ADDAC506 Stochastic Function Generator

What are people thinking about the ADDAC506 at this point? I’m interested in getting one instead of Maths. To me it just seems like there’s more potential for interesting moments with the EOR EOF and Sum / Averages.

Anybody replace Maths with the 506 and have any regrets? I guess at this point i have a fair ammount of attenuators so i think it would make sense… On the fence - sell me either way?

1 Like

i very nearly pulled the trigger on one but i can get a similar effect with an envelope, S&H and a slew. it’d save a bit of patching but i can’t justify that expense right now!

I know Keith Fullerton Whitman loves his.

1 Like

I like it well enough, and I’ve only just begun to employ it.

1 Like

I’ve noticed this one of several posts that proposes replacing Maths with the 506 and as an owner of the 506 since pre-order (and an owner of Maths) that I can deliberately say, it is not. I don’t think it’s an apt replacement for quad envelopes either, as the 506 brings pretty unique functionality with its variable duration (and response curve via the expander).

If you are seeking a Maths replacement, I would highly recommend the Frap Tools Falistri. It features an equally confusing aesthetic (upon first glance, it’s actually quite intuitive, like Maths), and several options for polarization and attenuation.

The 506 does feature both AVG and SUM functions in the vein of the Maths mixer, but lacks the 2 channels of input attenuation and polarization that Maths offers. Falistri does offer a 4 quadrant multiplier which matches some of Maths mixer functionality (there is some convenient normalization occurring within Falistri as well, but I don’t want to make this Falistri focused).

If your primary use for Maths is envelopes, then the 506 can in fact replace it, and you’ll get twice as many envelopes. I feel that envelopes are only half the function of Maths and to achieve the attenuation and polarization functions with channels of the 506 using its slew functionality is possible, the density of the controls within the confines of locking and unlocking channels makes it fairly difficult.

Maths isn’t necessarily a functionally dense module in comparison to the 506, not by 2020 standards anyhow, but there’s a reason it’s the most popular eurorack module. A lot of that has to do with its UI. It’s spacious and fairly straightforward in use. 506 is straightforward, like most ADDAC designs, but it’s not particularly comfortable. I also have sausage fingers.


Looking just at functionality, is the main difference just that it doesn’t have 2 dedicated attenuverter channels? While it might not be as clean as maths, is there anything you couldn’t achieve on the 506 if utilizing attenuverters in a separate module? Looks like 506 doesn’t have inverse output.

Also how do you personally use the 506 differently than you would use maths?

I use the 506 for variable shape and duration envelopes, usually with more CV directed at fall time and using the main length toggle to define the individual envelopes approximate duration.

I tend to use the Maths envelopes for what I feel they excel at: snappy exponential attack/logarithmic decay envelopes.

The lack of inverse output on the 506 has never been an issue for me, nor would I imagine it be for most…and for dedicated signal mixing I prefer Blinds. The sum and average outputs of the 506 are rarely used for me, unless I just need a source to add some movement to waveshaping or PWM. The fact that each off the 506’s envelopes has dedicated offset and attenuation makes it functionally superior to most other EG’s in my rack.

I, perhaps as a creature of habit who needs to broaden his horizons, use the 506 as a source of semi-random modulation instead of feeding random CV into Maths rise and fall controls for envelope variation. Now Maths serves me LPG style functions while 506 is a mechanism for evolving soundscape.


Nice, great breakdown.

Can the 2 middle EG’s of the 506 be used as Envelope Followers when audio is sent to the inputs? I really miss my ADDAC envelope follower and regret selling it. Haven’t been able to find an EF with as accurate of a response as the ADDAC401.

1 Like

UGH why did I sell my Maths (well rent blah) :weary: Glad I still have a Function but Maths was the first Eurorack thing I ever had, and is the only module I can see my self re-purchasing at some point. It’s a bit cliche but yeah I loved it as a snappy envelope generator (serge patch programmers come at me!) Attenuverters were a nice bonus. Hahhh ADDAC506 looks lovely though too.


Regarding audio: I am not sure but will try it tomorrow and report back. I’m curious and wish I had the energy to do it now…and I think that even some of the functionally simple ADDAC modules are the best evidence of their extremely well thought out designs, especially for utilities.

I’m never thrilled about selling one, but it happens and I’m always re-impressed when I pack one up. They’re built so solidly with such high quality components. I sold my 601 filterbank recently and for a 20hp module it weighs like 15 ounces.

1 Like

edit: sorry @dianus, meant to reply to @sisyphus69

So, the 506 as an Envelope Follower:

It can be done, and it does it well under some circumstances, but I don’t think it wants to do it.

I patched Pressure Points common gate out to 506 ch4 (trigger mode) signal in and 506 ch4 signal out to Tangle Quartet ch1 cv in. I multed TQ ch1 out with one outputting to my mixer and another to 506 ch3 (slew mode) signal in and 506 ch3 signal out to TQ ch2 cv in. I had a triangle wave from one Dixie II patched into TQ ch1 and the inverting saw wave patched into TQ ch2 set several semitones apart (also it was easier for me to hear exactly how the inverting saw wave was responding to signal output from the 506).
With each channel of the 506 locked at slower rise and fall rates, I thought it was working flawlessly. the 506 was accepting the audio in on ch3 and outputting a corresponding positive cv with adjustable slew rates. Seemed stellar. When I backed down the rise and fall rates for faster amplitude envelopes and slew rates, ch3 started to have trouble reading the incoming audio and would produce envelope following voltage, but sometimes it would be negative and sometimes it would fluctuate quickly (not quite audio rate) between positive and negative voltages (which made me wonder if it was tracking the audio signal “too closely”).
After the simple oscillator test, I took a pretty simple sequence (but with a range of amplitudes) from the Digitone and ran it into a stereo channel and patched the mono-out to the 506 ch3 signal in and ch3 signal out into TQ ch1 cv-in with Dixie II as the input signal and TQ ch1 outputting directly to the mixer. In this case, even at fast slew rates, the 506 tracked really well and I was still able to tweak the offset and amplitude controls to adjust the “envelope follower” signal.

So, it does work, with a few caveats… but having an optional 4 channels of envelope following might tempt me to get another 506 or Fallistri :thinking:
Thank you @sisyphus69 for raising the question, I had a wonderful time investigating the possibilities!

1 Like

I can chime in if none of the current owners have yet - I’ve been struggling with finding a ‘usable’ random source, and I actually came to the 506 that way. I realised that if I had a buchla-style random, I would use it to cv things like envelope lengths, mod speed, etc. For me, the 506 should do most of what I wanted in that vein.

I’ll get the expander as well, because having a random stepped CV that’s somehow related to the envelopes (which in turn are limited by max/min) seems incredibly useful. Basically, a different window into the same random modulation - which can be used on it’s own, slewed, quantised, etc. Also, being able to trigger envelopes adds another temporal aspect to the randomness - and it can be clocked or equally random.

To me, having those three seems like a really open and controllable random playground. Haven’t even started thinking about cross-patching, but I’m sure there’s something in that too.

1 Like

I’m starting to consider getting one of these, but I have a question that I haven’t been able to find an answer for so far: when you activate the LOCK switch on a channel, does the module lock the current rise/fall time “post-randomization”, or does it save the current min/max rise/fall settings and continue to randomize the times within those ranges while LOCK is engaged?

A locked enveloppe isn’t affected anymore by the randomization process. You can have one or two enveloppes / LFO pulsing at fixed length (sometime i use one as a master clock), and the other having different length at each cycle.


That was fast! I suppose that makes it a little less flexible (can’t have four envelopes with different degrees of randomization) but also less potentially confusing (less risk that you’ll lock an envelope and then forget what its settings were). Thanks.

True. The lock switch freezes the enveloppe one and for all. You get some happy accidents though when you unlock it after having mess with Min/Max…

1 Like

Holy mother of Krell, this thing is amazing. Took delivery this afternoon and have spent the last few hours having this thing (basically by ltself) driving my entire 12u Shared System. Such a well-designed and playable module + expander. Really happy I pulled the trigger on it.


The confusion is mind-opening feature, apparently.


Hey guys I’m super interested in this but was a little worried about buggy behaviour that has been reported in some places. Any idea if it got sorted out or if maybe it was just an initial batch that had some problems?

What sort of bugs are you thinking of? Ive got one, and i haven’t noticed anything, but that could have been me not noticing.

I’d be willing to test mine to see if any of the things you’re worried about is happening with my module if you let me know any specific issues you’re thinking of.