Anyone know of a way to set the amount of memory to a temposynced amount? Either pre-determined or, say, decided by duration of recording would work.
1 Like
I don’t know, but +1 for this! I like the memory system and this is a feature I would really like to be implemented. It will be very much more versatile as a collection.
1 Like
There is now the Fixation module. One of the STYLE’s allows you to send a CLOCK signal, with position set with the appropriate knobs+attn+input. Is that what you mean?
Here are a few videos I made while I was working on it, demonstrating a few techniques:
And glad you like them!
If I’ve misunderstood the question, do please rephrase what you are trying to make happen.
Thanks for the reply and the excellent modules! I’ve got some work coming up in the next months and have been looking for a new approach to live-looping setups, and this is the best lead so far. Exciting!
I know about the fixation module. And it’s super useful indeed. The feature I would love to temposync is the length of memory recorded (or loaded) into the Memory module itself though. As far as I can tell, the only way to set that length is with the dial on the memory module, and it snaps to integer number of seconds.
If it could be set to tempo-synced lengths, that’d be cool. If on loading a wav file, it would auto-adjust to exactly fit that file in memory, that’d also be nice. The latter would maybe be even better than the former, come to think of it.
The main reason I’d love this is for rhythmic workflows such as breakbeat chops. Fixation could really excel at that. Maybe it does already and I’m just missing something. But it seems to me like it would be much easier to set up if the Memory were exactly the length of the break. See what I mean?
1 Like
Perhaps not the primary intended use case of these modules. That’s ok if so, there are other options I think. And these are lovely for the freeform (and in-time manipulation of the freeform loops, which I love). But perhaps they can be great for both!
When you load a file, it fits the Memory to exactly that length; no padding out to an integer number of seconds or anything. So you could load pre-recorded .wavs at the tempo you want and it should work.
NOTE: I recently found a bug in Memory, where it trims off the end when the sample rate of VCV and the loaded .wav file don’t match! Fix in the next release (2.0.17).
A couple ideas:
- I’ve actually considered some sort of “trim” gestures that would let you set points at the beginning and end of a Memory to ignore. I haven’t thought too hard as to how a user would want to do that, though.
- But also, for live usage, some gesture (or alternate recording module) that says, hey, I’m going to record something, and when I stop, that should be the whole of Memory. I’ve sort of hacked things up to do that with BASICally at least once, but I wouldn’t call it rock solid.
I haven’t been working on Memory lately, as I’ve been distracted by this upcoming module. But I’m close to a first version of that being ready (working on more videos for it today), which will free up brain time for Memory.
Thoughts?
2 Likes
You can use the buffer_size parameter in Stochastic Telegraph to set the memory to a temposynced amount. Calculate the buffer size in samples based on the recording duration and sample rate. For example, for a 10-second buffer at 44.1 kHz, set buffer_size to 44100 * 10 = 441000 samples.
2 Likes
Oh that’s good news! I’m running VCV at 96k regularly so that bug must be what I observed when I loaded my breaks. 1.2.17 should suit the breakbeat-cutting case pretty much perfectly with that fix in place.
Both of those ideas also sound good, the second one especially. With automated triggers to start/stop recording, you could then sync recordings to an existing tempo. Or do the typical thing where the first loop dictates the tempo for the rest of a piece.
Thinking out loud, picture a module with stereo ins and outs which records to memory, and once recording stops, sets the memory length, immediately commences playback from time = 0, and won’t record again (until the memory is cleared). Imagine then it had something like a 0v-10v phasor* output showing the playback position. Maybe an EOC trig output, etc etc. You get it, that would let us start a loop and then sync sequences and such from it, and also manipulate loop playback with fixation/ruminate. Could be fun!
And yeah, I bet such things are already possible with Tipsy etc, thx art.sol for the tip. That’s also super cool, and I may mess around with that as well. But first I’ll look forward to the 2.0.17 release (is it in the queue?). Thanks again!
PS: Venn seems fun! I bet it’ll make a great companion for Bog Walk. 
*I’d love that since I use the Hetrick phasor sequencers all the time, but of course some would prefer a clock output with selectable divisions.
I welcome the ideas!
I am figuring out how to use the Memory Ensemble as the “tape-part” in a more complex multi delay setup.
Like it is used in the Strymon Magneto module for example: Here there is a tape-speed section, where I could use the ME, but I need to change the buffer length in sync with a clock in order for this to work…
A “trim” option (that can be adjusted on the fly) would be great. A sync option to a clock could work as well I believe.