Closed denjhang closed 3 years ago
Is there any document of this approach? I think I can study but I would like to avoid to invent the wheel and wanna save my time.
There is no related document. No one is currently conducting research on this. This VGM file was also discovered by accident. Before that, I always thought that FM could not emit white noise with variable frequencies.
I guess no more than 5 people are aware of this phenomenon.
I have investigated the vgm. First I thought it uses SSG-EG but it doen't. The noise is simply generated by using very very low frequency (fnum<90, blk=1 or 2). The voice parameter itself is not surprising, it may be well-known.
'@ N 000
AR DR SR RR SL TL KS ML DT AM SSGEG
'@ 031 000 000 015 015 000 000 000 000 000 000
'@ 031 000 000 015 015 000 000 001 000 000 000
'@ 031 000 009 015 015 000 000 000 000 000 000
'@ 031 000 000 015 015 000 000 000 000 000 000
AL FB
'@ 000 000
I think the parameter above can be used for AY noise n=31 or near, but I'm not sure if it is possible I think it is difficult to create an intermediate gradient for other noise frequencies.
I attach the test vgm for this noise on OPN2. opn_noise_test.vgm.zip
I did the same test with OPLL, but it seems that it can only generate white-like noise. opll_noise_test.vgm.zip
As for the noise of the middle gradient, please watch this video. I use multiple tones to superimpose white noise of the same frequency.It can also produce similar periodic noise.
https://www.youtube.com/watch?v=4c0PFs9nNEg https://www.youtube.com/watch?v=DWXR75rEbNA
I mean this is a well-known approach. not perfect but applicable for some music.
Please study this VGM file carefully. It uses YM2612 to generate variable frequency noise with a tone lower than C1, which is similar to PSG. I think this will help improve FM analog PSG. Although the VGM in the file uses OPN2, I actually think that the OPL series can also do this. Phobos.zip