Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[makott]

Problems with mini din 8 serial ports and MIDI interfaces
Hi guys, I need help, information, and spiritual comfort :-)

I'm trying to set up a MIDI to DMX setup with old Macintosh computers and MIDI interfaces.

I have an 8100 Maxpower G3 System 9.1 that works perfectly, and I have a PowerBook G3 Wallstreet 300 System 9.2 that also works. OK.

I have an MIDI Opcode Studio 4 interface and a MIDI Motu MTP AV interface, fine. I've tried a thousand ways but it doesn't work! The interfaces are not recognized.

8100 >Studio4 with OMS 2.3.8/2.3.5 nothing.

8100> MOTU with Freemidi 1.44 nothing.

Same thing with Powerbook G3.

With a cable, different from the one used to connect the interfaces, which has a mini din plug on one end and eight exposed wires on the other, I made a loopback by connecting the wire from pin 3 and the wire from pin 5, as described in many tutorials. Okay.

Computer serial port test with "loopback cable" and TechTool Pro 3 “passed.”
However, if I try to run the test with the normal cable connected to the computer and the interface, the test fails. The cable is not an original serial cable, but it is well known and has been tested with a multimeter by a technician who assured me of pin-to-pin continuity.

Where could the problem be? Non-original Apple mini din 8 serial cables? (But sold as compatible pinout pin to pin)

Problems with "stack AppleTalk" occuped serial port comunication?

MOTU/Studio 4 settings?

All the serial ports on the interfaces are dead?

Help me!

Thanks for your attention.

EDIT: I read this post https://macos9lives.com/smforum/index.php?topic=7542.msg58406#msg58406
 while searching the forum. Could it be that the cables need to be crossed ? I'm confused now!

[GaryN]

Don't feel bad. This is simple but has tripped up countless numbers of users over the years.

Loopback is NOT the answer. You need - to be precise - an RS-422 Mini-DIN8 crossover cable.
Long ago, they were also referred to as "null modem" cables.

I have a diagram that I can't find…… big help huh? However:
This protocol and connection is used when two-way communication is required, such as between two computers or maybe a computer and a CPU-equipped MIDI interface LOL…

The point is that there are two transmit pins and two receive pins on each connector.
The Tx and Rx lines are "crossed over" so the Tx goes to Rx and vice-versa.
There's also a signaling line that's crossed over too.

The pin connections go this way:

1>2
2>1
3>5
4>4
5>3
6>8
7>7
8>6

More info HERE: https://macos9lives.com/smforum/index.php?topic=2656.msg15948#msg15948

You can actually buy these but you have to be certain the seller knows what they're selling because the industry, in its infinite wisdom, didn't think is was necessary to actually require marking these cables somehow to differentiate them from the straight-thru types. More than a few people have bought MINI-DIN cables for their Mac-whatevers and found out the hard way that the cables didn't work for them.

Here's one that's correct:
https://www.ebay.com/itm/232343545580

[makott]

Hello... You don't know this, of course, but you saved my life! :-) :-) I've been going crazy for four months, buying interfaces, having my computer repaired (I have a beloved Classic in lab condition). After the 8100 thinking about problems with the serial ports, I bought a Wall Street PowerBook, cables, etc., etc., etc.  Anger, frustration, and disappointment were seriously dampening my initial enthusiasm for this project.

Now: as you can imagine, this is a delicate moment for me, a turning point, and I don't want to make any mistakes. Could you kindly explain to me exactly where to find a precise diagram? Like you would to a six-year-old child. It's not convenient for me to buy a cable in the US because I'm in Italy... I don't know how to solder wires, I'm not a technician, I need to take a CLEAR diagram to a technician to have it made for me. Could you help me?

I am very grateful to you.
Thank for you attention.
Mario


P.S. These crossed “null modem” cables are tested to work with the MIDI interfaces mentioned, right?
With the confusion in my head between rs422, rs232, geoport, crossed, pin-to-pin,
I just wanted to run a sequencer performer 2.7 or Cubase on Studio 4 or Motu with my beloved old Macintosh System 9...

I also found this diagram that says NOT to solder the “sock” to the body at both ends but only to one
I would have found this cable with exposed wires, which I would take to a technician with a plug and a clear diagram.

[GaryN]

Try not to overthink this.

If you have the cable you pictured with the bare ends:
The pic is showing you looking at the outside end of the plug. Therefore,

On the "new" plug the wires will attach as:

1---Brown
2---Black
3---Yellow
4---Orange
5---Red
6---Purple
7---Blue
8---Green

There should be tiny pin numbers inside the new plug

The shield (sock) can be attached on one end or both. Personally, I prefer both with these.

So:
1 and 2 are crossed over
3 and 5 are crossed over
6 and 8 are crossed over
4 and 7 are NOT crossed over

It's not that difficult to solder one of these plugs but it does require (1) a small vise to hold the plug, (2) a small fine-tipped soldering iron because the pins are so close together and (3) patience.

No sweat… right?

[makott]

...My project aims to revive old technology, Hypercard, which controls a MIDI-to-DMX converter via sequencer to recapture the multimedia spirit of the early 90s. It's as if Laurie Anderson and certain influences from Brian Eno had come together 35 years later.
Unfortunately, I've been bogged down in technical hardware mud for months.

And no, unfortunately I can't solder, I'm now an old history teacher.

If this fucking cable works, I'll put you in my personal credits :-)

Thank!
Mario

[marl3n3]

Don't feel bad. This is simple but has tripped up countless numbers of users over the years.

Gary, didn't you bail me out with this same info a few months ago?  You're Crossover Cable Batman!

Actually I've been in the weeds of a bunch of retro tech noobery this whole year and that cable problem was probably the worst thing.  It just really does not jump out at you if you're depending on the original 1st party manuals.

[GaryN]

...My project aims to revive old technology, Hypercard, which controls a MIDI-to-DMX converter via sequencer to recapture the multimedia spirit of the early 90s.

Wow… now I'm intrigued. You have a Hypercard stack with a sequencer… or that follows a sequencer to control a MIDI > DMX converter? Does that run on a Mac along with a sequencer app somehow?
I could have really used such a thing 25 years ago when I had a sequencer-augmented band.

You know, even an old teacher can learn to solder… old dog > new trick 

Let me know what happens. 

PS. Thanks to Marl3n3, I remembered what I had re: the saga of Mini-DIN cables.
More than you ever wanted to know is here:https://whitefiles.org/tec/pgs/h10b.htm

[makott]

Hi,  so  today I hope to be able to go  to the technician with cables and plugs to  have him solder  this  damned crossed cable according to your  invaluable suggestion  (without which I would still be stuck in quicksand).
Hypercard, in combination with HyperMIDI Stack, allows that wonderful tool that was Hypercard to communicate with a MIDI interface. (Among the many mistakes Apple has made, I would certainly put the abandonment of Hypercard development on the podium.)

However, HyperMIDI runs up to 7.5. So: a wonderful 1991 Mac Classic that I have jealously preserved (completely recapped recently)
runs Hypercard connected to an Opcode Studio 4  (the Opcode Studio 4 can communicate with Apple MIDI Manager without OMS on the Classic 68000, which is not compatible).

 The Studio 4 has one output to a DoreMIDI interface (which converts MIDI to DMX) and one output to 8100 with MOTU MTP AV, which runs Digital Performer 2.7.2 with a Roland JV1010.
I recently added a PowerBook G3 Wallstreet (the latest model with serial ports).

Classic and 8100 (or now Wallstreet) can act as either master or slave, as the interfaces are also equipped with SMPTE.

Interact with the Hypercard cards as an “oracle” that generates events, responses, sounds, and DMX commands for RGB moving heads. All in sync.
An ambient-retro multimedia experience, where the audience enters a living operating system. Sounds, lights, and interactions merge into a sensory narrative guided by vintage technology and contemporary vision.

The individual pieces all work, but without this fucking cable, the "chain is broken".

[GaryN]

Wow. I am officially impressed with your concept and your ambition!

A couple of notes…

* It may not matter BUT I started using Opcode Sequencer (forerunner of Vision) on a Mac Plus a century or so ago. I am almost certain I used a version of OMS to talk to an early Opcode MIDI interface and later a Studio Three, then Four, then 5LX. I never used Apple MIDI Manager. That OMS version was on a floppy disk and ran fine on the 68000 Plus. This may not matter at all to your setup but I thought I should mention it. The issue may be that the HyperMIDI won't recognize OMS and not so much it won't run on a 68000(?)

** The reason for attaching the serial cable shield at one end only is to prevent a ground loop from occurring when connecting multiple devices together. It's usually NOT an issue with a simple computer-to-interface setup BUT with all of the stuff you're interconnecting here I would change my recommendation to connecting the shield on one end only. It's there to ground out EMI "airborne" interference and will work fine connected on one (probably best the computer) end only.

*** I'm just a little concerned about sync here. Both interfaces are MTC-capable and the MOTU especially is pretty versatile but I doubt they'll sync with SMPTE. The SMPTE-outs on both are intended for striping audio tape rather than synchronizing multiple units together. The Studio Four typically would receive SMPTE from tape, convert that to MIDI Time Code and feed that back to the computer over one of the two serial lines to allow Vision or such to slave to the tape machine. That does NOT work in reverse with a computer generating MTC and the interface converting that back to SMPTE so that anything can be a master clock. It would seem you will nonetheless need to generate time code from one source and distribute it to where it's needed.

Perhaps (and I hope) you already have a handle on how to make everything play nice together here…
Progress reports are encouraged!

Gary