posted 27 August 2001 17:31            

---

How could I use the amount of activity of a changing value (ex: how often !x is tweaked A amount in T time from continuous samples of controller value).

So, I want to be able to use how frequently something is being modified A to create a changing value which expreses this activity.

Here is what I mean (in terms of splintery, phosphorescant slugs, a bionic nose, and a microbial photocopier).
A bionic nose is constructed syntheticly. It looks very realistic , to an untrained nasal technician, that is, because the small external nose hairs are placed all in a line exactly one mm. apart, rather than scatered acording to a biological criteria. So, a small slug, trying to get to the experimental elbow made out of snack food placed at the opposite end of the nose will travel strait down the row of hairs. A photo mechanism has been set up to record the distance traveled (in either direction). It takes a snapshot every 9 sec. If in one shot the slug has traveled 1 mm. in 9 sec. the value is 1, if in the next shot, it has travelled 3 more mm. (either forward or back) the vale will be three.

IP: Logged

posted 27 August 2001 17:49            

---

You already have a value that represents the amount of change - your distance? Another way would be to send triggers from the hairs. This would give you inter-onset (or delta) time. That information would provide the speed of movement of the slug, and could provide acceleration and deceleration information. Getting triggers from each haair on the nose would also give you positional information, rather than your current generic distance measure.

You must have a way of turning your incoming readings into MIDI controllers - ie. MAX, and then send them to the Capy. Just assign your distance number to a MIDI controller and off you go.

IP: Logged

posted 27 August 2001 17:52            

---

You could place the control signal into a Constant Sound, and then use the Constant as the Input to a SampleAndHold (with a HoldTime of 9 seconds). That SampleAndHold is used as an Input to a second SampleAndHold (with the same HoldTime).

Next, get a Difference Sound from the Prototypes, and Ctrl+drag the first SampleAndHold into the Plus input, and Ctrl+drag the second SampleAndHold into the Minus input.

Then the Difference has the signal you desire.

IP: Logged

posted 28 August 2001 09:44            

---

Hmmm, interesting problem. I'm not sure that your example, as stated, is exactly the same problem as in the first sentance.

Say that the slug is not fully enlightened; perhaps he's had a hard day or he's having some worries about that new project. In any event, he does not proceed monotonicaly toward the elbow/snack food. Instead, he shuffles back and forth somewhat (maybe it's a new slug dance from STV). So in the first nine seconds his accumulated travel distance is much greater than 1 mm though his endpoints are 1 mm apart.

Maybe you want to measure all the control changes that accumulate during time T, regardless of the beginning and ending values of the control?

IP: Logged

posted 29 August 2001 09:24            

---

slug.kym

Hey YerbaAdam, try out this example attachment. It seems to work but the slug is uncalibrated (i.e., I don't know it's units). It accumulates control changes over a 9 second interval rather than looking at the endpoint values.

(This is my first upload so I hope I did it right!)

IP: Logged

posted 29 August 2001 10:37            

---

Dennis-
You are right. After logging off I had realized how carried away I got with my slug story. It's problem deviates from what I am looking for. Which, as you correctly express, is a value assigned to the amount of controll CHANGES in T, not the amount of controll. Sorry for throwing you guys off. In the future I will refrain from overly dramatic complicated analagies.

IP: Logged

All times are CT (US)