Author |
Message |
Chet
Joined: Nov 19, 2004 Posts: 231 Location: Lititz,PA,USA
Audio files: 7
G2 patch files: 35
|
|
Back to top
|
|
|
Fozzie
Joined: Jun 04, 2004 Posts: 875 Location: Near Wageningen, the Netherlands
Audio files: 8
G2 patch files: 49
|
Posted: Sun Apr 30, 2006 8:56 am Post subject:
|
|
|
FANTASTIC!!! Although you really shouldn't post patches like this without warnings for sudden bouts of expansion board craving. I fear your awesome vibrato/leslie fx part will be cannibalized.
Thank you for this great patch. |
|
Back to top
|
|
|
Chet
Joined: Nov 19, 2004 Posts: 231 Location: Lititz,PA,USA
Audio files: 7
G2 patch files: 35
|
Posted: Mon May 01, 2006 7:21 pm Post subject:
|
|
|
Thanks, I'm glad you liked it. |
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18198 Location: Durham, NC
Audio files: 213
G2 patch files: 60
|
Posted: Tue May 02, 2006 3:11 pm Post subject:
|
|
|
Holy Tamole. This is a great Hammond! I love the distortion you can load in with it. Sounds very realistic - not just to the "idea" Hammond, but to a real one with all that distortion and noise in the Leslie.
This would make a great organ for accompanying those old radio soap operas. _________________ --Howard
my music and other stuff |
|
Back to top
|
|
|
Falk
Joined: Mar 07, 2006 Posts: 75 Location: Sweden
G2 patch files: 29
|
Posted: Mon Apr 30, 2007 3:59 am Post subject:
A couple of questions.... |
|
|
Hello Chet.
As a B3 noob I've got some questions for you. It is a wonderful patch so no criticism intended, I just don't understand some of it
According to Chet Smith in a Keyboard Mag B3 Shootout article, the percussion should be mixed in after vibrato/chorus. The G2 preset YetAnotherOrgan.pch2 does this. Do you know which is right?
Everything I read about Leslies talk about 800 Hz crossover filter. You have a 1.05 kHz crossover and the YetAnotherOrgan.pch2 has a LP filter at 390 Hz for the bass rotor and a HP at 930 Hz for the horn. Does it make any real difference (except to the real Hammond aficionados)?
There are some phase coherency issues. Are your implementation of foldback an attempt to try to solve this? Again, the YetAnotherOrgan.pch2 uses a different implementation using the pitch inputs to do the foldback. That solution would seem to need oscillators with sync to fix the foldback phase coherency issue.
When playing octaves you can get another phase issue that doesn't seem to be solvable at all. Even if you sync all osc within a note there is no way that I see to make oscillators of different notes to sync up. Have I missed anything?
Regards
Falk |
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18198 Location: Durham, NC
Audio files: 213
G2 patch files: 60
|
Posted: Mon Apr 30, 2007 12:43 pm Post subject:
|
|
|
I don't know about all of these B3 clones and stuff, but I do remember back in the sixties it was well accepted that no B3 sounded exactly like any other B3. That's why so many of the successful players lugged around their instruments to play at places where there were B3s already there. Similarly, some jazz Hammond players liked to play continually at the same club because the organ was to their liking.
I never had enough bread to buy my own B3, but I loved to play them. I enjoyed every one I played. The were all very much unique.
The perfect authentic B3 is sorta like the holy grail. It can never be found because there is no perfect authentic B3. Besides, if you have ever heard a real Leslie rotating speaker you'll realize that you can't emulate that sound from a static pair of speakers. Well, you can emulate anything, but it won't be convincing or perfect. Still, its amazing what the G2 can do, and you can tweak and modify this and other patches to your hearts delight.
I do know a bit more about Steinway pianos. No two sound or play the same. I've played 4 brand new ones that were identical on the show room floor. All different. How could one have a perfect emulation? I've heard impressive sampled pianos, but as the song says, "There ain't nothin' like the real thing baby." _________________ --Howard
my music and other stuff |
|
Back to top
|
|
|
Wout Blommers
Joined: Sep 07, 2003 Posts: 4529 Location: The Hague - The Netherlands
Audio files: 123
G2 patch files: 12
|
Posted: Mon Apr 30, 2007 1:01 pm Post subject:
|
|
|
mosc wrote: | ... I do know a bit more about Steinway pianos. No two sound or play the same. ... |
That's why some people make samples from different Steinways and use only those parts of the 'keyboard' which sounds the best, in their opinions. They always joke about their 'Frankensteinway piano'.
Wout |
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18198 Location: Durham, NC
Audio files: 213
G2 patch files: 60
|
Posted: Mon Apr 30, 2007 1:17 pm Post subject:
|
|
|
Wout Blommers wrote: | 'Frankensteinway piano'. |
_________________ --Howard
my music and other stuff |
|
Back to top
|
|
|
Falk
Joined: Mar 07, 2006 Posts: 75 Location: Sweden
G2 patch files: 29
|
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18198 Location: Durham, NC
Audio files: 213
G2 patch files: 60
|
Posted: Tue May 01, 2007 6:39 am Post subject:
|
|
|
Good points, Falk. You are right, of course. Correct modeling of the device is pretty important in coming up with a useful substitute. It also helps in understanding the sound. _________________ --Howard
my music and other stuff |
|
Back to top
|
|
|
Falk
Joined: Mar 07, 2006 Posts: 75 Location: Sweden
G2 patch files: 29
|
Posted: Tue May 01, 2007 11:50 am Post subject:
|
|
|
Hello again Mosc.
Didn't mean to come on all holier then thou
I just got a bit caught up in my research and programming
There is a distinctly ugly solution using 12 mono LFO's in the VA (voice area) to sync the sine oscillators. but that will (I think) push CPU and/or memory well above 50% for a Hammond (even without Chet's stepped drawbars). It will also only properly sync the 16', 8', 4', 2' and 1' oscillators.
At that performance point we need an expanded G2 to get just 7 voices with a Leslie.
With distortion and Leslie going full tilt I don't think anyone is going to notice the phase problems anyway. It might be a problem for a cleaner Hammond sound though.
At this point it is purely a programming/optimizing exercise for me.
/Falk |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24083 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Tue May 01, 2007 11:58 am Post subject:
|
|
|
Maybe syncing all oscillators with the gate output of the monokey module would work ? Didn't try, but if I remember correctly it will go up on the first key pressed to go down only when all keys are released. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
Falk
Joined: Mar 07, 2006 Posts: 75 Location: Sweden
G2 patch files: 29
|
Posted: Tue May 01, 2007 12:55 pm Post subject:
Re: syncing on monokey |
|
|
Hello Jan.
Sorry, won't do. At least not in the general case.
No keys pressed. You take a new cord and all oscillators sync up. In theory.
One of the keys will be the first. At that point the oscillators sync at the last frequency that that poly voice had. Now the next key in your chord is processed and the oscillators for that poly voice change frequency. Not in phase anymore
Man, I wish it was that simple....
Thanks for your nice mod-n building block by the way.
/Falk |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24083 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Tue May 01, 2007 1:16 pm Post subject:
Re: syncing on monokey |
|
|
Falk wrote: | Now the next key in your chord is processed and the oscillators for that poly voice change frequency. Not in phase anymore
|
Ok ... all the wheels sync ... so that would need at least 12 times the amount of oscs ...
Quote: |
Thanks for your nice mod-n building block by the way.
|
Hope it's useful for something _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24083 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Tue May 01, 2007 1:21 pm Post subject:
|
|
|
Hey .. that should also mean that there have to be integer relations between the tones ... which then in turn means that a Hammond has some non equal tempered tuning ... so how is it tuned ? _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24083 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Tue May 01, 2007 1:36 pm Post subject:
|
|
|
Ok, I found :
Quote: | The pitches approximate even-tempered tuning, (it's done with integer (rational?) math after all). |
Apparently the approximation is close enough, it's not an issue to give many hits in Google - still for the low notes I'd expect it to be not so exact. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24083 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Tue May 01, 2007 2:00 pm Post subject:
|
|
|
Also I found :
Quote: | The deviation of the Hammond's tuning from exact equal temperament is sufficient that irregularities in the progression of beats of fourths and fifths would be noticeable, if these beats were not avoided through a some distinctive features of the Hammond. |
and this is very instructive as well, it explains how the thing works and that 91 wheels are needed and 24 gears.
The gears could mabe be done using delay lines, but hell .. 24 ... _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18198 Location: Durham, NC
Audio files: 213
G2 patch files: 60
|
Posted: Tue May 01, 2007 4:41 pm Post subject:
|
|
|
If the phase shifts are static, why worry about it? You'll never hear it. _________________ --Howard
my music and other stuff |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24083 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Tue May 01, 2007 5:14 pm Post subject:
|
|
|
mosc wrote: | If the phase shifts are static, why worry about it? You'll never hear it. |
You are right, why do I keep forgetting that all the time
Only in the special case that the tones have the same pitch the absolute phase relation matters for the perceived volume. It changes just the phase of the beating frequency which is a bit special at 0 Hz as it will not cycle anymore.
BTW for those interested I posted a Hammond tuning in the g2 scala thread. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
Falk
Joined: Mar 07, 2006 Posts: 75 Location: Sweden
G2 patch files: 29
|
Posted: Wed May 02, 2007 4:45 am Post subject:
|
|
|
Thank you Jan for those links.
I'll try to get a proof of concept patch posted tomorrow for the 12 mono LFO sync approach. It will not be pretty and I don't think it will be paticullary useful either.
/Falk |
|
Back to top
|
|
|
Chet
Joined: Nov 19, 2004 Posts: 231 Location: Lititz,PA,USA
Audio files: 7
G2 patch files: 35
|
Posted: Wed May 02, 2007 10:26 am Post subject:
Re: A couple of questions.... |
|
|
Falk wrote: | Hello Chet.
As a B3 noob I've got some questions for you. It is a wonderful patch so no criticism intended, I just don't understand some of it
According to Chet Smith in a Keyboard Mag B3 Shootout article, the percussion should be mixed in after vibrato/chorus. The G2 preset YetAnotherOrgan.pch2 does this. Do you know which is right?
Everything I read about Leslies talk about 800 Hz crossover filter. You have a 1.05 kHz crossover and the YetAnotherOrgan.pch2 has a LP filter at 390 Hz for the bass rotor and a HP at 930 Hz for the horn. Does it make any real difference (except to the real Hammond aficionados)?
There are some phase coherency issues. Are your implementation of foldback an attempt to try to solve this? Again, the YetAnotherOrgan.pch2 uses a different implementation using the pitch inputs to do the foldback. That solution would seem to need oscillators with sync to fix the foldback phase coherency issue.
When playing octaves you can get another phase issue that doesn't seem to be solvable at all. Even if you sync all osc within a note there is no way that I see to make oscillators of different notes to sync up. Have I missed anything?
Regards
Falk |
Hi Falk,
Thanks for noticing the patch. I've always liked Hammond organs. In the late 70's I used to be a traveling electronic technician, fixing organs of all types in the Philly area (no kidding). I visited a lot of downtown churches that had Hammonds in them, some of them meticulously maintained.
I'll try to answer your questions.
I had never heard about the percussion being mixed in after the vibrato/chorus. I just assumed it came beforehand, since it's generated by the 9th bus on the keyboard, which is usually where the 1' signal comes out. But I could easily be wrong.
The crossover frequency was adjusted to my ears. No attempt was made to actually sound exactly like a Hammond, since I don't own one to compare it to.
Foldback is not an attempt to deal with phase issues. It's purposes are to minimize expense and to soften shrillness on high notes. It mimics what pipe organs did for a long time: suppose you want an 16' flute rank to have 8', 4', 2', and 1' couplers. That means you need 61 + 12 + 12 + 12 + 12 pipes, or 109 pipes. That's a lot of pipes. You can save the expense of 12 pipes by building only 97 pipes, and then wiring up the 1' coupler so that when it plays notes in the top octave it actually triggers the pipes that are one octave lower. You can save the expense of 12 more by repeating the process for yet another octave, which then affects one octave on the 2' coupler and two octaves on the 1' coupler. That's what the Hammond does, and it's called 'foldback'. Without foldback, you'd need 109 tonewheels.
Phase coherency is near impossible to achieve with the kind of architecture used by a typical synthesizer. I try to lessen the effect by detuning the oscillators very slightly, so that they beat slowly instead of suffering complete cancellation. But it's only partially effective.
Hope this helps. Good luck on your ideas.
Chet |
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18198 Location: Durham, NC
Audio files: 213
G2 patch files: 60
|
Posted: Wed May 02, 2007 11:45 am Post subject:
|
|
|
What is it that make the Hammond organ "growl"? I love that sounds but don't understand it. By adjusting up the treble and bass drive on Chet's patch, I hear that sound. _________________ --Howard
my music and other stuff |
|
Back to top
|
|
|
Falk
Joined: Mar 07, 2006 Posts: 75 Location: Sweden
G2 patch files: 29
|
|
Back to top
|
|
|
FlowerP
Joined: Jan 01, 2007 Posts: 38 Location: Sweden
Audio files: 1
G2 patch files: 73
|
Posted: Wed May 02, 2007 2:17 pm Post subject:
|
|
|
Hey Falk! (eller kanske Hej på dig Falk!)
It's very interesting to see your oscillator phase-locking solution. I'm developing a Solina emulation patch, and have come across the same phase issues there. The G2 oscillators are in many aspects good analogue VCO emulations, but for this application you want a phase-locked top-octave-divider oscillator mode instead.
However, the lack of selectable key assign mode on the G2 is part of the problem. The G2 always assigns notes to voices in a "rotate" mode. If a sound has more than minimal release time and you repeat a single note, each successive note will trigger a new voice with undefined phase relation to the previous voice, and this will certainly be audible. I wish the G2 also had a "reassign" mode, where a repeated note would be assigned to the same voice. Such a mode would also allow more accurate envelope handling when emulating e.g. the Solina.[/b] |
|
Back to top
|
|
|
Chet
Joined: Nov 19, 2004 Posts: 231 Location: Lititz,PA,USA
Audio files: 7
G2 patch files: 35
|
Posted: Wed May 02, 2007 6:07 pm Post subject:
|
|
|
Falk wrote: | Hello Chet.
Thanks for your long reply.
What I meant by asking about the foldback and phase coherency was that you use X-Fade modules to map the correct oscillator to the drawbar instead of changing the frequency of the folded oscillator.
By doing that you avoided a phase problem at a small price in CPU. I guess I wondered if it was a conscious descicion or if serendipity guided your hand
Anyway, here is my partial solution to the phase coherency problem. It is a cut down Hammond (only 8 drawbars, think of it as a Hammond with Percussion permanently engaged) and the foldback is totaly off.
To really make it right you need 4 more Mux 8-1 and 2 more S/H to be able to sync the 5 1/3', 2 2/3', 1 1/3' and the 1 3/5' correctly. This patch only syncs the 16', 8', 4'. 2' and 1' drawbars correctly.
One drawback is that the gate signal is slightly delayed, worst case 0.012 second. My keyboard chops are terrible but it might matter to you.
/Falk |
You're welcome. Yes, the foldback design was a conscious decision; I knew the patch had phase problems, and didn't want to make it any worse.
Congratulations on a very clever and innovative way to address the phase issue! I've never seen that done before. |
|
Back to top
|
|
|
|