Both the S-YXG50 and the SCVA are extremely close to hardware. Not sure where you get it from that the SCVA is not accurate enough. It's not *perfectly accurate*, but it's 99% there for all practical purposes, which is playing games and enjoying the music.
Sure, if you know them really well, you can pick them apart in A/B tests... However, if I just wake up in the morning then someone switches my hardware to the software emulator without telling me, I could use it happily for days without noticing there is a "problem"...
I've done an extensive test here on 46 DOS games, you can download the volume-matched FLAC audio recordings and hear the difference for yourself!
https://blog.johnnovak.net/2023/03/05/grand-m … -midi-showdown/
To me, the difference is there, but it's very very small, and I've been producing music for 30+ years. I'm sure for a normal person who just wants to play a few games these differences are completely insignificant, especially if they don't have very good audio equipment.
And yes, technically the Yamaha sample set of the S-YXG50 is closer to the MU80 (but the flute sample is waaaaay off; it's painful to listen to that damn flute sample in the S-YX50, e.g. check out the Azrael's Tear recording or the Larry 6 tunes). But ironically, the *aural* difference between the S-YXG50 and the MU80 is much greater than in the case of the Roland modules. The MU80 sounds deeper, richer, fuller, "more 3D", more spacious than the S-YX50, while the aural difference between the SCVA and the SC-55 is much less noticeable.
To me, IMHO, to my ears, etc 😉
Then if you compare them on a small desktop speaker or a $100 headphone, you can't tell them apart.
DOS: Soyo SY-5TF, MMX 200, 128MB, S3 Virge DX, ESS 1868F, AWE32, QWave, S2, McFly, SC-55, MU80, MP32L
Win98: Gigabyte K8VM800M, Athlon64 3200+, 512MB, Matrox G400, SB Live
WinXP: Gigabyte P31-DS3L, C2D 2.33 GHz, 2GB, GT 430, Audigy 4