2029

Depends a bit on the interface, but yeah. I’ve run straight out of several mixers to my interfaces (Presonus Firestudio Mobile and Babyface Pro) and also to Digitakt and Octatrack with no problems. I believe only XAOC Praga of my mixers was advertised as having line outs. I’ve found gainstaging inside the modular with different elements to be much more important than what the output module “officially” puts out.

2 Likes
2030

I guess mimeophon will be a good match at it is a really good module. I have one paired with a Tiptop Z5000 / clouds / nebulae and this is offering a ton of sonic possibilities

By the way nebulae is a great module not only for granular but also for the different instruments mode you can load

1 Like
2031

you can. Rings is in every rig out there (mine included), so don’t feel obliged to hold on to it. Be sure to explore the alternative modes (flashing LED), but ultimately if you find that the resonator part does not appeal to you, remove it. I personally like to explore the edges of its parameters - feeding it audio-rate stuff and momentary hi/lo signals as modulation. But it sounds like it sounds. It kind of needs to be sent into another module. Granular is an obvious choice, but e.g. a Mimeophone pairs quite well with Rings.

Rene v2 offers a lot more in the same form factor, but if you’re not into programming sequences, even on the fly, then that’s not-insignificant hp space that you can give to e.g. a Morphagene, Mimeophone, and Wogglebug…

1 Like
2033

Layering and mangling sounds is exactly what I do with eurorack. I can’t speak for the results but the process to me looks like what you are describing.
From my observations sample based modules, like Morphagene, Lubadh or Nebulae and multi input/output filters are best for this, at least in terms of speed and flexibility. Essentially splitting/cross-fading audio into different processing chains and then reassembling them.
I too struggle to find use for synthesis modules, as I find myself unable to produce quality results a lot of the time. Often times the sound is really stale or very chaotic.
There is one thing I gather from your posts about eurorack. It’s the desire for it to be quick and immediate in its results. My personal reality is that its immediate in its process and unpredictable in results. As the complexity of the patch grows it gets harder to switch things fast, and at some point it’s easiest to start anew.
Also I found it really hard for myself to interact with a lot of devices at the same time. Since getting octatrack I used it with modular a couple of times, but most of the time both of them are just too much for me :slight_smile:

3 Likes
2035

I think best tools are shortcuts to what makes us “click”. Your words sound like you search for the exact opposite of “shortcut”, while also wanting for it to be compact or contained:) I honestly think that you need either a big case of modules or a nice way to integrate pc into your setup.

2 Likes
2036

So I’ve been using an ES-8 for a bit now, and it’s great for so much of what I do. The thing is, I’m always, always, wanting more outputs. My heart paused a beat when I saw the ES-9 thinking that it had 14 outputs, or when reading the (“16-in/16-out”) in the description only to be disappointed to see that it was still only 8 outputs, with 14 inputs.

I’ve looked the Expert-Sleepers page a bit and beyond find all the expansion names/functions confusion, think that nothing will let me expand the ES-8 if I’m using the onboard ADAT I/O to connect it to my audio interface already. (Happy to be corrected here if I’ve misread that (actually, can you use an ES-9 + ES-5 + ESX-8CV to have 16 CV outs + 8 trigger/gate outs?))

So at the moment I’m looking to see what the options are for having “loads” of CV outputs coming from my computer. Ideally this would be at audiorate, but curious to know what other options are available. I’ve seen a bunch of MIDI-to-CV things, but wondering if there’s anything in the i2c/crow landscape that would do allow for something similar. Like a crow + some expanders (or multiple crows) or something. From the crow page it’s 16bit which is great, though doesn’t mention the CV rate.

2037

You can use 6 of the expansions ports of the ES5 for a large number of CV/gate/midi outputs

I use ES5, to connect ESX-4CV AND 3X ESX-8MD = 30 plus outputs, also use ES6 and ES7 for inputs - endless possibilities!!!

OS is on here Rodrigo and is most helpful

couldn’t live without expert sleepers :slight_smile:

2038

So I guess an ES-5 can connect to an ES-8, but then I would lose two of my outputs from the ES-8?

The ES-5 itself is just binary outputs, so that’s handy for that kind of stuff, but with additional ESX-8CVs, would that then gobble up additional “real” outputs from the ES-8 in order to add more CV-rate outputs from ESX-8CVs?

So, if I had an ES-8 + ES-5, I could then have, um, 4 ESX-8CVs in order to have 32 low-res-ish outputs with no functional outputs from the ES-8?

And if you go from an ES-9 does the piggybacking of an ES-5 take one of the minijack outputs, or can it absorb from the headphones or wherever else?

2039

So this “make everything myself” thing I TOTALLY get! I have a similar sort of… “false image” of what it means to create that I tend to hold myself to.

The issue is that I’m not Isaac Newton, Tesla, Don Buchla, and Bill Evans all wrapped into one. Gotta stand on the shoulders of giants and use whatever tools you have. The other anecdote here that helps assuage that feeling is: a painting is done when you can stand back and say “I did that.” Spoiler: that never happens.

I’ve nourished a focus on “what moves me?” over “what can I do?” – it’s not about me and my abilities and my expression, but about the connection between my ears, hands, and heart, and about casting a moment I perceive as beautiful in the best light.

It’s sometimes hard to maintain this mindset, but it’s MUCH better for learning, and for having the confidence to press record – it’s just the most beautiful thing you saw today, not the most beautiful thing you’re capable of rendering. Always witnessing.

I know this is sort of an abstract, mind set thing that isn’t really what you asked, but I identify with the attitude you expressed and realized, over years and years, that it was hurting me.

7 Likes
2040

I think it makes more sense when you learn what is actually happening with the expanders.

Each individual output on the ES-3, ES-8 or ES-9 are 24-bits, running at audio rate (they are just DC coupled audio outputs).

The ES-5 takes two of those 24-bit channels, and allows you to repurpose the total 48-bits for other data. Specifically, it allows for the connection of up to 6 expansion modules (though the initial ES-5 counts as 1 of the 6, if you want to use the ES-5 outputs). Each of those expansion modules uses 8-bits of data (24-bit stereo pair x 2 = 48-bits, 48-bits / 6 expanders = 8-bits per expander). For the gate output modules like the ES-5 itself, or the ESX-8GT, the 8-bits are used to sent gates / triggers at single sample precision (audio rate). For the CV output expander, the ESX-8CV, each 8-bit sample is instead time multiplexed across the 8 CV outputs, which are each 12-bits but individually run at a fraction of the audio rate of the project.

Another thing to keep in mind, at least in my experience using Ableton Live, is that while it is stated that the ES-5 uses two channels of the ES-3 / ES-8 / ES-9, if you are only using <= 3 expanders including the ES-5, you only have to sacrifice one output. This is because the first of the output pair is used to encode the first 3 expanders, and the second of the output pairs is used to encode the final 3 expanders.

But for example, if you give up two outputs on your ES-8, you could gain 8-triggers / gates at audio rate (the ES-5 outputs) as well as 40 12-bit CV outputs running at a fraction of audio rate (5 ESX-8CVs attached as expanders 2-6 to the ES-5, each with 8 CV outputs).

2 Likes
2041

Wow, thanks for the super detailed response! Os pointed to a useful thread on muffs that breaks some of those same points down.

One of the main points that I didn’t understand was the 24bit->8bit part of the process. Which makes what happens with the expands more understandable.

Does this 24/8 thing sitll hold true if you use an ES-8 in conjunction with another audio interface, and happen to be running at 16bit? Does it just 16/8 instead in that circumstance or does it bit reduce all the existing architecture?

Lastly, if I’m understanding you correctly, if I have an ES-8 + ES-5 + (1x) ESX-CV8 I would have:
7x audio rate CVs
8x audio rate gates
8x (1/8th (5.5kHz @ 44.1k)) 12-bit (4096) audio rate CVs

2042

I’m not sure I’ve got this right, but I’ve been thinking that envelope clipping is what some envelope generators (e.g., Quadigy) simulate as “punch”, and the resulting effect on volume on a module that responds to less than 10v might be why a 10v envelope generator like Maths sometimes stands out?

2020-10-12 at 10.02 AM
2020-10-12 at 10.02 AM1260×728 36.3 KB

Envelopes are of same duration and ‘exponentiality’. In addition to the extended duration at maximum extent where the clipping occurs, note the difference in volume (solid green) between the envelopes.

9 Likes
2043

That’s a good question about running at 16-bit. I personally use an ES-3 as an ADAT expander from my main interface, it sounds like you use the ES-8 in a similar way, rather than running from USB. I think the ADAT connection might always be 24-bits? I’m not sure how the DAW handles that if your main interface is at 16-bits. I also see that Expert Sleepers lists the ES-8 output as 24-bits, can you run the ES-8 at 16-bits if using as your only interface directly over USB?

On your setup, what you described is basically the case, except that the ESX-8CV sample rate is adaptive depending on how many outputs you are using. From the Expert Sleepers website:

In order to squeeze eight 12 bit channels through one 8 bit connection, the channels are time-multiplexed and so effectively run at between one-third and one-twenty fourth of the audio sample rate. For example, if the audio interface/DAW is running at 48kHz, the ESX-8CV’s outputs are running at between 16kHz and 2kHz, depending on how many channels are active.

I should also note that I use the ES-5 with Ableton while only consuming one output from the ES-3, however I’m not sure how that works on other DAWs (Ableton lets you route two tracks to the same physical output where I assume they are mixed)

1 Like
2044

For my purposes, I’d only be using two total expanders, so the 16-bit would cover it one way or another. I just wonder how that works, because even if the ADAT connection is 24-bit, presumably you still have to pump those bits out from your computer (firewire/USB).

To be fair, I only ever use it ES-8 when I’m in Max, so it could be that I’m in 24-bits anyways.

Yeah, that’s what I was trying to make sense of. So presuming I’m running all 8 outputs on the ESX-8CV, would it be 44.1k (in my case) divided by however many outputs? (so in my case, the 5.5kHz rate)

What’s a bit confusing from the description is that it says between 16kHz and 2kHz, where 16kHz is 1/3 of 48k, so not sure how that factors into things.

So I guess if I have all 8 running, it would be at 2kHz each?

edit:
And I guess because of the time-multiplexing, all 8 channels of CV from the ESX-8CV would happen across a period of (in the case of using all 8 channels at once) 4ms?

edit edit:
As confirmed by Os on mw:

Lastly, if using an ES-8 + ES-5 + ESX-8CV, as I presently understand it I would have (presuming all 8 outputs going on the 8CV):
7x audio rate CVs
8x audio rate gates
8x 2kHz 12-bit audio rate CVs (time-multiplexed/staggered)

I guess the 2kHz would be slightly slower if I’m running at 44.1k, so 1.84kHz.

And @DMR, you were correct about the 24-bit ADAT thing happening independent of the DAW/computer settings.

2046

I’ve looked at this like three times and it should have been obvious:

that’s the same way the amplitude would be impacted by compression.

2047

I don’t understand how. If you somehow applied compression to the envelope itself, I would think that would simply attenuate the envelope (scale it down) to some degree relative to its original amplitude. If you’re compressing the audio somewhere downstream, I would think that would merely attenuate the audio signal in a relative way, not reproduce the timbral- and relative amplitude effects baked in by an envelope “clipping” (the punch) at the CV input of the VCA or filter or whatever.

2049

I feel in a bit of a bind, because I’m sure you understand exactly how compression works, but I don’t know how to clarify my analogy without… explaining how compression works. I hope this doesn’t offend such a distinguished PhD in Cold Mac Studies :stuck_out_tongue: Maybe there’s something in my picturing of it that is mistaken, or that could help reveal angle… Gonna ignore attack and release (as I suspect this added layer may just obfuscate the thought).

Compression takes everything above a threshold and reduces it according to the ratio. If your threshold is 10, your ratio 2:1, and your signal peaks at 12, then your signal will now peak at 11, and all the sound below 10 will be unaffected (amplitude will increase at half its previous rate above the 10 threshold, and so will look “less steep”). An infinite ratio (limiter) will flatten your shit above the threshold – anything over 10 just stays at 10. The graph you posted looks, to me, identical to graphs I saw when first learning this stuff that would be a side-by-side comparison of a compressed and uncompressed signal (but at the envelope scale, rather than at the waveform scale, where compression is often working its magic – maybe this parenthetical contrast of scale is what makes them feel not-at-all similar).

I will say that this understanding is heavily colored by over-use of pretty extreme settings on compressors, and that I am far from an authority in this area. Also I should have said limiter with make-up gain in my initial post, probably.

3 Likes
2050

Turns out I only understand compression and limiting at an elementary level, and have very limited experience working with either. I can see if you had a DC-compatible hard limiter (does such a thing exist?) and use it to hard-clip an envelope, you could thusly flatten the peak of the envelope it in a manner like on the graph. But that seems like an improbable scenario.

The way I understand it, if CV is “turning a knob” for you, the 10v envelope—when sent to some modules—is trying to turn the “knob” further than it can actually go (depends on the module), so for the duration the envelope remains high enough, the “knob” is pinned at its maximum value.

1 Like
2051

Yeah I don’t think a DC-compatible hardware limiter is a thing. I was thinking about envelope following the compressed sound – its “amplitude silhouette” looks like what you showed in the above graph.

Understanding why this CV phenomenon manifests as “punch,” for me, was what made compression a useful analogy (since “flattening” your signal via too much “punch” – too high a ratio, too low a threshold – is the standard amateur mistake, and one I’ve made a million times). Because, initially, I was thinking pinning the knob just gives you an ASR/D envelope, and it wasn’t clear why that is “punchier,” but relating it to compression is like “yeah… ‘louder for longer and articulated in less detail’ is precisely what people mean by ‘punch’ for a compressor.” (this picture is, again, a bit complicated by attack and release settings – slow attack lets more unreduced signal through and that momentary ‘click’, analogous to the noise impulse on a kick drum, is also a feature of real ‘punch’).

cool experience to share a new perspective with someone who has taught me so much <3

ps: the parenthetical thought at the end of the last paragraph got me thinking about how to accomplish that more complex shape via CV without a compressor. And, mapping it in paint to show off my grand discovery, I found it’s LITERALLY AN ADSR. Time is a flat circle.

ADSR

EDIT II: If these CV concepts were initially explained via analogy to gear one uses when sampling (like a compressor), I think it would be clearer to more people moving from sample-based music to synthesis (as I have in the past few years) that many parts of sample-processing / mixing can actually be accomplished directly via things like more nuanced envelopes – I now feel the urge to replace every envelope with a four-stage envelope, and to consider the kind of compression I would apply to a sound and just… do it via CV, which then makes the pseudo-compression portion of the sound-design itself performable / CV-able without the need for an external hardware/software compressor. This kind of faith in synthesis not as another instrument to process (like the switch from guitar to trumpet), but as a unique discipline in and of itself with different tools available to accomplish similar ends, is precisely what’s missing in the mindset of many aspiring sound-wizards (myself included). Here we see just another manifestation of the same fundamental problem that exists in discussing generative sequencing techniques vs “writing” music (I’ll stop here, as hard as it is…).

4 Likes
2052

This would be an interesting experiment to try. I’m trying to think of a way to combine an envelope and a constant DC voltage using logic to make a limiter but the ol’ gray matter is coming up short.

In my head what @mdoudoroff is saying about the envelope being clipped producing the punch makes sense and I’d be curious if there would be a noticeable difference in something like running an envelope to a vca and a “clipped” version of that envelope to filter cutoff at the same time, for instance. This might just be an exercise in tedium, though :wink:

There is the DPW limiter which cuts at 5v (which may not be enough for maximum punch in some cases? I guess if it’s the shape and not strictly the amplitude then this should be ok)

Also, of course, looking at the disting manual… in mono compressor mode you can set the ratio to infinity and the threshold level with another parameter which should do the trick I believe.

2 Likes