Quantcast
EMU Proteus 2000, sample start, loop modulation - Gearspace.com
The No.1 Website for Pro Audio
EMU Proteus 2000, sample start, loop modulation
Old 6th January 2019
  #1
Lives for gear
EMU Proteus 2000, sample start, loop modulation

I noted in the manual for the Proteus 2000 that you can actually modulate the sample start, loop and retrigger for each layer in a patch. None of my synths have that feature. Very cool possibilities here.

Have any of you worked with this feature in designing your own sounds?
Old 6th January 2019
  #2
Lives for gear
 
🎧 5 years
Quote:
Originally Posted by AxxeMan317 ➡️
I noted in the manual for the Proteus 2000 that you can actually modulate the sample start, loop and retrigger for each layer in a patch. None of my synths have that feature. Very cool possibilities here.

Have any of you worked with this feature in designing your own sounds?
yeah it's useful for varying the attack transient with velocity etc.
check out XFade as well: you can use the beginning of either
primary or secondary and fade into the body of the other, and its filter.

then go crazy and start linking presets.
Old 6th January 2019
  #3
Lives for gear
Thanks for the info. Have you messed with the loop and retrigger mod destinations? I assume loop actually moves the loop point during playback? I wonder what happens if the sample start is modulated past the loop point. Reverse playback would be awesome but I doubt that it does that.
Old 6th January 2019
  #4
Lives for gear
 
xanderbeanz's Avatar
 
🎧 5 years
I believe the EMU’s loop modulation is a keyed event, ie you need to press a new note for the value to be updated. Nothing to stop you sending it loads of fast MIDI notes though.

It was only really the Ensoniqs that could modulate the loop position on a held down note.
Old 6th January 2019
  #5
Lives for gear
Bummer limitation, but thanks for the info! I assumed you might know the answer. :-)
Old 6th January 2019 | Show parent
  #6
Lives for gear
 
🎧 5 years
Quote:
Originally Posted by AxxeMan317 ➡️
Bummer limitation, but thanks for the info! I assumed you might know the answer. :-)
i've only got an old Orbit so i don't have those features. have to upgrade
really. it's definitely worth messing round with Start and XFade parameters.
then you start swapping waveforms and filters, copying presets onwards
until you find something you like.
Old 6th January 2019
  #7
Lives for gear
 
Plaeground's Avatar
 
🎧 10 years
Look up the emu command station thread for all the good tips
Old 6th January 2019
  #8
Lives for gear
Thanks! Will do.
Old 6th January 2019 | Show parent
  #9
Lives for gear
Quote:
Originally Posted by xanderbeanz ➡️
I believe the EMU’s loop modulation is a keyed event, ie you need to press a new note for the value to be updated. Nothing to stop you sending it loads of fast MIDI notes though.

It was only really the Ensoniqs that could modulate the loop position on a held down note.

You may well be correct, but I noticed this on the XL-1 manual. It specifically denotes sample start can only be modified at note-on (duh) but doesn't mention that for loop or retrigger. Maybe it's more flexible than we think.

From modulation destinations chart

‘SStart (Sample Start) -note-on)
SLoop (Sample Loop)
SRetrig (Sample Retrigger)

Unfortunately, this is the kind of esoteric feature that the engineers design in but very few users ever look at, so it's always hard to find usage examples. I think I must be in some small group of nerds who dig into a synth manual looking for mod-matrix source and destination charts. :-)

The mod matrix in the Proteus is a monster to be sure.
Old 6th January 2019
  #10
Lives for gear
So it looks like sample loop modulation is actually a realtime modulation feature!

Check out this thread. Just search for Balma's explanation using SLoop. Really powerful stuff. I certainly don't have this feature on any other synth.

The E-mu Command Station XL-7/MP-7/PK-7 and Proteus 2500 tutorials

Thanks again Plaeground for sending me looking in that direction.
Old 6th January 2019 | Show parent
  #11
Lives for gear
 
xanderbeanz's Avatar
 
🎧 5 years
Quote:
Originally Posted by AxxeMan317 ➡️
So it looks like sample loop modulation is actually a realtime modulation feature!

Check out this thread. Just search for Balma's explanation using SLoop. Really powerful stuff. I certainly don't have this feature on any other synth.

The E-mu Command Station XL-7/MP-7/PK-7 and Proteus 2500 tutorials

Thanks again Plaeground for sending me looking in that direction.
Well that’s good news. I’m gonna put the 2500 command station back on my list, I kinda want it for a couple of other reasons too.
Old 7th January 2019
  #12
Lives for gear
It's unfortunate that these really techie nerd features are rarely well documented. It shouldn't take 20 years for us to discover a cool feature like this.

2500 is a cool machine. I would go that way but the P2000 is much more compact and I'm already out of room. :-)
Old 8th January 2019 | Show parent
  #13
Gear Maniac
 
🎧 5 years
Quote:
Originally Posted by AxxeMan317 ➡️
You may well be correct, but I noticed this on the XL-1 manual. It specifically denotes sample start can only be modified at note-on (duh) but doesn't mention that for loop or retrigger. Maybe it's more flexible than we think.

From modulation destinations chart

‘SStart (Sample Start) -note-on)
SLoop (Sample Loop)
SRetrig (Sample Retrigger)

Unfortunately, this is the kind of esoteric feature that the engineers design in but very few users ever look at, so it's always hard to find usage examples. I think I must be in some small group of nerds who dig into a synth manual looking for mod-matrix source and destination charts. :-)

The mod matrix in the Proteus is a monster to be sure.
On EIV samplers the maximum amount of modulation of loop position is very limited, barely enough for a very lo-fi wavetable. However, I’ve never tried combining sample start modulation and retrigger, something to do!
Old 9th January 2019
  #14
Lives for gear
Unfortunately we really have just about no information on how this actually works at least in the EMU docs. I may eventually get out my scope and play around with a simple wave to try and determine just how flexible this is. Is it just a minor tweaking of the loop point or something close to Ensoniq transwave modulation? Who knows at this point.
Old 9th January 2019
  #15
Lives for gear
 
rom play thru
Old 9th January 2019
  #16
Lives for gear
Could you elaborate on that?
Old 6th April 2019 | Show parent
  #17
Gear Head
 
🎧 5 years
Quote:
Originally Posted by AxxeMan317 ➡️
Could you elaborate on that [rom play thru ]?
I just stumbled onto this thread. Apologies if the discussion has been brought up or moved elsewhere.

A few years ago, I played with some of the ROM sample parameters. In fact, I might still have some experimental User Presets where I played with the sample-based virtual patch cords destinations. I don't recall the specific differences between the E-Mu Sampler and the different E-Mu ROMpler OSes, but here's the concept as I came to understand it (someone correct me if I am wrong):

The samples/multisamples exist (more or less) in contiguous space. In the ROMplers, these are the ROM Instruments. The OS uses pointers. Somewhere along the line, E-Mu users might have been able to jump across/over the place where one sample started and one ended. I don't know if that was ever possible in any of the OSes (an earlier-than-the-last release or a beta version) with the original Sound SIMMs.

However, somewhere along the way some newer SIMMs had the rom-play-thru feature built in--not sure if the sample boundaries were coded differently, if the maps to the sample-space were treated differently in the baby flsh or what.

The idea (as I understand it) is that certain specified ROM Instruments allowed E-Mu users to cross the sample boundaries to have extra fun with the sample-based patchcord parameters.

I recall reading something about this, but I cannot find it in several manuals and Addenda I have looked at. I thought it was as a margin tip, but maybe not. I do see that the MP-7 ROM has 0758 rom:Thru Memory and 0759 rom:Vowels.

Also, Preset 062^2 on the MP-7 ROM is rom:Play Thru and features the 0758 rom:Thru Memory ROM Instrument on all 4 layers.

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

TEST of 'SStart patch destination using rom:Thu Memory:

1. Select a bts preset.
2. Change ROM Instruments on L2, L3, L4 to off.
3. Change L1 ROM Instrument to 0758 rom:Thru Memory
4. Turn Headphone volume to a low setting.
5. Press audition.
(If you pressed audition earlier, maybe you also turned down the volume already.)

6. Set up a layer-level patchcord (I used #24 since it was unused):
DC ->'SStart +000
7. Step the patchcord amount to +100 one-by one.

Verification that sample-space borders are not crossed with other ROM Instruments:

8. Change L1 ROM Instrument to something else.
9. Go back to the DC -> 'SStart test Patchcord page and vary the amount.

Well, that's (1) as far as I recall/know and (2) have tested. Thanks for raising the question, I have been wanting to get back to the concept of testing sample space boundaries for a few years. If anyone else has done testing, developed presets using this feature, I'd be interested in reading the details.

Oh--yeah, as mentioned, for randomized samples being triggered XfadRand -> 'SStart is wild! (JMO). Adding an LFO -> Lag patchcord (I used #23 , also previously un-used) further varies the glitchy wildness. With some mondo reverb (I used panning reverb to test with max decay) you can make something that sounds like a 1950s era sci-fi flick with a run-amok computer.

OK--Set the BPM way down 0013 and manually varied Release1 (Rate = 20 to 127) so some random sounds would sustain and some would have have short envelopes. Hmmmm. Works at BPM = 0001, too, though its very slow--like the sci-fi character having a hallucinogentic experience while the aliens have the character sedated with visual effects of trying to escape but in slow-mo--stuck in alien quagmire, maybe.

OK--saved as XFy:0000-000.

/me tries duplicating L1 onto L2. And L3. And L4.

Saved as XFy:0000-0000a.

If others are interested in extreme wild randomess, maybe I will make some additional variants and post.

Super thanks for starting this thread!!! (I feel like a kid again! :-D)
Old 7th April 2019
  #18
Lives for gear
Thanks for the details. Unfortunately, the Composer, Protozoa and Techno Construction SIMMs have no waves listed with the word THRU. :-(
Old 7th April 2019 | Show parent
  #19
Lives for gear
 
xanderbeanz's Avatar
 
🎧 5 years
Quote:
Originally Posted by AxxeMan317 ➡️
Thanks for the details. Unfortunately, the Composer, Protozoa and Techno Construction SIMMs have no waves listed with the word THRU. :-(
These thru waves would be the entire Rom as one waveform right? Does the sloop stuff still sound cool and glitchy on regular waveforms?
Old 7th April 2019 | Show parent
  #20
Gear Head
 
🎧 5 years
Thanks for the interest! Excellent questions and things to explore!!!

Quote:
Originally Posted by AxxeMan317 ➡️
Thanks for the details. Unfortunately, the Composer, Protozoa and Techno Construction SIMMs have no waves listed with the word THRU. :-(
True; nor the XLEAD, XL-7, or P2500 that I can see. There are some other wild-and-crazy randomness techniques with non-Thru Memory ROM Instruments but if memory serves me well, the ones I tried don't have the hardcore glitch-factor.

Quote:
Originally Posted by xanderbeanz ➡️
These thru waves would be the entire Rom as one waveform right? Does the sloop stuff still sound cool and glitchy on regular waveforms?
As for playing the data as one waveform, I am not sure; I haven't gotten that with my testing, but I have not done systematic minimalistic testing--yet.

So far I am really only using 'SStart with some Amp Env tweaks and FXA settings. With the parameters and settings I have tested, it is more like the randomness modifiers choose a >>starting point<< with each note trigger.

I had too much fun yesterday to be systematic! Today I have branched out a little--looking for specific modifiers and parameter settings that work well with yesterday's presets which primarily use just the 'SStart randomness and the Amp Env and FX parameters so BPM can go from 001 to 300 and continue endless sounds.

One possibility is L1 with settings to create mid to higher range sounds and L3 with settings to create lower range/muffled/filtered/ambient sort of sounds--with the P2500's 4x4 knobs (and/or external sliders) controlling the modifiers (direct and indirect use of patchcord amounts); some tweaking L1, others tweaking L3. L2, L4 and links reserved for additional layering.

Just did some quick tweaks using the patch on non-Thru Memory ROM Instruments. With the front panel on L1 INSTRUMENT ROM, manually dialing through the instruments is similar.

More experimentation is warranted!! SLoop and SRetrigger to be explored as well.

Last edited by Dr. Steve; 7th April 2019 at 07:57 PM.. Reason: to fix typos
Old 8th April 2019
  #21
Gear Head
 
🎧 5 years
2019-04-10: Please see Addendum at the bottom regarding attempts to isolate the parameters involved.

I have experimented a bit more with my basic test presets and have some insights that lead in the direction of how the 'SStart parameter works esp. with the Thru Memory ROM Instrument. The layer-level TRANSPOSE feature [+/- 127 on my P2500 with OS 2.00] produces an amazing assortment of randomness + glitchiness. I only wish it were a virtual patchcord destination!!! Four layers of the same preset with slight variations in the Amp/Vol Release 2 envelop and different layer-level transpose setting has been producing an always changing "Noodle" for hours, despite the fact that the trigger is the same audition riff!!!!!!

The concept is simple and well-known for traditional ROM Instruments:
(1) With pitched, sustaining instruments [> samples from the same instrument/waveform <], you can create a thick preset by using the same ROM Instrument (or even different/related ROM Instruments) on different layers in different ways including by using a different transpose value. Well-known technique.
(2) With un-pitched and often non-sustaining > multisamples < transpose can be used to slide the multisample up and down the keyboard. Again, a well-known technique.

With the Thru Memory ROM Instrument, as triggered by a BTS audition riff for example, the XfdRand -> 'SStart patchcord seems to be randomly grabbing samples from the whole array of samples (ergo, "contiguous sample space")-regardless of whether the sample/multi-sample is of type (1) or type (2) above.

So, single sounds (homogeneous samples/multisamples from the same instrument but with different key origins) get pitched up or down by the layer-level transpose amount. To my ear, the effect sounds like the rate at which the sample is played is offset from the original sample.

On the other hand, multisamples where different instruments have been assigned to different keypresses retain the original sample rate because it has been assigned properly with respect to they keyboard layout.

L1/L2/L3/L4 TRANSPOSE is nothing new; however, with the way Thru Memory and XfdRand allows different types of samples/multisamples to be treated, I now have a deeper appreciation of the following (from one of the manuals):

Quote:
The Transpose parameter lets you transpose the key of the current layer’s Instrument. Transpose works by shifting the keyboard position in semitone intervals relative to middle C. Use this parameter to transpose different layers apart by semitone intervals. For example, by transposing one layer by +7 semitones, it will track other layers at a perfect fifth interval.

The transposition range varies for each instrument. The software limiters have been removed in order to provide extreme transpositions. The upper and lower transposition range is now limited by the original pitch, the sample rate of the instrument’s sample and the G-chip interpolation hardware.
Quote:
The Tuning parameter changes the pitch of the key in semitone and 1/64 semitone intervals. Use the Coarse field to shift the tuning by semitone intervals. Use the Fine field to shift tuning by 1/64 semitones (or 1.56 cents) intervals.

Background: Transpose vs. Coarse Tuning
Transpose works by shifting the keyboard assignment of the Instrument (as if you were sliding the keyboard up and down with the Instrument remaining in the same position). Coarse Tuning keeps the instrument placement on the keyboard and actually tunes the samples up using a digital process called interpolation. Use Course Tuning on drum instruments to change the pitch while keeping sample placement constant.

Coarse Tuning can also be useful to slightly change the timbre of the instrument.
2019-04-10 Addendum:
In trying to isolate some of what's going on with the complex extreme glitchy randomness preset and especially in trying to see if Transpose worked in these two ways when using the rom:Thru Memory ROM Instrument, I explored different combinations of (1) ROM Instruments, (2) Transpose, (3) Tuning, (4) Sound Start, and (5) 'SStart.

At this point, it now seems that with the rom:Thru Memory Instrument Transpose does not slide the individual samples in a multi-sample up and down the range of notes played. Transpose seems to change the pitch of the sample selected [i.e., the rate the sample is played] just as it does with with the ROM Instruments in (1) pitched, sustained instruments.

At some point, I will explore the other parameters I was using to try to determine what caused the preset to sound like it was stepping through sequences of different samples. Possibilities include some combination of SLoop, SRetrigger, 'SStart, a Lag Processor, the Audition Riff, etc.

Update: The ROM Instrument itself plays through the samples one-after-the-other, such as by playing a note with sustain on. Also, it is possible using various settings to control where in the sample space a trigger will start (as well as the pitch/sample playback rate). So, with rom:Vowels it is possible to trigger I-O-U or just U.

Starting a trigger at a specific sample can be controlled with a number of different parameters. I found using multiple patchcords gave finer precision is selecting a starting point than using other parameters alone--e.g., MidiA -> Amount of MidiE -> 'SStart gave greater precision than Sound Start, etc.

Last edited by Dr. Steve; 10th April 2019 at 03:26 PM.. Reason: To add additional details and make a possible correction
Old 9th April 2019
  #22
Lives for gear
That does sound very cool. I'm really wishing one of my ROMS had that option. :-(
Old 1st September 2019 | Show parent
  #23
Lives for gear
 
psionic11's Avatar
Quote:
Originally Posted by Dr. Steve ➡️
At this point, it now seems that with the rom:Thru Memory Instrument Transpose does not slide the individual samples in a multi-sample up and down the range of notes played. Transpose seems to change the pitch of the sample selected [i.e., the rate the sample is played] just as it does with with the ROM Instruments in (1) pitched, sustained instruments.

Update: The ROM Instrument itself plays through the samples one-after-the-other, such as by playing a note with sustain on. Also, it is possible using various settings to control where in the sample space a trigger will start (as well as the pitch/sample playback rate). So, with rom:Vowels it is possible to trigger I-O-U or just U.

Starting a trigger at a specific sample can be controlled with a number of different parameters. I found using multiple patchcords gave finer precision is selecting a starting point than using other parameters alone--e.g., MidiA -> Amount of MidiE -> 'SStart gave greater precision than Sound Start, etc.
Thanks for verifying this. If I understand correctly, it's really a pretty straightforward implementation -- there is one contiguous block of ROM that holds all the Instrument samples. Therefore there are separate sections in memory that define how each Instrument will be rendered as a playable sound (like the data found in soundfonts, keymaps, multisample zone definitions, etc). Transpose and Tuning operate on the samples' pitches (your Example 1), rather than on an Instrument keymaps (your example 2, which is like shifting a drum kit to play the same rhythmic pattern across different drums, a fun technique to get more mileage out of stock beats). The sample info is separate from the instrument info.

In a way, it's like picking up and dropping a needle in specific places on spinning vinyl.

So you can implement rough shod pattern sequencing using waveROM scanning via rom:Thru Memory. I imagine it sounds somewhere between wave sequencing and scanning through a wavetable.

With the I-O-U triggering, I'm curious how you are determining where the sample start times are. Is there a parameter for sample start Index you can use in patch cords? In other words, how do you know where exactly each new sample (like the vowel I) begins? Or perhaps modulation simply drops the needle anywhere in the middle of a sample. You didn't mention pops and clicks, so there must be some intelligent method of only starting a new run at a zero crossing point.

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

After I finish balma's original thread, I'm heading back to the MP7 manual. Both for the bird's eye view as well as for the detailed parameter listings. It's almost as fun learning the machinations of a new synth's possibilities, and experimenting to hear and flesh out the theory, as much as it is making actual music.

Almost.
Old 15th September 2019 | Show parent
  #24
Gear Head
 
🎧 5 years
Quote:
Originally Posted by psionic11 ➡️
I imagine it sounds somewhere between wave sequencing and scanning through a wavetable.

With the I-O-U triggering, I'm curious how you are determining where the sample start times are. Is there a parameter for sample start Index you can use in patch cords? In other words, how do you know where exactly each new sample (like the vowel I) begins? Or perhaps modulation simply drops the needle anywhere in the middle of a sample. You didn't mention pops and clicks, so there must be some intelligent method of only starting a new run at a zero crossing point.
I will try to find the test recordings I did--an audio file is worth a million words!

But, yeah; scanning through a wavetable. I called the demo streams I did Electro-Music from Hell or something like that.

IIRC triggering a note event started in the same spot (like the start of a record), so I played with parameters to get the needle to drop in different places, to change rpms, etc.

Much of what I did had knobs (CCs) change values, including amounts of randomness. So, I started with the knobs I had assigned at zero, tweaked them, and then could return to the original starting point.

From the manual, you probably learned that the layer level patchcords have a control source, a control destination, and an amount. IIRC I probably used DC to 'SStart and tweaked the amount to find the starting point. I believe I had chains of virtual patchcords where one knob essentially controlled the range of the second knob.

I took down paper and pencil notes to map the preset as I was building it; not sure if I wrote down the final versions I saved. If you get a good, working E-Mu* with one of that PHATT roms in it, I would be happy to dig those out, and reconstruct what I did. *Either a Command Station/P2500 or a smaller P2K sized variant.

Just remembered, I used BTZ riffs which also determined sounds used in the preset.

Thanks for your interest.

Steve
Old 17th October 2019
  #25
Lives for gear
 
xanderbeanz's Avatar
 
🎧 5 years
I have just purchased a Mo-Phatt in good condition. It will arrive next week.

So I will be able to try all of this SStart Sloop fun for myself, with the rom-playthrough waveform.

Looking at forums, it would appear that SStart can sweep through the entire sample, and needs note ons to be effective...but I can set up a really fast arp for that anyway.

Sloop seems to only move the loop up to 128 samples forward...not very much at all, but apparently you can mess about with the GAIN X4 multipliers.

We’ll see!
Old 17th October 2019 | Show parent
  #26
Lives for gear
Quote:
Originally Posted by xanderbeanz ➡️
I have just purchased a Mo-Phatt in good condition. It will arrive next week.

So I will be able to try all of this SStart Sloop fun for myself, with the rom-playthrough waveform.

Looking at forums, it would appear that SStart can sweep through the entire sample, and needs note ons to be effective...but I can set up a really fast arp for that anyway.

Sloop seems to only move the loop up to 128 samples forward...not very much at all, but apparently you can mess about with the GAIN X4 multipliers.

We’ll see!
I'm interested in your results. 128 bytes is certainly limited but maybe those modulation processors can get you beyond that.

I have a romless Mo-Phatt basically as a hardware backup for my 2000. Unfortunately I don't have any ROMs that have the play-thru wave selection.

It will be interesting to see just how close these features can get to the Ensoniq ASR/TS capabilities.
Old 17th October 2019 | Show parent
  #27
Lives for gear
 
xanderbeanz's Avatar
 
🎧 5 years
Quote:
Originally Posted by AxxeMan317 ➡️
I'm interested in your results. 128 bytes is certainly limited but maybe those modulation processors can get you beyond that.

I have a romless Mo-Phatt basically as a hardware backup for my 2000. Unfortunately I don't have any ROMs that have the play-thru wave selection.

It will be interesting to see just how close these features can get to the Ensoniq ASR/TS capabilities.
I think it’ll be cool in a different way from the Ensoniqs. I might nab a 2000 or even 2500 at some point and whack the Rom in.

I paid ROM prices for the whole complete rack, lol.
Old 17th October 2019
  #28
Gear Addict
 
Muied Lumens's Avatar
 
🎧 5 years
The EMU Morpheus could scan through the entire memory if you wanted. It was great for chaotic sounding chords, resuting in an anarchy of instrument noise. It reminded me of Naked City tracks at times.

You could of course simulate this by creating a long string of multi-samples (notes at different intervals, and drums) and loading them into any sampler, then hit a massive chord and just hold it down.
Old 17th October 2019
  #29
Lives for gear
Is it only the Phatt rom that has the play thru feature? Oh wait XL7 does it as well right?
Old 18th October 2019
  #30
Lives for gear
I'm guessing that modifying the loop length is going to result in obvious pitch shifts as the Proteus does not have anything like the really cool Ensoniq LOOPSTART-X modulation mode that actually does realtime pitch correction as the loop length is modified.

Then again, does anything else have that? :-)

Ensoniq was so cool. CURSE YOU CREATIVE!!!
📝 Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 57 views: 43648
Avatar for Bender412
Bender412 24th October 2016
replies: 199 views: 38044
Avatar for wavejockey
wavejockey 30th October 2018
replies: 4 views: 2695
Avatar for CholoMarijuano
CholoMarijuano 30th October 2013
replies: 0 views: 455
Avatar for BHW
BHW 13th June 2018
Post Reply

Welcome to the Gearspace 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…

Forum Jump
Forum Jump