Jump to content
IGNORED

MIDI Interface


ivop

Recommended Posts

Another suggestion is to see what the program doing in its busy loop before starting the game in an emulator and try to determine which key combinations are acted upon in one way or another. Probably have to hack around the check for the MIDI clock in the beginning, because as far as I know, none of the emulators implement that.

 

ATM I have no time to do that myself, because I'm going to watch the WC Allround Speedskating in Calgary on the telly and have a beer ;)

 

I won't do that either, with my only excuse being that my brain would hurt if I did that, and then I'd have to drink a beer to get over it :) .

 

Thanks for the suggestions.

Link to comment
Share on other sites

SUCCESS :thumbsup: :thumbsup: :thumbsup:

 

https://youtu.be/qC0T-YM5PMc

 

The initial problem was a bent CLOCK-IN pin on the SIO jack of the 1088XEL. Once that was fixed, we were in business :) .

 

I was able to quickly diagnose the problem by running Midi Track III which will throw an error if it doesn't see the external clock signal. I probably should have seen this on ivop's MIDIMON program as well, but maybe it just wasn't as obvious, or I was smoking too much at the time ;) .

 

post-42561-0-73568400-1551653511.jpg

 

Midi Track III.atr

  • Like 7
Link to comment
Share on other sites

Well kick ass. Once I bring up the 1088XLD, I'll do what I can to get a working network setup. I'd love to demo this at VCFMW 2019.

I don't mind sharing the ones I make. I'm going to ship them to Bill Lange right after VCF Southeast for the folks to use at VCF East the following weekend.
  • Like 2
Link to comment
Share on other sites

Ok, so what happens if you try the Midi ring with PAL and NTSC computers?

 

I haven't tried that yet. But from what I've heard there are timing issues.

 

-----------------

 

Here's something that'll help with making the SIO cable connections to the SIO2MIDI board.

 

post-42561-0-15389600-1551656271.png

 

These color codes matched up with the wire colors on my cable, but I would suggest disassembling the plug end to see where things really go (not all factory made SIO cables are the same).

 

-----------------

 

For anyone that wants to make one of these, here's the OSH Park BUY LINK.

 

post-42561-0-28692500-1551813535.png

 

These are just under $13 for 3 bare boards (minimum order), which includes FREE international shipping.

 

Schematic (includes BOM)

PIC Firmware (includes JOY2PIC flashing ATR)

MIDI-MAZE (XEX and ROM)

 

Note: The LED (D2) is a required item. It provides a pull-down for the motor control enable, and if left out there is a good chance that SIO communications will get blocked for other devices on the bus.

  • Like 4
Link to comment
Share on other sites

I've been digging for information on MIDI and it's association with the Atari 8-bit computer, and I came across a few interesting tidbits that I thought I would share.

 

The Birth of MIDI on the Atari

 

We helped keep Atari alive for its last year or more, by convincing them that there was a Music market for personal computers.

 

Read article: http://www.atarimuseum.com/articles/thebirthofstmidi.html

 

 

The PROFESSIONAL MIDI MACHINE

And How It Grew

 

In April 1983 a small Los Angeles company named Hybrid Arts was formed that set out to bring MIDI to the Atari 8-bit computer line. Founder Bob Moore stated that they chose the Atari because it was the sturdiest of the lightweight personal computers...

 

Read article: https://www.atarimagazines.com/startv3n4/profmidimachine.html

 

 

The Prophet 600 (1982)

Produced by Sequential Circuits

Very first MIDI based synthesizer

 

800px-SCI_Prophet_600_%28angled%29.jpg

At the Goodman Music Exposition in 1983, Hybrid Arts and Robert Moore demonstrated Atari’s first personal computer, the Atari 800 (8 bit with 48k RAM), communicating with a MIDI equipped Prophet 600.

 

Read article: https://funtabulious.wordpress.com/2015/07/05/sequential-circuit-design/

 

 

Atari Protos: MIDI-MAZE

 

The secret to MIDI Maze was the MIDI interface (Musical Instrument Digital Interface). This port found on all Atari ST computers, was normally used by musicians to hook up keyboards and synthesizers to the computer. However some clever individuals figured out that it could also be used to network up to 16 computers together for multiplayer gaming (hence the name of the game). However since the Atari XE doesn't have a MIDI port, the programmers graciously implemented several new networking options including the MIDIMate...

 

Read article: http://www.atariprotos.com/8bit/software/midimaze/midimaze.htm

 

https://www.youtube.com/watch?v=WF6IYOEOlbo

  • Like 2
Link to comment
Share on other sites

Hey guys, I've been casually following this thread, not being an XEL owner and all... But seeing the standalone version, and realizing that this is compatible with the Hybrid Arts MIDImate has piqued my interest.

In response to mytek's posting of MIDIMATE III for testing, I dug out the one I have that I acquired from original owner who purchased it from B&C in 2002. It included two disks, which I just imaged to ATR and confirmed are exact duplicates. (How courteous!)

mytek's ATR indicates date March 22, 1986 on boot, this "130XE Version": April 16, 1986

EDIT: mytek ATR compares same as one found on pigwa.
I also just checked Farb's 2019-02-03 Atari 8bit Preserved Software Torrent, and the version there identifies itself as "December 6, 1985" even though the ATR/ATX is labelled 1986. :)

EDIT2: The version of the manual uploaded to Internet Archive even has searchable OCR'd text: https://archive.org/details/MidiTrackIIIManual

MIDITRACK III 130XE VERSION @BAREFOOT (April 16, 1986).atr

post-53052-0-88464000-1551764837_thumb.jpg

  • Like 5
Link to comment
Share on other sites

Thank you Nezgar for uploading that variation of MIDI Tracker III :thumbsup: . Looks to be a more current version, then what's floating around the net.

 

BTW, this isn't an XEL only thread. It actually started out life as ivop's MIDI Interface design which later morphed into the MIDI-MUSE, incorporating the S2 Dream Blaster module. Now I think it's become a place for all things new with MIDI on the the A8. And without ivop starting this thread, creating his design, and sharing how it worked, I don't think I would have gotten MIDI fever like I so obviously have :grin: . Anyway what I'd like to see happen is to take this thing full circle at some point and get back to the MIDI-MUSE system which was aimed at all A8's, irrespective of their model number, but perhaps add the MIDI Controller chip to his board to simplify the overall logic design. I know he's been working on a DIY wave blaster module, which would be cool to integrate that as well.

 

Some of the stuff I was working on for the XEL is very specific to that system, and also some of it was done as a precursor to putting MIDI into the 1088XLD. I released all of what came out of that experimentation, but it isn't everyone's cup of tea.

 

The recently released SIO2MIDI board is an extreme simplification that was requested to enable doing up to 16 multi-player bouts of MIDI-MAZE at some of the upcoming Vintage Computer Festivals being held around the country. You can think of it as the bare minimum version of a MIDIMATE. And in it's current form it is universal, and will work on any A8 system via the standard SIO port. What I'm hoping will come out of these things touring the show circuit, is a much stronger Atari 8-bit presence, and the envy of all those C64 owners ;) .

 

Ivop's MIDI-MUSE is above and beyond MIDIMATE or the SIO2MIDI, adding a MIDI-Thru port, on-board synth, stereo output jacks, while still retaining the single board aspect of the original MIDIMATE.

 

Now it's time for me to turn this thread back over to the OP :) .

  • Like 4
Link to comment
Share on other sites

Maybe I will be the first guy to have an Atari 8-bit drive a midi drumset?

 

That would be cool! I have had it drive a couple of synths (real ones) and QSynth running on a Linux box.

 

Ah, there's another idea. Running a softsynth on a Pi, either connected to MIDI-OUT or morph it into an internal WaveBlaster compatible board ;)

  • Like 4
Link to comment
Share on other sites

Ah, there's another idea. Running a softsynth on a Pi, either connected to MIDI-OUT or morph it into an internal WaveBlaster compatible board ;)

Aaaah... Now I need to have one (sio2midi/Midi Muse) too! Finally something that can compete (volume wise) with that c64/Amiga/MSX noise in Maarssen :)

  • Like 1
Link to comment
Share on other sites

I finished mytek´s interface-pcb. :-) Because there´s no place left for the three DIN-jacks in the relean-housing, I put the interface in an external housing and connect it via a SUB D-connector. There was one breakthrough left in the H80 housing:

post-18285-0-31614500-1551888453_thumb.jpg

 

post-18285-0-50271700-1551888612_thumb.jpg

post-18285-0-39849100-1551888633_thumb.jpg

 

The pcb fits nearly perfect into the housing; just some little changes needed:

post-18285-0-10033900-1551888743_thumb.jpg

Housing from Hammond manufacturing.

 

The switch is not populated; if the interface is connected to the 1088XEL, a relay automaticly turns off the waveboard.

post-18285-0-71969200-1551888937_thumb.jpg

post-18285-0-02961300-1551888979_thumb.jpg

post-18285-0-89193200-1551889014_thumb.jpg

 

Relay is powered if external interface is connected; in idle state it connects pin 3 and 4 of J1 / MIDI-controller and activates the waveboard.

 

Sucessfully tested with a Yamaha-keyboard and MidiMaze. :-)

 

Sleepy

Edited by Sleepy
  • Like 4
Link to comment
Share on other sites

Sleepy what a great idea!

 

And your end result looks very professional. So based on what was required, what could be done with the design of the DIN interface board to make things better? I'm game to create a new and improved board, since that first one was originally aimed at the 1088XLD project, and has been displaced even in that role with something else.

  • Like 1
Link to comment
Share on other sites

Some spontaneous ideas:

 

- a hole in the S2-board to fix it with a screw if there is no place to put it on the mainboard as provided. Practically on the place where actual the LED is placed; both pcb (waveboard and S2-board) could be mechanicilly merged and fixed to the housing.

- Put the pullup R5 (10k) from the interface-board to the S2-board. If the interface-board is not connected and there is no jumper on J1 (e.g. if I use a relay or similar trick ;-), "Synth in" has a defined level in each case.

- Maybe there´s a better way to to the automated switching I realised with the relay, which can be added to the S2 board.

 

The changes I did on the interface-board was only mechanical nature to fit it into the housing; so it depend on the housing you wish to use. For the housing I linked above, I cut of the electrical connections of the parts BEFORE soldering flush to the pcb (to save space). I have to file off a little bit of the DIN jacks and milled a little slot to the housing´s cover.

post-18285-0-79455400-1551894241_thumb.jpg

 

The only electrical change was not to fit SW1 because switching the waveboard off is realised with a relay. But for this, there´s no need to do changes on the pcb. :-)

 

Sleepy

  • Like 1
Link to comment
Share on other sites

Sleepy thanks for the feedback and your suggestions :) .

 

Looking at your Dsub9 connection has given me some new ideas. Since a lot of people that went the Realan case direction picked the one with the 3 openings on the back (H80), I was thinking perhaps a 2nd version of the S2 carrier board would be in order. Essentially make it with a Dsub connector integral to the design, so that when mounting the connector it would automatically act as the mount for the carrier board at the same. I will play around with this concept and see where it takes me. Stay tuned for an update.

 

I will also look at adding a hole inline with the one on the S2 module, while moving the LED, and perhaps add a 2 pin header for a remote LED as well. I'm not keen on eliminating the manual synth source switching as you've done with the relay, since I've found it nice to be able to decide what I want to send to the S2 board, and when. However the automatic switching I originally implemented will get removed, since it gets confused if an input is active. However I might be able to come up with a compromise that works like your relay, but also supports the manual switch, and make it electronic in nature.

  • Like 4
Link to comment
Share on other sites

Very good idea; I´m looking forward! :)

 

...so that when mounting the connector it would automatically act as the mount for the carrier board at the same.

 

...like this?

 

post-18285-0-78083000-1551951397.jpg

post-18285-0-47795700-1551951439.jpg

 

Doing the layout, keep in mind that the SUB D - opening is very near to the top. Maybe it´s better to do the soldering side of the S2-board upside, so that the components and the waveboard are oriented upside down.

 

Sleepy

  • Like 2
Link to comment
Share on other sites

 

Starting at 7:50 there's Bob Moore from Hybrid Arts Inc.

http://atariage.com/forums/topic/240934-activisions-music-studio/

Jim Patchell dug up some goodies for us, of course mostly ST but useful none the less.

and a review

https://www.atarimagazines.com/v5n5/musicstudio.html

 

would be good to have more 8 bit centric stuff though. I wonder how hard it would be to port midi back into the XL/XE version of the activision music construction set or even mouse support, it would be nice wouldn't it?

Edited by _The Doctor__
Link to comment
Share on other sites

Very good idea; I´m looking forward! :)

 

 

...like this?

attachicon.gif12039-Terminalblock-VGA-Stecker.jpg

 

Doing the layout, keep in mind that the SUB D - opening is very near to the top. Maybe it´s better to do the soldering side of the S2-board upside, so that the components and the waveboard are oriented upside down.

 

Sleepy

 

Yes I'm thinking of the 2nd example you showed (assuming there's enough room). And definitely the S2 board would have to hang underneath.

 

Also been thinking of that relay replacement and I believe a PNP transistor would work with it's base normally biased to ground with a high resistance attached. That would default to the transistor being in an ON state, thereby passing the MIDI-OUT data thru to the S2 module's input. Then I would have another resistor of a lower resistance value connected from the base of the PNP to a pin on the DB9 connector. Subsequently if +5V were applied to that same pin, the PNP transistor would be turned OFF, thus ceasing to pass thru the signal.

 

Since you created the method of interfacing the two boards via a female DB9 connector on the Realan case side, could you describe what you used for a pinout on that connector?

 

 

Front View looking into a Female DB9 Connector

1432204827_rs232_db9_f.gif

 

I figure if I copy what you did, then later we can test the new board on your system without having to make any changes.

  • Like 2
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...