did anyone say beeper?

utz · April 2011

some people where asking for a beeper demo... well i don't know anything about demo making so for now i've put together this musicdisk with some old tracks of mine from last year.

http://milkcrate.com.au/berlinhq/utztape1.zip

WARNING: i don't know how to program or how to make cool graphics, and the music could be better, too. so don't take this too seriously. better stuff coming up in the future, i hope ;)

Mister Beep · April 2011

utz wrote: »

some people where asking for a beeper demo... well i don't know anything about demo making so for now i've put together this musicdisk with some old tracks of mine from last year.
http://milkcrate.com.au/berlinhq/utztape1.zip
WARNING: i don't know how to program or how to make cool graphics, and the music could be better, too. so don't take this too seriously. better stuff coming up in the future, i hope ;)

Aaah, so it is you, IP. I didn't know, heh. ;)
Quite nice job, as for the first time. I mean the BASIC code and the graphic. Seems like you found attributes pretty easy to bridle. ;)
Anyway, I like those songs the most: 7, 1, 3 plus some of your sound experiments with hi-speed songs, like - 2, 4 and 5.
You have to remember that SpecialFX and Orfeus have problems with very high octaves, you know that. ;)
Thanks a lot for the program.

FrankT · April 2011

This frequency table does a much better job with higher octaves on the SFX Engine.

	db	255,241,227,214,202,191,180,170,161,152,143,135
	db	128,120,114,107,101, 96, 90, 85, 80, 76, 72, 68
	db	 64, 60, 57, 54, 51, 48, 45, 43, 40, 38, 36, 34
	db	 32, 30, 28, 27, 25, 24, 23, 21, 20, 19, 18, 17
	db	 16, 15, 14, 13, 13, 12

To use it, replace the section FREQ_TABLE in the compiled asm song output with this.
A lot of the very high notes are in perfect tune. The lowest octave is less in tune than the original, but its less noticeable on low notes. So I thought this a good alternative.

Edit:
Or it is an option in my SFX Song Arranger Demo maker.
http://dl.dropbox.com/u/9498358/sfxsongarranger_alpha.zip
Instructions are in the readme.txt

Arda · April 2011

utz wrote: »

some people where asking for a beeper demo... well i don't know anything about demo making so for now i've put together this musicdisk with some old tracks of mine from last year.

Hmm, I like them. Especially amelie's soundtrack sounds like Ay. Thank you.

utz · April 2011

thank you for checking it out!

@FrankT: oh cool, i've been meaning to ask about your modified frequency table! thanx alot for posting this stuff! btw when's wotw2011 coming?

@mister beep: hehehe yup it's me :D

yeah there's a lot of detuning going on in this demo. remember these are all old songs - back then i didn't give too much thought about accurate pitch ;)

a note about the 1st song - if you listen on real hardware you can hear a wrong note in the beginning (off by 1 octave). interestingly no emulator renders it that way, so here we have another proof that such thing as accurate beeper emulation doesn't exist.

also my apologies for the 6th song... however to my defence i can say it was made with Music Synth :D

btw no one found the easter egg yet? ;)

Woody · April 2011

utz wrote: »

a note about the 1st song - if you listen on real hardware you can hear a wrong note in the beginning (off by 1 octave). interestingly no emulator renders it that way, so here we have another proof that such thing as accurate beeper emulation doesn't exist.

Maybe because they're all sloppy coders?

utz · April 2011

now i wouldn't say that :) i think the problem is rather that the beeper circuitry produces some unique hardware effects (namely reverb) which haven't been researched sufficiently yet. some high precision measuring is needed - i'd do it but my audio equipment is too crappy to get accurate results.

utz · May 2011

and #2 is done. enjoy.

http://milkcrate.com.au/berlinhq/utztape2.zip

Mister Beep · May 2011

utz wrote: »

enjoy.

I did. :)

kphair · June 2011

Woody wrote: »

Maybe because they're all sloppy coders?

Or more possibly because they lose the higher-frequency harmonics and edge resolution because they use a fixed sampling rate of 22 or 44 KHz to play the beeper data. What you end up with is a smoothed-out waveform with a maximum frequency of 11 or 22 KHz. Fine for pure tones, but where the frequencies are changed so quickly the limitations become apparent. Perhaps some speaker modelling is needed in the next generation of emulators?

I think I'll try this beeper demo on a DOS machine in work where the emulator will drive the PC speaker directly.

Fred · July 2011

kphair wrote: »

Or more possibly because they lose the higher-frequency harmonics and edge resolution because they use a fixed sampling rate of 22 or 44 KHz to play the beeper data.

For the record they don't all do that (in fact I suspect the majors mostly don't).

kphair · July 2011

Fred wrote: »

For the record they don't all do that (in fact I suspect the majors mostly don't).

Just to clarify in case you thought I was implying that the emulators were quantising the beeper. I don't believe they are - What I meant was that at the output stage the beeper data has to be resampled to the fixed playback rate of the output device so, when using PWM or frequency hopping effects there's naturally going to be some fidelity lost. It's just a consequence of the PC hardware. There's nothing we can do about it since direct access to the sound card's DAC is a thing of the past.

Fred · July 2011

kphair wrote: »

Just to clarify in case you thought I was implying that the emulators were quantising the beeper. I don't believe they are - What I meant was that at the output stage the beeper data has to be resampled to the fixed playback rate of the output device so, when using PWM or frequency hopping effects there's naturally going to be some fidelity lost. It's just a consequence of the PC hardware. There's nothing we can do about it since direct access to the sound card's DAC is a thing of the past.

Fuse at least attempts to maintain the harmonics through band-limited synthesis, see http://www.slack.net/~ant/bl-synth/ for more details.

utz · July 2011

kphair wrote: »

I think I'll try this beeper demo on a DOS machine in work where the emulator will drive the PC speaker directly.

please do let me know if you're successful with that, as it would be a good possibility to make polyphonic music on pc speaker. unlike with our beloved speccy there are only a handful of tools to synthesize music on pc speaker (read: live synthesis, not digi playback), and most of them are crap.

btw you are absolutely right about insufficient sample rates on pc. a soundcard with a resolution of 44 khz is naturally unable to reproduce frequencies above 22 khz. accurate beeper emulation would require about 10 times that.

did anyone say beeper?

Comments