Well, I guess I might be right with what I am thinking of, in that Samples of the Genesis are stored normally in the ROM and just lowered in quality for use ingame via hardware. Or is that wrong?
I actually don't know much about Z80 and 68000, but what I do know is that if you play sound test FF on Sonic 3, it sounds like crap, compared to the startup.
_______________________
I have turned into a sausage.
I can provide you with sound examples.
First, the noise (sn-noise.wav)
DefleMask's highest noise frequency is played first, then the highest in VGM Music Maker is played.
Second, the weird fading wave channel on GameBoy (weird_fading.wav)
The volume of the wave channel seems to be affected by all other channels in DefleMask. In this demo, it is the noise. The track should not sound like this at all.
SN76489 noise has only 4 frequencies to use by default. The lowest 3 can be used while using Square 3 and the noise frequency will remain. The highest frequency is altered by all input in Square 3. After altering the noise with Square 3, there is no way to attain the highest possible frequency again, however, there is a way to "trick" the noise into going into a higher freqency than allowed after altering the pitch with Square 3 by using upwards pitch bend.
I'm not really sure what you're asking.. But SN76489 noise differs from 2A03 noise in that the former is just shit. Basically they have very little in common and you can't really make either one sound like the other.
With that being said, the best way to get somewhat decent sound out of the SN76489 noise is probably to use the highest possible frequency. But then you only get 2 squares to play with.
I can provide you with sound examples.
First, the noise (sn-noise.wav)
DefleMask's highest noise frequency is played first, then the highest in VGM Music Maker is played.
Second, the weird fading wave channel on GameBoy (weird_fading.wav)
The volume of the wave channel seems to be affected by all other channels in DefleMask. In this demo, it is the noise. The track should not sound like this at all.
Ohh, I see. Both bugs are fixed now, please update it to "Ver. 8 u8".
It's still incorrect. Do you have VGM MM by any chance? A visual representation would be easiest to demonstrate. I will attach a VGE file for reference.
This behavior all seems to work as the SN76489 is supposed to as far as I know, so I will explain what I did here. The first 4 notes are the noise alone. They do not need square 3 to function, and square 3 cannot alter the frequency of the first 3 notes. The highest possible frequency (represented as "CH3" ) is the only note affected by square 3, so I have created a blank instrument to alter the frequency after the first 4 notes. The 5th note is lowest frequency that the noise can go. 6th note is the highest possible AFTER altering with the square (and note now that the square has altered the pitch of the noise, it can no longer achieve the highest pitch of the "CH3" note that played on note 4). Now, last note, the 7th note is a trick to achieve a slightly higher frequency than just a CH3 and B-8 note combined by using quick pitch bend upwards (I have used this trick in a Crystal Snail cover I made and it has a better hat sound than just CH3 + B-8 alone).
I noticed, but I was demonstrating that because you have all notes as highest frequency and this behavior is incorrect. VGM MM behavor is mostly correct, so that is why I used it as an example.
Btw, you should try MAME emulation if you can. Afaik it is the most accurate =P
