Loading ...
Sorry, an error occurred while loading the content.
 

Re: [smartsockets] Fourplay ?

Expand Messages
  • Derryck Croker
    ... I bought my kit with the express purpose of this, so I ve definitely got my hand up. As for features, you ve mentioned some great ones, but I d definitely
    Message 1 of 14 , Jan 14, 2012
      On 12 Jan 2012, at 09:32, fixitsan2 wrote:

      > I have no experience of living with or operating a Four Letter Word type device so I'm looking for some input from anyone who has one or who would like one, regarding what features to include.

      I bought my kit with the express purpose of this, so I've definitely got my hand up.

      As for features, you've mentioned some great ones, but I'd definitely add an "X-rated" switch to avoid those 4 letter words which might not go down well in a family environment!

      Any time-span on this? I've just emailed you asking about the 2.3 upgrade you were going to apply to the PIC I sent at the beginning of the year. I'd be happy to wait for this!

      Cheers

      Derryck
    • fixitsan2
      ... Hi Derryck ! I ll definitely be including a censorship function to keep it clean . Timescale ? I don t have a great track record there, however, I am very
      Message 2 of 14 , Jan 14, 2012
        --- In smartsockets@yahoogroups.com, Derryck Croker <derryck@...> wrote:
        >
        > On 12 Jan 2012, at 09:32, fixitsan2 wrote:
        >

        >
        > Any time-span on this? I've just emailed you asking about the 2.3 upgrade you were going to apply to the PIC I sent at the beginning of the year. I'd be happy to wait for this!
        >
        > Cheers
        >
        > Derryck
        >


        Hi Derryck !

        I'll definitely be including a censorship function to 'keep it clean'.

        Timescale ? I don't have a great track record there, however, I am very interested in coding word selection algorithms which will keep me interested.

        Thankfully all of the display control coding is done and dusted !

        Due to space restrictions, because the current 18F25k22 is at about 24kB of it's 32kB capacity I'll have to move over to the 18F26k22, with 64kB capacity for all those 4000+ four letter words and word maps !

        I've just disabled autobaud selection. I've noticed some difficulty in getting the devices to reinitialise together after a power down when the backup battery is installed. A cold reset works great, but a warm start from a sleep condition was being too troublesome. The baudrate is now fixed at 9600baud.
        (nobody say "I told you so" ;0 )

        I hope your IC arrives in the next couple of days, it was posted at the start of the week from a post office in the shopping centre which was nominated to be Britain's worst looking town centre, Cumbernauld
        http://en.wikipedia.org/wiki/Cumbernauld
      • Derryck Croker
        Hi Chris, ... Neil sent me a doc from a 4LW device, which offers frequency of swear words from often to never, which is possibly a good compromise. ... The
        Message 3 of 14 , Jan 16, 2012
          Hi Chris,

          >> On 12 Jan 2012, at 09:32, fixitsan2 wrote:
          >>
          > I'll definitely be including a censorship function to 'keep it clean'.

          Neil sent me a doc from a 4LW device, which offers frequency of swear words from often to never, which is possibly a good compromise.

          > Thankfully all of the display control coding is done and dusted !

          The Hangman effect from the above doc would be a nice addition to the "Fourplay".

          > I hope your IC arrives in the next couple of days, it was posted at the start of the week

          Not yet I'm afraid :-(

          Cheers

          Derryck
        • fixitsan2
          ... Maybe yoiu could share the document or give a description of the hangman effect. There are a lot of transition effects in the Smartsockets, most of which
          Message 4 of 14 , Jan 16, 2012
            --- In smartsockets@yahoogroups.com, Derryck Croker <derryck@...> wrote:
            >
            > Hi Chris,
            >
            > >> On 12 Jan 2012, at 09:32, fixitsan2 wrote:
            > >>
            > > I'll definitely be including a censorship function to 'keep it clean'.
            >
            > Neil sent me a doc from a 4LW device, which offers frequency of swear words from often to never, which is possibly a good compromise.
            >
            > > Thankfully all of the display control coding is done and dusted !
            >
            > The Hangman effect from the above doc would be a nice addition to the "Fourplay".
            >
            > > I hope your IC arrives in the next couple of days, it was posted at the start of the week
            >
            > Not yet I'm afraid :-(
            >
            > Cheers
            >
            > Derryck
            >


            Maybe yoiu could share the document or give a description of the hangman effect. There are a lot of transition effects in the Smartsockets, most of which work from a blank display.

            A transition can take upto 4 seconds and added with one of the delays so that digits change sequentially a much longer effect can be created. I'm happy to look at new ideas anyway.

            I can only shrug my shpoulders about the speed of UK post recently. I know it was Christmas recently but nevertheless....lets give it a couple more days.

            I've come across a new problem for the next version of Smartsockets, which if anyone has any experience of please chime in, because it has got me, the authours of PIC Basic Pro, and Microchip, scratching our heads.

            Code which currently runs well on the 18F25k22, which increments a 'Seconds' counter in the interrupt, sometimes randomly increments the seconds by 2, when run on the 18F26k22. Nobody seems to understand why.

            This doesn't affect V2.30 code or any Smartsockets I have available, it is just a quirky thing which is stopping me from pushing ahead with the 4-play device on the 18F26k22, for now.

            Chris
          • Terry
            ... Amazon UK shipped a disc to me (in the US) on December 8th. After waiting until December 30th for it to arrive, I contacted Amazon. They said it must have
            Message 5 of 14 , Jan 16, 2012
              --- In smartsockets@yahoogroups.com, "fixitsan2" <fixitsan@...> wrote:
              > I can only shrug my shpoulders about the speed of UK post recently. I know it was Christmas recently but nevertheless....lets give it a couple more days.

              Amazon UK shipped a disc to me (in the US) on December 8th. After waiting until December 30th for it to arrive, I contacted Amazon. They said it must have been lost in the post, and shipped another copy that day. On the 9th of January, the disc arrived. On close inspection, it was the one that shipped on December 8th - over a month for delivery.

              > I've come across a new problem for the next version of Smartsockets, which if anyone has any experience of please chime in, because it has got me, the authours of PIC Basic Pro, and Microchip, scratching our heads.
              >
              > Code which currently runs well on the 18F25k22, which increments a 'Seconds' counter in the interrupt, sometimes randomly increments the seconds by 2, when run on the 18F26k22. Nobody seems to understand why.

              This sounds like the interrupt isn't being cleared properly in either the hardware or the software, so your routine is executed twice for the same event. I'm not familiar with PIC programming, but the normal procedure is to mask interrupts (or at least this one) on entry to the interrupt service routine, service it, then tell the hardware to dismiss the interrupt and then re-enable interrupts and exit.

              If the hardware hasn't gotten around to handling the request to dismiss the interrupt by the time you re-enable interrupts, your routine will be called again, thinking it is a new interrupt.

              You could add a global counter for "inside timer interrupt routine" which gets incremented as the first thing on entry to the routine and decremented as the last thing before exit. Then test for a value other than 1 in the body of the interrupt service routine and error out if not 1, and likewise in the main loop of the code for a value other than zero.

              It might be possible to hide the problem by adding a do-nothing delay loop between requesting the hardware to dismiss the interrupt and when you re-enable interrupts (assuming that the combination isn't an atomic operation on the PIC).
            • fixitsan2
              ... I hope that isn t normal. I just ordered a Kindle for my dad. I m in the UK, he is in Spain, and accordingly, i can only have the item shipped from the US.
              Message 6 of 14 , Jan 17, 2012
                --- In smartsockets@yahoogroups.com, "Terry" <terry+yg@...> wrote:
                >

                > Amazon UK shipped a disc to me (in the US) on December 8th. After waiting until December 30th for it to arrive, I contacted Amazon. They said it must have been lost in the post, and shipped another copy that day. On the 9th of January, the disc arrived. On close inspection, it was the one that shipped on December 8th - over a month for delivery.


                I hope that isn't normal. I just ordered a Kindle for my dad. I'm in the UK, he is in Spain, and accordingly, i can only have the item shipped from the US. no joke. They assure me a maximum of 4 days via express courier. $20 shipping and handling and $72 for importation tax in Spain. We're in the wrong business !


                >
                > This sounds like the interrupt isn't being cleared properly in either the hardware or the software, so your routine is executed twice for the same event.

                Thanks for your help Terry, I've gone over everythig like that already. the pic masks it's own interrupts when it enters the ISR and 4 lines of code after I test for the interrupt in PIR2,1, I reset it.

                It definitely seems that the ISR randomly runs twice. I have tried processing interrupts sequentially, by checking for any outstanding interrupts before leaving the ISR, as well as controlling the ISR so that it runs only one time for each interrupt and the results are always the same.
                I've relocated variables, to rule out a memory location issue too.

                The 18f26k22 is an 18f25k22 with another 32k of codespace, everything else is identical, including the locations of all SFR's and interrupt vectors, so in theory the same code which works on the 25k22 will work on the 26k22 without modification.

                It is obviously a strange thing because it happens randomly. Sometimes a while second is added in the first minute, sometimes nothing is added for 30 minutes, or several can be added in ten minutes.

                I would suspect noise but I wouldn't be able to explain how the only code which gets affected is only that small section of code, and only on the 26k22 and not with the 25k22.

                Microchip are looking at it, after now confirming my code and configuration settings are legitimate.

                The only thing left is that although I write intrrupts with assembly, for the general structure I fnd it quicker to write using PicBasic Pro. The pic has two priority levels for interrupts, high and low, but picbasic seems to only implement high level ints. In saying that, I think it is easy to use low level too, but the standard default from microchip is all interrupts are high level unless you change something in setup. I haven't added any code to change int priorities and if it were there it should have a continuous effect, not the variable effect I am seeing.

                Now that I'v written this, I wonder if picbasic uses TMR3 in the background for anything, such as for timing delays and regulating peripherals like the time spent reading ADC channels etc.
                Again, it won't explain why the 26k22 fails and the 25k22 doesn't though, because they have the same configuration files and the same hardware.

                I'll let you know if/when I get to the bottom of it

                Chris
              • John Rehwinkel
                ... It starts blank, then lights the letters up one (random) segment at a time (this is happening to all the letters together). I can make a video of it if
                Message 7 of 14 , Jan 17, 2012
                  > Maybe yoiu could share the document or give a description of the hangman effect. There are a lot of transition effects in the Smartsockets, most of which work from a blank display.

                  It starts blank, then lights the letters up one (random) segment at a time (this is happening to all the letters together). I can make a video of it if you like.

                  - John
                • fixitsan2
                  ... Thanks for adding detail John. It sounds simple enough to replicate. I have 20 tubes (5 x smartsockets) lined up before me as I type this so it will be fun
                  Message 8 of 14 , Jan 17, 2012
                    --- In smartsockets@yahoogroups.com, John Rehwinkel <jrehwin@...> wrote:
                    >
                    > > Maybe yoiu could share the document or give a description of the hangman effect. There are a lot of transition effects in the Smartsockets, most of which work from a blank display.
                    >
                    > It starts blank, then lights the letters up one (random) segment at a time (this is happening to all the letters together). I can make a video of it if you like.
                    >
                    > - John
                    >

                    Thanks for adding detail John.
                    It sounds simple enough to replicate.

                    I have 20 tubes (5 x smartsockets) lined up before me as I type this so it will be fun to play with that
                  • fixitsan2
                    ... it won t explain why the 26k22 fails and the 25k22 doesn t though, because they have the same
                    Message 9 of 14 , Jan 17, 2012
                      --- In smartsockets@yahoogroups.com, "fixitsan2" <fixitsan@...> wrote:
                      >

                      <regarding 26k22 and 25k22 code operation differences>

                      it won't explain why the 26k22 fails and the 25k22 doesn't though, because they have the same configuration files and the same hardware.
                      >
                      > I'll let you know if/when I get to the bottom of it
                      >
                      > Chris
                      >

                      I have an update on this issue.
                      In my ISR I have interrupts created by timer 1 and timer 3. Timer 1 is used as a one-shot timer and timer 3 is used as the low frequency 1Hz generator.

                      Timer 1 and timer 3 are identical peripherals, so I have simply renamed timer 1 registers to timer 3 registers and vice versa, and now the code works well on the 26k22 as well as the 25k22, with timer 1 running the 1 second epoch code.

                      The only difference between timer 1 and timer 3 is that they use different peripheral interrupt enable registers, maybe there is an issue with that ?

                      Chris
                    • Quixotic Nixotic
                      Chris, I have been unable to get to your project for a while but I see that will I have been inattentive there has been some issue with the baud rate or
                      Message 10 of 14 , Jan 17, 2012
                        Chris,

                        I have been unable to get to your project for a while but I see that
                        will I have been inattentive there has been some issue with the baud
                        rate or something in the PICs you sent out. Can you remember what the
                        version was that you sent to me? I have a little board built up using
                        a 16F688 with which I am hoping to send data to the smartsockets.

                        Regards,

                        John
                      • fixitsan2
                        Nice to hear from you John. Your version will most likley be V2.24 It seems the most likley baud rate to try first is 38400. please pop it in the post to me at
                        Message 11 of 14 , Jan 17, 2012
                          Nice to hear from you John.

                          Your version will most likley be V2.24
                          It seems the most likley baud rate to try first is 38400.
                          please pop it in the post to me at any time and i will put the latest software on.
                          Autobaud has been removed because of conflicts with chip reset procedures when recovering from a power off, while operating from backup battery, and so a fixed baudrate of 9600 is all I am using now.

                          You also have the 18F26k22 on that board, which is the one that has caused problems thanks to some sort of compiler bug or an undocumented feature by Microchip, the only problem which it seems to cause is random incrementation of the seconds count by two, instead of one.

                          I'm warming to the idea of using Atmel devices next, I am not happy about this newest timer/interrupt malfunction, and the previous issue regarding using the MOVFF command in interrupts. Mind you, can there be many more nasties left to discover ?

                          If anyone else would like to upgrade to the latest V2.30 code please drop me a line and we can arrange a way to do that quickly

                          Chris


                          --- In smartsockets@yahoogroups.com, Quixotic Nixotic <nixcited@...> wrote:
                          >
                          > Chris,
                          >
                          > I have been unable to get to your project for a while but I see that
                          > will I have been inattentive there has been some issue with the baud
                          > rate or something in the PICs you sent out. Can you remember what the
                          > version was that you sent to me? I have a little board built up using
                          > a 16F688 with which I am hoping to send data to the smartsockets.
                          >
                          > Regards,
                          >
                          > John
                          >
                        • michail1@aol.com
                          Actually, I need to know your selling site / price list. Yes, I know you have given it (prices) out before. :( No harm in putting out the updated list again.
                          Message 12 of 14 , Jan 17, 2012
                            Actually,
                             
                            I need to know your selling site / price list.
                             
                            Yes, I know you have given it (prices) out before.  :(
                            No harm in putting out the updated list again.  :)
                             
                            Michail
                             
                             
                            In a message dated 1/17/2012 10:13:00 A.M. Pacific Standard Time, fixitsan@... writes:
                            If anyone else would like to upgrade to the latest V2.30 code please drop me a line and we can arrange a way to do that quickly

                            Chris
                          • fixitsan2
                            ... I don t have a site, yet, Michail. I have been looking at e-commerce solutions recently though. Otherwise, I will continue selling via eBay, or emal
                            Message 13 of 14 , Jan 17, 2012
                              --- In smartsockets@yahoogroups.com, michail1@... wrote:
                              >
                              > Actually,
                              >
                              > I need to know your selling site / price list.

                              I don't have a site, yet, Michail. I have been looking at e-commerce solutions recently though. Otherwise, I will continue selling via eBay, or emal requests.

                              prices will probably continue to be as mentioned in message 797, for a while at least.
                              http://groups.yahoo.com/group/smartsockets/message/797


                              Chris
                            Your message has been successfully submitted and would be delivered to recipients shortly.