Re: Glockenspiel Oboe

Hi Paul,

Re glissandi and noise - you can get strange effects when you play very high
MIDI notes with a good soundcard that uses wave tables or the like.

I'm not sure why and would be interested to find out. Perhaps it
is something to do with using the wave samples well outside the
natural range of the instrument sampled.

Here is a new version of the glock. oboe, this time using 14 partials
and they are measured more accurately using a wave of about 10 seconds
duration to find the partials, and cut off at MIDI note 103.

http://homepage.ntlworld.com/robertwalker/fts/glock_oboe_v5_max_103.mid
Hopefully this will sound okay - let me know if it doesn't!

You will probably get the same effects in the
comet fly by, whenever one of the planets reaches really high MIDI notes.

What one hears is the note itself, together with a
lower note, which is often louder than the intended note.

When played fast, you get glissandi effects below the
tune.

With the comet fly by, you may get rather attractive descending
continuous and discrete glissandi towards the end

Then sometimes, you may get noise instead, as you say.

One doesn't often notice this, because most MIDI
clips keep reasonably within the natural range of
instruments.

For instance, default flute on my SB Live! soundcard is okay
up to a'''. It first gets a quiet undertone note at b''' flat (E7, 106),
but that note is basically okay. From c'''' upwards it plays all
sorts of pitches and undertones, but the basic pitch is still there
above them, though often much quieter, until MIDI note 113.

Piccolo has first quiet undertone at f''' (F7, 101).

The recorder is okay up to g'''' (G8, 115)'.

Koto has first quiet undertone at d''' (D7, 98),
and plays all sorts of under notes after that - and I have often
noticed this effect when playing fractal tunes high on the koto voice.
Actually can be rather pleasant.

Glockenspiel is okay up to 110.

My non wave table synth on the soundcard plays okay up to
a certain point, then just cuts off altogether.

Presumably the designer of the soundfont has to make
decisions about this, and I would be interested to hear more
about how it works.

Re the Slendro Alit mode:

According to the SCALA mode list, Slendro Alit is an
unequal four tone mode of Pelog:

degrees
[2] 1 1 2 3 Slendro Alit

However the modes list of FTS has it as
1 1 2 3 Slendro Alit

Sorry about that. It would have happened when I auto sorted
all the modes lists according to numbers of notes and
variation in step size, and must be a bug in that auto sort
causing it to leave out the [n] at start of mode.

I'll have to fix the auto sort, and FTS modes lists, and so
thanks very much for pin-pointing this bug.

However, it may also sound Pelog like in "atmosphere" because of the widely
varying step sizes, since Pelog is an unequal seven tone scale,
and this is a _very_ unequal four note scale.

With the Pelog scale I'm using, this mode has steps:
237 154 352 457

Here is a typical slendro scale:
slendro_pliat.scl | Gender wayang from Pliatan, South Bali (Slendro), 1/1=305.5 Hz
steps: 235.4 218.1 251.2 222.7 272.5

To within 20 cents
8/7~ 8/7~ 7/6~ 8/7~ 7/6~
-4.245 -13.03 -15.64 -8.507 5.676

Most of the Slendro scales in the SCALA archive have roughly
equal steps of around 8/7 or 7/6 (though one of them
has a step of 156 cents)

These are the other modes of Pelog in the SCALA mode names
list that have Slendro in the name:

[1] 1 1 2 1 2 Patet Barang with Pelog, Manangis, Slendro Gede, Pergenter 2
[2] 1 1 2 3 Slendro Alit
[4] 2 1 2 1 1 Slendro 1
[1] 2 1 2 1 1 Slendro 3
[3] 2 1 2 1 1 Slendro 4

The actual Pelog scale used was this one from SCALA:

pelog1.scl | Gamelan Saih pitu from Ksatria, Den Pasar (South Bali). 1/1=312.5 Hz
1/1 153 cents 315 cents 552 cents 706 cents 848 cents 1058 cents 2/1

Anyway, I think there are some gamelan experts on the TL, so over to them.

Robert

--- In tuning@y..., "Robert Walker" <robert_walker@r...> wrote:

/tuning/topicId_18465.html#18465

>
http://homepage.ntlworld.com/robertwalker/fts/glock_oboe_v5_max_103.mi
d
> Hopefully this will sound okay - let me know if it doesn't!
>

Just as a continuation of this discussion, as most people probably
know, patch editors generally have a "morph" function that will
create similar kinds of effects, although not as systematically. In
the TX81Z editor, for example, all of the data factors for envelopes
and frequencies can be gradually "morphed" to create such hybrid
effects...

_______ ____ ____ _____
Joseph Pehrson

--- In tuning@y..., "Robert Walker" <robert_walker@r...> wrote:

/tuning/topicId_18465.html#18465

>
> For instance, default flute on my SB Live! soundcard is okay
> up to a'''. It first gets a quiet undertone note at b''' flat (E7,
106),but that note is basically okay. From c'''' upwards it plays all
> sorts of pitches and undertones, but the basic pitch is still there
> above them, though often much quieter, until MIDI note 113.
>

Hello Robert...

Isn't this "undertone" phenominon simply the attempt by the soundcard
synthesizer to transpose things one or two "octaves" below after it's
driven out of range??

______ ____ ____ ____
Joseph Pehrson

Hi Joseph,

Isn't this "undertone" phenominon simply the attempt by the soundcard
synthesizer to transpose things one or two "octaves" below after it's
driven out of range??

Not quite.

If you play the flute, increasing in pitch from Midi note 104, you get a
rather attractive continuous descending and ascending glissandi
(and in fact most of the time, both simultaneously, so three
simultaneous notes, fast descending, fast ascending, and
the top slowly ascending note)

http://homepage.ntlworld.com/robertwalker/fts/flute_104.ra
(134 Kb)
http://homepage.ntlworld.com/robertwalker/fts/flute_104p.mp3
(930 Kb)

Starts at MIDI note 104.

This clip goes all the way up to MIDI note 127, in increments of 5 cents.

Actually the top note goes all the way up to MIDI 127, as one hears
easily listening to it this way.

The regular clicks are because I'm using the default pitch bend
range, so FTS keeps resounding the notes, which gives an
interesting percussive plink up high.

If you pause it at a single note, for some of the notes you get another
rather attractive wah wah wah effect:

http://homepage.ntlworld.com/robertwalker/fts/flute_119.mp3
(63 Kb)

Here is a single Koto playing an ascending scale in 31-tet from MIDI note
90 upwards, so starts as a normal ascending scale for a few notes:

http://homepage.ntlworld.com/robertwalker/fts/koto_90.ra
29 Kb

and the nice part, where it starts playing the extra notes, in mp3 format:
http://homepage.ntlworld.com/robertwalker/fts/koto_97.mp3
145 Kb

Robert

--- In tuning@y..., "Robert Walker" <robert_walker@r...> wrote:

/tuning/topicId_18465.html#18492

>
> If you play the flute, increasing in pitch from Midi note 104, you
get a rather attractive continuous descending and ascending glissandi
> (and in fact most of the time, both simultaneously, so three
> simultaneous notes, fast descending, fast ascending, and
> the top slowly ascending note)
>
> http://homepage.ntlworld.com/robertwalker/fts/flute_104.ra
> (134 Kb)
> http://homepage.ntlworld.com/robertwalker/fts/flute_104p.mp3
> (930 Kb)
>

Hi Robert...

OH! Well, this really is a spectacular effect. Most peculiar...

>
> Here is a single Koto playing an ascending scale in 31-tet from
MIDI
note 90 upwards, so starts as a normal ascending scale for a few
notes:
>
> http://homepage.ntlworld.com/robertwalker/fts/koto_90.ra
> 29 Kb
>
> and the nice part, where it starts playing the extra notes, in mp3
format:
> http://homepage.ntlworld.com/robertwalker/fts/koto_97.mp3
> 145 Kb
>

AND, these are practically "ready made" Morton Subotnik!

_______ ____ ____ ___
Joseph Pehrson

Hi,

I found this wild effect on preset 66 of a Suzuki Keyman. On the highest
notes, after cranking volume, pitch, and transposition - single or
multiple notes, but especially with pitch bends result in sounds that
aren't much related to the original (I forget what it is - I'm pretty
sure the thing is General MIDI) and much louder. Definitely not
appropriate for a piano showroom!

Clark

Clark wrote:

> I found this wild effect on preset 66 of a Suzuki Keyman. On the highest
> notes, after cranking volume, pitch, and transposition - single or
> multiple notes, but especially with pitch bends result in sounds that
> aren't much related to the original (I forget what it is - I'm pretty
> sure the thing is General MIDI) and much louder. Definitely not
> appropriate for a piano showroom!

You can get some really weird stuff with inharmonic FM timbres. It's
something to do with aliasing. Any partials that would be above the
Nyquist frequency get bounced below it. I suspect the same thing's
happening with these samplers, but less dramatic. My feeling is that
you're safest recording high notes and bending them down. It means you
lose some partials, but you don't get spurious tones from the aliasing.

Graham