Jump to content

jimhenry

Frequent Writer
  • Posts

    303
  • Joined

  • Last visited

Everything posted by jimhenry

  1. The parallel inputs of a chain of MB128s may not be elegant but it is as solid as a rock and about as easy as can be to troubleshoot. I've seen a lot of people successfully complete very impressive projects rather quickly with this approach. There is a lot to be said for using something simple, even if it uses more hardware, when you are doing a one off project. I've also seen quite a few people spend a lot of time trying to be clever by using matrix input without ever completing their project, myself included.
  2. There is no "mature" matrix input software for the 8 bit MidiBox. You have to be pretty comfortable with assembly language coding and be prepared for a lot of work to do matrix input on the 8 bit platform. I would recommend using the 32 bit platform and the newest MIDIIO128 software for matrix input. 60ms seems slow if you are talking about opening a valve under a pipe in a pipe organ. You'll notice the delay once it starts getting above 20ms.
  3. It depends on the software but it can be very low, i.e. not noticeable.
  4. jimhenry

    AIN

    It is possible to make a small patch to MIDIIO128 to support a few analog inputs connected directly to J5. I don't recall where the patch is posted at the moment.
  5. Yes, it is that simple to use NC rather than NO switches. The problem most people have with DIY switches on keyboards is providing a switch that will switch at about the midpoint of key travel without impeding further key movement.
  6. Not interested in doing the programming for this. I don't have the hardware and I am not interested in acquiring it. By the time you spend the money for the SAMs and the power supply required, the cost of a commercial solution for the control electronics is not unreasonable. I don't think you appreciate just how complex this seemingly simple system is. I have seen some very good engineers come close to going up in flames, literally, building combination actions due to very small errors in the code and/or hardware. There is a massive amount of amperage being controlled here. This is not a project that is forgiving of mistakes.
  7. What Thorsten is saying is that this project is sufficiently complex that the programmer will have to have direct access to the hardware. It is not something that can be accomplished is a reasonable amount of time by a programmer who has to rely on someone remote to test the software.
  8. All the more reason to avoid touching the hardware if you can. If you take it out it sounds like you should plan on redoing the "shortcuts" at that time. Can you get to anything that allows you to test the operation of the analog inputs at any point?
  9. I would try to avoid taking the modules out of the enclosure. There is a chance you will create problems if you don't have them already. Do you have a multimeter so you measure resistances and other electrical values?
  10. Converting it to a C project rather than an assembler project would probably make the changes you want to make simpler. But converting the project to C would probably be harder than making the fixes you want in assembler. In other words, no. Very good on figuring out the sustain value. Is everything that is based on an on-off switch working to your satisfaction? I wouldn't move on to the analog inputs until you have done everything with the digital (on-off) inputs.
  11. The really big thing you've accomplished is to make a change to the code, rebuild it, load it, and verify that it does what you want. Now the real fun starts, learning how to do the magic. Do you understand why the change to the code made the sustain send 127? Once you understand that little change you need to identify ONE issue to work on next. It probably should be something involving note on/off. Most of your issues seem to involve things that are not from the note on/off switches. That stuff is probably highly custom and I might not be able to offer much help. So try to stick to easy stuff while you are learning. Remember it took time to learn how to play the keyboard too.
  12. I would guess that should be "MIOS V1.5", a report of the version of the MIDIbox Operating System that is loaded. Maybe there is a language issue that is causing it to appear as "MEOS Z1.5" on your system. If you have a backup copy of the version of MIOS currently installed available, then I would try updating MIOS to the current version, which is V1.9g. I don't think that will cause any problems but I can't make any promises. I think it would be a good idea to bring everything up to date but I can pretty much guarantee that you will run into at least one issue that will take some effort to resolve in doing so. My guess is that updating the project code to work with the current tool chain is where that issue will be. As I said before, you will find it harder to get help if you leave the project as an old project.
  13. It doesn't have to be one or the other. Start by working on the issues you have with the keyboard as is. When all that is to your satisfaction you can decide if you want to go further. You'll know a lot more then and you can make a better decision.
  14. I would suggest trying the current version of MIOS Studio and all the tools first. You may find that there are things in the project code that need to be updated in order to build successfully with these tools. If you are really stumped trying to use the current versions of the tools, then fall back to the tools that are the same vintage as the project code. The downside is that it will be more difficult to get help if you need help with the tools. I haven't built a MidiBox project for a long time so I can't offer more specific help. Hopefully someone else will add more up to date advice.
  15. Your second step, the first one where you will actually make a change, is to find the sm_fast.inc file and then find this section of that file: COL_12_other movlw 0x0f cpfseq SM_BUTTON_ROW goto SEND_PRG_CH SUSTAIN_PROCEDURE movf SM_CH_STATUS, W addlw 0xb0 movwf MIDI_EVNT0 movlw 0x40 movwf MIDI_EVNT1 movlw 0x01 CPFSEQ MIOS_PARAMETER2 goto SEND_SUSTAIN_40 movlw 0x00 goto SEND_SUSTAIN SEND_SUSTAIN_40 movlw 0x40 SEND_SUSTAIN movwf MIDI_EVNT_VALUE call MIDI_EVNT_Send return Under SEND_SUSTAIN_40 change "movlw 0x40" to "movlw 0x7f". Build and load the project. Test your sustain pedal. If it now sends 127 when pressed, you can do the following cleanup: change "goto SEND_SUSTAIN_40" to "goto SEND_SUSTAIN_7F" change "SEND_SUSTAIN_40" to "SEND_SUSTAIN_7F" These two changes don't affect how the project works. They are just to keep the code clear. Congratulations! You've just taken the first step up the mountain. Have a glass of Polish vodka.
  16. That code is based on Thorsten's experimental fast switch matrix demonstration project. A few people have gotten things sort of working based on it but it has never advanced to the point of being a solid project. I'm not sure but MIDIO128 v3 might include support for a switch matrix. If it does, switching to that project probably would be a good idea. There will be more people who can help if you are using a mainstream project. [[Edit: I just looked and MIDIO128 v3 uses a different Core board. I wouldn't switch the Core for now since you have something that is basically working.]] The biggest challenge is to get switch matrix code that handles velocity sensitivity. CuBase has reported the greatest success in this area and your code is probably based on his efforts. Unfortunately what you are trying to do is not a beginner project. This type of thing presents challenges even to more experienced MidiBox users. You need to be very patient as you climb a steep learning curve. You probably will need to do several things that are less than all you want to do just to learn. One beauty of MidiBox is that it is easy to replace the software so you can start with easier, less functional software and then build on it. This type of project should be discussed under "Midification" where it is slightly more likely that people with experience in this type of project will notice you. This area is for questions about assembly language programming. You might have those types of questions along the way but you are asking bigger questions right now. Your first step is to be able to build the code you have, without changes, and successfully load it on the MidiBox. Have you done that yet?
  17. Does the lever on the end of the bus rod press against the felt when it is the position where the conductive part of the rod contacts the wipers? If so, you might be able to tie the rod in that position with a piece of wire that is looped through those holes to the right in the picture.
  18. Does the lever on the end of the bus rod press against the felt when it is the position where the conductive part of the rod contacts the wipers? If so, you might be able to tie the rod in that position with a piece of wire that is looped through those holes to the right in the picture.
  19. Can you post a picture of the mechanism that rotates the bus rods?
  20. Can you post photos of the bus bars and explain what it is you want to "lock down"? It is likely that no one here has ever even seen a Conn 632. I know I haven't.
  21. Why do you want to use six bus bars? One should be enough. Maybe two for redundancy.
  22. I hope you looked at the WIKI to be sure the LCD you picked is compatible with the Core. I assumed you had so I didn't say anything. There is a link on that page to my guide for connecting an LCD.
  23. Plenty of people have managed to bring up a Midibox without the LCD display. However, I still think it is handy to have, especially on your first Midibox. MIOS Studio requires that you have a computer and a MIDI connection to the computer. The LCD display gives you an independent way of seeing what the Core is doing. For a $1 more I'd go with the 2x40.
  24. Yes, ICD (or IDC, insulation displacement connector) connectors for sure at least on the Midibox end. MIOS will already be loaded on the Core8 PIC that you get from SmashTV.
  25. As I understand the new version of MIDIO128, it is still limited to 128 inputs unless you are using a switch matrix. Based on the drawing you've made, I think the PIC8 boards will be more cost effective. BTW, the DIN chains should connect to J8/J9.
×
×
  • Create New...