Two cores, one board...

[/tilted/]

So, I plan to do a SEQ and FM on one board, and as much SMD as possible (cause I'm a sucker for punishment)...

I figure that I can connect pin TX of SEQ direct to pin RX of FM, no need for opto, as there are no "potential" problems.

OK, so, if I also want a duplicate MIDI OUT from the SEQ without needing to do any IIC tinkering, can I simply do this:

[tt]

(SEQ)TX]---+--------[RX (FM)

          |

          +-/\/\/--[Midi Out

              220R

[/tt]

Problems?

[moxi]

TK suggested in another post to always use an optocoupler when analog audio circuitry is mixed with digital one (it's your case), I don't remember exactly why, but do a search...

[Pascal]

At the MIDI-Merger Shematic, you can see that a Optocopler is not a must when the VS from the same Powersuply. (See the Shematics and the Merge to Core Pics)

The MIDI-Splitting would i do with a 74HC14 (LS14 etc.) as a thrue.

[Wilba]

No problems with that design... This is exactly how MB-SID V1 worked, where up to three slave PICs were controlled from a master PIC with direct Tx->Rx connections, plus the MIDI Out via 220 ohm resistors.

However, I would suggest headers and jumpers that lets you switch the MIDI In/Out connections between the two PICs, i.e. some way of connecting each PIC directly to MIDI In/Out sockets (via an optocoupler). This will make uploading new firmware easier.

Also, have you considered how you would deal with separating SysEx dumps (for SEQ patterns and FM patches) to/from each PIC? Maybe a couple of SPDT switches near the MIDI In/Out ports are an option.