating the patent system would probably take a constitutional amendment. Article I Section 8: The Congress shall have power ... To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries;
I haven't personally seen any bad bugs, as in hard crashes or outright wrong behavior; have you? But I confess, I don't simulate a lot, maybe a couple of circuits a month, and mostly simple stuff. What I do see is unpredictable convergence, things that run in a second, or maybe never finish, when small changes are made somewhere.
The best simulator I've used was ECA, a text netlist DOS thing with primitive graphics. It always converged, was apparently always right, and had some cool features that made it easy to program changing component parameters. One reason it was so good was that its documentation was superb.
--
John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com
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.