The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Synths for sale  Search Gearslutz Go Advanced
The E-mu Command Station XL-7/MP-7/PK-7 and Proteus 2500 tutorials Keyboard Synthesizers
Old 28th October 2017
  #241
Gear Head
 

Quote:
Originally Posted by butter
As I understand the manual, that the first two ticks (0 and 1) of a pattern are reserved for "pre-sequence setup". When the loop repeats, it doesn't repeat those ticks - in fact from the diagram in the manual it almost looks like events and notes have different tick counters!]

- "event data" on Ticks 0 and 1 is ignored when the loop repeats
This is what I understood, too.
Old 28th October 2017
  #242
Gear Head
 

Quote:
and hello dr. Steve, welcome to Gearslutz
Thank you!

Last edited by Dr. Steve; 28th October 2017 at 09:50 PM.. Reason: I took the ] off of [Quote] by mistake
Old 28th October 2017
  #243
Gear Head
 

Quote:
Originally Posted by balma View Post
When I have defined in my mind, what I want to modulate on a kick during a realtime performance, I create from scratch a TEMPLATE specifically designed for kicks, save it as a category, and then, create other 15 kick presets using this template, and maybe just changing the selected kick samples inside the patch, and making some small changes on filters here and there you get a whole new sound.
Another excellent suggestion!
Old 28th October 2017
  #244
Gear Head
 

Quote:
Originally Posted by Dr. Steve View Post
So you might be able to insert sysex commands to set (for example) the initial controller amounts either in a setup sequence or within a sequence's track.
I just re-read what I had written; I think it gets at the reset-when-looped problem. By "setup sequence" I originally had in mind a pattern that just had setup stuff--including the embedded sysex to set parameters--in this case the initial controller amounts for the variants of the basic preset.

Looking at my now-booted P2500, I see that another part of this preset + differing initial controller amounts strategy would likely involve setting the initial controller amounts to "off."

I never did this, but either setting them to off says, (1) "Don't set an initial controller amount for this controller [e.g., MidiA, etc.] whenever the preset is called up or (2) "Don't use this controller [e.g., MidiA, etc.]. My guess is (1).

If so, then when that preset is re-booted as a pattern loops, it >theoretically< shouldn't reset that controller [e.g., MidiA] to the initial amount.

And if so, if you have a sequence with just initial parameters of your own choosing that is called before the pattern with notes and other stuff, then when the pattern loops maybe the parameters would be at the last-used state?

Again, just speculating here.

And if this works as speculated, templates could be created as Balma suggests with certain (or all) initial controller amounts set to "off" where the intention is to have initial controller amounts come from a pre-sequence that is played just once before looping patterns play.

Of course, maybe "off" tell the E-Mu box not to use a particular controller [e.g., MidiA].

I will try to test some of this.

After Test # 1:
Yes: "off" means (1) "Don't set an initial controller amount for this controller [e.g., MidiA, etc.] whenever the preset is called up." And it reminded me of something I used to know (I think): each of the 32 midi channels has its own set of CC values for MidiA -> MidiP.

So, it would seem you could have presets with all all initial values off and set the values with an initializing sequence that you start before your looping patterns. I think the implication is that you could then tweak the controllers in a live performance using presets in such a way that your MidiA -> MidiP values don't get changed from what you last had them set to on each of the 32 midi channels--so long as you did not called up another preset or something else that re-initialized or changed the last-state values of MidiA -> MidiP.

I will try to test some more.

After Test # 2:
I did not test putting this concept into a sequence; I did some more development borrowing from the concept of what I call !FX presets. The idea came to me from Aaron Eppolito who used to work for E-Mu. Here's (1) the concept and (2) how it applies to the topic.

The concept: Have an empty preset [all Instruments = 000 none]; give it the FX settings you want. (That's why I start mine with the prefix !FX as in !FX:MondoVerb.) When you call or dialup the !FX preset on a designated control channel (for example, C16A), you effectively change the FX settings. And since there is just one set of FX, dialing up the preset acts as a global FX parameter switch.

How it applies here: Take a preset; set all initial controller amounts to "off." I named mine with the prefix <&> as in <&>:Life. Create a bunch (or a family or a series) of these <&> presets. I tested on C01A with audition. So, when switching from one <&> preset to another on the fly, each preset keeps MidiA->MidiP from where it was previously. Since CCs in multimode function on a per channel basis (so far as I can recall from previous experiments), we have 32 sets of MidiA -> MidiP (one for each midi channel).

Borrowing from the !FX preset idea, we can create different <&> presets to set the initial CC values for different variations of a basic preset. I created 10 < >:I-CC = off presets. The next step might be a <!>:I-CC A001 preset that has starting CC values to be used with a family of presets to be used on C06A (for example) in Song A, and then a <!>:I-CC B001 preset that has starting values for a family of presets to be used on C04B (for example) in song B.

I think it should work with the sequencer, but more testing is needed.

Last edited by Dr. Steve; 29th October 2017 at 01:11 AM.. Reason: To add the results of another test I did.
Old 29th October 2017
  #245
Gear Head
Quote:
Originally Posted by Dr. Steve View Post
I don't use grid mode, though I did play around with it a few times. What you are trying to do is the kind of puzzle I like to work on when I have time (maybe Thanksgiving?). Help me understand what you want to do.

Intuitively (inexperienced as I am in RT grid live performance), grid modes seem to me to be designed so you can place/manipulate data on specific beats and on various specific beats in between. To me that seems antithetical to swing--where you have the machine place/manipulate data according to swing settings.

The trouble I am having is, if you want to be able to place/manipulate data at specific timings, why do you also want the OS to bobble the timings?

Not being critical here at all; just trying to get past my personal mental block as a non-grid mode RT performer so I can attempt the challenge you have identified.

Steve
https://www.youtube.com/watch?v=auHgAfonj6Y

@ 2:55

Basically every second 16th on the grid is delayed by a percentage, which changes the feel of the rhythm even when the tempo itself is rigid and machine-like. (Though on many vintage models, including the TR-909, the base tempo may be a bit wonky too!)

I'll try re-creating "the TR-909 experience" on my CS at some point. It'll take some effort to separate the instruments into individual outputs (which is something I definitely want to do) as that's not (I think) possible with the pre-mapped kits.
Old 29th October 2017
  #246
Lives for gear
 
balma's Avatar
Quote:
Originally Posted by Dr. Steve View Post
I don't use grid mode, though I did play around with it a few times. What you are trying to do is the kind of puzzle I like to work on when I have time (maybe Thanksgiving?). Help me understand what you want to do.

Intuitively (inexperienced as I am in RT grid live performance), grid modes seem to me to be designed so you can place/manipulate data on specific beats and on various specific beats in between. To me that seems antithetical to swing--where you have the machine place/manipulate data according to swing settings.

The trouble I am having is, if you want to be able to place/manipulate data at specific timings, why do you also want the OS to bobble the timings?

Not being critical here at all; just trying to get past my personal mental block as a non-grid mode RT performer so I can attempt the challenge you have identified.

Steve
I normally set quantization to off when recording most of the tracks of a sequence. Is is normal to end with some notes out of place due to timing error, so I apply quantization, but reducing the percentage to 66%. That realigns the notes your recorded on realtime but not estrictly to the clock, producing a swinging effect over the track.
Another method can be applied using grid mode:
While recording notes on grid mode, you can use either the 16 trigger buttons or the knobs to input notes. For example, when recording a percussive track, I input some notes, and then go the screen SCALE/SHIFT START TIMES, and shift the notes I just recorded to -00:032. - You don't even have to abandon the GRID record mode to do it-... then, go back to the record screen, input more notes in the track, go back to scale/shift screen and shift to +00:16. Rercord more notes, and go back to scale shift and increase the shift again to +032.

This method allows you to produce all kind of swing effects in a detailed way with some practice.
Old 29th October 2017
  #247
Gear Head
 

Quote:
Originally Posted by Yammy GS1 View Post
Basically every second 16th on the grid is delayed by a percentage, which changes the feel of the rhythm even when the tempo itself is rigid and machine-like.
This definitely explains the goal and the reason for doing it. Thanks. It seems like the purpose is to place a percentage delay, almost like quantization but (1) a delay, not a timing alignment to "the beat" and (2) just for a specified part of the measure--with the goal being RT tweaking of the percentage amount just for the part of the measure set up to swing.
Steve
Old 29th October 2017
  #248
Gear Head
 

Quote:
Originally Posted by balma View Post
I normally set quantization to off . . . . [snip]

While recording notes on grid mode, you can use either the 16 trigger buttons or the knobs to input notes. For example, when recording a percussive track, I input some notes, and then go the screen SCALE/SHIFT START TIMES, and shift the notes I just recorded to -00:032. - You don't even have to abandon the GRID record mode to do it-... then, go back to the record screen, input more notes in the track, go back to scale/shift screen and shift to +00:16. Rercord more notes, and go back to scale shift and increase the shift again to +032.

This method allows you to produce all kind of swing effects in a detailed way with some practice.
Being familiar with "traditional" quantization [from a few decades ago], but not being very experienced with grid mode recording, your explanations were helpful to get me thinking of possibilities. So, I booted my P2500, put it into grid mode recording, pressed a number buttons, and found that my randomness had a constant, non-tweakable swing to it. (Just a very short pattern.)

I tried testing a few things and might do some more in the future, although I don't feel very comfortable with grid mode (maybe that will change).

One thing I found interesting (at least to me) was the ability to place multiple notes in a grid cell, for example, slightly ahead of a down beat--e.g. 3 for triplet, 5 for quintalet, etc. up to 9 (nonelet?).

I did that by pressing the insert/repeat button. It seems that the repeated notes get placed in the location where the trigger/mute button placed a note event in the grid.

I then wondered if I went to the quantization/swing menu if I could insert some sort of quantization in the location of the last selected trigger/mute button. Unfortunately, (1) that seems to do quantization/swing for the whole pattern (I think) and (2) after pressing the "Do It" button, the sequencer stops and kicks out of record mode. That would have been a great solution if it had worked. Sorry, I tried!

Steve

Update: I went to the SCALE/SHIFT START TIMES menu, but again, when I pressed the "Do It" button to apply the settings, the sequencer kicked out of grid mode recording mode and stopped. Is that supposed to happen or should I be trying it in a different way? Thanks.
Old 30th October 2017
  #249
Lives for gear
 
balma's Avatar
Quote:
Originally Posted by Dr. Steve View Post
Update: I went to the SCALE/SHIFT START TIMES menu, but again, when I pressed the "Do It" button to apply the settings, the sequencer kicked out of grid mode recording mode and stopped. Is that supposed to happen or should I be trying it in a different way? Thanks.
You can perform the scale shift start times function while playing, but it will kick you out from grid mode.
Old 30th October 2017
  #250
Lives for gear
 
balma's Avatar
Aaron Appolito is one of my synth heroes

....

Sometimes I try to see the sequences or notes, as any other parameter/property of a sound, together with filter, reso, volume, etc, and all those elements you can control or manipulate. A sequencer is just like any other section of sound synthesis. Sound programming and sound sequencing are equally important and one does not serve the other.
But the command station's patch-sequencer relationship, forces the user to have the sounds in one side, and the notes on the other one.


´MAybe DSI Tempest's users know what I'm talking about. A pattern has its own set of sounds. Every single modification you perform to a sound, becomes permanent. Patterns does not share a common patch database, but each track is a patch, edition mode is always active for all the tracks on all the patterns. Such configuration results in a very fluid sound editing/song composition, both fully available and interactive on realtime.
When you load a project on the Tempest, you immediately have 512 sounds localed on 16 sequences of 32 tracks, each one on realtime edition mode.


In the command station is harder to evolve sounds and sequences. If you are editing a sound, moving to another track and entering edition mode, will reset the previous patch you were editing. Any change must be saved before exit. You can have only one sound on edition mode.

But in the command station, "notes" are not necessarily inherent to sequences or tracks.

There are multiple functions inside patch edition wich allow to create movement and variations in multiple parameters, and synchronize them with the bpm.

START TIME
the start time of each one of the 4 layers inside a patch can be individually delayed, and sync with the tempo/bpm.

So if we set these values on the L1, L2, L3 & L4 respectivelly:
L1: DELAY 0
L2: DELAY 1/16
L3: DELAY 1/8
L4: DELAY 1/8d

the result will be a quite basic 4 note 1/16 sequence of the 4 layers, one after another....

The tempo-based Envelope mode function:

The 3 ADSR envelopes on the CS have 6 stages. And each stage/section can be sync with the BPM. Also, envelopes can loop. That means, you can create modulations with the stages, and stick to the sequence's BPM.
The only hard-wired ADSR is the Amp Envelope. Filter and Aux envelopes can be freely assigned to any destination, and produce modulations while keeping a strict relation with the tempo beat.
The fact that each one of the layers have its own set of envelopes increase the amount of possible variations you can produce inside a single patch.


The "CLOCK DIVIDERS"

Clock dividers are simple YES/NO (peak/valley) operators and can modulate multiple things. They are 1/16, 1/8, 1/4, 1/2, 1/1. 2/1, 4/1 and 8/16 timers. When combined, they can produce bpm modulations on truly elaborated ways, and you can assign knobs to get control over them on
realtime.
Also when combined with the RETRIGGER destination, they can produce repetitions of layers, kind of roll effect, and always with an individual setting for each one of the layers. Knob programming tricks, allow to go from a 1/16 note retrigger, to 1/8 or 1/4 in the same sound, just tweaking to active/deactivate the patchcords where the clock dividers are assigned to retrigger function.

Also, the start of each LFO can be delayed and also syn with the BPM. Rember we can use 8 different LFOs by patch (from a selection of 15) and some of these LFOS, are really sequence-oriented, like the "octaves", "fifth+Oct", "Sus4 trip" and "Neener". They are even named as "pattern LFOS".

Such huge arsenal of features of controllable and semi automated functions strictly related to the bpm of the sequencer, allows you to literally create note/formant sequences inside a patch, just holding a note.

Is like the programmers tell you to move some of your song composing routines to the patch programming arena....

The combination of different types of sounds wich can produce sequences, with their own knob setup to control such variations in a really accurated way, can release a lot of creative power when combined with patterns created to spefically sequence these sounds.
... will continue....
Old 30th October 2017
  #251
Lives for gear
 
manalishi's Avatar
Question for the E-mu experts: are there some 2000-series ROMs that are incompatible with each other?

I've got XLEAD which doesn't seem to work in the Orbit 3 - which has both Beat Garden and Tecno ROMs. I get a message on boot-up - something to the effect that some patches may be unavailable. And then the XLEAD ROM does not appear. And - weirdly - the same ROM worked fine in the Orbit 3 up until I reinstalled the Beat Garden.

Same XLEAD ROM works just fine in the Vintage Pro.

Is it possible that the ROMs need to be in certain locations in the 4 slots?
Old 30th October 2017
  #252
Lives for gear
 

Quote:
Originally Posted by manalishi View Post
Question for the E-mu experts: are there some 2000-series ROMs that are incompatible with each other?

I've got XLEAD which doesn't seem to work in the Orbit 3 - which has both Beat Garden and Tecno ROMs. I get a message on boot-up - something to the effect that some patches may be unavailable. And then the XLEAD ROM does not appear. And - weirdly - the same ROM worked fine in the Orbit 3 up until I reinstalled the Beat Garden.

Same XLEAD ROM works just fine in the Vintage Pro.

Is it possible that the ROMs need to be in certain locations in the 4 slots?
I don't know, but I bought my PX-7 empty, without any ROMs, got the XLEAD card and no matter what slot I place it, the PX-7 appears to have no ROM installed when I turn it on, and I have to manually go and highlight and choose the ROM for it to work.
Old 31st October 2017
  #253
Gear Head
 

Quote:
Originally Posted by manalishi View Post
Question for the E-mu experts: are there some 2000-series ROMs that are incompatible with each other?

I've got XLEAD which doesn't seem to work in the Orbit 3 - which has both Beat Garden and Tecno ROMs. I get a message on boot-up - something to the effect that some patches may be unavailable. And then the XLEAD ROM does not appear. And - weirdly - the same ROM worked fine in the Orbit 3 up until I reinstalled the Beat Garden.

Same XLEAD ROM works just fine in the Vintage Pro.

Is it possible that the ROMs need to be in certain locations in the 4 slots?
I am only a bit of an expert on the things I have tried and explored obsessive / compulsively, but I will throw in what I have picked up over the years.:

I heard of one person who got a weird error message with one of the ROMS and was looking for an error code guide. The message about some patches not being available might have been a symptom. I don't remember where that was, but it was within the past year. I will try to find it (one of the E-Mu Yahoo groups, I think). Your experience sounds similar--trying a ROM and finding it works in some configurations but not others. It would be good to see if we can discover some patterns to this.

I know there were some ROM conflicts but my understanding is that OS2.26 for the single ru E-Mus and OS2.0 for the CS/P2500 solved those conflicts.

I haven't ever heard that the number of the slot made any difference.

I did trash the P2K preset section of a P2500 composer ROM by putting it in the wrong slot on an ultra sampler. The samples/multisamples [used for the ROM INSTRUMENTS are intact, so it works in an ultra sampler.

I will have to look up the diagnostic test for ROMS. I believe one of the tests will give you the CRC checksum. That was bad on my trashed P2500 Sound ROM. Somewhere I have a list of the checksums either I have or people have told me about. Having a valid checksum is no guarantee the ROM will work. On the other hand, I think that having a bad checksum indicates a problem.

Not sure this solves the issue, but it it might lead to some discoveries.

Steve
Old 31st October 2017
  #254
Gear Head
 

Quote:
Originally Posted by HUBA View Post
I don't know, but I bought my PX-7 empty, without any ROMs, got the XLEAD card and no matter what slot I place it, the PX-7 appears to have no ROM installed when I turn it on, and I have to manually go and highlight and choose the ROM for it to work.
I am trying to understand this. For any P2K-based module I have used, I have to manually dial up presets (user and ROM-based presets)--unless I select them externally via midi.

OH--wait a sec. If you have user presets installed that were designed for one ROM (e.g., User Presets designed for the PX-7) and you don't have that ROM, they will not work because the pointers in the user presets are looking for resources that are on the ROM you don't have. Is that what you are experiencing?

If so, there are several options. I have a utility I wrote that will let me change the ROM ID in a syx file bank of presets. It's not converting the presets completely for use by a different-than-intended ROM; it just replaces the pointers to the ROM (mostly ROM INSTRUMENTS).

The XLEAD ROM is a 12-controller sound ROM; the XROM is a 16-controller sound ROM. MidiA -> MidiL v. MidiA -> MidiP. The ROM Instruments are the same, but the ROM IDs are different. So by changing the ROM ID in a preset bank designed for an XL-7 to the XLEAD ROM ID, it will sort-of work. However, the XLEAD ROM doesn't have all the other resources (arps, riffs, etc.).

Not sure if I am on the right track; that's the only thing I can think of that fits what you described. You might want to back up your User Presets if they are for ROMS you don't have yet, and load the XLEAD presets into User memory so you can fiddle with them.

Steve

Last edited by Dr. Steve; 31st October 2017 at 02:35 AM.. Reason: I left out an important comma and a small suggestion.
Old 31st October 2017
  #255
Lives for gear
 
Spectralwaves's Avatar
Quote:
Originally Posted by balma View Post
Good willing and the love for music are intact! So not too much has been lost, really....
Hey Balma,

hard luck on losing some of your stuff in the hurricane.

Im glad you are ok & your Command Stations & valuable synths etc..

If the lost data was on discs/ hard discs etc, it is quite likely retrievable when cleaned & dried. (I mean load the HD into a usb caddy & it hopefully will read on another computer)
A data retrieval service will nearly always be able to get your data back as long as the discs are not burned or broken, but usually expensive eg often @ $ 7-800 per service, which is something to bear in mind for future..ie dont throw away the HDs .

Good luck with repairs etc
Old 2nd November 2017
  #256
Lives for gear
 
balma's Avatar
Quote:
Originally Posted by Spectralwaves View Post
Hey Balma,

hard luck on losing some of your stuff in the hurricane.

Im glad you are ok & your Command Stations & valuable synths etc..

If the lost data was on discs/ hard discs etc, it is quite likely retrievable when cleaned & dried. (I mean load the HD into a usb caddy & it hopefully will read on another computer)
A data retrieval service will nearly always be able to get your data back as long as the discs are not burned or broken, but usually expensive eg often @ $ 7-800 per service, which is something to bear in mind for future..ie dont throw away the HDs .

Good luck with repairs etc
Really thanks a lot for that man! Not just for the advice, but also for taking the time to provide it.
Unfortunately, I've already disposed of most of it, so f*ked up. That was a hard lesson to learn. Always have a backup of your most valuable work, in a diferent location. Anyway, lately, nobody hires me to play on gigs, there must be a reason, so who knows?... maybe Nate Storm made a favor and took away all the s*it music I was doing, forcing me to reestart things from zero.

Love how this thread has capture the attention of cool people, hope to provide help and ideas to other Command Station owners and even reconsider its enormous value as a creative tool.


I'm gonna slowly evolve some of the things I want to share, starting on preparations before reprogramming the sound of the Command Station, in order to bring to the surface in clear terms, what you can do with it. Is a fascinating and extremely complex musical instrument.


Evolve your sounds:

A command station is like a LEGO BASIC. The BASIC series from Lego, are sets of pieces not aimed to build a specific model, like others like Lego Space, or those cool Ninja does. They have basic structural pieces and forms to create whatever comes to your mind. They were my favorites as a kid since their versatility allowed me to build robots, houses, spacecrafts, anything, after I collected several Lego Basic boxes. I use to mix all of my legos ann the classify the part by colors, or shapes, or specific functions.
Quite similar to the Command Station's concept. It is good to create any musical genre. You can play bossa nova in a wedding, compose a film scoring, classic instrumentals or atmospheric techno, salsa criolla, cheesy 90ies trance, any kind of pop, or big room techno. The more ROMS you have installed, the better the range of possibilities of the sound palette.
It can adopt many forms depending of your intentions, and understanding of a unique instrument.
You can make it sound like the most vulgar rompler, and stay with those crappy factory sounds that do NOT reflect at all its true power, or accept the challenge, and redesign its sound. It's a synth made of clay.






I began creating a blank template patch, to be used as a blank page to start programming sounds from scratch,

While using the LA (All layer) edition mode:
Reset the 24 patchcords to off on both sides and set the amount to 0. Also, you can perform the reset of all settings on one layer and then copy it to the L2, L3 and L4 respectively.
Set all the ADSR envelopes to tempo-based, turn the loop function ON, and set the values for the amp envelope decay2 to 100 (most of the factory presets have this value set to 30).
I also set the attack level to 2 or 3, because the default Zero value produces clicking on some of the samples.
I also, enlarge the release time a little bit increasing the value of rel1 and rel2.
Change the fine tune of each layer with some discrete values, like -3,0,+5,+2, avoiding exactly the same tune on them.
Set one of the LFOs with clock sync on, and the second one with free run.
Set the filters to OFF.
Now, why not using your own tuning scales on your user patches?

Keyboard Tuning section
The command station has 12 fully programmable keyboard tuning scales, and 12 from factory. They are hard-wired and stored apart from the user sound section, on the global settings. If you modify one of your user scales, the changes will affect all the patches using it. Each scale has 128 notes.

Most of the samples, sound good on the middle range. from C1 to B6. Higher or lower than that normally produces no sound or useless stuff. So it would be useful, tochange the pitch values of notes from C-2 to B0, to C7-G8 to some odd tune combinations as an extra resource for your patches. Saved this user tuning scale on User1 position, and select it on the Blank Patch. Choices made at the beginning of a long term process, will slowly spread and reproduce. Same applies for mistakes and omissions...
This blank user preset will be the empty vessel for many sounds.
I set ALL the knobs to the left, to 0 value. and then save it on the first user location. The patch 000 is a blank preset. It will serve to create sounds from scratch, and also when using the sequencer to play external synths and you want to avoid having a sound on those tracks (even if you set it to EXT on MIDI settings). Also using this patch on MIDI OUT tracks, let's you know these tracks are for external devices and not the internal sound engine.

Knob calibration: the range of the knob values goes from 0 to 100, and such polarity can be inverted to -100 Is not binary like most of the engine.

If you modulate the filter with a knob (0 to 255), with its full range of 100 each increment with a knob equals to 2.55 on the filter.
In the case of classic LPF Filters, a value of 0 produces no sound.
Not all the filters on the E-mu present such behavior. The filters are divided in Types, and they go from lower to higher "Ord" (values 2,4,6 and 12).
The higher the Ord, the more complex the filter is, and also uses more polyphony.

The first 8 filters are type LPF. Smooth, Classic, Steeper and Millenium won't sound with a 0 value. But MegaSweepz, EarlyRizer,Klubklassik and BassBox-303 do not reduce the level volume when close to zero, and have a quite smooth transition, adding peculiar tones depending of the value.

I mention this, because knowing the behavior of the different filters, helps to better calibrate the knob settings and manipulate them.

For example, the classic LPF filter becomes audible after 16 (in the range of 0-255). I like it to have its minimum level at Freq: 64 on the Filter Menu .
So now not only the Filter Knob in the mod matrix is determining the filter amount, but also the freq value of the Filter Settings, so I divide 64/2.55 and I get the percentage of the knob range I have to reduce, wich is 25. Then I set the range of the knob to 75.
Or easier: 64 is the 25% of 255. I rest 25% to 100 on the knob calibration.

MidiA -> FiltFreq +75.

Depending of the filter selection, knob's ranges should be set accordingly to its behavior.
Also, some of the filters, can't be used on some samples. For example, the filters of the type DST, like the FuzziFace, will produce a horrible reso saturation feedback on some type of samples, like the first ones on the wav: category of the XL-7. Samples with high ring modulation or bells, sent to FuzziFace could blow your ears or speakers. Many of the Z plane filters, emphatize and boost specific frequencies of the audio spectrum, and they can produce peaks when applied to some samples.

So, with this brief introduction, I 'm trying to explain how important is to practice with the filters, and learn how they behave.
Most of the factory sounds, have a single filter knob, KnobA, to modulate the filter. They just tried to make things easier to the customers, and programmed the 16 knobs to emulate the market tendencies.
knobA for filter, B for resonance, C & D for the filter envelope.
Knobs E to F to modulate the 4 stages of the ADSR
I for velo to Amp, J to send the filter to velo, K and L for the LFOs
M is for PITCH, N for tune level, O & P for arpeggio stuff.

And such configuration repeats on almost all the factory presets.
With some exceptions, most of the sounds, use only 1 or 2 layers...

Such a crime.... really.

I will be updating this "tutorial" during these days...

And of course, a little bit of 100% command station music:





I hope to be able to expose most of the methods I´ve used to compose this stuff....

Old 2nd November 2017
  #257
Here for the gear
 

Great stuff Balma, so many cool trix and ideas from you, thank you...this machine is such a deep beast. Sorry for the loss on all your old files...I think it is best to try to find something positive about it...you could see it as a way to get a new fresh start, clean out the old system so to say...the most important things are still in your head.

So you are saying the filters have 256 steps when i sweep it with the build in knobs? but via external midi it will just get 128 values right?

I would be glad if someone could explain the E-mu multistage envelopes in a way that makes it more like a normal ADSR if possible...i find it really confusing, maybe is it not so snappy, or maybe i dont understand it.
Old 3rd November 2017
  #258
Lives for gear
 
balma's Avatar
[QUOTE=mastaman;12937317]Great stuff Balma, so many cool trix and ideas from you, thank you...this machine is such a deep beast. Sorry for the loss on all your old files...I think it is best to try to find something positive about it...you could see it as a way to get a new fresh start, clean out the old system so to say...the most important things are still in your head.

Quote:
Originally Posted by mastaman View Post
So you are saying the filters have 256 steps when i sweep it with the build in knobs? but via external midi it will just get 128 values right?
Did I?

I'm not yet messing with CC MIDI transmision, but trying to illustrate why the classic filter knob configuration results inadequate to manipulate the Z-Plane Filters in most of the sounds.

I have almost all my sounds with the knob positions totally to the left.
In a synth where each knob performs a different function depending of the patch, I need to be sure that I'm not going to screw the sound with a value drastic jump if I touch something. Also, having different knob positions for all the sounds, makes the LINK function harder to use.

The LINK function, can stack three patchs into one single patch.
TRIGGER 14, "LINKS"

Is like the sound browsing screen, but there you can browse and select any sound of the user/ROM patchs banks and save it on the patch you are editing. There are two links, so, depending of the patches linked, you can layer a max of 12 samples inside a patch.
Is like subordinating two sounds to a third one. The sounds linked, keep all their properties, except one: the knob positions

So, let's say you are editing a sound where the INITIAL CONTROLLER AMT ( trigger button 12: (INITIAL CONTROLLERS) value for the KnobA is 74, and you want to use the LINK function, to bring another sound into this one you are creating. And it happens, that the Init Cont Amt for the knobA for this sound you want to add, is not 74 but 0. A knob can't be in two different places at the same time The sound you are linking, will use the knob positions of the sound where it has been linked.
You can test by yourself with a simple experiment, how knob positions affect sounds when linked
Use the BLANK PRESET I just explained, and, move all the 16 knobs to the midle (12 O'clock position).
Go to the LINKS screen, and start browsing the sounds of the ROM. The sounds will immediately adopt the 12 O'clock on all knobs, no matter wich are their INITIAL CONTROLLER values. Sometimes you get cool results, most of the times, sounds like s*hit.

That't the main reason I began to use the same INITIAL CONTROLLER AMT to zero in all my patches.
But if you do not change the knob configuration and ranges depending of the used filter, a zero value on the filter knob, could result in no sound at all depending of the filter, as I explained before with the case of the LPF filters.
Therefore, in those cases where the used filter is the "Classic" or "Smooth", I go to the Filter Freq Screen (trigger 7) and change the frequency to 64 (25% of 256 wich is the filter range), and then, I rest that 25% to the knob range, and set the FILTER Knob to 75
MidiA -> FiltFreq +75. So I will use the whole tweak from left to right, to increase the filter from 65 to 255.

But, not all the 50 Zplane Filters work in the same way, and some ranges of the filters, mostly the 12 ORd types, can produce a terrible loud feedback in some sounds, even if the Q value (wich plays the role of Resonance in some of the filters) is zero.

Another experiment, select a simple SAw waveform from the sample list. In the XL-7 is the 0002 wav: saw. In the MP7 Model is the 0276 wav:sawtooth.
You must make this, from a blank preset as described before. So no other source like an ADSR or an LFO or velocity is affecting the sample.
Use any knob to move the sample from 0 to 100, get sure the Freq is 0 on the filter screen, and select the FuzziFace Filter, with the volume of the synth to almost zero.
The result will be a loud feedback in some notes in most of the sounds with the Filter knob totally at the left. But it sounds cool in a few sounds.
Also, if you move the knob to around 66 percent of its whole turn and select the ZoomPeaks Filter, it could blow your speakers. Such effect does not happens when the ZoomPeaks freq Filter value is under 32. It sounds like a cool chatbox effect in some samples
There is another valid reason to use a differnt knob range configuration depending of the filter to use, and furthermore, to use more than one simple Knob to tweak the filters.
If the patch you want to create uses the 4 layers, it would be nice to use 4 knobs, to tweak the filter. Or maybe 2, tweaking the filter of the samples in pairs, and assignate a different knob range amount depending wich Zplane filters you want to use, so you won't get an annoying peak when turning the knob, or no sound output...
The Z plane filters are quite unique. Several of them behave like effects, so the controllers must readapt their modulation parameters to obtain satisfactory results with some of them.
Old 4th November 2017
  #259
Gear Head
 

Lots of good discussion of techniques here. Much to talk about, but let me throw in a few small things.
Quote:
Originally Posted by balma View Post
Keyboard Tuning section
The command station has 12 fully programmable keyboard tuning scales, and 12 from factory. They are hard-wired and stored apart from the user sound section, on the global settings. If you modify one of your user scales, the changes will affect all the patches using it. Each scale has 128 notes.
A cool trick I picked up in one of the user groups is to create scales in specific keys for use with arps. Arp patterns are based on 1/2 steps. So if you play a C with a pattern that goes: +00 +05 +07 +05 etc. you get C-E-G-E. If you play D, you get a D major chord--D-F#-A-F# and so on. But suppose you want your arps to play according to a I-ii-iii-IV-V-vi-vii[dim] scale. You can set a tuning table so that when an arp calls for F#, it play an F, all across the keyboard. I'd give credit for this tip, but I don't remember who mentioned it.

Also, there is a free program called Scala. You can use it to define all kinds of scales. There are also several hundreds of scales already made. In fact some softsynths use scala files as tuning tables. You can also have Scala create a tuning table file that you can send to your P2K-based synth.

Balma's methods of creating templates zero-out, with certain values as a starting point for creating your own presets is excellent. When I make some time, I plan to use this technique to made a basic template for a family of free-floating presets with the initial controllers set to off, so I can switch them around on-the-fly and have the CC values remain in their last state [i.e., no initializing of the CC values when I change.

Related to that there's the idea that you can set up a controller to basically mix between 4 layers to simulate things that otherwise can't be done. Balma spells out the setup with the copying of layers. Create a 1 layer preset; copy it to the other layers [possibly use [A] - all-layer editing if you prefer. Put a different filter type on each of the 4 layers, Lastly, create a knob that has 4 zones so you can sweep smoothly from L1->L2->L3->L4 - the same preset but with a different filter on each. Actually, I think I would build that concept into a family template--maybe 4 knobs that each sweep a specific parameter through four different settings on L1->L2->L3->L4.

About links: Somewhere Aaron answered a question I had about what parameter settings in the links were inherited from the base preset and which parameter setting came from the linked presets. If you work with links, that is good to know. (And someday I will find where I put that!)

As for the A1A2D1D2R1R2 envelopes, I think the manuals explain this a bit in the margin comments (tips?) somewhere. I seem to recall it functions basically as a 6-stage envelope, although if [I think its] the level is the same on say for example A1 and A2, you really only have one stage.

I seem to recall its a pretty standard operation--pressing a key [or having some other trigger] starts the envelop beginning with A1. If you keep the key depressed [or have a wide gate from another source] long enough, the envelope goes through A1->A2->D1->D2. If I remember correctly, the D2 level is what an ADSR envelope calls Sustain [level]. When you release the key [or close the gate] R1 and the R2 shape the parameter [Volume, Filter, or whatever else you are sending an envelop to].

When sorting out how envelopes and specific settings work, I like to use pitch--just because I can better hear the contour of the envelope and the interaction of the rate and level settings at the various stages.

I am tired and don't have my gear actually doing this, so I apologize if any of this is unclear, leaves out steps, or is just plain wrong.

Again, excellent discussions here about the E-Mu/P2K synth engine. They are making me want to ignore all the things I HAVE to do, so I can start playing around again!

Thanks for hosting this excellent thread!

Steve

PS: There are a number of typos in the above; I will re-read this with fresh eyes over the weekend and will fix as many typos as I can.

Also, a warning about using layers (if you haven't found it out yet): they can eat up polyphony very fast!!

Let's say you have 128 voices. Each layer that has a ROM Instrument on it uses at least one voice. So if you have 4 ROM instruments on the base preset, 4 on link 1, and 4 on link 2, that's 128 divided by 12 note polyphony at most. I emphasize at most because some parameters use up 2 voice per layer. IIRC 12th order filters cut polyphony in 1/2 - for each layer. Using layer-based Chorus with any setting but "off" cuts polyphony it 1/2, as I recall, for each layer.

I am not 100% sure, but I think chorus and filter type are parameters that come from each layer in each of the links--that is, they might not be inherited from the base preset. I think it might even be possible to create one mondo preset that uses up most if not all of the 128 voices--so you can only trigger one or two "notes" without dynamic voice stealing happening.

Of course, this might a desirable effect, say for example, live CC tweaking a single preset on on one e-mu, while another is doing beats, playing accompaniment, etc.

Last edited by Dr. Steve; 4th November 2017 at 01:34 AM.. Reason: To point out that I intend to fix typos and to add a PS related to linking and the potential to use up 128 voice polyphony v
Old 5th November 2017
  #260
Lives for gear
 
balma's Avatar
That was great Steve.
Actually I think If you link 3 sounds into one, and use 12 Order Filters and turn on the chorus on all the 12 layers, the poly will be reduced to a few notes.
I constantly have cuts for exceeding the 128 poly. If you use the sounds of the roms, you won't notice any problem even with 16 tracks played at the same time, because most of the factory sounds use 1 or 2 layers only, and thin filters


Must reply with music

I just designed this "sound" just holding a note, and then, creating oscillations of the volume, repetitions, and all kind of crazy stuff, using combinations of the Clock Dividers, LFOs and Envelopes, interacting between them.
The effect of "sequencing" can be easily created on the modulation matrix.
Clock Dividers retrigger the layers on rythmical patterns when sent to the SRetrig destination.
Then, use and pattern LFO to activate and deactivate the clock divider.
And finally, use an aux envelope to modulate the rate of the LFO....

Knobs control both individual and global parameters of the layers, and control multiple patchcors of clock dividers. I´ll put here all the patchcords and data info to create sounds like this one:




So much you can do with a single sound, holding a note during 3 minutes and enjoying the possibilities with the use of some basic math inside the patch cord section

Sasme concept for the second sound demo. A single note holded,. while tweaking patchcords with the knobs: oscillating interactions between the 4 layers:

Last edited by balma; 5th November 2017 at 08:57 PM..
Old 8th November 2017
  #261
Lives for gear
 
balma's Avatar
Some of the sounds in electronic music, do not necessarily have multiple notes across a chromatic scale, or their characteristic variations or "states" are associated with the combination of notes.
By default, harmonic sounds have a chromatic distribution of notes across scales, increasing pitch from left to right in semitones.
There are two screens for this purpose in the patch edition mode: -Transponse: changes semitones in a range of -127 to +127. Transponse changes the position of sample notes across the scale.
The second one changes the notes of the scales.
The difference between TRANSPONSE and TUNING can be understood easily with a kit sample wich does not have a pitch distribution across the keys, but a different sound per key.
Most of the kit samples, have the kick at key C2, and snare is at D2.
If you set tranpose to -2 the kick will move to D2 and snare to E2.
But if you set tuning to -2, the notes of all the sounds of the kit sample, will decrease 2 semitones.
There's also a function wich deactivates the pitch scale: NONTRANSPOSE.
If nontranspose is set to ON, the same note will play on all notes. This function affects only normal samples, not kit samples or those ones wich have different sounds per key.

Now take a BLANK preset, and set nontranspose to ON. Select the note you want with COARSE.

Go to the patch cord sectio and create the following patchcords:

LA PATCHCORDS #01
key+- -> Gain4X +000

LA PATCHCORDS #02
Gain4X -> Pitch +100

LA PATCHCORDS #03
MidiA -> C01Amt +100

Now you can switch from having the same note across all over the keyboard, to the regular chromatic scale moving the knob A from left to right

If you want the opposite:


LA PATCHCORDS #01
key+- -> Gain4X +100

LA PATCHCORDS #02
Gain4X -> Pitch +100

LA PATCHCORDS #03
MidiA -> C01Amt -100

Now you have by deffect a chromatic scale on the keyboard with the knob A totally to the left. but moving it to the right, will turn the pitch scale off.

Combining these tricks with arpeggios, can produce really interesting stuff, furthermore if you program user arpeggios with a wide distribution of different notes across several octaves. Remember you can compose arpeggio user patterns, with a max of 32 notes.
You can hold an arpeggio, and go from a flat note , to a complex sequence of different pitch notes , just turning the knob.


Using the GAIN4X lag processor is the key for this trick. If you don't use the gain4x between the key+- and Pitch connection, moving the knob won't change enough the pitch of each key. Gain4X amplifies the signal of a Source.

You can also use other destinations, not only Pitch, but anyone like filter, resonance.
Sometimes I use this feature on sounds like percussions. For a snare, I don't need a whole chromatic scale, normally a snare has always the same pitch.
Why not, having different filter values for each note, instead semitunes? higher notes will produce brighter filters and viceversa.

For even more complex modulations, you can use 4 layers, and change the pitch scales of each one of them by separated.
The possibilities for reprogram this instrument to your will, are only limited by your imagination
Old 8th November 2017
  #262
Gear Maniac
great, this will help me design a patch where i can change the pitch by turning the knob

-i got a great sequence, but i want to transpose it, and do it 'online' while it's running

so i was thinkin (a year or two...) that i could copy the patch and the sequence and add some knobs to pitch, now it turns out that i should add the gain4x to it too

(that modifier turn out to be handy for quite a few destinations)

anyway, maybe combine these (numbers) with this table i found in the E3 advanced manual, so you can tune and relate/align certain numbers with certain transpositions :
-you can see that each semitone takes a value of 3.125 in the cords-

Attached Thumbnails
The E-mu Command Station XL-7/MP-7/PK-7 and Proteus 2500 tutorials-pitchamounttable.jpg  
Old 10th November 2017
  #263
Gear Head
 

Quote:
Originally Posted by balma View Post
The difference between TRANSPONSE and TUNING can be understood easily with a kit sample wich does not have a pitch distribution across the keys, but a different sound per key. [snip]
Excellent example of using kits/multisamples to show transpose v. tuning and explanations of some cool techniques.It really shows nicely increasing levels of complexity we have with the P2K synth engine.

As for using knobs to control pitch, the basics involve patchcords -> pitch [fine and course] directly. For me the next level of complexity is to use some sort of indirect control, such as one patchcord controlling the >amount< of pitch change produced by another patchcord, (as Balma shows).

There's also the use of modifiers such as gain (which Balma also demonstrates) to control the range), as well as others such as quanitzation, lag, etc.

Since I first got my P2500, I have just loved its creative potential to explore these increasing levels of complexity and its great to have a place (this tutorial thread) to share and expand on them.

Steve
Old 10th November 2017
  #264
Lives for gear
 
balma's Avatar
Quote:
Originally Posted by tiger001 View Post
great, this will help me design a patch where i can change the pitch by turning the knob

-i got a great sequence, but i want to transpose it, and do it 'online' while it's running

so i was thinkin (a year or two...) that i could copy the patch and the sequence and add some knobs to pitch, now it turns out that i should add the gain4x to it too

(that modifier turn out to be handy for quite a few destinations)

anyway, maybe combine these (numbers) with this table i found in the E3 advanced manual, so you can tune and relate/align certain numbers with certain transpositions :
-you can see that each semitone takes a value of 3.125 in the cords-


That conversion table is useful if you want to change the pitch of a sound with a knob. But reaching specific notes in the chromatic scale is not easy, as you can see in the table, most of the notes do not have an accurate value in the knob's range of +100 -100. But it can be done if you create a second layer over the knob and modulate fine tune,

The Gain4X processor isn't necessary for what you want to do. Just assign a knob to the pitch on the mod matrix, and determine the value according to the table you showed before.
You can use 4 knobs, -I normally use M N O P for pitch stuff- to change each one the pitch in a different way. For example:

To jump a 12 notes scale with a the knobM's full turn:

LA PATCHCORDS #01
MidiM -> Pitch +37

LA PATCHCORDS #02
MidiM -> FinePtch +50

To jump 17 notes from with the knobN's full turn:

LA PATCHCORDS #03
MidiN -> Pitch 53

LA PATCHCORDS #04
MidiN -> FinePitch +15

And so on.... You check the table, too see if you must add extra fine tune to reach a specific pitch note with the knob's full tyurn, and create a second patch cord for the same knob.

turning off the LA (All Layers Edition Mode) makes possible to change the values of those patch cords for each one of the layers if you are working on a patch with the 4 layers active, so you can morph from a flat note to a chord with a single knob, when assigning a single knob to produce a specific pitch change on each one of the layers, for exaple
L1 Pitch +2
L2 Pitch +5
L3 Pitch +7
L4 Pitch +9
and looking on the manual's table, wich knob value is the closest to the pitch you want to reach for each layer.

Knobs can control the pitch in a large number of ways. ....

For example you can use the clock dividers to produce a squace oscillation of the pitch in specific clock times.

LA PATCHCORDS #05
ClkHalf -> Pitch +000

LA PATCHCORDS #06
ClkQtr -> Pitch +000
LA PATCHCORDS #07
MidiO -> C05Amt +037

LA PATCHCORDS #08
MidiP -> C06Amt -025


more complex pitch variations can be produced on realtime from flat sounds to elaborated pitch sequences, if you use several knobs to control patchcords wich contain dormant associations between different clock dividers sent to the pitch. And you can always go further, and use specific knob values for each one of the layers.

Also, remember the crappy ribbon controller's pitch range can be changed for each patch, and you can specify a different pitch for each active layer.

this patter is totally based on knob tweaking the pitch of each layer on a different way, while a track is playing the sound on realtime. What I do, Is to send BPM synchronized LFOs to patchcords wich are modulatint the pitch, and leave such associations dormant.

LA PATCHCORDS #01
LFO1 - C02Amt +000

LA PATCHCORDS #02
Clk8th - Pitch +000

LA PATCHCORDS #03
MidiA C01Amt +100

With the previous patchcord sequence, you will send the LFO1 to control the association between a clock divider and the pitch.

A 4th & 5th patchcords, could added to send one of the ADSR to modulate the LFO1's depth for further complex variations.

LA PATCHCORDS #04
AuxEnv+ -> C01Amt +000

LA PATCHCORDS #05
MidiB -> C04Amt +100


This is combination of several patches programmed in such ways. What you hear across the track, are not arppegios or tracks, but simple latched notes being modulated with knobs specially programmed to create pitch variations on different ways. The combination of different note positions, is like creating note sequence variations over the holded note of a patch:


When applied over voice samples, depending of the LFO used and the peaks and valleys of the ADSR envelopes, you can produce chatbox effects, vocalizations, choir combinations, etc.


Last edited by balma; 10th November 2017 at 12:12 PM..
Old 11th November 2017
  #265
Gear Head
 

Real-time MIDI Looping Techniques

Now that I have a P2500 setup at my "Wooden Desk" workstation, I am looking for suggestions to implement a real-time performance modality I test drive 4-5 years ago. The concept is much like recording tracks one-at-a-time (i.e., stopping in between, setting up the next track/layer, testing it, practicing notes, recording the next track/layer, etc.). I have no problems doing this non-realtime layering modality, switching/chaining patterns, etc.

Now, I want to see if I can cultivate that into a real-time midi looping technique: put the sequencer into real-time record (pattern mode) and doing track layering on the fly.

I know I can turn tracks I don't want any more off by simply muting a track; but what I want to do is be able to erase a track on the fly, advance to record on an open track, etc. without stopping the transport, with as few button presses as possible.

I also know I can keep recording over a track and I can erase certain notes as the sequencer loops.

It might be easy--once I attempt it, but in case others have already mastered this performance modality, know how to do what I want to do and have other techniques for real-time midi looping, I thought I'd ask.

Steve

PS: One of my goals is to implement this with families of presets that have the initial controller values set to "off" as discussed above.

Last edited by Dr. Steve; 11th November 2017 at 05:11 PM.. Reason: (1) to correct punctuation, (2) to clarify I mean MIDI looping, (3) to add a relevant PS.
Old 11th November 2017
  #266
Here for the gear
 

Quote:
Originally Posted by Dr. Steve View Post
Now that I have a P2500 setup at my "Wooden Desk" workstation, I am looking for suggestions to implement a real-time performance modality I test drive 4-5 years ago. The concept is much like recording tracks one-at-a-time (i.e., stopping in between, setting up the next track/layer, testing it, practicing notes, recording the next track/layer, etc.). I have no problems doing this non-realtime layering modality, switching/chaining patterns, etc.

Now, I want to see if I can cultivate that into a real-time midi looping technique: put the sequencer into real-time record (pattern mode) and doing track layering on the fly.

I know I can turn tracks I don't want any more off by simply muting a track; but what I want to do is be able to erase a track on the fly, advance to record on an open track, etc. without stopping the transport, with as few button presses as possible.

I also know I can keep recording over a track and I can erase certain notes as the sequencer loops.

It might be easy--once I attempt it, but in case others have already mastered this performance modality, know how to do what I want to do and have other techniques for real-time midi looping, I thought I'd ask.

Steve

PS: One of my goals is to implement this with families of presets that have the initial controller values set to "off" as discussed above.
i remember getting quite cool results doing this (i think) in the dreaded song mode..(or just by changing patterns in pattern mode can not remember now)...if i remember correctly i could have it in record and change pattern and it would wait (que up) until the bar was finished then switch to the new pattern...(maybe the record was disabled...and one could quickly just re-enable it...fuzzy memory...but now that you talk about it i want to go back and test it again)...i remember there was something going on with like a record buffer...it would keep 1 pattern in memory even if i switched to another pattern (i think) but if i switched to a third pattern the first would be gone (i think) but you could use that to your advantage...save on the fly if you want to keep it or choose to loose it....not sure about how good the results where...it was a long time since i experimented with it...but it was quite interesting. i had to set up some tracks ahead of time like the drums so that there was similar sound from pattern1 when switching to pattern2...so then i could liveloop a bass in 2 different patterns... i wanted to do some A/B looping thing...also had an idea about using the first row of tracks as A and the second as B part...one can just swipe quite fast to switch between parts on the mutes...tried to find out if there was a sysex for the mutes and control them from a external sequencer or controller (never found a way to externally control the mutes)...for example i wanted to use a roland RC-300's tracks and when i switched tracks on the rc-300 it would also switch pattern/or mutegroups on the command station...even had a plan to mod the buttons with CV control...haha. but so much to do and so little time. Hope you can make it work!
Old 11th November 2017
  #267
Gear Head
 

Quote:
Originally Posted by mastaman View Post
i remember getting quite cool results doing this (i think) in the dreaded song mode
I thought it was just me that test drove song mode and didn't find it the most rewarding feature of the E-Mu Sequencer!!

Thanks for sharing your recollections. It triggered some ideas I hope to try out.

Quote:
Originally Posted by mastaman
even had a plan to mod the buttons with CV control...haha. but so much to do and so little time.
I can certainly relate! Years ago (it's been tucked away in a box for about 2 years or so), I dabbled with an Arduino and tested some ideas for a super midi controller with as many sliders as I could handle. About 8-10 months ago, I made more progress by buying about 90 sliders (now tucked away in boxes, too). With regard to the E-mu, having extensively gotten into parameters and real-time sysex control, the "plan" is to be able to have as many sliders as possible to be able to control parameters on the fly. I think my first idea was to have 256 sliders!!! Also, at least 32 buttons in part to toggle 32 arp patterns laid out in front of me at once.

If I ever get around to it, and assuming it will work the way I imagine it, having CV to midi might be an excellent extension.

Steve
Old 11th November 2017
  #268
Gear Head
 

Quote:
Originally Posted by mastaman
...tried to find out if there was a sysex for the mutes and control them from a external sequencer or controller (never found a way to externally control the mutes)
Let me know if you get back into it. I have some resources to show how to do it (browser-based html+ stuff using the jazz-soft plug-in) that I was developing but never pushed to my website because (1) most people couldn't make use of it (i.e. of limited, highly specialized use), (2) developing the pages just to explain how to do it was getting to be too tedious, and (3) putting it on my site as it was would probably mean I might have to answer more questions than I had time for. Also, around the time I was working on it, (4) web-midi was being developed/catching on and I wasn't sure if the jazz-soft plug-in would be replaced.

That being said, using a web browser and the jazz-soft plug in, I believe I have it set up with virtual buttons that will send every possible button presses [and maybe button releases--it's been at least a year since I worked on it] from 0x00 to 0x7F. As I recall, not everything has a virtual [midi via sysex] counterpart, and the logic of the control values is only partially logical, but I think all of the mute/trigger button presses can be emulated.

It is complex--for example, you can turn the leds on and off without actually simulating the button presses and button release, as I recall. Not for the non-technical, non-obsessive user of the E-Mu sequences, but if you get back into it and want the sysex coding to emulate button presses, I can try to revisit the non-released web-pages I have and maybe push it to my site with a private password--in part so the various internet bots don't suck up my bandwidth downloading the pages day after day!!!!

Footnote: it is possible to embed sysex commands in E-Mu sequencer patterns by editing *.mid files using software and then ploading them via e-loader. I think it might also be possible to remove the "pre-roll" data from E-Mu patterns. As I said, not for the casual gear user!!!

Last edited by Dr. Steve; 11th November 2017 at 07:11 PM.. Reason: (1) to add a missing ')' and (2) to add reason (4) for not making the web-pages I developed public
Old 12th November 2017
  #269
Here for the gear
 

Quote:
Originally Posted by Dr. Steve View Post
Let me know if you get back into it. I have some resources to show how to do it (browser-based html+ stuff using the jazz-soft plug-in) that I was developing but never pushed to my website because (1) most people couldn't make use of it (i.e. of limited, highly specialized use), (2) developing the pages just to explain how to do it was getting to be too tedious, and (3) putting it on my site as it was would probably mean I might have to answer more questions than I had time for. Also, around the time I was working on it, (4) web-midi was being developed/catching on and I wasn't sure if the jazz-soft plug-in would be replaced.

That being said, using a web browser and the jazz-soft plug in, I believe I have it set up with virtual buttons that will send every possible button presses [and maybe button releases--it's been at least a year since I worked on it] from 0x00 to 0x7F. As I recall, not everything has a virtual [midi via sysex] counterpart, and the logic of the control values is only partially logical, but I think all of the mute/trigger button presses can be emulated.

It is complex--for example, you can turn the leds on and off without actually simulating the button presses and button release, as I recall. Not for the non-technical, non-obsessive user of the E-Mu sequences, but if you get back into it and want the sysex coding to emulate button presses, I can try to revisit the non-released web-pages I have and maybe push it to my site with a private password--in part so the various internet bots don't suck up my bandwidth downloading the pages day after day!!!!

Footnote: it is possible to embed sysex commands in E-Mu sequencer patterns by editing *.mid files using software and then ploading them via e-loader. I think it might also be possible to remove the "pre-roll" data from E-Mu patterns. As I said, not for the casual gear user!!!

Haha, not many seem to like the song mode (I have not used it so much but the time I used it …it seemed it was not fully developed) but it could be a hidden gem for looping…and if you get fast with that x-mix (not sure the name is correct, but the feature where you could import tracks from other patterns) feature that could maybe also be useful…

also the loop feature in pattern mode could be used…(esp if you have access to the sysex commands and can move the loop points in realtime with sysex!!) this would be super cool for looping
Example:
Make a 8 bar long pattern
use it as 2, 4bar loops (A and B) and have your sysex web code switch start end end points to 4bar loop A or 4bar loop B!

I remember it was a little fiddly to setup with the shift buttons…
Shift + stop to enter loop mode
Then I you had to hit the SELECT button (in time with the music, maybe + the shift button i forget now) to set the start point of the loop
And the EDIT button to set the endpoint of the loop..then shift + stop again to quite looping mode…but if this can be done via sysex!! Wowowow…there you have something for a live looper!

I can relate 100% with mods and stuff that pile up in a box somewhere :D , also got an Arduino to test and build some midi controller or Mido to CV thing…But the Adruino uno I got is not ideal for analog CV…good for triggers…but not to many digital writes, so I guess I need multiplexing….hahaha…90 sliders!! THAT IS ALOT!!! Maybe look at the BCR2000? it has presets

I was thinking also to control it from the modular with CV…vactrols on the knobs and transistors on the buttons should work..I think the ting I found that is a little limiting is that midi is not processed in the matrix…so I think I will just use a few of the audio outputs to make random gates/triggers and stuff from the clock dividers and noise and all the other goodie processors from the mod matrix…I have big plans for using the arps also with the modular…have a few dedicated arp tracks, and combine them divide/mult them out summed to a audio out and so on for interesting triggers to the modular.

WOW, impressive! that jazz-soft stuff sounds great! I will let you know when I go back to setting up the command station. Super interesting for me!

Do you have any sysex for the set pattern loop start and end feature?

To be honest I have not done so much Sysex programming…dabbled a little with som hex tables on a midi event programmer box. And some with the BCR2000…there is a pc editor for the BCR2000 to input custom Sysex strings...so i am not so advanced in that regard…I wonder if that would work to map out custom commands on the command station??? That could be a real winner!

I will let you know when I get back into it…please report back your findings!

Thanks
Old 13th November 2017
  #270
Gear Head
 

Quote:
Originally Posted by mastaman View Post
Do you have any sysex for the set pattern loop start and end feature?
I have read about that feature many times, but never used it. Sequencers parameters are totally undocumented if they even exist. The only thing I could do is simulate button presses with remote control sysex commands. I used that technique to generate from cakewalk software the creation and storing of a full bank of randomized presets.

When the time comes, if you can make a list of the manual button presses to make it happen (starting from a specific/constant starting point), I could test the virtual keypresses to see if it would work.

If anyone else is interested, remote control sysex is described on pdf page 40 [numbered page 36] of the E-Mu Sysex manual version 2.2" The events listed are just a start.

Steve
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump