On Tuesday 10 June 2014 17:21:38 a k did opine And Gene did reply: > Basically I have a small dictionary (link to HAL) and I need to write > – very good poetry – story and without mistakes – otherwise it > – program will not work. > > If one can read dictionary and grammar book than he can write poetry/ > story/novel book in any language!!! > > This is Great attitude! > > Need close down all language school. Dictionary and grammar book there > – just read book and what ever else available and that is it. > > From one radio program I found out that it took many years of hard > work and help from famous tutor to Australian artist to learn how to > speak like American (US English). > > In that radio talk I did not found anything about “use your grey > cells†and you can speak any language. Maybe because people was > different type. > > Does this sounds familiar situation? > No, but its beginning to look a lot like a troll from here. But please read on, it could be quite educational. > > //////////////////////////////////////////////////////////////// > > > > 1.1 loadrt > > I have that loadrt scale count=1 > Somewhat Correct. 1 instance, starting from base 0. But IIRC scale is not a base_thread module as its floating point, so it must be servo_thread. Only modules that do not do floating point math can be added to the base_thread execution loop. > > 1.2 add > > I have that addf scale servo-thread
Wrong. "scale" s/b scale.0 > > 1.3 load usr > > what need here ? > > loadusr scale ? servo-thread? > loadusr scale count=1 Later in the file: addf scale.0 servo_thread For this, analyze the sequence of operations, then list the addf's is the top to bottom order because this is the order they will be executed in. Done in the wrong order, it might be quite a few "servo_threads" time slices before an error value actually gets to where it can adjust things. In the right order, each module addf'd will do its calculations and have them ready for the next module to absorb, and the propagation delay from start to finish will be a fraction of the normal 1 millisecond servo_thread. Servo systems are many times more stable when these unwanted delays are minimized to the extent possible. > 1.4 Net net starts a linkage in the following format: net; nameofsignal; source of signal; destination of signal; repeat as needed. Skip the semi-colons I used, a space separator is all thats required. net nameofsignal can be reused later but will take all arguments as additional output fanouts. > Something > > Net scale –percent scale.0.out=> motion.adaptive feed s/b more like net nameofsignal scale.0.out motion.adaptive_feed.in(check the manpage for motion to make sure the dots and underscores used match the manpage. My aged wet ram doesn't always remember such details) The => isn't needed as its the position in the space or tab separated line that truly counts. Basically "net" tells hal how to parse the line. Same for setp or addf or one of the loads. The next (2nd) argument is the name of your signal, totally arbitrary, use a label you will understand 6 months down the log. The 3rd argument is the source of the signal, in this case scale.0.out The 4th, and all subsequent arguments on this same line are where the scale.0.out data is to be sent. All of this is in the integrators manual, so I always highly recommend downloading the pdf of it and making it into dead tree format, punched & mounted in a cheap 3 ring binder so it can be studied and learned from in a comfortable sitting position with minimal distraction. I'll finish be stating that there are quite a few people here who are far more adept at this than I, but they can't get anywhere until you are willing to do the testing exactly as you have been requested to do, in this case by Andy, but rarely actually do. Without that level of co- operation on your part you are forcing the folks that can probably help you to do a lot of guessing and wheel spinning. That is not productive use of anyones time. These people, none of them paid, are taking the time to try and lead you, and correct whatever it is you may have done wrong, doing it hopefully in a manner that will teach you. > > 1.5 setp > > setp scale.0. gain 0.000344 Junk the space, it is scale.0.gain > setp scale.0. offset 0.417 Ditto.> > > > 1. 6 sets > > do not have that do I neet it? Its a slightly obsolete keyword, but still parsed by hal I believe. Use is not encouraged. > > On Tue, Jun 10, 2014 at 1:44 PM, Viesturs Lؤپcis > <viesturs.la...@gmail.com> > > wrote: > > 2014-06-10 21:17 GMT+03:00 a k <pccncmach...@gmail.com>: > > > hm2-servo.hal:206: addf failed > > > > This thread has been around for more than a week and basically is not > > moving forward, because lots of different answers are provided > > basically for the same question over and over again. This reminds me > > the saying about giving a man a fish versus teaching him, how to > > catch a fish. > > > > Aram, did you know that those error messages are useful and contain > > meaningful information? Please, do try to read that whole stuff next > > time. > > For example, if you pay 30 seconds, you can find the line that says > > "addf failed". > > I do not know, how about you, but to me it tells that there is a > > mistake with this line > > addf scale servo-thread line > > > > How do I know? Because error message says that it failed! I do not > > understand approximately 95% of error message output, but in 98% > > cases there is one line, which contain some words that I have in my > > config and that line tells, where should I look for the cause. > > > > If you then try reasonable thinking and actually using your existing > > config as an example, you can find that hal components need instance > > number in addf line. > > > > For example: > > loadrt and2 count=2 > > > > addf and2.0 servo-thread > > addf and2.1 servo-thread > > > > Please, use your grey cells to understand, what is missing in your > > line: addf scale servo-thread. > > > > And also do not forget to correct the pin names of scale component, > > if you will figure out, what is missing > > > > > > Viesturs > > Viesturs is also correct. He is trying to teach you how to decode the error messages you are getting. They really do tell you what you did wrong. These folks have helped me immensely because I'm a somewhat late arrival to LCNC, only been using it for about a decade. What I really am is the lists genuine old fart, a retired television engineer, a Certified Electronics Technician whose license cards are fading with age because I'll be 80yo in the next October. Cheers, Gene Heskett -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) Genes Web page <http://geneslinuxbox.net:6309/gene> US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS ------------------------------------------------------------------------------ HPCC Systems Open Source Big Data Platform from LexisNexis Risk Solutions Find What Matters Most in Your Big Data with HPCC Systems Open Source. Fast. Scalable. Simple. Ideal for Dirty Data. Leverages Graph Analysis for Fast Processing & Easy Data Exploration http://p.sf.net/sfu/hpccsystems _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
