Fons! Let's make a bargain: We collect the info and you design and implement the code. I'll do the help-function again. :-) Aeh... Well perhaps still a little unfair... :-( Pity! Warm regards Julien -------- Music was my first love and it will be my last (John Miles) ======== FIND MY WEB-PROJECT AT: ======== http://ltsb.sourceforge.net the Linux TextBased Studio guide ======= AND MY PERSONAL PAGES AT: ======= http://www.juliencoder.de

You're right. But a couple of us have tried and it didn't work. We might do a kind of a petition and send him a kind plea with the petitioners names attached. Might that help? I've heard it, it sounds great. Kindest regards Julien -------- Music was my first love and it will be my last (John Miles) ======== FIND MY WEB-PROJECT AT: ======== http://ltsb.sourceforge.net the Linux TextBased Studio guide ======= AND MY PERSONAL PAGES AT: ======= http://www.juliencoder.de

Garry Ogle wrote: His license includes a reason it's not open source: Note the last sentence. It sounds like he'd be more open to community development than forking his code. -Scott

Hello Gaz! I reached him allright. But nothing much came of it. Well perhaps it might be worth a trial to do a well-reasoned petition to ask him round. I suspect there would be programmers willing to write JACK support and a couple of other things (like making the samplerate a selectable value, rather like a static value). From what I remember, there were still a few items on the todo-list, which I THOUGHT shouldn't be too dificult if one or the other could be interested in helping. :-) Kindest regards Julien -------- Music was my first love and it will be my last (John Miles) ======== FIND MY WEB-PROJECT AT: ======== http://ltsb.sourceforge.net the Linux TextBased Studio guide ======= AND MY PERSONAL PAGES AT: ======= http://www.juliencoder.de

Julien Claassen wrote: 2 years ago Frederik was subscribing to LAU. He may still be lurking here. I sent Frederik a patch in summer 2007 that implements JACK support, ALSA-midi-seq and runs the engine at different sample-rates.. His reply was short and he said that he'll respond once he has time to look at it properly. I never heard back after that. I still have the patch sitting around, but can not release it with the current license of Beatrix. However I managed to write the patch additive only. So a non-uniform diff does not contain a single line of code from beatrix! I guess I could distribute that and it's YOU who patches beatrix that will actually break the license. But I'd rather not go down that road. I'll contact him again, and see what can be done. robin

On Tue, May 12, 2009 at 01:29:43AM +0200, David Adler wrote: IIRC, I was involved in that thread back then. I have been using AZR3-JACK ever since that discussion two years ago, on my own tracks, with live bands, in the studio, on recordings, and now even on a commercial rock record. I love it. Problem solved, AFAIAC. The Leslie emulation could be a tiny bit more accurate, and the tonewheel shapes are kind of "flattened" sines (with several different settings) rather than a 100% faithful representation of a Hammond tonewheel, but, hey, it works and it's free and open-source and it sounds great. -ken

Julien Claassen wrote: I don't know but I doubt if azr3-JACK can come close to Beatrix... \r

On Tue, 12 May 2009 01:34:05 -0700 Ken Restivo <ken(a)restivo.org> wrote: Maybe this one? git://git.savannah.gnu.org/r/ll-plugins/azr3-jack.git Regards, Philipp

On Tue, May 12, 2009 at 11:47 AM, becks <becks(a)msound.org> wrote: and here another nice implementation in csound, with maybe useful information http://www.goodeveca.net/RotorOrgan/index.html -- cb

Quoting Josh Lawrence <hardbop200(a)gmail.com>om>: Coincidentally, I mailed Fredrik yesterday to ask if there was any news. Back in Oct 2007 I did the same and he responded that he'd been occupied with a commercial release and was hoping to be able to release a GPL version in 2008. I haven't got any reply back yet.

Quoting Grammostola Rosea <rosea.grammostola(a)gmail.com>om>: Nope, no JACK. Make sure you have snd_rawmidi, snd_seq_midi_event and snd_seq_midi loaded in your kernel (lsmod | grep midi) and you have rights to use /dev/snd/midiC0D0. Edit the default.cfg file and change midi.device to /dev/snd/midiC0D0, change any other value you feel like and start it. It'll display the currently loaded "patches" which you can use by changing the sound on your MIDI keyboard to that value.

Peder Hedlund wrote: How do you connect vkeybd or midi keyboard to Bea? \r

Peder Hedlund wrote: Thanks for info, but got no sound yet. $ lsmod | grep midi snd_virmidi 3548 0 snd_seq_virmidi 4664 1 snd_virmidi snd_seq_midi 5448 0 snd_rawmidi 18544 5 snd_usb_lib,snd_seq_virmidi,snd_mpu401_uart,snd_seq_midi snd_seq_midi_event 6072 3 snd_seq_virmidi,snd_seq_oss,snd_seq_midi snd_seq 42120 7 snd_seq_virmidi,snd_seq_dummy,snd_seq_oss,snd_seq_midi,snd_seq_midi_event snd_seq_device 6108 5 snd_seq_dummy,snd_seq_oss,snd_seq_midi,snd_rawmidi,snd_seq snd 48420 21 snd_usb_audio,snd_hwdep,snd_virmidi,snd_seq_virmidi,snd_ice1712,snd_ak4xxx_adda,snd_cs8427,snd_ac97_codec,snd_pcm_oss,snd_mixer_oss,snd_pcm,snd_i2c,snd_mpu401_uart,snd_seq_oss,snd_rawmidi,snd_seq,snd_timer,snd_seq_device I've an maudio keystation usb midi keyboard... How should I connect it? Just USB or also MIDI cable? How do I know if I have the rights for /dev/snd/midiC0D0 \r

Quoting Grammostola Rosea <rosea.grammostola(a)gmail.com>om>: I don't know if USB-MIDI works in linux, and in this case in particular, so use the old round MIDI jacks/cables. 'ls -la /dev/snd' should say something like crw-rw---- 1 root audio 116, 16 mar 16 08:56 midiC0D0 so in this case you should either be root (not recommended) or member of the audio group to have the rights (rw) to use the device. If you see your username instead of "root" there, you're fine. If not, run 'groups' to see if you're a member of the audio group and if you aren't run 'usermod -G audio -a your-username' as root

Peder Hedlund wrote : You can use beatrix with vkeybd :-) : - modprobe the ALSA "snd-virmidi", - set "midi.device=/dev/midi1" in "default.cfg" (beatrix's config file), - launch vkeybd and connect it to the first virtual MIDI card created by the virmidi module. You can use for example the "ALSA" tab of qjacktcl. Read the vkeybd manpage if you need more options (channels, octaves, etc.) T.

Grammostola Rosea wrote : OSS (emulation) is already being used by something else (jack ?). Anyway, you can use other tools such as aconnect, aconnectgui, KAconnect... T.

Julien Claassen wrote: Well, here's a bit of research to get a start: http://b3world.com/hammond-techinfo.html (there are three pages worth of info here) Here's an interesting site, aimed at providing the technical detail needed for soft-synth makers: http://www.electricdruid.net/index.php?page=info.hammond I can't seem to find one article I had read once upon a time, that pointed out that the upper harmonics (the smaller tonewheels) are slightly out-of-tune with the rest of the wheels. That last link basically covers that detail in a different way, though. To emulate the exact waveform shapes (the tonewheels never made perfect sine waves, after all), you might consider using Bezier curves oscillators (I have a paper on it around here somewhere), but of course, you'd need to know what the shape for each tonewheel needs to be. Anyone have an old Hammond they wouldn't mind tearing apart? You could run some audio analysis on the output of the tonewheels if you could tap their output before they hit the preamp. For me, the most interesting part would be emulating the sound of the old preamps, plus the amp (you want the tube sound, right?). Then, there's the rotary speaker, the most interesting part of all. Look at lumped modeling for the tube preamp/amp parts of the chain. It's all quite doable, but non-trivial, I think. The Linux community in general (and this one as well) seems to me to have a bad track record when it comes to cooperating on large monolithic apps. A better approach might be different people making different parts as LV2 plug-ins which can then be brought together in something like Ingen. I think I'll stop there before I start pretending to be an expert. This is all just stuff I learned from my own research in wanting to do a Hammond emulator myself. -- Darren

MarcO'Chapeau wrote: Hey, I had a fine time running the output of my old band's Hammond through a wahwah pedal and adding a homemade fuzz box ... I have an old recording of that combo sounding like a distorted screaming guitar instead of an organ. Burned out a lot of wahwahs that way, too! -- David gnome(a)hawaii.rr.com authenticity, honesty, community

Julien Claassen wrote: A nice scripting language would do. I presume the LV2 standard has some way of exposing the controls of plugins to outside applications, yes? -- David gnome(a)hawaii.rr.com authenticity, honesty, community

On Mon, May 11, 2009 at 07:02:55PM -0400, Darren Landrum wrote: The tube amos can be emulated using non-linear convolution, same for reverb (probably springs), and this can be done quite accurately if we could measure the real thing. Recordings of individual tone wheels would be essential as well, to find out what was the exact waveform. And then there is the general structure, and details such as how the 'percussion' effect (allowing an envolope on some of the harmonics, very nice and a classical sound if you use the third) worked. As far as I can remember this was not polyphonic. So what were the parameters and how was it triggered (by each new note or only the first after releasing all keys...) etc. etc. Ciao, -- FA Io lo dico sempre: l'Italia è troppo stretta e lunga.

On Tue, May 12, 2009 at 04:52:26PM -0400, Darren Landrum wrote: It's not very common, but it's not high priest magic either. There not so much available in the public domain. A good intro can be found in some of Angelo Farina's papers: http://pcfarina.eng.unipr.it/Public/Papers/list_pub.htm Look at #211 and maybe also #241 there. Ciao -- FA Io lo dico sempre: l'Italia è troppo stretta e lunga.

Percussive was only available on 2 harmonics: 2 2/3' and 4' and only on the upper manual. The cheesy later models give more options even eventually on all the harmonics as a type of synth. Lots of softsynth version give this capability too, to avoid talking about other more important limitations of their emulations: percussion on upper and lower mauals, all drawbars, even on the pedals. Percussion bypassed the drawbar, strength is with a 'Soft' switch. Decay lengths are something like 8 seconds (Long) 250ms (short). Percussion bypassed the VibraChorus, definite requirement for the well know B3 ping percussive, it loses a lot if it trills along with the rest of the harmonics. Most emulations sidestep this routing. It plays legato style, as you say, first key only and people played it that way. Again, later cheesy models introduced this per key and a lot of 'softsynths' do the same but if you listen to keyboardist playing the percussive it can be used a lot more expresively due its legato trigger - the keyboardist decides when the percussive is heard. Other relevant stuff: Tonewheels were in compartments of 4 wheels each, between them they demonstrated slightly higher crosstalk but they were all harmonics of each other. Not all compartments were full, some of them did not contain all four pickups and the associated wheels were just discs to maintain balance of the compartment. There is foldback of the lowest octave and the upper one and half (roughly). The top octave were out of tune since it was not possible to engineer toothwheels accurately with 256 teeth. Hammond took 192 teeth and regeared the upper notes resulting in them being out of tune by cent or two. Its not a great amount but it is above the audible thresholds, especially for pedantics who like to sit and listen to two sinewaves for an hour. The lower frequency toothwheels were not sine formed or even curved for that matter. Some of there were almost square with rounded edges which helped generate the deformations. Some of the tonewheels just cut sections off a circle, consider placing a rectangle of width r over the centre of a circle of the same r and cutting off equal parts from the top and bottom of the circle. The tonewheels tended to be shaped in groups, ie, an octave of harmonics night have the same shape however the output waveform would be adjusted separately by the filter circuit: Different models corrected the distortions of each of the wheel forms to different degrees using filtering baluns on each pickup and depending on the values of these filters then different models exhibit different harmonics, that's a large part of why B3 sound different to C3 for example: filtering especially of the lower frequencies affected the odd harmonic content noticably. There is different crosstalk from the gearboxes tonewheels compared to the filtering baluns: the compartments had two pairs of opposing wheels which were harmonic however these arrived on opposing filter panels that typically had non-harmonic adjacencies. The tonewheels produced very different output signal levels, these were equalised using a set of 'tapering' resistors however the net output signal levels was never flat and it also varied by different models. The tonewheels are not fixed in position, they can move over time - they were under spring pressure contact (for good reasons) and in transport they could shift resulting in no certain phase relationship between the tonewheels. It does not generally shift during playing, only during transport. The spring loading was a fix to prevent the motor burning out if one of the pickups moved (they were already close to the tonewheel) and jammed the wheel - it would jam the motor unless it moved under the pressure of the spring. The motor could still jam if a sufficient number of pickups jammed, even with the sping mounting. If the same tonewheel is tapped multiple times its output does not show linear gain - there is a degree of damping of the signal level since it passes through a changing resistor circuit for each note and drawbar that is played/pulled. There was a lot of crosstalk from the cable looms however there is no documentation of where they came from, ie, it depended on how the loom was built and which harmonics signal were in relative proximity. The drawbars demonstrated a small amount of crosstalk and a larger amount of leakage. Keyclick is very complex. Each key is a contact bar and 9 spring wires. As the bar touches each contact it produces capacitive discharge and hence noise however none of the bars connects at the same moment, the spring wires tend to creep with time so no two notes actually have the same click sequence and any given drawbar harmonic will not sound until its contact is made. If you move a key down very, very, slowly you are usually able to make each contact start independently, each with its own click. Emulating that with MIDI is a pain, the clicks have to be extrapolated from the velocity with a small amount of randomisation. That is just the gearbox and wiring stuff, there are other points regarding the vibrachorus and reverb. I have some stuff of the vibra circuit but not a great deal. It was an 8 stage filtering capacitive delay line. There was a rotating spindle of pickup membranes, it was linked to the main drive shaft so ran at a constant speed however the speed of the vibra could be adjusted by selecting which membrames were fed to the output and which air capacitor they were tapped off. Don't have input on the reverb, I think it was a spring delay line but could be wrong. The leslie is for somebody else to comment on, some salient points: Contained a bass speaker horizontally mounted over a rotating drum. Had two rotating HF horns Most people say the bass rotation exhibits little to no noticable filtering, phasing or dopler due to the low frequency of the cutover, it is mostly just change of volume from the appeture. The horns probably do exhibit filtering effects of the rotation and some dopler. I can't comment on dispersion and echo from the sounds, nor the effects of the paneling material which came up on this list last year. The bass drum and horns had separate motors. They accelerated/decelerated at different rates. They also span at slightly different rates due to independent motors. Roughly the low speed was 1/2 Hz and the high frequency was 8Hz but that depends on your source. A lot of players would disconnect the bass drum belt as its inertia annoyed them and also had less influence than the HF rotation. A lot of players would disconnect the HF motor and link it to the bass spindle so they were always in phase and exhibited the same inertia. Many systems included a brake to speed up deceleration. Speaking of disconnecting motors, some hammond owners added a friction brake to the main spindle to add a rather awesome if sometimes disturbing glide to the output. I am sure there are loads of people who could add to this list and expand upon its content - none of this is 'definitive' fact. Beatrix implements most of these features, the percussive routing, crosstalks, tapering, waveform distorts, leakages, etc. Keyclick last time I looked was based on a single contact but had multiple signal options and the reverb was a pretty nice but clean multiple chained delay lines. To be honest I thought beatrix had been 'sold' to an italian company making 'Hammond Sound' keyboards which was largely why it is no longer maintained publicly, its maintained privately - what I want to say is that is not an orphaned app by any means, its one of the more successful ones perhaps. Regards, nick. Apologies if this is a duplicate submit, my firefox hung when I pressed send. _________________________________________________________________ More than messages–check out the rest of the Windows Live™. http://www.microsoft.com/windows/windowslive/

On Tuesday 12 May 2009 5:53:55 pm Nick Copeland wrote: One other element was if the tubes were shot. The old blown out tubes would give it a really dirty, but soft texture which for some things was desirable. The best synthesis of a Hammond I've heard is a hardware keyboard. Nord makes a really great sounding Hammond synth that beats out anything else I've heard. (It's Rhodes, Claves and Whirlitzer are great too, piano kinda sucks though) -Reuben

On Wed, May 13, 2009 at 12:53:55AM +0200, Nick Copeland wrote: <a *lot* of very interesting info about Hammonds> Fascinating ! [*] Could you tell us more about the VibraChorus ? From what you've told already it seems to use 'mechanical' variable capacitors... Ciao, [*] I've seen the latest Star Trek today (in Italian !). the term 'fascinating' every five minutes. So where did he pick up that habit ? Kond of liked the movie, until it degraded into Stars Wars like nonsense such as completely improbable man-to-man fights between the main good and main bad one. -- FA Io lo dico sempre: l'Italia è troppo stretta e lunga.

Fons Adriaensen wrote: Some progress on this? Already contact with the author? BTW Beatrix is not GPL, but open source right, so you can use the code as example how to make things!?? \r

On Tue, May 19, 2009 at 11:10:00AM -1000, Joel Roth wrote: Studying the way some software emulates a system does not necessarily provide correct information on what is being emulated, unless it is a complete and accurate physical model, which is unlikely. You may be able to find out some interesting parameter values but there it ends. For example the way Aeolus generates an organ pipe sound (additive synthesis and some postprocessing) has no relation at all to what happens in a real organ pipe (which is a complex non-linear and sometimes partially chaotic combination of an oscillatir and filter that are tightly coupled and influence each other). As an author of such software you make choices on how to do certain things, for all sort of reasons ranging from conceptual to very pragmatic. The latter may obfuscate things considerably, and provide the author's signature in a sense. If someone would write an organ synth by re-implementing the processing in Aeolus in entirely new code I would probably have no problem in detecting that quite soon, given the new code. Ciao, -- FA Io lo dico sempre: l'Italia è troppo stretta e lunga.

On Tuesday 19 May 2009 16:22:34 Grammostola Rosea wrote: I have a vague memory of it not being open source from something I read in this recent go around. What makes you think it is open source? What license is it using? all the best, drew

On Wed, 2009-05-13 at 00:53 +0200, Nick Copeland wrote: <snip> A good read! Some will see this as heresy, but I think that most of the Hammond sound comes from the Leslie speaker. The sound just from the console speaker on my L100 is almost a different instrument. With the right patch on a DX7 through a real Leslie you can get a pretty authentic Hammond sound.

Jörn Nettingsmeier wrote: Good story :)

Hi All, I am dealing with an issue with Bristol where the B3 emulator uniquely exhibits stuck notes. None of the other emulators seem to show this problem and all debuging to date indicates that this is not an external issue such as lost MIDI messages, etc. The B3 has a lump of code that emulates all 92 tonewheel whenever a key is pressed, it chews up a nice amount of CPU and this may be related to note event management sequences that might be buggy. Now the guy who reported the issue sees it regularly, he just needs to play a sequence of notes quickly and eventually one will stick on until pressed again. For the life of me I cannot reproduce this on my dev system, after hours and hours of playing. The bristol B3 is perhaps not the most popular example of hammond emulators but has anybody seen this? Regards, Nick. _________________________________________________________________ Drag n’ drop—Get easy photo sharing with Windows Live™ Photos. http://www.microsoft.com/windows/windowslive/products/photos.aspx

I would be more than happy to finger ALSA here but from the debugging done so far the problem happens with no evident lost note_off events, just one spurious duplicated note_off debug message from bristol that definitely did not result from an event delivered by the Seq interface. Debuging is an awkward operation since each time I want to concentrate on selected code sections I have to provide updates and wait for the output - its a bit like programming with punchcards. The difference in ALSA versions could naturally be relevant and will review it. Regards, Nick. _________________________________________________________________ Show them the way! Add maps and directions to your party invites. http://www.microsoft.com/windows/windowslive/products/events.aspx

On Thu, 2009-05-21 at 23:42 +0200, Nick Copeland wrote: Yep, I just tried version 0.40.2 and it happened almost immediately. It happens pretty regularly to me. I can help debug it if you like. (I've recently tracked down several bugs like this in my own synth.) Damon

El 05/11/2009 11:57 PM, Fons Adriaensen escribió: http://www.google.com/patents?id=NUlkAAAAEBAJ Do you need something more than that? :-)

Hello David! I've only heard demos of hogand so far, but it doesn't reach beatrix. Hogand sounds like an electric organ, but not close to hammond. It sounds somehow more sterile, thus giving an 80s impression. This could be easily explained to my ears, because - from what I remember - hogand uses FM synthesis. I very well-known synth in this category is the Yamaha DX7 (1983-87). I wouldn't discard hogand, if I could use it, but as an organ in its own right. Kindest regards Julien -------- Music was my first love and it will be my last (John Miles) ======== FIND MY WEB-PROJECT AT: ======== http://ltsb.sourceforge.net the Linux TextBased Studio guide ======= AND MY PERSONAL PAGES AT: ======= http://www.juliencoder.de