I already tested driver + the cypress dev board, flashed with the right firmware,
and the printer port driver options looks good :

This is on windows 7 64 bit.

Once my board prototype is ready (in a few days, waiting for components but I have the pcb's),
I will be able to test if it works for our purpose.

Are you a messenger from the computer gods?

Sounds like a very interesting concept, and the driver looks quite good. The speed of the USB does seem like it might be a concern, though mostly for the games that do sampled sound using the Adlib or such. I think there was a post on the OPL2LPT thread with someone benchmarking their parallel port in the order of 2us, and the Adlib needing ~3us + ~23us delay between writes. Perhaps a solution would be to pack multiple bytes into each micro-frame, each a division of the micro-frame timing?

I assembled the first prototype, see photo :

It seems to work, basically, the printer port is detected, driver installed,
and I when I set one of the data lines high, it effectively goes high.
A quick test with adlipt driver in virtualbox did not work,
this needs more investigation.

Nice to hear someone's working on this once again, I did a ton of research into the original on which it was based as a solution for a client back in the day, which inspired this post in 2008: http://forums.xkcd.com/viewtopic.php?p=944602#p944602

[EDIT] Seems I already told you about it tho: Re: CVX USB Parallel Port experiments 😁

Stiletto, great, do you have a similar device to test with ?

dreamblaster wrote:

Stiletto, great, do you have a similar device to test with ?

Unfortunately I do not. A decade ago I had all sorts of weird LPT dongles for commercial apps and things still, but that time has passed.

I might have a parallel scanner kicking around.

did you ever use such USB2LPT device ?
(especially the high speed version? )

dreamblaster wrote:

did you ever use such USB2LPT device ?
(especially the high speed version? )

No, but I consulted for a client and suggested they purchase one for their use case, many years ago.
I don't actually know if they ever did or not. 😢

if you are interested in trying this one, pm me.

Any further development on this? I used to have somewhere a USB2LPT converter for use with a parallel port eprom programmer, but can't find it. I was looking at the original USB2LPT page (ver 1.6) https://www-user.tu-chemnitz.de/~heha/basteln … PT/ul-16.en.htm and it shouldn't be hard to build one from scratch even on a protoboard with an Atmega328 instead of the Atmega8 and a few other components. But I'd be up for buying a premade one too, especially a newer version that doesn't have any quirks to be worked around on x64 Windows. I've ordered a couple of the CY7C68013A-56 dev boards to try and make the hi-speed version (1.7).

Why can’t you cut out LPT interface completely? If the usb driver can map to 0x378 why wouldnt you just map to 0x388 or 0x220 directly?

The LPT interface is useful for other things too. Covox for instance.

stamasd wrote:

Any further development on this?

Not really, I made the prototype and it basically worked (setting outputs 1 by 1), but not with opl2lpt.
I could not find time to do more testing.

stamasd wrote:

I've ordered a couple of the CY7C68013A-56 dev boards to try and make the hi-speed version (1.7).

Yes my prototype is also based on the high speed version.
I will pm you, maybe we can join forces