That indeed doesn’t look like Yamaha chip just based on size and shape.

I do think it’s some kind of audio chip and that is a legit cheapo sound card. Based on the shape of the chip, it could be some Opti chip or even just Analog Devices Soundport/etc. Windows Sound System chip.

It is quite interesting that they bother with these kind of fakes, YMF-7xx sound cards aren’t that expensive after all.

Those tiny-flimsy PCBs of those sound cards look like something made with Crystal chips... Never seen such cards with YMF chips.

GigAHerZ wrote on 2023-03-10, 15:21:

Those tiny-flimsy PCBs of those sound cards look like something made with Crystal chips...

Found the Crystal card photo with exactly the same pcb. Yep, that's relabeled Crystal.

this seems weird
have you installed the card in a system? what do drivers say ? crystal or yamaha?
PCBs might look the same but the routing is what matters

This is the same thing as they did later with C-MEDIA (CMI) chips and called them Yamaha xg724 ( the chips where fake marked like this one , but the actual size was that of the CMI) . Even the cd driver was Yamaha branded , but the drivers from it were working on CMI cards and the ones from CMI cards were working on them , they only difference was how the device was listed in device manager . I had a card like this back in the day and took me a while to understand why it was not working with the original ymf724 drivers . Anyone knows this soundcard? .

So, first trying the UNISOUND on it for general detection. The card is found and configured. The report is shown as:

The string read from EEPROM is "[CDC4835] OPL3-SAx" is meant to convince inexperienced user he got the Yamaha.
But see the last line - "Crystal Mixer ...". It's Crystal all right, with edited EEPROM content, and this ugly FM synthesis of CX4235 (listening to Dyna Blaster).

For the sake of test completeness I've ran the Yamaha utilities:

Wow rare Crystal-SAx soundcard! Lol 🤣 🤣

Yeah, these are Crystal based cards. I have seen them on ebay few times and I could immediately tell they got the wrong pinout and connectivity but it happened to match that of Crystal parts which use I2C EEPROM over µW and single 16.9344MHz crystal rather than 33.8688+24.576MHz(although 24MHz one is missing on the lowest end cards, which produces some problems like wrong 32 and 48KHz sample rate).