Musical potential of the NEXT.

This section is for discussing everything about Next hardware and latest updates.
SamusDrake
Posts: 250
Joined: Mon Jun 26, 2017 10:11 pm
Contact:

Musical potential of the NEXT.

Postby SamusDrake » Thu Sep 28, 2017 3:14 pm

Now, I've probably got the wrong end of the stick, but with three AYs and an SID...what is the NEXT capable of compared to say...a Megadrive, SNES or an Amiga?

Would there be any demos at this time?

imprimis
Posts: 7
Joined: Fri Sep 22, 2017 9:51 pm

Re: Musical potential of the NEXT.

Postby imprimis » Thu Sep 28, 2017 9:49 pm

Until someone a little more clued up comes along I will say the little bit I do know.
The three AY chips and SID offer 12 channels of audio.

AY is square sound and noise which is generated noise rather than prerecorded instrument samples, which may sound like not a lot but you will have heard some of the great music demos on the Speccy, they are using just THREE channels only.
So imagine three demos playing at once each dealing with different sound waves while the SID chip offers true waveform sounds meaning more synth style bass lines and sampled instruments (from what I have gathered.
The Amiga offered 4 channels albeit used sample banks, the NES 5 split between square, triangle, noise and a channel barely used for tiny samples so it will at least trounce all over the NES.

Not sure about channel wise for the SNES and Megadrive but they use dedicated chips rather than 'computer noise' which is pretty much like sticking the sound chip from a keyboard in which is why they had their distinctive sound effects and I'm assuming they had a small reserve for samples loaded from the cartridge for voice samples and the odd dedicated instrument sample.

Anyway someone should be along to give a much better explanation than I, I just thought I'd fill in a bit of the basics from my limited knowledge to keep you tied over.

three of these
https://www.youtube.com/watch?v=9osyYMZNT2o
with a little bit of this
https://www.youtube.com/watch?v=8fusdhM082E
I think we're safe to say sound is most definitely NEXT worthy :)

SamusDrake
Posts: 250
Joined: Mon Jun 26, 2017 10:11 pm
Contact:

Re: Musical potential of the NEXT.

Postby SamusDrake » Fri Sep 29, 2017 6:35 pm

Thanks for the info, matey. ^_^

User avatar
varmfskii
Posts: 137
Joined: Fri Jun 23, 2017 1:13 pm
Location: Albuquerque, NM USA

Re: Musical potential of the NEXT.

Postby varmfskii » Fri Sep 29, 2017 9:41 pm

The details aren't exactly right.

The SID only allows one to chose from 4 fixed waveforms: square wave (the same as the AY), triangle wave, saw-tooth wave, and noise.

Amiga audio uses 4 8-bit D/A converters, 2 each on the left and right channels. On each channel the output of the two D/A converters are effectively multiplied (one is the volume and the other is the audio signal) by carefully using them you can get the effect of 14-bit audio. The voice/channel description often used with sound chips likel yhe AY-3-8912 and SID don't really apply to D/A audio.

If you are looking for high quality stand alone audio for the ZX Next, there is also an 8-bit D/A which will be CPU intensive to use well.
Backer #2741 - TS2068, Byte, ZX Evolution

PiyoTaro
Posts: 111
Joined: Thu Jun 01, 2017 11:13 am

Re: Musical potential of the NEXT.

Postby PiyoTaro » Mon Oct 02, 2017 9:05 am

varmfskii wrote:
Fri Sep 29, 2017 9:41 pm
The details aren't exactly right.
The SID only allows one to chose from 4 fixed waveforms: square wave (the same as the AY), triangle wave, saw-tooth wave, and noise.

Amiga audio uses 4 8-bit D/A converters, 2 each on the left and right channels. On each channel the output of the two D/A converters are effectively multiplied (one is the volume and the other is the audio signal) by carefully using them you can get the effect of 14-bit audio. The voice/channel description often used with sound chips likel yhe AY-3-8912 and SID don't really apply to D/A audio.

If you are looking for high quality stand alone audio for the ZX Next, there is also an 8-bit D/A which will be CPU intensive to use well.
In order to play uncompressed PCM sound source, continuous memory space is required. Most ADPCM/PCM sound chip has its own memory bus.
If SpectrumNext 's DMA can directly access the address of the expanded memory, I think that I can do something like PCM sound source playback.

However, in the CPU of this machine, in the scroll game which rewrites the graphics of Layer 2 with high frequency, is there any computational power to do things like PCM sound source playback?

PiyoTaro
Posts: 111
Joined: Thu Jun 01, 2017 11:13 am

Re: Musical potential of the NEXT.

Postby PiyoTaro » Mon Oct 02, 2017 9:08 am

SamusDrake wrote:
Thu Sep 28, 2017 3:14 pm
Now, I've probably got the wrong end of the stick, but with three AYs and an SID...what is the NEXT capable of compared to say...a Megadrive, SNES or an Amiga?

Would there be any demos at this time?
1.
"ZXSpectrum 128" can play MIDI instruments in the MML macro language of the "PLAY" command of the BASIC language.

In order to create "game BGM" in a new environment, you need driver software to play music.
On the SID 6581 chip and Yamaha's FM tone generator chip, key on / key off commands like MIDI instruments are activated to generate sound.(*To put it more precisely, in the SID, the GATE bit of the control register is raised)
However, to generate sound, PSG tone generator chips such as AY need to simulate sound by software such as volume attenuation and pitch fluctuation.

2.
In the past, I have posted on "What chips are adopted for FM tone generator chip?"
In "TURBOSOUND-FM"(and Masakrator FM?), two OPN (YM2203) are adopted instead of AY. So I imagined a sound chip of Sega megadrive.

I think that it is up to implementing the new function on the emulator and then making it into FPGA.
I want the emulator to have a simulation DLL for the tone generator chip of OPM (YM2151) or OPN (YM2203, YM3438, YM2608).

Additional notes
Mr.Yuzo Koshiro is famous as a musician who makes game music with FM sound chip.
‘Etrian_Odyssey_Battle_Theme_2 (OPN,OPNA)’ by Yuzo_Koshiro(Retro_Style) on #SoundCloud? #np https://soundcloud.com/yuzo_koshiro/etr ... soundcloud
‘Flash_Flash_Flash (OPNA)’ by Yuzo_Koshiro(Retro_Style) on #SoundCloud? #np https://soundcloud.com/yuzo_koshiro/fla ... soundcloud
Last edited by PiyoTaro on Fri Oct 06, 2017 5:58 am, edited 1 time in total.

User avatar
varmfskii
Posts: 137
Joined: Fri Jun 23, 2017 1:13 pm
Location: Albuquerque, NM USA

Re: Musical potential of the NEXT.

Postby varmfskii » Mon Oct 02, 2017 4:16 pm

PiyoTaro
Depends on the game. I suspect that most audio in games will be done using the sound chips. It is easier, portrays the classic 8-bit sound, and doesn't consume that much of the CPU. If the dma can be set up for a specific rate, it could handle the audio, if not it isn't helpful for this particular job. The hardest thing is not shoving the data, but rather the timing. If you are using the d/a audio, you want to send a value every 2-4 scan lines. For this you can either count cycles (great for demos or pure audio, not so much for most applications in games) or handle an interrupt every few lines (inefficient).
Backer #2741 - TS2068, Byte, ZX Evolution

PiyoTaro
Posts: 111
Joined: Thu Jun 01, 2017 11:13 am

Re: Musical potential of the NEXT.

Postby PiyoTaro » Mon Oct 02, 2017 4:52 pm

varmfskii wrote:
Mon Oct 02, 2017 4:16 pm
PiyoTaro
Depends on the game. I suspect that most audio in games will be done using the sound chips. It is easier, portrays the classic 8-bit sound, and doesn't consume that much of the CPU. If the dma can be set up for a specific rate, it could handle the audio, if not it isn't helpful for this particular job. The hardest thing is not shoving the data, but rather the timing. If you are using the d/a audio, you want to send a value every 2-4 scan lines. For this you can either count cycles (great for demos or pure audio, not so much for most applications in games) or handle an interrupt every few lines (inefficient).
Indeed, if we consider the PSG tone generator chip to be a "4-bit DAC", I think that it is possible to reproduce 2 seconds with the sound quality of "4 bits 32 KHz" with the 8 KB data of the MMU 1 bank.
By the way, "SpectrumNext" also has "8 bit DAC" output. For background playback of PCM sound, there is a demonstration that places a buffer on the Z80 address bus and plays it with DMA.
I think that it is not suitable for "the new hardware that will appear in the future" how to use the scan line interrupt frequently.

By the way, I am thinking of PCM sound to sound a short time tone of "gate snare drum" and "orchestra hit", a long time tone such as cymbal and explosion sound. How about the sound quality?

seedy1812
Posts: 51
Joined: Tue May 30, 2017 11:31 am

Re: Musical potential of the NEXT.

Postby seedy1812 » Mon Oct 02, 2017 5:34 pm

The 4 bit DAC is logarithmic so you would need specially prepared 4 bit samples. I think the scan line interrupt only occurs where the screen gets drawn so you can't use value of 192 ( would like to be proved ) so that would rule out changing the pitch etc ( digi drums ) every N scan lines

User avatar
varmfskii
Posts: 137
Joined: Fri Jun 23, 2017 1:13 pm
Location: Albuquerque, NM USA

Re: Musical potential of the NEXT.

Postby varmfskii » Mon Oct 02, 2017 6:45 pm

4 bits at 32kHz is roughly 0.5 sec for an 8k bank. 4 bits at 8kHz (which should be sufficient for most purposes, accurate up to about 4kHz, or near the top end of a piano) is 2 sec. CITT Delta encoding is better than 4 bit ulaw (logarithmic), and the next uses 9-bits for the line number and the line interrupt value so is should work past the 192 rows for your normal screen and even beyond the 256 rows for sprites.
Backer #2741 - TS2068, Byte, ZX Evolution


Who is online

Users browsing this forum: No registered users and 1 guest