oh wow, now I have to get a pcb when you sell them shock__ , even though I don't have a real covox 🤣

Can you also post schematics (with componen values too) if you want to have a review?

I just want to see how the FULL signal handling is now. If one of them is full, the other one should also be full, or they are one or more samples out of sync and this is terrible anyway. Therefore I think it is sufficient for the full signal to come out of one chip only, if it simplifies anything. What do you think?

What would be really great is if you could make one that is toggle switch selectable between Covox and Disney compatibility instead of having to make a daughter board to go with it.

Jepael wrote:

Can you also post schematics (with componen values too) if you want to have a review?

That would be very welcome 😀

Jepael wrote:

I just want to see how the FULL signal handling is now. If one of them is full, the other one should also be full, or they are one or more samples out of sync and this is terrible anyway. Therefore I think it is sufficient for the full signal to come out of one chip only, if it simplifies anything. What do you think?

I agree a NOR or just a single input would be either more safe or sufficient, going with NAND for now just to make sure they remain in sync.

sliderider wrote:

What would be really great is if you could make one that is toggle switch selectable between Covox and Disney compatibility instead of having to make a daughter board to go with it.

It's just a loose design for now (like I said, it's intended to get integrated into a full featured mono/stereo-on-1/stereo-on-2 board in the end). Making a standalone mono covox/dss switchable plug shouldn't be an issue (but would increase the pcb footprint considerably [adding at least one 8bit latch]).

keropi wrote:

oh wow, now I have to get a pcb when you sell them shock__ , even though I don't have a real covox 🤣

All glory to crapvox

(cheapass R2R Covox made on breadboard ... looking not so fabulous)

Also not sure, if I'll sell anything - I prefer to release schematics/layouts so people can still find the files in 2 years and get their own. Also don't enjoy distributing stuff to much. If the design remains as it is, I may have 3 spare prototype boards I could give away for cheap tho. If I will even produce those ... a very first prototype will be made on breadboard with loose wires and stuff 😁

Also some old redrawing/measured/reversed stuff based upon the original dongle I once had:
http://i.imgur.com/sMTdFhD.jpg (500kb)

So the parallel port pin is low only if both FIFOs have room, and if either FIFO goes full (output low), it turns off either transistor, and this lets parallel port pin high. Sounds reasonable and can easily be prototyped if the indication is wanted from only one FIFO.

One final note, each IC should have preferably a 100nF ceramic bypass capacitor over its power pins, and somewhere on the PCB there should be one larger bulk capacitor like 10uF electrolytic on the supply pins. Just saying, as it seems the dongle has no capacitors whatsoever on power supply pins. Otherwise, it looks fine to me.

Still waiting for the parts to arrive (pre-ordering for pickup seemed to have been a stupid choice)

But I got bored today, enjoy an awesome 90s rendering of the emulator and a covox plug rotating ... in Dolby GIF (where available - warning 2MB file)
http://i.imgur.com/8tfrQQ0.gif

Got the missing parts today, wired up a hastly prototype on breadboard, adjusted the variable resistor by some guesswork (was too lazy/excited to measure) and on the second try ... SUCCESS

http://i.imgur.com/rsttAeY.jpg
Now off to wire the actual data lines and the connector to the Covox.
(yes, that's a Thinkpad T20 without a screen connected to a TV)

^ awesome as always

Good news everyone!

After changing the LPT ports adress to a more standard one (0x378 instead of 0x3bc) and changing the mode from "bi-directional" to EPP - I can now confirm the sound in Wolf3d as working 😀
Guten Tag!

NiteRaid & Duke3D (which actually does quite a lot [mixing various channels, effects + digis, volume control]) confirmed working.

EDIT: ModPlay Pro 2.19 confirmed working (even tho 4 channel mods at 7kHz don't sound too good)

Congratulations, great stuff, good to know it works! I always considered doing DSS emulation with a modern MCU or even CPLD or FPGA, but I never bothered to think off-the-shelf TTL logic FIFOs could do the trick.

Thanks 😀

Going with a MCU might actually be a better approach, considering the 74ALS232 bills in at $5-$9 each these days making the clone as it is not very cost effective.
Seeing how i.e. the ATMega8 costs ~$2 and supports 2x8bit I/O along with various control pins it might probably be a better option.

Cleaning up the schematic and writing a bit of documentation for now and planning to release them along with an example layout later on.
Or would anyone actually be interested in actual PCBs (I'm gonna stick with my slightly chaotic prototype for now)? I could do a small batch if there's interest.

EDIT: Project files attached.

Someone just poked me into the direction that the expensive 74ALS232 chips might actually be substituted with CD40105 chips (which seem almost identical, but might need inverting on some pins).
If that works that would be awesome - lowering the price from ~$20 to ~$5 per clone.

I'll keep you updated.

Would anyone of you have the Windows95 driver for the Sound Source by chance?
I can only seem to find the Windows 3.11 one for some reason. Would like to test if my circuit works with Windows (don't have a Win3.11 system at hand).

here ya go 😊

Great news everyone! Using cheap and available CD40105 FIFOs ($0.50 instead of $6) works as well 😀 Just needs an additional inverter.

wow excellent!!!

The Windows driver seems to be a piece of crap tho.
Bluescreens after every sound. But hey ... using a 5 minute long startup sound I could play back a converted MP3 😉

Pretty much as shown here:
https://www.youtube.com/watch?v=1OKNKwPQ3Gg

hmmm maybe the one I have is crap , doesn't look all that official to me :\
will search around and see if I can find anything else 😊

edit: bah, I only saw a couple of comments on how the driver is unstable... maybe there is nothing more 😢

Nah, I think the original driver is pretty much the same crap 😉 And pretty much intended for Win3.11 anyways.
Seeing how the guy in the youtube video has the same issue I can at least strongly assume it's not the hardware's fault.

Adapting the schematic/PCB layout to use the CD40105 chips now, over 3 forums combined I have 4 persons interested in PCBs now.

yeah, "unstable" are the comments I get from searching too...
you better run a batch of them 😁