• Please review our updated Terms and Rules here

ISA OPL2 Card

I personally hate hearing it when the audio dies down during breaks in game play. I'm sure the original game sound engineer probably hated it too but it was an artifact of most cards. A few beads and caps would eliminate much of what's left, but I suppose it boils down to personal preference.
 
Chuck(G) said:
I'm not a PC sound guy, so obviously I'm missing something.
Click to expand...

Well, I've seen people talking about the "special" sound of Adlib card. Or even wanting to get an original Adlib instead of using a SoundBlaster that has exactly the same OPL2 chip... Definitely there are differences in the audio stage in these cards, but I doubt they are significant enough for anyone to hear. Audiophile effect in action?! :)

(There might be some other valid reasons for using Adlib, for example if the program supports and autodetects Sound Blaster, but you just want the FM sound)

The original intention of my project was just building an OPL2 card for myself, mostly for fun (you know, I build some other early IBM PC like things too). So now I can say that I build my own sound card. Another important thing is that this card doesn't require PnP (most of my other ISA sound cards do), and it doesn't use -5V as my SoundBlaster 2.0 does (I don't have -5V on my small PS).

But since I've already done that, why not to offer this project to others?
 

Attachments

  • b9639c9eb38c4c38a1d5a184e96a4b16_измен.размер.jpg
    b9639c9eb38c4c38a1d5a184e96a4b16_измен.размер.jpg
    29.9 KB · Views: 2
Wow, look at that russian ICs! They look cool ;)

BTW, those tracks are awesome... I can't understand why games didn't had so great tunes
 
Last edited:
Sorry to drag up a thread from so long ago; but I figured it's better than creating a new one.

I wanted an OPL2 card for my XT clone as I used to have a genuine Adlib back in the day (which I lost...) and I see that they are now selling for ridiculous prices of $200+. Rather than spend a ridiculous amount of money I downloaded the gerbers from Sergey's site, sent them off to OSHpark who fabricated 3 boards (which is the minimum quantity) and ordered all the parts from element14 and Mouser.

OSHpark did a great job with the board, all that was required is to file down a few breakaway points and also slightly round / chamfer the edges of the ISA tongue so it slides easily into the slot. Don't file too much as the tolerances are very tight and you do not want to misalign the 'fingers'.

Everything assembled easily as Sergey has provided a good BOM and numbered all the components logically and sequentially. The only difficult part was it uses a 270pF ceramic capacitor and I forgot to order one (I thought I had one in a box of assorted ceramics). I ended up 'making one' using 3 different ceramics in parallel which works fine.

The only issue I ran into was R13 and R4 have had their labels transposed. R13 on the component list shows as 10k, whereas on the board it has '2.2k' printed on it. Likewise R4 on the component list is 2.2k, whereas on the board it is '10k'. A quick check of the schematics shows R13 and R4 form a voltage divider. From there it was easy to determine that the *printed values* on the board were correct, however the *component label* is not. As such I installed a 2.2k into R13 and a 10k into R4.

One question I do have is why does the OPL2/Adlib have jumpers for IRQ? From what I have read they operate by using reads/writes to a port (388h or something); does the board actually use the IRQ for anything?

Thanks again to Sergey for this project. The sound quality is fantastic, even better than some of my older soundblasters (they probably need recapping). If your in the market for a OPL2 board this is definitely the way to go.
 
3pcedev said:
The only issue I ran into was R13 and R4 have had their labels transposed. R13 on the component list shows as 10k, whereas on the board it has '2.2k' printed on it. Likewise R4 on the component list is 2.2k, whereas on the board it is '10k'. A quick check of the schematics shows R13 and R4 form a voltage divider. From there it was easy to determine that the *printed values* on the board were correct, however the *component label* is not. As such I installed a 2.2k into R13 and a 10k into R4.
Click to expand...

Good catch, strangely enough nobody had reported it before. I hope most of the people followed the printed values and not the component reference labels. Got to check my own boards :)

I've updated the PCB silkscreen, and I'll upload the updated file to my site shortly.

One question I do have is why does the OPL2/Adlib have jumpers for IRQ? From what I have read they operate by using reads/writes to a port (388h or something); does the board actually use the IRQ for anything?
Click to expand...

The OPL2 (and the OPL3) have two built-in timers that can generate periodic interrupts. The INT output of the OPL2 is connected to one side of the header. So theoretically if the header is soldered and a jumper is installed, that could be used to generate a periodic IRQ. Yet, the original Adlib came without the installed header, and I am not aware of any application that would use this functionality.
 
3pcedev said:
The only issue I ran into was R13 and R4 have had their labels transposed. R13 on the component list shows as 10k, whereas on the board it has '2.2k' printed on it. Likewise R4 on the component list is 2.2k, whereas on the board it is '10k'. A quick check of the schematics shows R13 and R4 form a voltage divider. From there it was easy to determine that the *printed values* on the board were correct, however the *component label* is not. As such I installed a 2.2k into R13 and a 10k into R4.
Click to expand...

Didn't notice this when I built mine... I guess I got lucky that I was following the printed values rather than the component names!
 
@eeguru I'd be interested in what exactly you'd recommend to remove noise. I don't care about the "vintage sound", I want to hear it the way the sound engineers would have wanted me to hear it, sans noise.
 
You must log in or register to reply here.
Back
Top