Maker Pro
Maker Pro

My *Soundcard* MIDI Synth Dream

R

Radium

Jan 1, 1970
0
Hi:

Creative Music Synth -- my favorite *soundcard*-based MIDI synth-- is
the chip-based, hardware, digital FM synth present on the SoundBlaster
16 ISA card. Creative Music Synth usually has and I/O address of 220
-- hence, Creative Music Synth [220]. However, I have seen Creative
Music Synth set at 240 on some PCs. Creative Music Synth does not --
to any extent -- playback samples from memory. It freshly generates
its tones. It is physically built to generates its tones from scratch
without stealing any energy from the PC's main CPU. As such, Creative
Music Synth does not have or need any moving parts -- no discs, no
fans, etc.

Notes on out-of-phase stereo signals in Creative Music Synth:

1. I try playing Creative Music Synth, through my so called 'karaoke
voice canceller' -- which inverts the phase of one stereo channel
[right or left] and then combines it the other channel -- which
results in anything identical in both the left and right channels
being removed. I get a mono of what was different in the left and
right channels. When I play the Creative Music Synth audio through the
voice-canceller, it sounds more treble, sharper, brighter, warmer, and
crisper [qualities that I like] than when I don't use the voice
canceller. The waveforms with sawtooth-wave characteristics are phased
differently in the L and R channels, while those with sine-wave
characteristics are phased identically in the L and R channels.

2. The voice-cancellor remove whats phased similarly in L and R while
preserving what phased differently in the L and R channels. Sawtooth
waves tend to give a much brighter and warmer feel than sine waves do.
If anyone has a valid email address, I can email him/her two WMA audio
files from the MIDI audio of Creative Music Synth. File A is converted
to mono without inverting the phase of either channel. File B is
converted to mono after inverting the phase of the left channel. IOW,
file A is not "voice-cancelled", while file B is. After listening to
them, you certainly will notice the difference I describe.

3. The signals that are phased similarly in L and R are removed, while
signals phased differently are not removed. I am sure -- though I
could be wrong -- that Creative Labs deliberately made these phase
differences in the stereo Creative-Music-Synth. The sounds that are
phased similarly in L and R sound like sine-waves [such as a "whine"
or "hum"] while the tones that are phased differently in L and R sound
like sawtooth waves [such as a "buzz" or "fiz"]. AFAIK, thats just how
the company designed the synth.

Due to the above, my MIDI synth dream would be a mono, 64-bit-
resolution, 2.88-Ghz-sample-rate, 40,000-voice, 4,000-operators-per-
voice, 10,000-channeled**, version of Creative Music Synth based on
the signals that were phased differently in the original Creative
Music Synth. Creative Music Synth is a stereo FM synth who left and
right signals are phased differently, this is why I get a different
sound when I play it through voice-canceller. The signals of Creative
Music Synth that have the same phase for both L and R have a more
cheesy sine-wave quality which I don't care for. The signals in
Creative Music Synth that are phased differently in L and R tend to
resemble a fresher sawtooth-wave quality which I like. That is why I
want my version of Creative Music Synth to be based on the sounds that
were phased differently in the original Creative Music Synth. I want
my version to be monoaural because I want all speakers to give out the
same signal.

**Yamaha's OPL3 has 18 channels


Regards,

Radium
 
B

Bob Myers

Jan 1, 1970
0
And right about here is where the insanity kicks in...
Due to the above, my MIDI synth dream would be a mono, 64-bit-
resolution, 2.88-Ghz-sample-rate, 40,000-voice, 4,000-operators-per-
voice, 10,000-channeled**, version of Creative Music Synth based on
the signals that were phased differently in the original Creative
Music Synth.

You still have absolutely no idea WHY you want any of the above,
except for some completely absurd "bigger numbers are always
gonna be better!" assumption on your part. You don't even know
why the above numbers are silly to the point of causing hours and
hours of uproarious laughter on the part of anyone reading your
post.

For just one example - how many of the supposed "64 bits" in
the samples you're talking about would you expect to contain
valid and/or significant information? Why?

Bob M.
 
R

Rick Massey

Jan 1, 1970
0
Bob Myers said:
And right about here is where the insanity kicks in...


You still have absolutely no idea WHY you want any of the above,
except for some completely absurd "bigger numbers are always
gonna be better!" assumption on your part. You don't even know
why the above numbers are silly to the point of causing hours and
hours of uproarious laughter on the part of anyone reading your
post.

For just one example - how many of the supposed "64 bits" in
the samples you're talking about would you expect to contain
valid and/or significant information? Why?

The four thousand operator part gets me. Not to even contemplate the
processing power to keep all that straight, at that point the benefits of FM
are far outshone by additive. A fundamental understand of acoustics would go
a long way in figuring out why a lot of the above specifications are absurd
beyond measure.
 
R

Radium

Jan 1, 1970
0
For just one example - how many of the supposed "64 bits" in
the samples you're talking about would you expect to contain
valid and/or significant information? Why?

64 bit-resolution give 384 dB of dynamic range and
18,446,744,073,709,551,616 possible levels of loudness.

x-bit-resolution gives 6x dB of dynamic range.

6(64) = 384

x-bit-resolution gives 2^x levels of loudness.

2^64 = 18,446,744,073,709,551,616
 
H

HellPopeHuey

Jan 1, 1970
0
Um, guys... Radium is a numbnutted troll and is only on here to bait
the crowd with bandwidth-clogging drivel. Stop encouraging him or he'll
just keep staining the carpets.

--

HellPope Huey
If the neighbors play rock, beat 'em with sticks;
If they play rap, kneecap them;
but if they play opera, shoot 'em in the EYE.

I can name the newscaster on The Simpsons,
but I can't name my own congressman.
And that's what makes America great.
~ Mark Hoppus (Blink-182)

"Its disheartening to know that you live
in a country that's just teeming
with semi-literate, mediocre psychos."
~ Henry Rollins
 
64 bit-resolution give 384 dB of dynamic range and
18,446,744,073,709,551,616 possible levels of loudness.

x-bit-resolution gives 6x dB of dynamic range.

6(64) = 384

x-bit-resolution gives 2^x levels of loudness.

2^64 = 18,446,744,073,709,551,616

And... I suppose that you can tell the difference between each and
every level, yes?
 
Why particularly that one?

Because many FM sounds, say on a DX7, have up to 6 operators - if
that.

I'd love to see you program a 4,000 operator FM sound with decent
results and make it musically useful, without being random noise.
 
R

Radium

Jan 1, 1970
0
I'd love to see you program a 4,000 operator FM sound with decent
results and make it musically useful, without being random noise.

Why would 4,000-operator FM synth contain anymore random noise than a
6-operator FM synth?
 
D

Dave Platt

Jan 1, 1970
0
x-bit-resolution gives 2^x levels of loudness.

2^64 = 18,446,744,073,709,551,616

And... I suppose that you can tell the difference between each and
every level, yes?[/QUOTE]

#chuckle#

This same sort of proposal came up on one of the audio newsgroups a
decade or so ago... somebody was dissatisfied with the dynamic range
of CDs and wanted a "future-proof" digital system spec.

It was soon pointed out that if a system of this nature was adjusted
so that the amplitude of the least-significant bit was set just at (or
even somewhat below) the quietest sound that the human ear could hear
against a background of silence in a perfectly noise-free room, then a
full-amplitude signal would instantly release enough energy to blow
the entire neighborhood (and perhaps the continent) into a ball of
incandescent plasma.

The general consensus was that such a system had only a very limited
commercial appeal.
 
M

Mr.T

Jan 1, 1970
0
Radium said:
My *Soundcard* MIDI Synth Dream

Probably akin to Granny Clampett, would make most men throw up, but there is
always someone who will have a wet dream over it.:)

MrT.
 
M

Mr.T

Jan 1, 1970
0
Bob Myers said:
And right about here is where the insanity kicks in...

Not so, he's been posting the same insanity for years.

MrT.
 
M

Mr.T

Jan 1, 1970
0
And... I suppose that you can tell the difference between each and
every level, yes?

Sure can, for the first 90dB or so. It's the other ~300dB that is a bit of a
waste :)
But not to worry, nothing can reproduce over 130dB DNR in any case.

MrT.
 
D

Daniel Mandic

Jan 1, 1970
0
Dave said:
This same sort of proposal came up on one of the audio newsgroups a
decade or so ago... somebody was dissatisfied with the dynamic range
of CDs and wanted a "future-proof" digital system spec.

Yeah, like DVB-T :))))

I am sure it will last for fifty years, as PAL/NTSC satisfied...


What you can do if you Receiver gets broken (and they do, very fast) is
out of my imagination, but. That's business :), more for less.... not
to mention the thrilling uparts (update, new version, service pack...)
to come. Even more business. Energy economy is also profitting, or
wasting, as you wish... Superb :) The eggs laying wool-milk sow,
perfect! At least for the business type...

And that all for a more crappy Picture. The best, they are!!! Wow, I am
so impressed!!!! What a technical insight of living digital being...
wow!



Best Regards,

Daniel Mandic
 
R

Rick Massey

Jan 1, 1970
0
Radium said:
Why particularly that one?

FM is a shortcut. It's a synthesis method developed to simulate, in rough
terms, the same effect you get through Additive synthesis, as at the time
Additive was expensive. (Hardware intensive, as well as computation
intensive, in a time when you couldn't digitally create things in the chip
realm like we can now) Once you exceed about 128 oscillators, (which, thanks
to the typical arrangement of FM, means 64 operators) you've reached a point
where additive is less hardware intensive than your theoretical FM synth,
and also far more flexible.

It just shows me that you don't know what you're really dealing with, don't
know much about acoustics and how sound is created, and are just
overinflating your numbers because you want some super synth that you really
don't know anything about. That's why when you post stuff like that, no one
takes you seriously.
 
R

Rick Massey

Jan 1, 1970
0
HellPopeHuey said:
Um, guys... Radium is a numbnutted troll and is only on here to bait the
crowd with bandwidth-clogging drivel. Stop encouraging him or he'll just
keep staining the carpets.


You know, I don't agree. I have noticed some forward progress with him.
Normally, I'd agree with you wholeheartedly, but Radium seems to actually
want to know things, and the responses it (not sure of gender) makes
indicate that it is really interested in this topic, just highly confused
and not firing on all cylinders. If we can make it see that it's coming off
as a whackjob on this, and get it to actually learn a bit what it's talking
about, which it has already shown some progress in, then there will be a net
sum gain in humanity.
 
B

Bob Masta

Jan 1, 1970
0
64 bit-resolution give 384 dB of dynamic range and
18,446,744,073,709,551,616 possible levels of loudness.

x-bit-resolution gives 6x dB of dynamic range.

6(64) = 384

x-bit-resolution gives 2^x levels of loudness.

2^64 = 18,446,744,073,709,551,616

Please stop and consider waht you are suggesting.
More bits will indeed give more dynamic range ("levels of
loudness"). Presumably you want to be able to experience
the full range, or else why bother about it? The softest
sound you can hear is typically 0 dB SPL, so you'll be
working up from there.

Let's say you have a typical speaker system that can put
out 90 dB at 1 meter for 1 watt input. The power doubles
for each 3 dB increase, so to put out 384 dB you need
384 - 90 = 294 dB above 1 watt, which is 98 doublings
or 3.169 * 10^29 watts. This is roughly 3 times the total
power output of the Sun. (I think you will need Monster
speaker cables for that!)

But back to your earlier question about the OPL3 and
phase cancellation: You wonder what is happening in the
OPL3 that causes the results you get when you run it
through a "vocal cancellation" process. But why should
you think this has anything at all to do with the OPL3,
instead of the vocal cancellation circuit? I don't know
anything about the particular device you are using,
but it's common to allow some phase adjustment in
case the vocalist isn't dead-center. You may be
hearing the results of cancelling more low-frequency
sounds than high-frequency due to this.

Best regards,




Bob Masta

D A Q A R T A
Data AcQuisition And Real-Time Analysis
www.daqarta.com
Scope, Spectrum, Spectrogram, Signal Generator
Science with your sound card!
 
B

Bob Masta

Jan 1, 1970
0
Please stop and consider waht you are suggesting.
More bits will indeed give more dynamic range ("levels of
loudness"). Presumably you want to be able to experience
the full range, or else why bother about it? The softest
sound you can hear is typically 0 dB SPL, so you'll be
working up from there.

Let's say you have a typical speaker system that can put
out 90 dB at 1 meter for 1 watt input. The power doubles
for each 3 dB increase, so to put out 384 dB you need
384 - 90 = 294 dB above 1 watt, which is 98 doublings
or 3.169 * 10^29 watts. This is roughly 3 times the total
power output of the Sun. (I think you will need Monster
speaker cables for that!)

Oops! That should have been "3 orders of magnitude
more than the total power output of the Sun", which is
about 3.86*10^26 watts. Hey, it's only 29 dB!

Best regards,


Bob Masta

D A Q A R T A
Data AcQuisition And Real-Time Analysis
www.daqarta.com
Scope, Spectrum, Spectrogram, Signal Generator
Science with your sound card!
 
Why would 4,000-operator FM synth contain anymore random noise than a
6-operator FM synth?

Well, personally, I don't think you could program that many operators
in the first place.

After about the 10th or so, I can see you stacking 100s on each point.
 
L

Leslie Sanford

Jan 1, 1970
0
Bob Myers said:
Sure, but just imaging the advertising possibilities....

Or the military applications. Perhaps someone should build this Radium
Soundcard and use it to take over the planet.

I, for one, welcome the evil soundcard overlords.
 
Top