What's new


Staff member
Jun 12, 2015
Reaction score
I've been thinking if people are really satisfied with the FPGA implementation on the CPS1. It's a digital implementation that runs code and transforms sound data. The data is the same and the code run is the same, so actually it does the same as the Qsound chip, but it's in an FPGA. There is an option that I'd like to investigate and that's running the same code on a WE-DSP16A chip, which is the chip that hides behind Qsound, so people have a feeling more that it's like the original. What do you guys think? Would that be more compelling to you? Or it wouldnt make any difference at all?

Please let me know your thoughts.

These are DPS16A chips. Exact same chip, but they run different code.

WhatsApp Image 2023-12-08 at 14.28.25(2).jpeg
WhatsApp Image 2023-12-08 at 14.28.25(1).jpeg
WhatsApp Image 2023-12-08 at 14.28.25.jpeg
I really like your work and the fact that you can provide alternatives to the same issue, we should be really grateful for that.
And as to me I would like both you know, if there is pcb mount option we users can have the original replacement and the replacement that will last longer in time, and we can decide wich we like better and in this hobby that's the best part of the fun.

Edit: sorry it doesn't help you determine one way or the other but I'm sure the forum will speak for the best option.
Last edited:
Is it possible to make a bracket that is similar to the "CPU" bracket for computer. This way, whatever is available at the time or if one find one what they like in the future can switch out. If possible, that would be an awesome solution. :)
I don't think this would make a difference to anybody who cares about the existing original vs FPGA qsound implementations. It certainly wouldn't get me to change up my CPS1 multi setup.
Thanks for the info guys. My intention was to know if people would buy one only if it had an original QSound chip or the original chip with memory running externally. As I said, there is no difference as it's a processor executing code based on a clock. Keep posting please. I want to know what others think.
I’d personally opt for the QSound option even if no noticeable difference can be heard.

I know there’s issues sourcing chips but think I suggested a BYO approach to any potential buyer in previous discussions in the CPS1 multi thread.

(There’s plenty of cheap CPS2 or dead ZN1/ZN1 boards out there to take QSound chips from 😂)
I'm selling empty sound modules with a socket to put an original Qsound. That is an option that many people can use, but they need to source their own Qsound chip. To me in this specific case there are no differences, because it's a single chip. It took a couple of iterations to take the code of this specific chip and put it into an FPGA, but now is like the original IMO.
For context, when I originally bought my CPS1 multi, I wanted the Q-sound chip version. However, this is still a 30+ year old chip and things do become internally disconnected over time. So, I bought the FPGA sound board as a backup. Could have saved myself a little money and just gone that route to begin with.

There is not really a strong argument for an original chip on a multi if your concern is originality. Yes, I kept all my original CPS1/1.5 boards after the multi dropped, but I can't honestly see any reason to actually plug them in now.
No difference. Usually, in an arcade the speakers are shit anyway. Maybe at home with a 100k sound system it is different.