Hammond Temperament in GO?

[gblatt]

Hi everbody,
question of a newbie:
is it possible to implement the original
Hammond temperament

http://www.electricdruid.net/index.php?page=info.hammond#calculations

into the temperaments set
of GrandOrgue?

Thanks for help!

Günter

 

[e9925248]

I must admin, that I don't know Hamond organs and therefore don't know, if they require something special.

GO temperaments only specify the pitch of the various notes,eg:

c - 0 cent relative c of equal temperament
c# - 1.96 cent higher than c# of equal temperament
d 1.96 lower than d of equal temperament

You can simulate it via the OrganDialog.

For further details:
http://www.magle.dk/music-forums/16573-go-midi-output-2.html

If you can provide the necessary number, we can add it to GO.

 

[gblatt]

Thank you so far, I will go for the numbers and tell you.
The Hammond Organ was/is a famous electronic organ
with a unique temperament due to the very special
way of tone generation (the tonewheel, see link above).
I play some church organ sample sets with GO and would like
them "sound" like a Hammond. Eg to play "A whiter shade of pale" Procol Harum
on a virtual church organ with Hammond temperament, which I did on a real church organ
in school times many years ago ....

 

[gblatt]

Here are the numbers (I hope)

C	8,8005624959
C#	10,9287919419
D	-3,4273726208
D#	1,7262149258
E	2,7837006998
F	13,5241088364
F#	-0,5886585127
G	-0,3139512068
G#	17,012132906
A	0
A#	7,8216822552
B	17,0492118517
C	17,5011249918

 

[L.Palo]

Hi!

Those numbers are unfortunately not correct according to the link given above. I get the values for the octaves (the first seven) with regular increments to:

C -0,57648076882506
C# -0,68380450083854
D 0,19961552854962
D# -0,08832376157198
E -0,14473747456411
F -0,68088386412621
F# 0,02609638998286
G 0,0201567086584
G# -0,70698025410888
A 0
A# -0,28526108384531
B -0,59256010255413

Then for the last (highest) octave it should be:

C 1,27411700126128
C# 1,9810972553706
D 1,97515757404616
D# 1,2480206112783
E 1,95500086538723
F 1,66973978154221
F# 1,36244076283315

The irregular last octave cannot be put into the normal temperament system of GO but could be individually added in the organ settings dialog.

All the information and calculations are based upon the source given above but since I've no experience with the real instrument I cannot really validate the source... It does seem a bit odd to me to have a F# as the highest note, but it's a harmonic so...

Kind regards

Lars P

 

[e9925248]

Adding support for a irregular last octave would not be the problem. Hammond seems really to be different, so could you please try it first:

* Load a properly annotated organ (Pitea) and switch to equal temperament.
* In the organ take a few stops and fill in for all pipes the apropriate number (from abvoe) for the tone into the tuning field.

You should probably try with a set of pipes (eg. 8', 4', 2', a mixture/aliqute) to be able to test, if it sound correct. If we want to implement the differnt last octave, we need to find factors starting with G.

The only requirement, that a temerament is implementable in GO is, that that pipes behind one key use the same tuning value.

 

[scush]

Hi
To model the Hammond in GO would require 20 enclosures for the drawbars, if this is possible ?.

To get the sound for "a whiter shade of pale", sine waves, or very pure flutes at 16', 5_1/3', 8' and 4' at the same amplitude are required.

I believe it's the lack of harmonics that give the sound, not so much the tuning.

PS

The Barton organ will not load at revision 1203.

 

[L.Palo]

20 enclosures? Please explain. There are 9 drawbars that actually are volume controllers for the pitches they represent, then there should be a master volume controller... That should be 9 enclosures within one enclosure? Also from the link above I understand that a hammond really just uses 91 sine wave pipes from which the individual drawbars will set the playback volume for that pitch range (stop)? They cannot be borrowed with REF then, so the drawstops/bars would need to re-use the pipes instead...

This is a bit beyond my primary interest, but I guess that GO as a software should make it possible to accurately simulate different organ types in a sensible way. One should remember though that GO is a sample based VPO, thus one should perhaps really record a real hammond instead of generating the sounds in a synth...

Kind regards

Lars P

 

[scush]

Hi

9 for each manual 2 for the pedals so thats 20 + 1 for the swell pedal.
If the enclosures could be placed anywhere on the gui, images of drawbars could be used for them.
Each drawbar could need it's own rank.
Using wave samples to model such an organ has it's limitations, but it could be worth having a go.
The last time I checked, the max amount of enclosures was 14, this limitation seemed to be caused by the available space in the floating manuals panel.

john.

PS correction barton will not load at rev 1191

 

[e9925248]

The maxiums are usually derived from the HW1 Spec - most can be increased, if it is really necessary.

PS correction barton will not load at rev 1191

I was not aware, what kind of things people try to use:

  [Panel002Coupler004] ; Solo to Great2T Sub

A single search/replace (insert a line break between ] and ; ) can be used to fix the ODF.
As the Barton ODF contains some known bugs (CombinationsStoreNonDisplayedDrawstops=N, a few incrorect coupler), maybe everyone will benefit from the release of an updated ODF.

 

[scush]

I agree, a new barton odf is required before the next release of grandorgue.

 

[ggoode.sa]

Hi Guys,

Yes, a new Barton ODF with all the fixes and updated functionality will be worked on sometime this month - so if you have specific things that you remember need updating or changing please send them to me.

Thanks

GrahamG

 

[e9925248]

1) Seperate comments from content lines
2) Store invisible objects (maybe also the store tremulant setting should be changed)
3) There are 1-2 incorrect couplers (sorry, I don't remember which ones)
4) The toy buttons should be converted to effect stops, while the associated manuals should go away. This allows a flexible MIDI assignment (currenty they only support note on).

Effect stops are stops with exactly one pipe. An example from Lars:

[Stop401]
Name=Motor noise
GCState=-1
NumberOfLogicalPipes=1
NumberOfAccessiblePipes=1
FirstAccessiblePipeLogicalPipeNumber=001
FirstAccessiblePipeLogicalKeyNumber=001
Percussive=N
Displayed=N
Pipe001=.\Noises\Motor\Motor.wav
........

5) If still time is left: Look at the warnings (and report incorrect warnings from GO)

 

[scush]

Hi
I have spent a couple of days cleaning up the barton to get it to work in revision 1203.
One thing I noticed was the number of image's for the swell etc,needs to be 1 higher than the amount of bitmaps or the last one is not used.

As this has taken so long I think that I should put it here.
There are 2 remaining warnings I can't resolve.
These are associated with general 10

 

[scush]

Hi again
This one has no warnings

 

[e9925248]

One thing I noticed was the number of image's for the swell etc,needs to be 1 higher than the amount of bitmaps or the last one is not used.

This looks like a (low priority) bug in GO. If you want 12 bitmaps, specify 12 in the ODF.

Regarding the warnings:
GO currently warns about anything, it does not use. If you think, that a reported entry would be useful or the warning is incorrect, please report it.

 

[scush]

Hi
There were that many warnings with this one that GO hung before whatever the fatal error was, could be seen.
If there is a required value missing it seems that many warnings are generated, this occurs with an odf that does not give warnings.

The reported line numbers did not correspond to the actual one's seen in ms word.
the difference in the 2 files I have uploaded demonstrate this.
But I would have never found all the error's without this new logging.


Thank's


john

 

[e9925248]

This looks like a (low priority) bug in GO. If you want 12 bitmaps, specify 12 in the ODF.

Should be fixed. I had to hand craft a missing image for the default GO enclosures (src/grandorgue/images/Enclosure*.png).

If somebody does not like it, feel free to submit improved versions, but please be aware, that some metrics of the image and its content are not changeable.

 

[scush]

Hi

Rev1204

Your image looks very good to me.

After changing the number of images from 13 to12 in the 4 required place's barton loaded ok.
However warnings were given that stop 133 in all of the genenerals was not used, these warnings did not occur after the cmb file had been deleted.

I then changed one of the image number's back to 13, all sorts of warnings are now given about unused items,but the cause of the crash is not shown.

It seems the loging prevents the cause of fatal errors from being seen, example, [SetterElement002] does not = type=GC

 

[e9925248]

However warnings were given that stop 133 in all of the genenerals was not used, these warnings did not occur after the cmb file had been deleted.

Bug confirmed: If a CMB file is in use and an element (divisional/general) contains a content in the ODF, incorrect warnings about the combination content can be reported.

I then changed one of the image number's back to 13, all sorts of warnings are now given about unused items,but the cause of the crash is not shown.

It seems the loging prevents the cause of fatal errors from being seen, example, [SetterElement002] does not = type=GC

I changed it in my Barton version: It prints a error message box, that the bitmap sizes does not match for the enclosure section.
This is correct, since the default bitmap are the builtin GO enclosure images (see GO reference) - so the last bitmap is taken from it, which result in the size error.