Yahoo Tuning Groups Ultimate Backup tuning Frequency comparisons between Yamaha S-YXG50 and Peterson AutoStrobe 590

Here is an interesting frequency measurement experiment with the
YAMAHA S-YXG50 SoftSynth. This elusive software synthesizer was hard
to hunt down on the internet because it was discontinued since some
years. Unfortunate! Since S-YXG50 is one of the sweetest sounding
SoftSynths you can utilize under Windows XP... a far cry from that
cheap trash most will recognize as "Microsoft GS Wavetable SW Synth".

First, because I think you ought to try S-YXG50, I must tell you how
to obtain and install it.

YAMAHA S-YXG50 can be downloaded here:

http://www.nodevice.com/driver/S-YXG50/get43254.html

The driver also requires its own wavetables to be seperately installed
from the Add/Remove Hardware application in the Control Panel. The
wavetables can be downloaded here:

http://en.wikipedia.org/wiki/Software_synthesizer

Once this neat little SoftSynth is fully installed with its proper
wavetables, "YAMAHA XG WDM SoftSynthesizer" must be selected as the default MIDI playback device from within the Sound Devices in Control
Panel.

I decided to use S-YXG50 not only because of its beautiful wavetablesounds, but also because that trashy Microsoft GS Synth generates MIDI
note frequencies 1/10th an equal semitone higher than specified!

That is to say, MIDI note A4 ordinarily set and shown as 440 Hz yields
a 443 Hz frequency from the speakers. Strange!

Or else, this is a phenomenon restricted to my Parallels Desktop
configuration. The PD team was at a loss, however, to understand what
was going on...

Anyway, I immediately tried out some of the new sounds of S-YXG50 in
SCALA with the tuning chosen as Yarman-36c. I found out that Clarinet
and Reed Organ are the most dependable samples. Other instruments
sound too vibrato and their timbres makes the pitch very ambiguous.
Measurements with the AP Tuner confirmed this.

AP Tuner can be downloaded here:

http://www.aptuner.com/cgi-bin/aptuner/apmain.html

Alas, AP Tuner also confirmed that S-YXG50 pitches did not exactlymatch SCALA's tuning data. Evidently, there was something awry with
the pitch-bends.

Therefore, I decided to measure the pitches of S-YXG50's Reed Organ
tuned to as 12-tone subset of Yarman36c with the PETERSON AutoStrobe
590 tuner unit.

I completed the pitch measurements at about three o'clock in the
morning (GMT+2) in a very silent environment. My experimental setup
was as follows:

1. SCALA was running in Windows XP Virtual Machine (Parallels Desktop
3.0),
2. The 12-tone subset of Yarman-36c quantized to 12000-EDO was the
chosen tuning,
3. PETERSON AutoStrobe 590 was up and running with the same tuning set,
4. A Line-Out from my Macbook Pro was connected to Alesis M1 Active
Mk2 speakers,
5. One CAD25 Microphone resting on its own stand was connected to
PETERSON AutoStrobe 590.

Here are the results of my measurements:

C in S-YXG50 is sounding -0.8 cents compared to PETERSON
C# in S-YXG50 is sounding -0.7 cents compared to PETERSON
D in S-YXG50 matches D in PETERSON
Eb in S-YXG50 is sounding -0.4 cents compared to PETERSON
E in S-YXG50 is sounding -0.7 cents compared to PETERSON
F in S-YXG50 is sounding -0.4 cents compared to PETERSON
F# in S-YXG50 is sounding -0.8 cents compared to PETERSON
G in S-YXG50 is sounding -0.5 cents compared to PETERSON
G# in S-YXG50 is sounding -1.2 cents compared to PETERSON
A in S-YXG50 is sounding -0.25 cents compared to PETERSON
Bb in S-YXG50 is sounding +0.15 cents compared to PETERSON
B in S-YXG50 is sounding +0.2 cents compared to PETERSON

I skip the octave, because there is no way to insert stretched octave
cent offset into AutoStrobe 590. Suffice it to say, C in AutoStrobe
must be raised by +3.5 cents for it to match the c of S-YXG50.

The Yarman-36c 12-tone subset quantized to 12000-EDO in cents was:

0.0
99.1
200
302
400
502
599.9
702
802.2
901.9
1004
1101.9
1203.9

Cent offsets entered into PETERSON AutoStrobe 590 was:

C -1.9
C# -2.8
D -1.9
Eb 0.1
E -1.9
F 0.1
F# -2.0
G 0.1
G# 0.3
A 0.0
Bb 2.1
B 0.0

Note that A was set as 440 Hz in both SCALA (901.9 cents) and PETERSON.

Measured pitches in cents in reference to D were:

3.5
100.8
200.0
301.6
403.2
500.9
603.2
700.7
803.1
902.6
1003.3
1105.8
(1207.4)

Comparison of SCALA's Yarman-36c 12-tone quantized subset with the
measured pitches in reference to D as follows:

SCALA meas. SCALA is higher by

0 -0.8 0.8 cents
99.1 98.4 0.7 cents
200 200.0 0.0 cents
302 301.6 0.4 cents
400 399.3 0.7 cents
502 501.6 0.4 cents
599.9 599.1 0.8 cents
702 701.5 0.5 cents
802.2 801.0 1.2 cents
901.9 901.7 0.3 cents
1004 1004.2 -0.1 cents
1101.9 1102.1 -0.2 cents
1203.9 1206.6 -2.7 cents

I am at a loss to explain these discrepancies. At first I thought I
was dealing with steps of 768-EDO (64 equal divisions of 100 cents),
but higher pitches deviate from it. What do you think is going on with
the pitch-bends of S-YXG50?

Cordially,
Oz.

✩ ✩ ✩
www.ozanyarman.com

🔗Danny Wier <dawiertx@...>

I used to use the XG softsynth! Actually I had S-YXG70, which you got with Final Fantasy VII for the PC. That alone was worth the price of the game.

And yes, it was a whole lot better than Microsoft GS.

It looks like you can download the .cab files for the top-of-the-line S-YXG100 from Microsoft itself; the links are in Wikipedia for now:

http://en.wikipedia.org/wiki/Software_synthesizer#S-YXG_series

I assume it's safe and legal, since it's hosted at Microsoft and it's probably abandonware anyway.

As for tuning capabilities: I don't remember if it handles tuning tables, so you may have to do it the dirty way through pitch bends, but at least you get all those XG sounds.

~D.

----- Original Message ----- From: "Ozan Yarman" <ozanyarman@...>
To: "Tuning List" <tuning@yahoogroups.com>
Sent: Friday, 16 October, 2009 20:37
Subject: [tuning] Frequency comparisons between Yamaha S-YXG50 and Peterson AutoStrobe 590

First, because I think you ought to try S-YXG50, I must tell you how
to obtain and install it.

YAMAHA S-YXG50 can be downloaded here:

http://www.nodevice.com/driver/S-YXG50/get43254.html

The driver also requires its own wavetables to be seperately installed
from the Add/Remove Hardware application in the Control Panel. The
wavetables can be downloaded here:

http://en.wikipedia.org/wiki/Software_synthesizer

Once this neat little SoftSynth is fully installed with its proper
wavetables, "YAMAHA XG WDM SoftSynthesizer" must be selected as the
default MIDI playback device from within the Sound Devices in Control
Panel.

I decided to use S-YXG50 not only because of its beautiful wavetable
sounds, but also because that trashy Microsoft GS Synth generates MIDI
note frequencies 1/10th an equal semitone higher than specified!

That is to say, MIDI note A4 ordinarily set and shown as 440 Hz yields
a 443 Hz frequency from the speakers. Strange!

Or else, this is a phenomenon restricted to my Parallels Desktop
configuration. The PD team was at a loss, however, to understand what
was going on...

AP Tuner can be downloaded here:

http://www.aptuner.com/cgi-bin/aptuner/apmain.html

Alas, AP Tuner also confirmed that S-YXG50 pitches did not exactly
match SCALA's tuning data. Evidently, there was something awry with
the pitch-bends.

Therefore, I decided to measure the pitches of S-YXG50's Reed Organ
tuned to as 12-tone subset of Yarman36c with the PETERSON AutoStrobe
590 tuner unit.

I completed the pitch measurements at about three o'clock in the
morning (GMT+2) in a very silent environment. My experimental setup
was as follows:

Here are the results of my measurements:

The Yarman-36c 12-tone subset quantized to 12000-EDO in cents was:

0.0
99.1
200
302
400
502
599.9
702
802.2
901.9
1004
1101.9
1203.9

Cent offsets entered into PETERSON AutoStrobe 590 was:

C -1.9
C# -2.8
D -1.9
Eb 0.1
E -1.9
F 0.1
F# -2.0
G 0.1
G# 0.3
A 0.0
Bb 2.1
B 0.0

Note that A was set as 440 Hz in both SCALA (901.9 cents) and PETERSON.

Measured pitches in cents in reference to D were:

3.5
100.8
200.0
301.6
403.2
500.9
603.2
700.7
803.1
902.6
1003.3
1105.8
(1207.4)

Comparison of SCALA's Yarman-36c 12-tone quantized subset with the
measured pitches in reference to D as follows:

SCALA meas. SCALA is higher by

Cordially,
Oz.

✩ ✩ ✩
www.ozanyarman.com

------------------------------------

You can configure your subscription by sending an empty email to one
of these addresses (from the address at which you receive the list):
tuning-subscribe@yahoogroups.com - join the tuning group.
tuning-unsubscribe@yahoogroups.com - leave the group.
tuning-nomail@yahoogroups.com - turn off mail from the group.
tuning-digest@yahoogroups.com - set group to send daily digests.
tuning-normal@yahoogroups.com - set group to send individual emails.
tuning-help@yahoogroups.com - receive general help information.
Yahoo! Groups Links

🔗Ozan Yarman <ozanyarman@...>

Dear Danny, I own both a very antique YAMAHA MU80 unit and its PCI
counterpart SW1000XG in my Desktop PC in the corner. I swear that the
wavetables downloadable from the link you and I gave feature almost or
exactly the same sounds.

Oz.

✩ ✩ ✩
www.ozanyarman.com

On Oct 17, 2009, at 5:19 AM, Danny Wier wrote:

> I used to use the XG softsynth! Actually I had S-YXG70, which you got
> with Final Fantasy VII for the PC. That alone was worth the price of
> the
> game.
>
> And yes, it was a whole lot better than Microsoft GS.
>
> It looks like you can download the .cab files for the top-of-the-line
> S-YXG100 from Microsoft itself; the links are in Wikipedia for now:
>
> http://en.wikipedia.org/wiki/Software_synthesizer#S-YXG_series
>
> I assume it's safe and legal, since it's hosted at Microsoft and it's
> probably abandonware anyway.
>
> As for tuning capabilities: I don't remember if it handles tuning
> tables, so you may have to do it the dirty way through pitch bends,
> but
> at least you get all those XG sounds.
>
> ~D.
>
> ----- Original Message -----
> From: "Ozan Yarman" <ozanyarman@...>
> To: "Tuning List" <tuning@yahoogroups.com>
> Sent: Friday, 16 October, 2009 20:37
> Subject: [tuning] Frequency comparisons between Yamaha S-YXG50 and
> Peterson AutoStrobe 590
>
>
> Here is an interesting frequency measurement experiment with the
> YAMAHA S-YXG50 SoftSynth. This elusive software synthesizer was hard
> to hunt down on the internet because it was discontinued since some
> years. Unfortunate! Since S-YXG50 is one of the sweetest sounding
> SoftSynths you can utilize under Windows XP... a far cry from that
> cheap trash most will recognize as "Microsoft GS Wavetable SW Synth".
>
> First, because I think you ought to try S-YXG50, I must tell you how
> to obtain and install it.
>
> YAMAHA S-YXG50 can be downloaded here:
>
> http://www.nodevice.com/driver/S-YXG50/get43254.html
>
> The driver also requires its own wavetables to be seperately installed
> from the Add/Remove Hardware application in the Control Panel. The
> wavetables can be downloaded here:
>
> http://en.wikipedia.org/wiki/Software_synthesizer
>
> Once this neat little SoftSynth is fully installed with its proper
> wavetables, "YAMAHA XG WDM SoftSynthesizer" must be selected as the
> default MIDI playback device from within the Sound Devices in Control
> Panel.
>
> I decided to use S-YXG50 not only because of its beautiful wavetable
> sounds, but also because that trashy Microsoft GS Synth generates MIDI
> note frequencies 1/10th an equal semitone higher than specified!
>
> That is to say, MIDI note A4 ordinarily set and shown as 440 Hz yields
> a 443 Hz frequency from the speakers. Strange!
>
> Or else, this is a phenomenon restricted to my Parallels Desktop
> configuration. The PD team was at a loss, however, to understand what
> was going on...
>
> Anyway, I immediately tried out some of the new sounds of S-YXG50 in
> SCALA with the tuning chosen as Yarman-36c. I found out that Clarinet
> and Reed Organ are the most dependable samples. Other instruments
> sound too vibrato and their timbres makes the pitch very ambiguous.
> Measurements with the AP Tuner confirmed this.
>
> AP Tuner can be downloaded here:
>
> http://www.aptuner.com/cgi-bin/aptuner/apmain.html
>
> Alas, AP Tuner also confirmed that S-YXG50 pitches did not exactly
> match SCALA's tuning data. Evidently, there was something awry with
> the pitch-bends.
>
> Therefore, I decided to measure the pitches of S-YXG50's Reed Organ
> tuned to as 12-tone subset of Yarman36c with the PETERSON AutoStrobe
> 590 tuner unit.
>
> I completed the pitch measurements at about three o'clock in the
> morning (GMT+2) in a very silent environment. My experimental setup
> was as follows:
>
> 1. SCALA was running in Windows XP Virtual Machine (Parallels Desktop
> 3.0),
> 2. The 12-tone subset of Yarman-36c quantized to 12000-EDO was the
> chosen tuning,
> 3. PETERSON AutoStrobe 590 was up and running with the same tuning
> set,
> 4. A Line-Out from my Macbook Pro was connected to Alesis M1 Active
> Mk2 speakers,
> 5. One CAD25 Microphone resting on its own stand was connected to
> PETERSON AutoStrobe 590.
>
> Here are the results of my measurements:
>
> C in S-YXG50 is sounding -0.8 cents compared to PETERSON
> C# in S-YXG50 is sounding -0.7 cents compared to PETERSON
> D in S-YXG50 matches D in PETERSON
> Eb in S-YXG50 is sounding -0.4 cents compared to PETERSON
> E in S-YXG50 is sounding -0.7 cents compared to PETERSON
> F in S-YXG50 is sounding -0.4 cents compared to PETERSON
> F# in S-YXG50 is sounding -0.8 cents compared to PETERSON
> G in S-YXG50 is sounding -0.5 cents compared to PETERSON
> G# in S-YXG50 is sounding -1.2 cents compared to PETERSON
> A in S-YXG50 is sounding -0.25 cents compared to PETERSON
> Bb in S-YXG50 is sounding +0.15 cents compared to PETERSON
> B in S-YXG50 is sounding +0.2 cents compared to PETERSON
>
> I skip the octave, because there is no way to insert stretched octave
> cent offset into AutoStrobe 590. Suffice it to say, C in AutoStrobe
> must be raised by +3.5 cents for it to match the c of S-YXG50.
>
> The Yarman-36c 12-tone subset quantized to 12000-EDO in cents was:
>
> 0.0
> 99.1
> 200
> 302
> 400
> 502
> 599.9
> 702
> 802.2
> 901.9
> 1004
> 1101.9
> 1203.9
>
> Cent offsets entered into PETERSON AutoStrobe 590 was:
>
> C -1.9
> C# -2.8
> D -1.9
> Eb 0.1
> E -1.9
> F 0.1
> F# -2.0
> G 0.1
> G# 0.3
> A 0.0
> Bb 2.1
> B 0.0
>
> Note that A was set as 440 Hz in both SCALA (901.9 cents) and
> PETERSON.
>
> Measured pitches in cents in reference to D were:
>
> 3.5
> 100.8
> 200.0
> 301.6
> 403.2
> 500.9
> 603.2
> 700.7
> 803.1
> 902.6
> 1003.3
> 1105.8
> (1207.4)
>
> Comparison of SCALA's Yarman-36c 12-tone quantized subset with the
> measured pitches in reference to D as follows:
>
> SCALA meas. SCALA is higher by
>
> 0 -0.8 0.8 cents
> 99.1 98.4 0.7 cents
> 200 200.0 0.0 cents
> 302 301.6 0.4 cents
> 400 399.3 0.7 cents
> 502 501.6 0.4 cents
> 599.9 599.1 0.8 cents
> 702 701.5 0.5 cents
> 802.2 801.0 1.2 cents
> 901.9 901.7 0.3 cents
> 1004 1004.2 -0.1 cents
> 1101.9 1102.1 -0.2 cents
> 1203.9 1206.6 -2.7 cents
>
> I am at a loss to explain these discrepancies. At first I thought I
> was dealing with steps of 768-EDO (64 equal divisions of 100 cents),
> but higher pitches deviate from it. What do you think is going on with
> the pitch-bends of S-YXG50?
>
> Cordially,
> Oz.
>
> ✩ ✩ ✩
> www.ozanyarman.com
>
>
>
> ------------------------------------
>
> You can configure your subscription by sending an empty email to one
> of these addresses (from the address at which you receive the list):
> tuning-subscribe@yahoogroups.com - join the tuning group.
> tuning-unsubscribe@yahoogroups.com - leave the group.
> tuning-nomail@yahoogroups.com - turn off mail from the group.
> tuning-digest@yahoogroups.com - set group to send daily digests.
> tuning-normal@yahoogroups.com - set group to send individual emails.
> tuning-help@yahoogroups.com - receive general help information.
> Yahoo! Groups Links
>
>
>
>
>
> ------------------------------------
>
> You can configure your subscription by sending an empty email to one
> of these addresses (from the address at which you receive the list):
> tuning-subscribe@yahoogroups.com - join the tuning group.
> tuning-unsubscribe@yahoogroups.com - leave the group.
> tuning-nomail@yahoogroups.com - turn off mail from the group.
> tuning-digest@yahoogroups.com - set group to send daily digests.
> tuning-normal@yahoogroups.com - set group to send individual emails.
> tuning-help@yahoogroups.com - receive general help information.
> Yahoo! Groups Links
>
>
>

🔗Danny Wier <dawiertx@...>

They probably are the same sounds; I had a Yamaha keyboard a decade ago that had pretty much the same sounds as the S-YXG50, in fact.

Also, to anyone interested: I'm not sure the wavetables (NOT "soundfonts" as I said earlier) work in Windows 2000, XP or Vista, and you do have to do some registry work if they do. At least you have to extract the files from the cab into a common folder then right-click the .inf file; using "Add Hardware" in Control Panel doesn't do it.

I'll give it a try at least.

~D.

----- Original Message ----- From: "Ozan Yarman" <ozanyarman@...>
To: <tuning@yahoogroups.com>
Sent: Friday, 16 October, 2009 21:35
Subject: Re: [tuning] Frequency comparisons between Yamaha S-YXG50 and Peterson AutoStrobe 590

Oz.

✩ ✩ ✩
www.ozanyarman.com

On Oct 17, 2009, at 5:19 AM, Danny Wier wrote:

🔗Ozan Yarman <ozanyarman@...>

O brother Danny,

Let me clear some confusion. No need to mess with cab files. Just
download both wavetables (2 and 4 Mbs), choose the 2 MB inf file from
the Add/Remove Hardware app, copy the .tbl files of the 4 Mb wavetable
package into the System32 folder under Windows and you are done. Yes,
it works with Windows XP. I don't know about Vista though. But as I
stated earlier, you need to install the SoftSynth program as well. No
need to mess with the registry if you do. Just click on
SoftSynthesizer>Detailed Settings>4MB and you have the best sounds for
Windows MIDI.

Oz.

✩ ✩ ✩
www.ozanyarman.com

On Oct 17, 2009, at 6:15 AM, Danny Wier wrote:

> They probably are the same sounds; I had a Yamaha keyboard a decade
> ago
> that had pretty much the same sounds as the S-YXG50, in fact.
>
> Also, to anyone interested: I'm not sure the wavetables (NOT
> "soundfonts" as I said earlier) work in Windows 2000, XP or Vista, and
> you do have to do some registry work if they do. At least you have to
> extract the files from the cab into a common folder then right-click
> the
> .inf file; using "Add Hardware" in Control Panel doesn't do it.
>
> I'll give it a try at least.
>
> ~D.
>
> ----- Original Message -----
> From: "Ozan Yarman" <ozanyarman@...>
> To: <tuning@yahoogroups.com>
> Sent: Friday, 16 October, 2009 21:35
> Subject: Re: [tuning] Frequency comparisons between Yamaha S-YXG50 and
> Peterson AutoStrobe 590
>
>
> Dear Danny, I own both a very antique YAMAHA MU80 unit and its PCI
> counterpart SW1000XG in my Desktop PC in the corner. I swear that the
> wavetables downloadable from the link you and I gave feature almost or
> exactly the same sounds.
>
> Oz.
>
> ✩ ✩ ✩
> www.ozanyarman.com
>
> On Oct 17, 2009, at 5:19 AM, Danny Wier wrote:
>
>> I used to use the XG softsynth! Actually I had S-YXG70, which you got
>> with Final Fantasy VII for the PC. That alone was worth the price of
>> the
>> game.
>>
>> And yes, it was a whole lot better than Microsoft GS.
>>
>> It looks like you can download the .cab files for the top-of-the-line
>> S-YXG100 from Microsoft itself; the links are in Wikipedia for now:
>>
>> http://en.wikipedia.org/wiki/Software_synthesizer#S-YXG_series
>>
>> I assume it's safe and legal, since it's hosted at Microsoft and it's
>> probably abandonware anyway.
>>
>> As for tuning capabilities: I don't remember if it handles tuning
>> tables, so you may have to do it the dirty way through pitch bends,
>> but
>> at least you get all those XG sounds.
>>
>> ~D.
>
>
>
> ------------------------------------
>
> You can configure your subscription by sending an empty email to one
> of these addresses (from the address at which you receive the list):
> tuning-subscribe@yahoogroups.com - join the tuning group.
> tuning-unsubscribe@yahoogroups.com - leave the group.
> tuning-nomail@yahoogroups.com - turn off mail from the group.
> tuning-digest@yahoogroups.com - set group to send daily digests.
> tuning-normal@yahoogroups.com - set group to send individual emails.
> tuning-help@yahoogroups.com - receive general help information.
> Yahoo! Groups Links
>
>
>

🔗Charles Lucy <lucy@...>

Being a Yamaha instrument, it might be using YAMS as its microtuning
resolution.
see this page:

http://www.lucytune.com/midi_and_keyboard/yamaha_01.html

On 17 Oct 2009, at 02:37, Ozan Yarman wrote:

> Here is an interesting frequency measurement experiment with the
> YAMAHA S-YXG50 SoftSynth. This elusive software synthesizer was hard
> to hunt down on the internet because it was discontinued since some
> years. Unfortunate! Since S-YXG50 is one of the sweetest sounding
> SoftSynths you can utilize under Windows XP... a far cry from that
> cheap trash most will recognize as "Microsoft GS Wavetable SW Synth".
>
> First, because I think you ought to try S-YXG50, I must tell you how
> to obtain and install it.
>
> YAMAHA S-YXG50 can be downloaded here:
>
> http://www.nodevice.com/driver/S-YXG50/get43254.html
>
> The driver also requires its own wavetables to be seperately installed
> from the Add/Remove Hardware application in the Control Panel. The
> wavetables can be downloaded here:
>
> http://en.wikipedia.org/wiki/Software_synthesizer
>
> Once this neat little SoftSynth is fully installed with its proper
> wavetables, "YAMAHA XG WDM SoftSynthesizer" must be selected as the
> default MIDI playback device from within the Sound Devices in Control
> Panel.
>
> I decided to use S-YXG50 not only because of its beautiful wavetable
> sounds, but also because that trashy Microsoft GS Synth generates MIDI
> note frequencies 1/10th an equal semitone higher than specified!
>
> That is to say, MIDI note A4 ordinarily set and shown as 440 Hz yields
> a 443 Hz frequency from the speakers. Strange!
>
> Or else, this is a phenomenon restricted to my Parallels Desktop
> configuration. The PD team was at a loss, however, to understand what
> was going on...
>
> Anyway, I immediately tried out some of the new sounds of S-YXG50 in
> SCALA with the tuning chosen as Yarman-36c. I found out that Clarinet
> and Reed Organ are the most dependable samples. Other instruments
> sound too vibrato and their timbres makes the pitch very ambiguous.
> Measurements with the AP Tuner confirmed this.
>
> AP Tuner can be downloaded here:
>
> http://www.aptuner.com/cgi-bin/aptuner/apmain.html
>
> Alas, AP Tuner also confirmed that S-YXG50 pitches did not exactly
> match SCALA's tuning data. Evidently, there was something awry with
> the pitch-bends.
>
> Therefore, I decided to measure the pitches of S-YXG50's Reed Organ
> tuned to as 12-tone subset of Yarman36c with the PETERSON AutoStrobe
> 590 tuner unit.
>
> I completed the pitch measurements at about three o'clock in the
> morning (GMT+2) in a very silent environment. My experimental setup
> was as follows:
>
> 1. SCALA was running in Windows XP Virtual Machine (Parallels Desktop
> 3.0),
> 2. The 12-tone subset of Yarman-36c quantized to 12000-EDO was the
> chosen tuning,
> 3. PETERSON AutoStrobe 590 was up and running with the same tuning
> set,
> 4. A Line-Out from my Macbook Pro was connected to Alesis M1 Active
> Mk2 speakers,
> 5. One CAD25 Microphone resting on its own stand was connected to
> PETERSON AutoStrobe 590.
>
> Here are the results of my measurements:
>
> C in S-YXG50 is sounding -0.8 cents compared to PETERSON
> C# in S-YXG50 is sounding -0.7 cents compared to PETERSON
> D in S-YXG50 matches D in PETERSON
> Eb in S-YXG50 is sounding -0.4 cents compared to PETERSON
> E in S-YXG50 is sounding -0.7 cents compared to PETERSON
> F in S-YXG50 is sounding -0.4 cents compared to PETERSON
> F# in S-YXG50 is sounding -0.8 cents compared to PETERSON
> G in S-YXG50 is sounding -0.5 cents compared to PETERSON
> G# in S-YXG50 is sounding -1.2 cents compared to PETERSON
> A in S-YXG50 is sounding -0.25 cents compared to PETERSON
> Bb in S-YXG50 is sounding +0.15 cents compared to PETERSON
> B in S-YXG50 is sounding +0.2 cents compared to PETERSON
>
> I skip the octave, because there is no way to insert stretched octave
> cent offset into AutoStrobe 590. Suffice it to say, C in AutoStrobe
> must be raised by +3.5 cents for it to match the c of S-YXG50.
>
> The Yarman-36c 12-tone subset quantized to 12000-EDO in cents was:
>
> 0.0
> 99.1
> 200
> 302
> 400
> 502
> 599.9
> 702
> 802.2
> 901.9
> 1004
> 1101.9
> 1203.9
>
> Cent offsets entered into PETERSON AutoStrobe 590 was:
>
> C -1.9
> C# -2.8
> D -1.9
> Eb 0.1
> E -1.9
> F 0.1
> F# -2.0
> G 0.1
> G# 0.3
> A 0.0
> Bb 2.1
> B 0.0
>
> Note that A was set as 440 Hz in both SCALA (901.9 cents) and
> PETERSON.
>
> Measured pitches in cents in reference to D were:
>
> 3.5
> 100.8
> 200.0
> 301.6
> 403.2
> 500.9
> 603.2
> 700.7
> 803.1
> 902.6
> 1003.3
> 1105.8
> (1207.4)
>
> Comparison of SCALA's Yarman-36c 12-tone quantized subset with the
> measured pitches in reference to D as follows:
>
> SCALA meas. SCALA is higher by
>
> 0 -0.8 0.8 cents
> 99.1 98.4 0.7 cents
> 200 200.0 0.0 cents
> 302 301.6 0.4 cents
> 400 399.3 0.7 cents
> 502 501.6 0.4 cents
> 599.9 599.1 0.8 cents
> 702 701.5 0.5 cents
> 802.2 801.0 1.2 cents
> 901.9 901.7 0.3 cents
> 1004 1004.2 -0.1 cents
> 1101.9 1102.1 -0.2 cents
> 1203.9 1206.6 -2.7 cents
>
> I am at a loss to explain these discrepancies. At first I thought I
> was dealing with steps of 768-EDO (64 equal divisions of 100 cents),
> but higher pitches deviate from it. What do you think is going on with
> the pitch-bends of S-YXG50?
>
> Cordially,
> Oz.
>
> ✩ ✩ ✩
> www.ozanyarman.com
>
>
>

Charles Lucy
lucy@...

- Promoting global harmony through LucyTuning -

for information on LucyTuning go to:
http://www.lucytune.com

For LucyTuned Lullabies go to:
http://www.lullabies.co.uk

🔗Petr Pařízek <p.parizek@...>

🔗Ozan Yarman <ozanyarman@...>

Dear Petr,

According to my measurements, S-YXG50's tuning system is awry. It does
not retune by cents, or by degrees of 768-EDO or 1024-EDO.

This experiment was done under conditions similar to the previous one
at about the same time hours before dusk. SCALA was running with 1200-
EDO set and the synth was selected as number 108 (YAMAHA XG - 12
tones).

Here are the results of going up a quarter-tone from MIDI note A4 (440
Hz).

1200-EDO versus "whatnot" increments

0) 0. [Reed Organ, A=440 Hz] -0.3 cents
1) 1. +1.1 cents
2) 2. +1.1 cents
3) 3. +2.4 cents
4) 4. +3.6 cents
5) 5. +5.0 cents
6) 6. +6.3 cents
7) 6. +6.3 cents
8) 7. +7.5 cents
9) 8. +8.9 cents
10) 9. +10.1 cents
11) 9. +10.1 cents
12) 10. +11.4 cents
13) 10. +11.4 cents
14) 11. +12.7 cents
15) 12. +14.0 cents
16) 13. +15.3 cents
17) 13. +15.3 cents
18) 14. +16.6 cents
19) 15. +17.9 cents
20) 16. +19.1 cents
21) 16. +19.1 cents
22) 17. +20.5 cents
23) 18. +21.7 cents
24) 19. +23.0 cents
25) 20. +24.3 cents
26) 21. +25.6 cents
27) 22. +26.9 cents
28) 23. +28.2 cents
29) 23. +28.2 cents
30) 24. +29.4 cents
31) 25. +30.7 cents
32) 26. +32.0 cents
33) 27. +33.3 cents
34) 27. +33.3 cents
35) 28. +34.6 cents
36) 29. +35.9 cents
37) 30. +37.1 cents
38) 30. +37.1 cents
39) 30. +37.1 cents (again!)
40) 31. +38.4 cents
41) 32. +39.7 cents
42) 33. +41.0 cents
43) 34. +42.2 cents
44) 34. +42.2 cents
45) 35. +43.5 cents
46) 36. +44.8 cents
47) 37. +46.0 cents
48) 38. +47.3 cents
49) 38. +47.3 cents
50) 39. +49.8 cents

Increments in cents:

1.4
1.3
1.2
1.4
1.3
1.2
1.4
1.2
1.3
1.3
1.3
1.3
1.3
1.3
1.2
1.4
1.2
1.3
1.3
1.3
1.3
1.3
1.2
1.3
1.3
1.3
1.3
1.3
1.2
1.3
1.3
1.3
1.2
1.3
1.3
1.2
1.3
2.5

Can anyone make heads or tails of this? I am at a loss to explain the
final 2.5 cent step.

And Petr, what do you know about Roland Virtual Sound Canvas's tuning
abilities?

Your composition is sounding vey good. I like it! But here is the
strange thing: QT Synth is starting to play the midi file from the key
of C# while the S-YXG50 Player is starting from the key of C! I am not
aware of having activated any transpositions. What do you think is
wrong here?

Oz.

✩ ✩ ✩
www.ozanyarman.com

On Oct 17, 2009, at 1:17 PM, Petr Pařízek wrote:

>
>
> Wow, finally a topic to which I can add some contribution fron. :-D
> I have a hardware synth which is fully XG compatible and therefore I
> was always interested in many XG-related things.
>
> The XG tuning format uses a cent resolution with a maximum allowed
> deviation of -64..+63 cents away from 12-equal. Each retuning
> command affects the particular key in all octaves, which means that
> you can use up to 12 commands if you want to retune all the keys in
> all octaves. Of course, the disadvantage is that you can't retune,
> for example, C1 by +20 cents and C2 by -40 cents at the same time on
> the same MIDI channel. Scala has the ability to send these commands
> through MIDI or to a file, unfortunately only for retuning the
> channel #1. AFAIK, Scala itself allows to do this only if a 12-tone
> scale is loaded. You should be able to do that by using "Set synth
> 108". Anyway, some years ago, I made a small QBasic utility which
> converts plain text numbers to this kind of MIDI file, which allows
> me to use various tunings in one MIDI file because each MIDI channel
> can be tuned differently.
>
> An example might be the MIDI file which I've made in august this
> year. There's diaschismatic on channels 1 to 4, negri on 5 to 9,
> porcupine on 11 to 14, and meantone on 15 to 16.
> So if you have something like the XG-50 or XG-100 or another similar
> driver, you may try this (there's also music there, not only tuning
> commands):
> www.sendspace.com/file/t9z3ar
> Petr
>
>
>
>
> __._,

🔗Ozan Yarman <ozanyarman@...>

Sorry Charles, that doesn't seem to be the case.

Oz.

✩ ✩ ✩
www.ozanyarman.com

On Oct 17, 2009, at 10:11 AM, Charles Lucy wrote:

>
>
> Being a Yamaha instrument, it might be using YAMS as its microtuning
> resolution.
> see this page:
>
> http://www.lucytune.com/midi_and_keyboard/yamaha_01.html
>
>
> On 17 Oct 2009, at 02:37, Ozan Yarman wrote:
>
>> Here is an interesting frequency measurement experiment with the
>> YAMAHA S-YXG50 SoftSynth. This elusive software synthesizer was hard
>> to hunt down on the internet because it was discontinued since some
>> years. Unfortunate! Since S-YXG50 is one of the sweetest sounding
>> SoftSynths you can utilize under Windows XP... a far cry from that
>> cheap trash most will recognize as "Microsoft GS Wavetable SW Synth".
>>
>> First, because I think you ought to try S-YXG50, I must tell you how
>> to obtain and install it.
>>
>> YAMAHA S-YXG50 can be downloaded here:
>>
>> http://www.nodevice.com/driver/S-YXG50/get43254.html
>>
>> The driver also requires its own wavetables to be seperately
>> installed
>> from the Add/Remove Hardware application in the Control Panel. The
>> wavetables can be downloaded here:
>>
>> http://en.wikipedia.org/wiki/Software_synthesizer
>>
>> Once this neat little SoftSynth is fully installed with its proper
>> wavetables, "YAMAHA XG WDM SoftSynthesizer" must be selected as the
>> default MIDI playback device from within the Sound Devices in Control
>> Panel.
>>
>> I decided to use S-YXG50 not only because of its beautiful wavetable
>> sounds, but also because that trashy Microsoft GS Synth generates
>> MIDI
>> note frequencies 1/10th an equal semitone higher than specified!
>>
>> That is to say, MIDI note A4 ordinarily set and shown as 440 Hz
>> yields
>> a 443 Hz frequency from the speakers. Strange!
>>
>> Or else, this is a phenomenon restricted to my Parallels Desktop
>> configuration. The PD team was at a loss, however, to understand what
>> was going on...
>>
>> Anyway, I immediately tried out some of the new sounds of S-YXG50 in
>> SCALA with the tuning chosen as Yarman-36c. I found out that Clarinet
>> and Reed Organ are the most dependable samples. Other instruments
>> sound too vibrato and their timbres makes the pitch very ambiguous.
>> Measurements with the AP Tuner confirmed this.
>>
>> AP Tuner can be downloaded here:
>>
>> http://www.aptuner.com/cgi-bin/aptuner/apmain.html
>>
>> Alas, AP Tuner also confirmed that S-YXG50 pitches did not exactly
>> match SCALA's tuning data. Evidently, there was something awry with
>> the pitch-bends.
>>
>> Therefore, I decided to measure the pitches of S-YXG50's Reed Organ
>> tuned to as 12-tone subset of Yarman36c with the PETERSON AutoStrobe
>> 590 tuner unit.
>>
>> I completed the pitch measurements at about three o'clock in the
>> morning (GMT+2) in a very silent environment. My experimental setup
>> was as follows:
>>
>> 1. SCALA was running in Windows XP Virtual Machine (Parallels Desktop
>> 3.0),
>> 2. The 12-tone subset of Yarman-36c quantized to 12000-EDO was the
>> chosen tuning,
>> 3. PETERSON AutoStrobe 590 was up and running with the same tuning
>> set,
>> 4. A Line-Out from my Macbook Pro was connected to Alesis M1 Active
>> Mk2 speakers,
>> 5. One CAD25 Microphone resting on its own stand was connected to
>> PETERSON AutoStrobe 590.
>>
>> Here are the results of my measurements:
>>
>> C in S-YXG50 is sounding -0.8 cents compared to PETERSON
>> C# in S-YXG50 is sounding -0.7 cents compared to PETERSON
>> D in S-YXG50 matches D in PETERSON
>> Eb in S-YXG50 is sounding -0.4 cents compared to PETERSON
>> E in S-YXG50 is sounding -0.7 cents compared to PETERSON
>> F in S-YXG50 is sounding -0.4 cents compared to PETERSON
>> F# in S-YXG50 is sounding -0.8 cents compared to PETERSON
>> G in S-YXG50 is sounding -0.5 cents compared to PETERSON
>> G# in S-YXG50 is sounding -1.2 cents compared to PETERSON
>> A in S-YXG50 is sounding -0.25 cents compared to PETERSON
>> Bb in S-YXG50 is sounding +0.15 cents compared to PETERSON
>> B in S-YXG50 is sounding +0.2 cents compared to PETERSON
>>
>> I skip the octave, because there is no way to insert stretched octave
>> cent offset into AutoStrobe 590. Suffice it to say, C in AutoStrobe
>> must be raised by +3.5 cents for it to match the c of S-YXG50.
>>
>> The Yarman-36c 12-tone subset quantized to 12000-EDO in cents was:
>>
>> 0.0
>> 99.1
>> 200
>> 302
>> 400
>> 502
>> 599.9
>> 702
>> 802.2
>> 901.9
>> 1004
>> 1101.9
>> 1203.9
>>
>> Cent offsets entered into PETERSON AutoStrobe 590 was:
>>
>> C -1.9
>> C# -2.8
>> D -1.9
>> Eb 0.1
>> E -1.9
>> F 0.1
>> F# -2.0
>> G 0.1
>> G# 0.3
>> A 0.0
>> Bb 2.1
>> B 0.0
>>
>> Note that A was set as 440 Hz in both SCALA (901.9 cents) and
>> PETERSON.
>>
>> Measured pitches in cents in reference to D were:
>>
>> 3.5
>> 100.8
>> 200.0
>> 301.6
>> 403.2
>> 500.9
>> 603.2
>> 700.7
>> 803.1
>> 902.6
>> 1003.3
>> 1105.8
>> (1207.4)
>>
>> Comparison of SCALA's Yarman-36c 12-tone quantized subset with the
>> measured pitches in reference to D as follows:
>>
>> SCALA meas. SCALA is higher by
>>
>> 0 -0.8 0.8 cents
>> 99.1 98.4 0.7 cents
>> 200 200.0 0.0 cents
>> 302 301.6 0.4 cents
>> 400 399.3 0.7 cents
>> 502 501.6 0.4 cents
>> 599.9 599.1 0.8 cents
>> 702 701.5 0.5 cents
>> 802.2 801.0 1.2 cents
>> 901.9 901.7 0.3 cents
>> 1004 1004.2 -0.1 cents
>> 1101.9 1102.1 -0.2 cents
>> 1203.9 1206.6 -2.7 cents
>>
>> I am at a loss to explain these discrepancies. At first I thought I
>> was dealing with steps of 768-EDO (64 equal divisions of 100 cents),
>> but higher pitches deviate from it. What do you think is going on
>> with
>> the pitch-bends of S-YXG50?
>>
>> Cordially,
>> Oz.
>>
>> ✩ ✩ ✩
>> www.ozanyarman.com
>>
>>
>
> Charles Lucy
> lucy@...
>
> - Promoting global harmony through LucyTuning -
>
> for information on LucyTuning go to:
> http://www.lucytune.com
>
> For LucyTuned Lullabies go to:
> http://www.lullabies.co.uk
>
>
>
>
>

🔗Daniel Forró <dan.for@...>

You and Petr are talking two different things: Petr mentioned
standard SysEx tuning parameters in XG standard, and you wrote about
Pitch Bend retuning (???) by Scala. Am I right?

Or which way have you got those results? If you want to use PB for
retuning, could you try to measure PB increments directly from any
sequencer, not controlled by Scala? Just set PB range to major second
by RPN. If software is OK, there must be always some increment in
Cents when you shift PB one step more.

This way you can isolate where problem is. I doubt it's Scala, so
more probably S-YXG50. There were more versions, some of them had
some issues, and newer versions were not necessarily better than
older ones. BTW, why do you not try S-YXG100?

XG standard doesn't use 768 or 1024 resolution like older Yamaha
synths. It retunes each of 12 tones by cents (as Petr explained),
similar way as Roland GS and Korg instruments.

Daniel Forro

On 18 Oct 2009, at 10:09 AM, Ozan Yarman wrote:

>
>
> Dear Petr,
>
> According to my measurements, S-YXG50's tuning system is awry. It
> does not retune by cents, or by degrees of 768-EDO or 1024-EDO.
>
> This experiment was done under conditions similar to the previous
> one at about the same time hours before dusk. SCALA was running
> with 1200-EDO set and the synth was selected as number 108 (YAMAHA
> XG - 12 tones).
>
> Here are the results of going up a quarter-tone from MIDI note A4
> (440 Hz).
>
> 1200-EDO versus "whatnot" increments
>
> 0) 0. [Reed Organ, A=440 Hz] -0.3 cents
> 1) 1. +1.1 cents
> 2) 2. +1.1 cents
> 3) 3. +2.4 cents
> 4) 4. +3.6 cents
> 5) 5. +5.0 cents
> 6) 6. +6.3 cents
> 7) 6. +6.3 cents
> 8) 7. +7.5 cents
> 9) 8. +8.9 cents
> 10) 9. +10.1 cents
> 11) 9. +10.1 cents
> 12) 10. +11.4 cents
> 13) 10. +11.4 cents
> 14) 11. +12.7 cents
> 15) 12. +14.0 cents
> 16) 13. +15.3 cents
> 17) 13. +15.3 cents
> 18) 14. +16.6 cents
> 19) 15. +17.9 cents
> 20) 16. +19.1 cents
> 21) 16. +19.1 cents
> 22) 17. +20.5 cents
> 23) 18. +21.7 cents
> 24) 19. +23.0 cents
> 25) 20. +24.3 cents
> 26) 21. +25.6 cents
> 27) 22. +26.9 cents
> 28) 23. +28.2 cents
> 29) 23. +28.2 cents
> 30) 24. +29.4 cents
> 31) 25. +30.7 cents
> 32) 26. +32.0 cents
> 33) 27. +33.3 cents
> 34) 27. +33.3 cents
> 35) 28. +34.6 cents
> 36) 29. +35.9 cents
> 37) 30. +37.1 cents
> 38) 30. +37.1 cents
> 39) 30. +37.1 cents (again!)
> 40) 31. +38.4 cents
> 41) 32. +39.7 cents
> 42) 33. +41.0 cents
> 43) 34. +42.2 cents
> 44) 34. +42.2 cents
> 45) 35. +43.5 cents
> 46) 36. +44.8 cents
> 47) 37. +46.0 cents
> 48) 38. +47.3 cents
> 49) 38. +47.3 cents
> 50) 39. +49.8 cents
>
> Increments in cents:
>
> 1.4
> 1.3
> 1.2
> 1.4
> 1.3
> 1.2
> 1.4
> 1.2
> 1.3
> 1.3
> 1.3
> 1.3
> 1.3
> 1.3
> 1.2
> 1.4
> 1.2
> 1.3
> 1.3
> 1.3
> 1.3
> 1.3
> 1.2
> 1.3
> 1.3
> 1.3
> 1.3
> 1.3
> 1.2
> 1.3
> 1.3
> 1.3
> 1.2
> 1.3
> 1.3
> 1.2
> 1.3
> 2.5
>
> Can anyone make heads or tails of this? I am at a loss to explain
> the final 2.5 cent step.
>
> And Petr, what do you know about Roland Virtual Sound Canvas's
> tuning abilities?
>
> Your composition is sounding vey good. I like it! But here is the
> strange thing: QT Synth is starting to play the midi file from the > key of C# while the S-YXG50 Player is starting from the key of C! I
> am not aware of having activated any transpositions. What do you
> think is wrong here?
>
> Oz.
>
> ✩ ✩ ✩
> www.ozanyarman.com
>
> On Oct 17, 2009, at 1:17 PM, Petr Pařízek wrote:
>
>>
>>
>> Wow, finally a topic to which I can add some contribution fron. :-
>> D I have a hardware synth which is fully XG compatible and
>> therefore I was always interested in many XG-related things.
>>
>> The XG tuning format uses a cent resolution with a maximum allowed
>> deviation of -64..+63 cents away from 12-equal. Each retuning
>> command affects the particular key in all octaves, which means
>> that you can use up to 12 commands if you want to retune all the
>> keys in all octaves. Of course, the disadvantage is that you can't
>> retune, for example, C1 by +20 cents and C2 by -40 cents at the
>> same time on the same MIDI channel. Scala has the ability to send
>> these commands through MIDI or to a file, unfortunately only for
>> retuning the channel #1. AFAIK, Scala itself allows to do this
>> only if a 12-tone scale is loaded. You should be able to do that
>> by using "Set synth 108". Anyway, some years ago, I made a small
>> QBasic utility which converts plain text numbers to this kind of
>> MIDI file, which allows me to use various tunings in one MIDI file
>> because each MIDI channel can be tuned differently.
>>
>> An example might be the MIDI file which I've made in august this
>> year. There's diaschismatic on channels 1 to 4, negri on 5 to 9,
>> porcupine on 11 to 14, and meantone on 15 to 16.
>> So if you have something like the XG-50 or XG-100 or another
>> similar driver, you may try this (there's also music there, not
>> only tuning commands):
>> www.sendspace.com/file/t9z3ar
>> Petr
>>
>>
>>
>>
>> __._,
>
>
>
>

🔗Ozan Yarman <ozanyarman@...>

Dear Daniel,

Sorry for not being specific. As the title of this thread implies, I
have measured the S-YXG50 Reed Organ output with Peterson AutoStrobe
590. The YAMAHA SoftSynth is, according to Petr, using the XG tuningstandard. But it does not appear to be so.

To be more precise, this is what I did:

I ran SCALA, set the synth model as 108 (XG family - 12 tones) and
chose 1200-EDO. The microphone was connected to AutoStrobe and the
speakers giving the sound output were Alesis M1 Active Mk2. Using the
Chromatic Keyboard in SCALA, I started from the 440 Hz A (900 cents
from C) and travelled up a quarter-tone by 1 cent increments. The
instrument was a Reed Organ patch from the S-YXG50 wavetable.
Measurements indicate that SCALA's one cent increments do not match
the increments measured. That is to say, SCALA says it is 1200-EDO,
but S-YXG50 cannot microtune to 1200-EDO. It so appears that S-YXG50
does not microtune by cent increments as Petr said it should.

Where can I obtained S-YXG100? I did not find it on the internet.

Oz.

✩ ✩ ✩
www.ozanyarman.com

On Oct 18, 2009, at 4:30 AM, Daniel Forró wrote:

> You and Petr are talking two different things: Petr mentioned
> standard SysEx tuning parameters in XG standard, and you wrote about
> Pitch Bend retuning (???) by Scala. Am I right?
>
> Or which way have you got those results? If you want to use PB for
> retuning, could you try to measure PB increments directly from any
> sequencer, not controlled by Scala? Just set PB range to major second
> by RPN. If software is OK, there must be always some increment in
> Cents when you shift PB one step more.
>
> This way you can isolate where problem is. I doubt it's Scala, so
> more probably S-YXG50. There were more versions, some of them had
> some issues, and newer versions were not necessarily better than
> older ones. BTW, why do you not try S-YXG100?
>
> XG standard doesn't use 768 or 1024 resolution like older Yamaha
> synths. It retunes each of 12 tones by cents (as Petr explained),
> similar way as Roland GS and Korg instruments.
>
> Daniel Forro
>
>
>
> On 18 Oct 2009, at 10:09 AM, Ozan Yarman wrote:
>
>>
>>
>> Dear Petr,
>>
>> According to my measurements, S-YXG50's tuning system is awry. It
>> does not retune by cents, or by degrees of 768-EDO or 1024-EDO.
>>
>> This experiment was done under conditions similar to the previous
>> one at about the same time hours before dusk. SCALA was running
>> with 1200-EDO set and the synth was selected as number 108 (YAMAHA
>> XG - 12 tones).
>>
>> Here are the results of going up a quarter-tone from MIDI note A4
>> (440 Hz).
>>
>> 1200-EDO versus "whatnot" increments
>>
>> 0) 0. [Reed Organ, A=440 Hz] -0.3 cents
>> 1) 1. +1.1 cents
>> 2) 2. +1.1 cents
>> 3) 3. +2.4 cents
>> 4) 4. +3.6 cents
>> 5) 5. +5.0 cents
>> 6) 6. +6.3 cents
>> 7) 6. +6.3 cents
>> 8) 7. +7.5 cents
>> 9) 8. +8.9 cents
>> 10) 9. +10.1 cents
>> 11) 9. +10.1 cents
>> 12) 10. +11.4 cents
>> 13) 10. +11.4 cents
>> 14) 11. +12.7 cents
>> 15) 12. +14.0 cents
>> 16) 13. +15.3 cents
>> 17) 13. +15.3 cents
>> 18) 14. +16.6 cents
>> 19) 15. +17.9 cents
>> 20) 16. +19.1 cents
>> 21) 16. +19.1 cents
>> 22) 17. +20.5 cents
>> 23) 18. +21.7 cents
>> 24) 19. +23.0 cents
>> 25) 20. +24.3 cents
>> 26) 21. +25.6 cents
>> 27) 22. +26.9 cents
>> 28) 23. +28.2 cents
>> 29) 23. +28.2 cents
>> 30) 24. +29.4 cents
>> 31) 25. +30.7 cents
>> 32) 26. +32.0 cents
>> 33) 27. +33.3 cents
>> 34) 27. +33.3 cents
>> 35) 28. +34.6 cents
>> 36) 29. +35.9 cents
>> 37) 30. +37.1 cents
>> 38) 30. +37.1 cents
>> 39) 30. +37.1 cents (again!)
>> 40) 31. +38.4 cents
>> 41) 32. +39.7 cents
>> 42) 33. +41.0 cents
>> 43) 34. +42.2 cents
>> 44) 34. +42.2 cents
>> 45) 35. +43.5 cents
>> 46) 36. +44.8 cents
>> 47) 37. +46.0 cents
>> 48) 38. +47.3 cents
>> 49) 38. +47.3 cents
>> 50) 39. +49.8 cents
>>
>> Increments in cents:
>>
>> 1.4
>> 1.3
>> 1.2
>> 1.4
>> 1.3
>> 1.2
>> 1.4
>> 1.2
>> 1.3
>> 1.3
>> 1.3
>> 1.3
>> 1.3
>> 1.3
>> 1.2
>> 1.4
>> 1.2
>> 1.3
>> 1.3
>> 1.3
>> 1.3
>> 1.3
>> 1.2
>> 1.3
>> 1.3
>> 1.3
>> 1.3
>> 1.3
>> 1.2
>> 1.3
>> 1.3
>> 1.3
>> 1.2
>> 1.3
>> 1.3
>> 1.2
>> 1.3
>> 2.5
>>
>> Can anyone make heads or tails of this? I am at a loss to explain
>> the final 2.5 cent step.
>>
>> And Petr, what do you know about Roland Virtual Sound Canvas's
>> tuning abilities?
>>
>> Your composition is sounding vey good. I like it! But here is the
>> strange thing: QT Synth is starting to play the midi file from the
>> key of C# while the S-YXG50 Player is starting from the key of C! I
>> am not aware of having activated any transpositions. What do you
>> think is wrong here?
>>
>> Oz.
>>
>> ✩ ✩ ✩
>> www.ozanyarman.com
>>
>> On Oct 17, 2009, at 1:17 PM, Petr Pařízek wrote:
>>
>>>
>>>
>>> Wow, finally a topic to which I can add some contribution fron. :-
>>> D I have a hardware synth which is fully XG compatible and
>>> therefore I was always interested in many XG-related things.
>>>
>>> The XG tuning format uses a cent resolution with a maximum allowed
>>> deviation of -64..+63 cents away from 12-equal. Each retuning
>>> command affects the particular key in all octaves, which means
>>> that you can use up to 12 commands if you want to retune all the
>>> keys in all octaves. Of course, the disadvantage is that you can't
>>> retune, for example, C1 by +20 cents and C2 by -40 cents at the
>>> same time on the same MIDI channel. Scala has the ability to send
>>> these commands through MIDI or to a file, unfortunately only for
>>> retuning the channel #1. AFAIK, Scala itself allows to do this
>>> only if a 12-tone scale is loaded. You should be able to do that
>>> by using "Set synth 108". Anyway, some years ago, I made a small
>>> QBasic utility which converts plain text numbers to this kind of
>>> MIDI file, which allows me to use various tunings in one MIDI file
>>> because each MIDI channel can be tuned differently.
>>>
>>> An example might be the MIDI file which I've made in august this
>>> year. There's diaschismatic on channels 1 to 4, negri on 5 to 9,
>>> porcupine on 11 to 14, and meantone on 15 to 16.
>>> So if you have something like the XG-50 or XG-100 or another
>>> similar driver, you may try this (there's also music there, not
>>> only tuning commands):
>>> www.sendspace.com/file/t9z3ar
>>> Petr
>>>
>>>
>>>
>>>
>>> __._,
>>
>>
>>
>>   
>
>
>
> ------------------------------------
>
> You can configure your subscription by sending an empty email to one
> of these addresses (from the address at which you receive the list):
> tuning-subscribe@yahoogroups.com - join the tuning group.
> tuning-unsubscribe@yahoogroups.com - leave the group.
> tuning-nomail@yahoogroups.com - turn off mail from the group.
> tuning-digest@yahoogroups.com - set group to send daily digests.
> tuning-normal@yahoogroups.com - set group to send individual emails.
> tuning-help@yahoogroups.com - receive general help information.
> Yahoo! Groups Links
>
>
>

🔗Daniel Forro <dan.for@...>

Hi, Ozan,

thanks for detailed explanation. As I don't know how Scala sends microtune data to that synth, and also I haven't a manual with SysEx implementation for S-YXG50, and don't work with Windows and that synth, I can't help more.

But always it's possible to send PB increments directly from sequencer and measure the reaction. And a way how to check if that instrument understands standard XG SysEx tuning data would be to play a XG MIDI song which uses such control messages. There's always a possibility S-YXG50 has reduced SysEx implementation, and it can be expected that such very special features (from the point of view of common users) like microtuning were sacrificed.

Also I don't know where to download S-YXG100. I wonder you couldn't find it. I suppose we both use the same internet, so when you try Google, you will get lot of possible places, but all of them look ... hm, let's say ... not so official... S-YXG100 has additional VL physical modelling engine implemented.

Daniel Forro

On 18 Oct 2009, at 11:04 AM, Ozan Yarman wrote:

>
> Dear Daniel,
>
> Sorry for not being specific. As the title of this thread implies, I
> have measured the S-YXG50 Reed Organ output with Peterson AutoStrobe
> 590. The YAMAHA SoftSynth is, according to Petr, using the XG tuning
> standard. But it does not appear to be so.
>
> To be more precise, this is what I did:
>
> I ran SCALA, set the synth model as 108 (XG family - 12 tones) and
> chose 1200-EDO. The microphone was connected to AutoStrobe and the
> speakers giving the sound output were Alesis M1 Active Mk2. Using the
> Chromatic Keyboard in SCALA, I started from the 440 Hz A (900 cents
> from C) and travelled up a quarter-tone by 1 cent increments. The
> instrument was a Reed Organ patch from the S-YXG50 wavetable.
> Measurements indicate that SCALA's one cent increments do not match
> the increments measured. That is to say, SCALA says it is 1200-EDO,
> but S-YXG50 cannot microtune to 1200-EDO. It so appears that S-YXG50
> does not microtune by cent increments as Petr said it should.
>
> Where can I obtained S-YXG100? I did not find it on the internet.
>
> Oz.
>

🔗Ozan Yarman <ozanyarman@...>

O Daniel,

Is it not quite evident at this point that the highest S-YXG50 can
microtune is by increments of about 1.3 cents, which strikingly
implies 77 equal divisions of the equal semitone or 924-EDO?

This I glean from the fact that neither the XG synth model nor the GM
synth model selected in SCALA makes a difference with the YAMAHA Soft-
Synth. The microtuning is not at all by 1 cent increments.

Something is very wrong, tuning-wise, with S-YXG50 indeed.

Having googled around, I discovered that no YAMAHA soft-synth except S-
YXG50 will run on Windows XP.

Tough luck!

What do you think of Edirol's Virtual Sound Canvas? They say it works
on both XP and Mac OS X.

Oz.

✩ ✩ ✩
www.ozanyarman.com

On Oct 18, 2009, at 5:37 AM, Daniel Forro wrote:

> Hi, Ozan,
>
> thanks for detailed explanation. As I don't know how Scala sends
> microtune data to that synth, and also I haven't a manual with SysEx
> implementation for S-YXG50, and don't work with Windows and that
> synth, I can't help more.
>
> But always it's possible to send PB increments directly from
> sequencer and measure the reaction. And a way how to check if that
> instrument understands standard XG SysEx tuning data would be to play
> a XG MIDI song which uses such control messages. There's always a
> possibility S-YXG50 has reduced SysEx implementation, and it can be
> expected that such very special features (from the point of view of
> common users) like microtuning were sacrificed.
>
> Also I don't know where to download S-YXG100. I wonder you couldn't
> find it. I suppose we both use the same internet, so when you try
> Google, you will get lot of possible places, but all of them look ...
> hm, let's say ... not so official... S-YXG100 has additional VL
> physical modelling engine implemented.
>
> Daniel Forro
>
>
> On 18 Oct 2009, at 11:04 AM, Ozan Yarman wrote:
>
>>
>> Dear Daniel,
>>
>> Sorry for not being specific. As the title of this thread implies, I
>> have measured the S-YXG50 Reed Organ output with Peterson AutoStrobe
>> 590. The YAMAHA SoftSynth is, according to Petr, using the XG tuning
>> standard. But it does not appear to be so.
>>
>> To be more precise, this is what I did:
>>
>> I ran SCALA, set the synth model as 108 (XG family - 12 tones) and
>> chose 1200-EDO. The microphone was connected to AutoStrobe and the
>> speakers giving the sound output were Alesis M1 Active Mk2. Using the
>> Chromatic Keyboard in SCALA, I started from the 440 Hz A (900 cents
>> from C) and travelled up a quarter-tone by 1 cent increments. The
>> instrument was a Reed Organ patch from the S-YXG50 wavetable.
>> Measurements indicate that SCALA's one cent increments do not match
>> the increments measured. That is to say, SCALA says it is 1200-EDO,
>> but S-YXG50 cannot microtune to 1200-EDO. It so appears that S-YXG50
>> does not microtune by cent increments as Petr said it should.
>>
>> Where can I obtained S-YXG100? I did not find it on the internet.
>>
>> Oz.
>>
>
>
>
> ------------------------------------
>
> You can configure your subscription by sending an empty email to one
> of these addresses (from the address at which you receive the list):
> tuning-subscribe@yahoogroups.com - join the tuning group.
> tuning-unsubscribe@yahoogroups.com - leave the group.
> tuning-nomail@yahoogroups.com - turn off mail from the group.
> tuning-digest@yahoogroups.com - set group to send daily digests.
> tuning-normal@yahoogroups.com - set group to send individual emails.
> tuning-help@yahoogroups.com - receive general help information.
> Yahoo! Groups Links
>
>
>

🔗Petr Pařízek <p.parizek@...>

🔗Daniel Forro <dan.for@...>

🔗cameron <misterbobro@...>

🔗Marcel de Velde <m.develde@...>

🔗hpiinstruments <aaronhunt@...>

Oz, I'm not sure if someone else replied about this, but I
can confirm that the Windows default synth is a bit
sharp on my Dell laptop. I never measured it precisely,
but A443 Hz sounds about right. I can't believe that
wikipedia article says Apple Quicktime synth is comparable
in quality to the Windows synth! There is no comparison.
That default Windoze synth blows, big time. Is it the same
in Windows 7? I know they have updated their media player
to support .mov files (to try and kill Quicktime for Windows).
The Apple synth has far superior sounds and effects, and
it supports Soundfonts.

Cheers,
Aaron
=====

--- In tuning@yahoogroups.com, Ozan Yarman <ozanyarman@...> wrote:
>
> Here is an interesting frequency measurement experiment with the
> YAMAHA S-YXG50 SoftSynth. This elusive software synthesizer was hard
> to hunt down on the internet because it was discontinued since some
> years. Unfortunate! Since S-YXG50 is one of the sweetest sounding
> SoftSynths you can utilize under Windows XP... a far cry from that
> cheap trash most will recognize as "Microsoft GS Wavetable SW Synth".
>
> First, because I think you ought to try S-YXG50, I must tell you how
> to obtain and install it.
>
> YAMAHA S-YXG50 can be downloaded here:
>
> http://www.nodevice.com/driver/S-YXG50/get43254.html
>
> The driver also requires its own wavetables to be seperately installed
> from the Add/Remove Hardware application in the Control Panel. The
> wavetables can be downloaded here:
>
> http://en.wikipedia.org/wiki/Software_synthesizer
>
> Once this neat little SoftSynth is fully installed with its proper
> wavetables, "YAMAHA XG WDM SoftSynthesizer" must be selected as the
> default MIDI playback device from within the Sound Devices in Control
> Panel.
>
> I decided to use S-YXG50 not only because of its beautiful wavetable
> sounds, but also because that trashy Microsoft GS Synth generates MIDI
> note frequencies 1/10th an equal semitone higher than specified!
>
> That is to say, MIDI note A4 ordinarily set and shown as 440 Hz yields
> a 443 Hz frequency from the speakers. Strange!
>
> Or else, this is a phenomenon restricted to my Parallels Desktop
> configuration. The PD team was at a loss, however, to understand what
> was going on...
>
> Anyway, I immediately tried out some of the new sounds of S-YXG50 in
> SCALA with the tuning chosen as Yarman-36c. I found out that Clarinet
> and Reed Organ are the most dependable samples. Other instruments
> sound too vibrato and their timbres makes the pitch very ambiguous.
> Measurements with the AP Tuner confirmed this.
>
> AP Tuner can be downloaded here:
>
> http://www.aptuner.com/cgi-bin/aptuner/apmain.html
>
> Alas, AP Tuner also confirmed that S-YXG50 pitches did not exactly
> match SCALA's tuning data. Evidently, there was something awry with
> the pitch-bends.
>
> Therefore, I decided to measure the pitches of S-YXG50's Reed Organ
> tuned to as 12-tone subset of Yarman36c with the PETERSON AutoStrobe
> 590 tuner unit.
>
> I completed the pitch measurements at about three o'clock in the
> morning (GMT+2) in a very silent environment. My experimental setup
> was as follows:
>
> 1. SCALA was running in Windows XP Virtual Machine (Parallels Desktop
> 3.0),
> 2. The 12-tone subset of Yarman-36c quantized to 12000-EDO was the
> chosen tuning,
> 3. PETERSON AutoStrobe 590 was up and running with the same tuning set,
> 4. A Line-Out from my Macbook Pro was connected to Alesis M1 Active
> Mk2 speakers,
> 5. One CAD25 Microphone resting on its own stand was connected to
> PETERSON AutoStrobe 590.
>
> Here are the results of my measurements:
>
> C in S-YXG50 is sounding -0.8 cents compared to PETERSON
> C# in S-YXG50 is sounding -0.7 cents compared to PETERSON
> D in S-YXG50 matches D in PETERSON
> Eb in S-YXG50 is sounding -0.4 cents compared to PETERSON
> E in S-YXG50 is sounding -0.7 cents compared to PETERSON
> F in S-YXG50 is sounding -0.4 cents compared to PETERSON
> F# in S-YXG50 is sounding -0.8 cents compared to PETERSON
> G in S-YXG50 is sounding -0.5 cents compared to PETERSON
> G# in S-YXG50 is sounding -1.2 cents compared to PETERSON
> A in S-YXG50 is sounding -0.25 cents compared to PETERSON
> Bb in S-YXG50 is sounding +0.15 cents compared to PETERSON
> B in S-YXG50 is sounding +0.2 cents compared to PETERSON
>
> I skip the octave, because there is no way to insert stretched octave
> cent offset into AutoStrobe 590. Suffice it to say, C in AutoStrobe
> must be raised by +3.5 cents for it to match the c of S-YXG50.
>
> The Yarman-36c 12-tone subset quantized to 12000-EDO in cents was:
>
> 0.0
> 99.1
> 200
> 302
> 400
> 502
> 599.9
> 702
> 802.2
> 901.9
> 1004
> 1101.9
> 1203.9
>
> Cent offsets entered into PETERSON AutoStrobe 590 was:
>
> C -1.9
> C# -2.8
> D -1.9
> Eb 0.1
> E -1.9
> F 0.1
> F# -2.0
> G 0.1
> G# 0.3
> A 0.0
> Bb 2.1
> B 0.0
>
> Note that A was set as 440 Hz in both SCALA (901.9 cents) and PETERSON.
>
> Measured pitches in cents in reference to D were:
>
> 3.5
> 100.8
> 200.0
> 301.6
> 403.2
> 500.9
> 603.2
> 700.7
> 803.1
> 902.6
> 1003.3
> 1105.8
> (1207.4)
>
> Comparison of SCALA's Yarman-36c 12-tone quantized subset with the
> measured pitches in reference to D as follows:
>
> SCALA meas. SCALA is higher by
>
> 0 -0.8 0.8 cents
> 99.1 98.4 0.7 cents
> 200 200.0 0.0 cents
> 302 301.6 0.4 cents
> 400 399.3 0.7 cents
> 502 501.6 0.4 cents
> 599.9 599.1 0.8 cents
> 702 701.5 0.5 cents
> 802.2 801.0 1.2 cents
> 901.9 901.7 0.3 cents
> 1004 1004.2 -0.1 cents
> 1101.9 1102.1 -0.2 cents
> 1203.9 1206.6 -2.7 cents
>
> I am at a loss to explain these discrepancies. At first I thought I
> was dealing with steps of 768-EDO (64 equal divisions of 100 cents),
> but higher pitches deviate from it. What do you think is going on with
> the pitch-bends of S-YXG50?
>
> Cordially,
> Oz.
>
> â© â© â©
> www.ozanyarman.com
>

🔗Petr Parízek <p.parizek@...>

🔗Ozan Yarman <ozanyarman@...>

I concur with you that Apple QT is hyper light years ahead of
Microsoft's decrepit GM synth. But what is the maximum tuning
resolution of Apple? Is it 4096 equal steps to the equal semitone?

I don't know the situation with Windows 7. I only use XP because of
Scala, FTS, Encore, Mus2okur and several other worthwhile gems.

Oz.

✩ ✩ ✩
www.ozanyarman.com

On Oct 19, 2009, at 1:24 AM, hpiinstruments wrote:

> Oz, I'm not sure if someone else replied about this, but I
> can confirm that the Windows default synth is a bit
> sharp on my Dell laptop. I never measured it precisely,
> but A443 Hz sounds about right. I can't believe that
> wikipedia article says Apple Quicktime synth is comparable
> in quality to the Windows synth! There is no comparison.
> That default Windoze synth blows, big time. Is it the same
> in Windows 7? I know they have updated their media player
> to support .mov files (to try and kill Quicktime for Windows).
> The Apple synth has far superior sounds and effects, and
> it supports Soundfonts.
>
> Cheers,
> Aaron
> =====
>
> --- In tuning@yahoogroups.com, Ozan Yarman <ozanyarman@...> wrote:
>>
>> Here is an interesting frequency measurement experiment with the
>> YAMAHA S-YXG50 SoftSynth. This elusive software synthesizer was hard
>> to hunt down on the internet because it was discontinued since some
>> years. Unfortunate! Since S-YXG50 is one of the sweetest sounding
>> SoftSynths you can utilize under Windows XP... a far cry from that
>> cheap trash most will recognize as "Microsoft GS Wavetable SW Synth".
>>
>> First, because I think you ought to try S-YXG50, I must tell you how
>> to obtain and install it.
>>
>> YAMAHA S-YXG50 can be downloaded here:
>>
>> http://www.nodevice.com/driver/S-YXG50/get43254.html
>>
>> The driver also requires its own wavetables to be seperately
>> installed
>> from the Add/Remove Hardware application in the Control Panel. The
>> wavetables can be downloaded here:
>>
>> http://en.wikipedia.org/wiki/Software_synthesizer
>>
>> Once this neat little SoftSynth is fully installed with its proper
>> wavetables, "YAMAHA XG WDM SoftSynthesizer" must be selected as the
>> default MIDI playback device from within the Sound Devices in Control
>> Panel.
>>
>> I decided to use S-YXG50 not only because of its beautiful wavetable
>> sounds, but also because that trashy Microsoft GS Synth generates
>> MIDI
>> note frequencies 1/10th an equal semitone higher than specified!
>>
>> That is to say, MIDI note A4 ordinarily set and shown as 440 Hz
>> yields
>> a 443 Hz frequency from the speakers. Strange!
>>
>> Or else, this is a phenomenon restricted to my Parallels Desktop
>> configuration. The PD team was at a loss, however, to understand what
>> was going on...
>>
>> Anyway, I immediately tried out some of the new sounds of S-YXG50 in
>> SCALA with the tuning chosen as Yarman-36c. I found out that Clarinet
>> and Reed Organ are the most dependable samples. Other instruments
>> sound too vibrato and their timbres makes the pitch very ambiguous.
>> Measurements with the AP Tuner confirmed this.
>>
>> AP Tuner can be downloaded here:
>>
>> http://www.aptuner.com/cgi-bin/aptuner/apmain.html
>>
>> Alas, AP Tuner also confirmed that S-YXG50 pitches did not exactly
>> match SCALA's tuning data. Evidently, there was something awry with
>> the pitch-bends.
>>
>> Therefore, I decided to measure the pitches of S-YXG50's Reed Organ
>> tuned to as 12-tone subset of Yarman36c with the PETERSON AutoStrobe
>> 590 tuner unit.
>>
>> I completed the pitch measurements at about three o'clock in the
>> morning (GMT+2) in a very silent environment. My experimental setup
>> was as follows:
>>
>> 1. SCALA was running in Windows XP Virtual Machine (Parallels Desktop
>> 3.0),
>> 2. The 12-tone subset of Yarman-36c quantized to 12000-EDO was the
>> chosen tuning,
>> 3. PETERSON AutoStrobe 590 was up and running with the same tuning
>> set,
>> 4. A Line-Out from my Macbook Pro was connected to Alesis M1 Active
>> Mk2 speakers,
>> 5. One CAD25 Microphone resting on its own stand was connected to
>> PETERSON AutoStrobe 590.
>>
>> Here are the results of my measurements:
>>
>> C in S-YXG50 is sounding -0.8 cents compared to PETERSON
>> C# in S-YXG50 is sounding -0.7 cents compared to PETERSON
>> D in S-YXG50 matches D in PETERSON
>> Eb in S-YXG50 is sounding -0.4 cents compared to PETERSON
>> E in S-YXG50 is sounding -0.7 cents compared to PETERSON
>> F in S-YXG50 is sounding -0.4 cents compared to PETERSON
>> F# in S-YXG50 is sounding -0.8 cents compared to PETERSON
>> G in S-YXG50 is sounding -0.5 cents compared to PETERSON
>> G# in S-YXG50 is sounding -1.2 cents compared to PETERSON
>> A in S-YXG50 is sounding -0.25 cents compared to PETERSON
>> Bb in S-YXG50 is sounding +0.15 cents compared to PETERSON
>> B in S-YXG50 is sounding +0.2 cents compared to PETERSON
>>
>> I skip the octave, because there is no way to insert stretched octave
>> cent offset into AutoStrobe 590. Suffice it to say, C in AutoStrobe
>> must be raised by +3.5 cents for it to match the c of S-YXG50.
>>
>> The Yarman-36c 12-tone subset quantized to 12000-EDO in cents was:
>>
>> 0.0
>> 99.1
>> 200
>> 302
>> 400
>> 502
>> 599.9
>> 702
>> 802.2
>> 901.9
>> 1004
>> 1101.9
>> 1203.9
>>
>> Cent offsets entered into PETERSON AutoStrobe 590 was:
>>
>> C -1.9
>> C# -2.8
>> D -1.9
>> Eb 0.1
>> E -1.9
>> F 0.1
>> F# -2.0
>> G 0.1
>> G# 0.3
>> A 0.0
>> Bb 2.1
>> B 0.0
>>
>> Note that A was set as 440 Hz in both SCALA (901.9 cents) and
>> PETERSON.
>>
>> Measured pitches in cents in reference to D were:
>>
>> 3.5
>> 100.8
>> 200.0
>> 301.6
>> 403.2
>> 500.9
>> 603.2
>> 700.7
>> 803.1
>> 902.6
>> 1003.3
>> 1105.8
>> (1207.4)
>>
>> Comparison of SCALA's Yarman-36c 12-tone quantized subset with the
>> measured pitches in reference to D as follows:
>>
>> SCALA meas. SCALA is higher by
>>
>> 0 -0.8 0.8 cents
>> 99.1 98.4 0.7 cents
>> 200 200.0 0.0 cents
>> 302 301.6 0.4 cents
>> 400 399.3 0.7 cents
>> 502 501.6 0.4 cents
>> 599.9 599.1 0.8 cents
>> 702 701.5 0.5 cents
>> 802.2 801.0 1.2 cents
>> 901.9 901.7 0.3 cents
>> 1004 1004.2 -0.1 cents
>> 1101.9 1102.1 -0.2 cents
>> 1203.9 1206.6 -2.7 cents
>>
>> I am at a loss to explain these discrepancies. At first I thought I
>> was dealing with steps of 768-EDO (64 equal divisions of 100 cents),
>> but higher pitches deviate from it. What do you think is going on
>> with
>> the pitch-bends of S-YXG50?
>>
>> Cordially,
>> Oz.
>>
>> âœ© âœ© âœ©
>> www.ozanyarman.com
>>
>
>
>
>
> ------------------------------------
>
> You can configure your subscription by sending an empty email to one
> of these addresses (from the address at which you receive the list):
> tuning-subscribe@yahoogroups.com - join the tuning group.
> tuning-unsubscribe@yahoogroups.com - leave the group.
> tuning-nomail@yahoogroups.com - turn off mail from the group.
> tuning-digest@yahoogroups.com - set group to send daily digests.
> tuning-normal@yahoogroups.com - set group to send individual emails.
> tuning-help@yahoogroups.com - receive general help information.
> Yahoo! Groups Links
>
>
>

🔗Ozan Yarman <ozanyarman@...>

Petr,

I work rather well with the Chromatic Keyboard in SCALA. I am not
comfortable with being restricted to 12-tones per MIDI session unless
everything is before my eyes. I like to see the whole layout in one
screen, with one simple interface. The Chromatic Keyboard does that. I
can see the cents, the intervals, play the chords, listen to the
microtonal harmonies with visual aid.

Hence, I am not using a Midi file, and yes, I work with pitch bends
over 16 channels simultaneously.

What has been confirmed up to this point is that S-YXG50 does not seem
to respond to the MTS protocol. Measurements show that the tuning
resolution is by about 1.3 cent increments. This fits nothing I know
of, but Cameron did provide a clue as to what it might be.

Anyway, I used to manipulate the XG cent offsets manually back when I
was using XG-Works. This brings back several old and unpleasant
memories.

What do you think of Timidity Synth's tuning resolution? Can you make
heads or tails of my measurements?

Oz.

✩ ✩ ✩
www.ozanyarman.com

On Oct 18, 2009, at 11:56 AM, Petr Pařízek wrote:

>
>
> Dear Ozan,
>
> I'm sorry to say this but Daniel was right.
> From what you're describing, I can see that you're actually sending
> pitch-bend messages to the MIDI file, not XG SysEx commands. If you
> want to try out XG SysEx commands for microtuning, then you can do
> the following:
> Load a 12-tone scale with a period of 2/1 and type "Set Synth 108".
> Then type "Send/screen" to see if the tuning deviations fit in the
> -64..+63 cent range.
> If they don't, type "Set adjustment" and try "Send/screen" again.
> If it still doesn't work, then load another 12-tone scale which does.
> Finally, type something like "Send/file example", which will create
> "example.mid" where the first MIDI channel will be retuned to the
> specified scale (I'm going to contact Manuel to tell him how to make
> it possible for the other channels).
> This means there won't be any notes in the MIDI file, only the
> tuning commands. So you can then connect an external MIDI keyboard
> to your sound card, run the MIDI file to retune the channel #1, and
> then try playing channel #1 on the external keyboard to hear if the
> microtuning works.
>
> BTW: To answer your question: I used an XG-specific SysEx command
> (that's why it didn't work in your other player) which shifts the
> channels 1 to 4 a semitone lower.
>
> And then, I don't know anything about Roland Virtual Sound Canvass
> tuning capabilities.
>
> Petr
>
>

🔗Daniel Forro <dan.for@...>

Maybe answer is Pitch Bend resolution and setting of a range. Problem can be limitation in synth engine, or in the way how it process incoming MIDI PB messages. I would suppose Scala should transmit tuning data properly. But...

According to MIDI protocol PB resolution can be 14 bit, that's 16384 steps. Even with PB range one octave we should get theoretical resolution 0.146484375 C for one step, with whole tone setting 0.02441 C. It's questionable if tone generator can retune with such accuracy, I doubt. All those GM/GS/XG machines were produced for the purpose of pop music mainly.

Problem is many producers use just 7 bit resolution only, which is 128 steps. Even with PB range whole tone we still get rough 3.125 C for one step. Some makers use 8, 9, maybe more bits, so one step can be 1.5625, 0.78125 C...

Anyway it would be interesting to know how Scala deals with such limitations... Can somebody who works with Scala PB tuning analyze what's transmitted? I don't use Scala or Pitch Bend for microtones...

Daniel Forro

On 20 Oct 2009, at 2:03 PM, Ozan Yarman wrote:

>
>
> Petr,
>
> I work rather well with the Chromatic Keyboard in SCALA. I am not > comfortable with being restricted to 12-tones per MIDI session > unless everything is before my eyes. I like to see the whole layout > in one screen, with one simple interface. The Chromatic Keyboard > does that. I can see the cents, the intervals, play the chords, > listen to the microtonal harmonies with visual aid.
>
> Hence, I am not using a Midi file, and yes, I work with pitch bends > over 16 channels simultaneously.
>
> What has been confirmed up to this point is that S-YXG50 does not > seem to respond to the MTS protocol. Measurements show that the > tuning resolution is by about 1.3 cent increments. This fits > nothing I know of, but Cameron did provide a clue as to what it > might be.
>
> Anyway, I used to manipulate the XG cent offsets manually back when > I was using XG-Works. This brings back several old and unpleasant > memories.
>
> What do you think of Timidity Synth's tuning resolution? Can you > make heads or tails of my measurements?
>
> Oz.