Pretty sure I've been pushing 32.7Hz out of a Game Boy for years now...
That is true. My mistake, should have checked it myself instead of relying on some false information.
chipmusic.org is an online community in respect and relation to chip music, art and its parallels.
You are not logged in. Please login or register.
ChipMusic.org / Forums / Posts by Tomy
Pretty sure I've been pushing 32.7Hz out of a Game Boy for years now...
That is true. My mistake, should have checked it myself instead of relying on some false information.
VGM Music Maker can also do GG (+ SMS & MD), but the program does have bugs left, some pretty critical when using NTSC speed (crashes, not working on hardware without fixes). As GG is NTSC only I'm not sure if this tracker would work at all with it, otherwise it's a very capable tool for the SN76489.
I want to point out the specs of the Master System / Game Gear audio:
- Channels 1-3 can only do 50% square waves, lowest frequency 109 Hz.
- Channel 4 by itself can do white noise using only three preset frequencies, notes A2 (110 Hz), A3 (220 Hz) & A4 (440 Hz). You can also change the white noise for a 6.25% square wave tone. That can only use those three frequencies as well.
- You can combine channel 3 & 4 together, CH4 will then use CH3 frequency data. This is where SN76489 gets interesting: The noise channel will become quite flexible being capable of playing 1024 different frequencies of white noise or 6.25% square waves.
Yes, the sound chip can do both 50% and 6.25% duty cycle square waves. It's just weirdly labeled as "periodic noise" everywhere.
If you do not input any frequency data to CH3 in the combined mode then CH4 will play a higher than normally achievable noise / tone. If you enter a note in CH3 then CH4 will change to play that same note and can't go back anymore. (Not 100% sure that you can't achieve the high frequency again, at least on the tracker you can't.)
It seems to me that the Game Boy can also do 6.25% squares using the Wave channel but the lowest possible frequency is 64 Hz?
Then SMS / GG outperforms that because it can play as deep as 6.8 Hz.
About the sound quality / difference I can say that SMS has much brighter sound than MD. The square tones seem shorter, more staccato on MD.
Now if the music doesn't use any special tricks, just the noise preset sounds then it will sound mostly the same between the systems. But there is a noticeable difference in volume balance when the periodic noise mode is in use. Particularly the thin noise bass & low white noise drums. Those are much louder on MD. A track like this will sound good on the system it was made for and bad on the other system.
https://tomy.bandcamp.com/album/fm-seri ut-sinulle
There was something.. Something I never told you about the FM cyber world.
~90 minutes of hardware recorded Mega Drive FM music. Expect eerie cyberpunk happiness.
Thank you Imaginary & ptesquad! I wasn't planning to do this release, but since I found new things from the SN76489 I knew I had to do it. So many hours spent tracking the music... But I'm very happy with the outcome. Exactly what I wanted to show - a different side of Master System music.
marcb0t, no problem. Just hit me with a message somewhere if you need any more hardware recordings, I can do SMS & MD (NTSC / PAL, shouldn't matter).
I tested this with MD1, MD2, SMS1 and SMS2. The MD's have the same balance between them (CCAM makes the MD2 PSG a bit brighter) and the SMS's have the same balance between them. The SMS PSG is super bright.
Haha, yeah your SuperPSG test version sounds wildly different. I'd definitely use it to make deeper tone drums in my music, that's one thing I miss.
Let's see.. I couldn't get the rom to work. But you had the .vgm file there, so I grabbed that and made another rom using Maxim's VGM Player. Works now. Let me know if something goes wrong, here's Tarzan Boy recorded from my SMS2 in .wav and .mp3 (I only boosted the volume):
https://www.dropbox.com/s/k5hatuod93utx
2.wav?dl=0
https://www.dropbox.com/s/2026ge9nw38cl
2.mp3?dl=0
I have now re-recorded the whole album from real SMS hardware.
Couple of things were found out the hard way while doing this. First of all, lower frequency noise isn't as audible on SMS than it is on MD. I have not found any emulation of SMS PSG to be accurate, either the noise balance is based on MD or the noise channel can't properly play the more advanced sounds. This means that I had to re-balance everything. The other thing is that also the PSG suffers from the usual pops/clicks, but emulation and MD (filtering?) mask those. You can't do too extreme frequency jumps. Just wanted to mention this information, so someone could learn from my mistakes. Check with hardware if you can. The good news is that the sound quality is very good, volume changing is smoother and instruments are "longer". It doesn't sound that obvious now that I'm interrupting sounds all over the place.
marcb0t, thank you once again for your awesome comment! About making the music lower in tone.. I actually use a trick in some of these tracks where I transpose everything below what is normally the lowest note in Master System with standard tuning, the note A. It can go little bit deeper than that, not enough for G# though. 5th Unit for example is done this way, also slightly lower frequency thanks to PAL format, it's about the lowest you can go with square tone bass. But it's painful to do, I have to transpose every single note by hand.
Very interested to hear more examples of that modulation technique, and most of all when hardware playback is possible. It's exciting to hear you're getting things working out in Deflemask too, I can't wait to hear more Master System music!
Thank you for the comments! Makes me happy to hear that there were some little surprises too. There are some things on the album that I wouldn't have believed to be possible. I got ideas that at first were like: "No way it would work, I'll just try it for fun"... And then the SN76489 actually pulled it off. This chip is really underused, not as bad as it's reputation seems.
marcb0t, many thanks for your kind words. It feels weird to hear that I've inspired someone with my weird music, but really, thanks! Trying to get the most out of any soundchip is about half the fun, I love to use synthesized sounds and create my own. One soundchip/synthesizer making every sound in a track is awesome. Great to hear you like my occasional dissonance, I've mostly heard negative feedback about that. That and the chromatic stuff is what I like too.
That is very cool that you're working with the PSG chip! The "DEMO Rob0t In Action" at your soundcloud is such a fun track. That square wave modulation is just killer, would be cool to hear more of that used on the hardware.
Hope that Series 4 doesn't disappoint you.
https://tomy.bandcamp.com/album/psg-series-4
Welcome back to PSG cyber world!
Please connect and be ready for the cold, harsh reality.
~90 minutes of hardware recorded Master System PSG music. I lied last time when I said I couldn't get anything more out of the SN76489..
This is more experimental, not everything is nice and clean as I really tried to push it far.
(I got better recording equipment too, so everything in my bandcamp site has been updated.)
I'm also very happy how Mind Of Machine turned out, if you're interested check out the original FM version from the FMS6 release. Thank you for your kind words, this really was the most difficult project out of all of my releases. Sometimes I felt like quitting and that 90 minutes (the album length I aim for) was unreachable. But at the same time I had to finish this to show what the Master System can do. I don't want it to be so underappreciated.
Thank you for the comments!
Chemical X is indeed the one using bass as a melodic lead. I thought I'd try it out, since the FM original features somewhat similar sounding lead. It was difficult to achieve as I needed the square to work out the bassline momentarily, and that had to be as low as possible. I also used another square to add some detune and phasing to it. Snare drum had to be a bit shorter as well as to not interrupt the noise channel too noticeably. In the end I'm happy that it barely holds together.
http://tomy.bandcamp.com/album/psg-series-3
Now is time to the SN76489 heart on fire!
~90 minutes of hardware recorded Master System music. This is something I worked extremely hard for and I'm so, so happy to finally release this monster project. It started when I was thinking how could I get the whole drum set working together with the noise bass. Bassdrums, snares, hihats and all. So.. I made a test and tracked at the highest speed possible. By using few frames for drums and switching quickly between the different instruments, it works!!
This is the closest I can push the SN76489 to it's limits, I just can't get anything more out of it. There's all the usual tricks and this time every single track has multi-channel drums. Includes various uses of the advanced noise mode. Even using the noise as a melodic lead, and to not make things too easy the noise goes out of tune at higher frequencies, so I had to tune it manually by myself..
Hope you enjoy!
That's great that you got it working!
Timing 2=6.
Yes, even rows have 2 frames. But that means only EC1 works, ECx has to be smaller than the number of frames, so EC2 won't have any effect.
And yes uneven rows have 6 frames, and you are correct in that EC1-5 do work.
I think I know why the ECx still doesn't work. Your speed is 1=4, that means that all even rows have 1 frame in them and all uneven rows have 4 frames. Let's take a closer look from the beginning with 1=4 for example:
00 | G-4 1F 01 ... | <-- 1 frame
01 | --- .. .. ... | <-- 4 frames
02 | G-4 .. .. ... | <-- 1 frame
03 | G-4 .. .. EC3 | <-- 4 frames
etc. etc.
In this case you can use ECx on those uneven rows with 4 frames, EC1, EC2 or EC3. With the even rows the *** cut off is the only option. If your speed would be 4=1 instead then the even rows have 4 and uneven 1. With a steady tempo like 4=4 all rows have 4 frames.
If your track is using 1=1 speed, then the ECx command doesn't do anything. You need to use the note cut effect which you place in the same place as the normal notes. For me the default key is "a". It would look like this in your example:
7C | G-4 .. 01 ... |
7D | --- .. .. ... |
7E | *** .. .. ... |
Oh, and check your Release Rate (RR) setting in your instrument, if it's set to 0 then the sound will keep playing forever after the note cut. The higher the RR, the more sharply the note will cut off. Keep in mind only the output operator RR matters in this case.
If you are tracking with any other speed than 1=1, then the ECx command is very useful. The x has to be a smaller number than the speed to work. For example a 5=5 speed means that every row has 5 frames in it. So if you put a EC4 there, it will cut off the note at the fourth frame inside the row. With this you can let the instruments play a little longer than with the "***" note cut, which will cut the note instantly at the beginning of a row.
YM2612 doesn't like changing certain notes / instruments straight into another without making a popping sound, but note cuts will eliminate most of that. I usually cut off a note just before the change. It will sound more "staccato" now but you can alter the RR to give it some time to fade out.
If you want realistic, it might be tricky. But for synthetic/80's sounding, FM will deliver. It's really nice to be able to change the parameters on the fly with FM, pitch-adjusting, tone changes, etc.
I don't use samples either, so if you want to check out what I've done with YM2612, for example this track: Partners In Crime. Check the intro and especially the outro for some tom drum fills.
4op algorithms. I think the best algorithms for making tom/snare drums out of those are: 4, 5 & 7. If the sound gets clicky, change the Attack Rate of the carrier/output operator(s) to fix that. Some of the best FM drums can be found from Hitoshi Sakimoto, TechnoSoft & the game Madou Monogatari I. Study the patches, see if they sound good "as-is" and/or add pitchbends to spice them up.
ChipMusic.org / Forums / Posts by Tomy