#203

Since I had to use Miasma as my min/max, I was sending the output of each T&H into it and applying the slew there, before the signals get compared for the min/max output. Using my Noise Tools and Disting for the T&H’s also meant I didn’t have any leftover slew limiters to run the final CV mix through.

I will say, I tried the Miasma as a min/max with the D0 based patch as well and so far that was my favorite. Being able to control the upward and downward slew of the dry and delayed signals, before they got compared for the Min/Max outputs, allowed me to get some interesting character from the “bowing” and created differences in upstrokes vs downstrokes. I think when I put this all together in a single system with the ribbon controllers it’ll probably be based around the D0 and Miasma working together, especially since then I also get all the outputs from the individual dry and delayed signals, and a ring modulated output as well. Lots of options for changing the character of the sound based on how you’re playing!

1 Like
#204

Oh man, do either of you have a Cold Mac? That could be really interesting to play with in this patch!

I’m thinking you could send your dry control voltage to the Survey input, and then take the Location out into the OR/AND input. That should give you the min/max based on your original CV and the slewed signal, and all the other outputs will give you weird stuff related to the input with the option to modulate them with other signals. I’m gonna ask folks in the Cold Mac thread if they can try this out.

2 Likes
#205

I’ve tried the ADDAC306 (which some people compare to the Cold Mac for some reason) as my bow slider and it’s killer as you not only move the “bow” up/down but also control other parameters up/down, but with different ranges.

I decided to add an Elements module in the system, and try to stack all the possible bow modeling options on top of each other in a single system. To be honest I want mostly want it for the bowing sound. Let’s see where this path goes …

2 Likes
#206

i just tried the patch with a c0 and it works pretty well too. since i don’t have a min/max module i am also interested in the answers from the cold mac people, that would be ideal! the addac306 looks also extremely handy for this patch!
what modules are good and compact for slope detection btw?

#207

I don’t have a comprehensive list but off the top of my head there’s the Ladik Derivator (4hp) that gives gate out for moving/static/rising/falling CV’s. And the soon-to-come Joranalogue Contour 1 (8hp) is a slew limiter that will have gate outputs for rise and fall. Mystic Circuits ANA looks pretty interesting for this stuff. It’s 6hp and has two attenuverted inputs that then give you sum outs, min/max, a 4qx output, difference, track and hold, and a weird comparator-based output.

It’s a bit bigger, but the Befaco Rampage fits right in with this. That’s what the Miasma module I keep mentioning is based on and it’s been adding a lot of interesting details to this patch.

1 Like
#208

thanks, didn’t know the rampage had these logic functions, really cool!

#209

Yeah it’s great. And if you can DIY or find a builder, the Sin Phi Miasma is the same thing but it has attenuverters for the Rise and Fall CV inputs and an extra 4 quadrant multiplier output. It’s 2hp larger than the Rampage but those extra features are super handy.

#210

Tried a version of the patch this morning only using only Miasma and an inverting mixer and got pretty good results.

Multed my control voltage to both slew inputs.
Channel A set to low, almost no slew and Channel B set to a slew of roughly 2 seconds.
Max out to mixer, Min out to mixer and inverted
I then multed the mix output to open the low pass gate and negatively effect the rise and fall times of the Channel B slew. So the higher the final output, the less slew that channel is applying to the incoming control. This helped to get the patch’s momentum going, but kept it from having too long of a release once movement stopped.

It’s still not as responsive as D0, but much better than when I tried the S&H and T&H methods. You could get a lot out of just this, or a Rampage, and a mixer.

#211

So you’re not sampling the signals at all, you’re just doing the subtraction using slewed signals, while changing the slew rate of one? Am I getting this right?

#212

I’m reading up on Elements. It’s been years since I beta tested for it; seems so long ago actually. Anyway, reading up on the back-end differences between that and Rings brought to my attention two posts from Emilie (https://www.muffwiggler.com/forum/viewtopic.php?t=167755) which I think are relevant to this thread and so I’m going to repost them here.

A significant difference between Rings and Elements is that Rings runs at 48Khz, while Elements runs at 32kHz. This fact might explain why some people find Elements more “woody” or “natural” than Rings, because it removes some of the highs - with Elements you can’t shoot yourself in the foot and get harsh/glassy tones when you heavily modulate the module or feed it garbage, because there’s not even the possibility of the hardware generating highs above 16kHz. I kind of like the idea of gear doing “physical” things (physical modelling, reverb…) running at 32kHz. But it takes guts to proudly display that on a spec sheet in 2017…

In the context of a module simulating a natural, physical process, it’s much more likely that highs above 16kHz would be byproducts of artefacts/approximations/quantization than something naturally occurring. To put things in perspective, some early, well respected reverbs also have low sample rates (20kHz for 224, 34kHz for 224XL).

6 Likes
#213

Yes, though one signal is passing through with almost no slew. So it acts pretty much like the shift register or delay patch, where the second signal is lagging behind the first a bit and producing that gap for the min/max outputs.

#214

so i’m again reading about the marimbaphone and its overtones, i really love this instrument and since i first discovered pinging bandpass filters i tried, more or less successfully, to “build” a marimba sound. there are very interesting information on the net about tuning a marimbaphone and on this site (http://www.lafavre.us/tuning-marimba.htm) there is a useful description about the first 12 modes (partial, overtone) of a specific marimbaphone.
i don’t have 12 bandpass filters but 8 so i tried to tune pinged bandpass filters which is quite difficult since the pitch isn’t that stable, it rises at first transient impact and then falls down a bit. anyway, with a tuner it’s at least possible to get close to the desired frequency. these are (for the C2 bar):

  • mode 1 - 65.70 Hz (First Transverse Mode)
  • mode 2 - 102.2 Hz (First Torsional Mode)
  • mode 3 - 262.8 Hz (Second Transverse Mode)
  • mode 4 - 595 Hz (Second Torsional Mode)
  • mode 5 - 662.2 Hz (Third Transverse Mode)
  • mode 6 - 786 Hz (First Lateral Mode)
  • mode 7 - 1203 Hz (Third Torsional Mode)
  • mode 8 - 1287 Hz (Fourth Transverse Mode)

the second mode apparently isn’t that important, it isn’t excited to any great degree which indeed makes quite a sound difference if one turns this partial down.

here’s the result (first the monophonic single bar and then pitchshifted and delayed copies with the help of a h9)

what’s clear is that the relative volume of the partials makes a huge sound difference and i think i had too much higher partials, it sounds a bit plastic and stiff (at the beginning of the soundfile i attenuated the first 4 partials to show the balance between the two 4bandpass groups (four lower, four higher = 2 cwejman modules)

a trigger is also not the right mallet sound for this, it needs to be something softer but then i fear that the higher modes won’t get excited at all, so this is pretty difficult to find a balance. also the lower the note the more ringing out, which is i think similar to an actual instrument but it’s too extreme with these bandpass filters. i need to find a way to modulate the bandwidth of the filters in a convenient way.

7 Likes
#215

Where do wavefolders come into your physical modeling patches, if at all? I haven’t found much yet but it seems like a good idea to have one around.

One example I found was “Use a wavefolder on the envelope controlling the vca controlling the excitation noise to create a more complex attack shape”.

Nothing to report yet in terms of examples, as I’m on the road for the past days. I really want to get home and work on string sounds.

#216

I’ve been working on refining the RoC patch using just my function generator and I’ve gotten it responding pretty much as well as the D0 on its own. Sending an attenuated copy of the final RoC voltage to the more slewed channel’s exponential CV input seems to be the trick. So the Rampage or Miasma are the best contenders for handling this function in a compact way, IMO, given that all you need apart from that is a voltage source and an inverting mixer.

Made a little patch diagram of what I think are the least possible patch points needed to get this working in a satisfying way. Hopefully it’s easy enough to read. Might be unclear but Rampage’s A output is patched to the B input. This way you can have minimal slew settings on A and pass the essentially unchanged input voltage onto B where you apply medium slew.

Not sure if I’ll put together a dedicated 44hp controller like this, or try doing something larger and having the sound making modules included, but this would be a pretty great little string-instrument control interface to plop down in front of any system.

2 Likes
#217

So you’re continuing down the Unslewed-Slewed signal path, instead of using Max-Min, am I correct? I don’t have the Rampage so I’m not fully aware of what the above patch entails, but thanks none the less, I’ll try to understand and try it out with my simple ADDAC slew.

According to the above you need two slew limiters, no? One for the subtraction and one to slew the RoC response.

#218

As fascinated as I am by the ongoing RoC patching challenge, wouldn’t a digital implementation be much easier and cheaper? I wonder if a Hemispheres app could be designed to this purpose. The math isn’t too hard, but certain types of math get very expensive in the analog realm… But perhaps this would take some of the joy out of it :man_shrugging:

#219

It’s hard to make out because the patch sketcher is so cramped, but I’m still using the min/max out of Rampage. So voltage > A, A out > B, Min/Max > inverting mixer. It’s a pretty handy one-stop-shop for the patch, especially since you have all the other outputs to send to different things.

@desolationjones I’ve no doubt a digital solution would be much easier, and probably better given that a program could be fine tuned to respond more precisely to how you’re moving, but I don’t know of any module that has this function built in. I’m not familiar with Hemisphere–looks like it’s a sort of virtual modular within O_c? I imagine that could work if it has the right sub-modules, but I’d be a lot more interested in software written specifically for the purpose. If @ParanormalPatroler and I can get SDS Digital interested then I bet she would be able to cook up something proper for the task.

2 Likes
#221

I got an Elements yesterday and tried a combination of Braids (in BOW sound), Elements, and Serge Resonant EQ. I did not implement the RoC patch at all and I have to say, soundwise, I’ve gotten the best results so far. I’m fairly certain that using the RoC on various parameters will enhance the sound.

Unfortunately it seems that Elements is not perfect for this type of sounds, but it brings you quite close. Some elements (pun intended) that I’m missing is 1) CV control over Bow which according to the manual is what increasing the pressure of the Bow on the string would sound like, and 2) CV control over the AR shape which makes a huge difference in how the attack of the bow on the string is perceived.

I’m also unsure whether the resonator is best utilized as a body or as a string in this patch, since Braids already provides the string. In the next days I will try both “options” and I’ll also experiment mixing the various sounds externally, in contrast to passing Braids through Element’s exciter section.

All in all, there’s a lot of ground to experiment with and I have to admit this duo cuts a lot of corners soundwise. That being said, I’m keeping the CG Delay in there, as well as a Comb filter and a LPF, and I’ll try to be as crazy as possible with the patching. I even tried having the Braids in PLUCK sound which was also interesting enough for bowed modeling.

Some aspects worth asking:

  • I’m still curious where Wavefolders could be utilized in this procedure, especially since I’m using very complex initial sound sources. I’m inclined to try utlizing the WMD Geiger Counter in there somewhere, but I don’t know why my instinct says it will be useful. :thinking:

  • As mentioned before I’m using a slider module to replicate the movement of the bow. Obviously for the RoC patch and via the Ladik Derivator. Any ideas on how else one could utilize the actual CV output of the slider?

#222

I’ve been using the direct output from my controller to effect my body filter, slightly tightening the delay time as it increases, which helps create the effect of the sound changing depending on where along the bow’s length length it’s making contact with a string. I wonder what you could get by sending it out to Elements’ Position or Brightness inputs.

#223

Very little information available, but it’s apparently an analog physical modeling synth built beneath CHAIR’s Haptic controller. Mods, feel free to move this into the Physical Modeling thread. This sound demo has me going :drooling_face:

5 Likes