Yes, unfortunately it seems quite difficult to get the original OS.
The interesting thing is that supposedly it’s an “open distribution” OS – yet it cannot be found anywhere, not even on torrent sites.
Anyway, I upgraded a CPU, and surprisingly it runs very well under Windows 10.
I can run Roland Canvas, vArranger2, and Visual Studio 2022 at the same time without problems.
I am writing my own arranger software for it, and I have already documented the commands for the potentiometers, buttons, and LED control.
The internal CRC-based communication gave me quite a challenge, and decoding the potentiometer states was also a pretty tough job.
The state of a given potentiometer was not read with a normal ADC, but instead I received multiplexed values depending on its position.
Now I have reached the point where the essential part of the software is working. For example, vArranger2, Giglad, Halion 6 & 7, Korg Triton, Triton Extreme, Roland Canvas, etc. – basically almost everything runs smoothly, as long as the motherboard can handle it.
It’s true that I haven’t yet fully implemented the LED states of the buttons, but that’s just a detail.
If I really want to go further, the FATAR keyboard can also be programmed using an Arduino board.
That way I could remove the original big red board with the built-in factory sounds, DAC, etc., and replace it with a professional studio audio interface (although that’s already possible even now).
The original touch screen and video card can also be removed, since they are PCI-based, and I could install at least a 7-inch IPS touch screen in their place (which physically fits the original space). I already ordered one.
One thing that has to be ensured is a PCI UARTx4 + 1 LPT card, in case the new motherboard doesn’t support it. The good thing is that it’s Windows 10 compatible. If the new motherboard won’t accept it, then it’s enough to use 2 USB–UART adapters, since the 1xLPT and 2xUART are not absolutely required anymore.
The PC motherboard can also be swapped, so that’s not an obstacle either.
But on the other hand, I’m thinking it may not even be worth doing too many modifications, since in its current form it works stably, and programming it will keep me busy for a long time.
By the way, the original “factory” sound sets are not useless either. With just two wires I can connect one unused UART directly to the SAM sound module on the red board. It doesn’t even need a driver, because it works with simple MIDI messages (31250 baud). Then I just connect its audio output to the Line-In or CD input of a new PC sound card, and it becomes perfectly usable.
So there are many solutions. For me the top priority now is the software.
About the sound card: The original Lionstracs audio card couldn’t be used under Windows without Linux, because Windows requires a proper driver.
But the Windows driver itself also needs the correct EEPROM content on the card – currently it’s just a 24c02 chip that contains only an identifier, nothing else.
Under Linux this EEPROM content is not required, since the card can be configured directly.
Basically this is why I don’t want to let this instrument go: with relatively little money it can still be turned into a very powerful instrument, with a sound quality that I couldn’t afford otherwise.
While programming, it’s not very comfortable to sit in front of the instrument, so I usually program it remotely from a laptop using Remote Desktop. Much easier this way.
Still, I would like to find the original OS, because during development I would like to have it on a separate HDD – one for making music, and another for development.
I’m not really a professional VB.NET programmer – I’ve mostly programmed microcontrollers and designed circuits before – so I’m progressing a bit slowly. But I see this whole project as a useful challenge. Honestly, if the original OS had been available, I probably wouldn’t have even started all this reverse-engineering, and I wouldn’t have discovered all the possibilities inside.
After all, it has become a complete and very flexible platform.
If the OS had been available, I wouldn’t have started any of this. But I really didn’t want to throw away such a well-preserved instrument…
About the market value of the instrument:
I saw some ads where they try to sell these instruments at shockingly high prices.
But if you think about it, since the “open” OS is almost impossible to obtain, one hard disk failure can turn the whole thing into a useless doorstop.
If the front panel fails, e.g. one of the processors, it’s 99.8% unrepairable.
The same applies to the audio card if the EEPROM with the minimal data gets corrupted.
There is no service background, and I don’t know anyone who would dare to repair it knowing exactly how it works inside. (Of course, if I’m wrong here, please correct me – and if you know someone who can program new processors for it, I’d be very happy to have their contact, I may need that someday.)
Also, there aren’t too many VSTs available that can be installed on the original OS.
So, in its original state, I don’t really consider it a valuable instrument.
For the same money they ask for these, you could buy a decent used Yamaha or Roland – lighter, better-sounding, etc.
For someone who sees it as a development platform, it might still be worth buying cheap, for personal use.
Its value will increase only if you invest your own work, time, and knowledge into it. But again, there’s no service background…
So basically, its real value is only for the person who develops it further. On resale, nobody will pay for that invested work.
I think the downfall of Lionstracs was mostly because the technology was ahead of its time. Many musicians back then couldn’t even handle computers properly, especially the “old school” musicians. Probably at first glance they also considered it too complicated.
Another reason could be the lack of proper marketing.
Sorry if my post became a bit long – it just reflects my disappointment that something which was advertised as an open system has ended up with almost no technical service documentation or OS available. Even the user manual doesn’t cover everything.
Of course, this could start a debate – but I don’t think it would have much point.