Propably not an easy way.. or readily available solution.
But if you (or your company) has sample code for using the card I think it should be possible to program midi device driver for it. (More or less doin’ ripping on public domain source code and providing some glue code)
MIDIO128 provides TTL level outputs, triggered by (for example) midi notes, this is propably ok, for midi side?!
For interfacing to (I’m assuming) mains powered equipment there is an example, using relays. (Check MIDIO128 page) Proper relays are good for switching (and driving) heavily loaded ac lines. But as being a mechanical components, relays are kind of slow. Â
Another way is using TRIACs. I’m not sure how much load TRIACs can handle. I found couple of links for reference:
It should be even possible to connect MIDIO128 direct to that breakout box, both use TTL levels.
Both relay and TRIAC methods are okay normal lamps, I guess.
What I meant with that “midi controlled dimming with AOUT”, was that then standard 0-10v dimming equipment can be interfaced with midibox.. (It could be done with TRIACs, by means of pulsewidth modulating triacs triggering thus affecting its on time -> ac duty cycle.. but that is with extra programming efford.)
In the electrical world, what is the difference between digital signals and analog signals ? I assume that digital signals uses a pulse waveform to mimic the on/off paradigm, but this is just a guess.
But problems ahead: WinNT/W2k/XP ->No DOS, no debug and if my memory serves me right: NT based Windows prohibits any “user space” programs directly accessing peripherals.
But you can overcome this (and writing device driver ) by using special I/O driver, check:
This is a dll containing functions to acess IO ports from programs written in C, C++, Vb and other. This might help me if I want to write an object in C for Max/msp. I guess that object in C can link to that DLL. (If I udnerstand what a dll is)
The thing I wonder now is will that be possible using a PCI version of the card, the technical support guy told me that I needed to talk to a PCI abstraction layer or something like that before I can send bytes to the card.
Oh yes, I checked that package too.. Just thought that PortTalk could be easier to use, as there is 'bout complete C-example included. (Uses LTP1, but easily adapted to different I/O range)
That “talk to a PCI abstraction layer”.. well, I’m not sure, but I think this means using operating system, and so called memory mapped I/O.. (but these things are getting way out of my knowledge (as most of this) )
But even as it’s a PCI device, there is a possibility, that it can be used with port I/O. If you have that PCI card handy give it a try, with debug first–
Ok, checked it out, it’s I/O mapped. Here’s link for document explaining it’s registers, in other words, what bytes to write and where in it’s i/o space:
Thanks, so If I understand that PDF well, the PCI board as a chip on it that will allow register level programming ?  If that’s the case I think I’m in buisness  That means It would work under xp and with a PCI card using porttalk or the io.dll file.
Edit : From the pdf : “Only experienced programmers should attempt register-level programming”
That sounds funny to me, I’m not a programmer and sending bytes directly to the card seems way more simple to me