Created: 2020-08-20T00:24:18.336940
This card pertains to a resource available on the internet.
Hey. I built the nogasm and it works.
You mentioned using an electromyography for detection instead of a buttplug. I was wondering how well this worked.
The graphs I got with the EMG shield looked very similar to the ones obtained with the pressure sensor.
Did u ever explore the possible idea of using recurrent neural networks for orgasm detection.
The other idea Ive been thinking of is instead of just measuring the strength of any one clench like the nogasm does, recording something as a “clench” if a certain threshold is reached. Then the length and intensity of the clench would be determined and it would be stored. Then maybe the number of recent clenches and their intensity could be used for an accurate gauge of arousal level
I decided to go the classic PID-controller route. It regulates the speed of a vibrator and/or the intensity of e-stim to keep the victim on the absolute edge without (so far) ever going over it unintentionally.
I don't keep an average pressure and check for crossings through it; instead, I track peaks and troughs of the signal. Swings from trough to peak exceeding a certain minimum value (for noise reduction) are considered a 'clench', and the magnitude of the upswing is simply added to the running arousal value. This way > I don't have to actually calculate or estimate the frequency; the faster the clenches occur and the stronger they are, the faster the arousal value rises - effectively proportional with the product of frequency and amplitude, as you suspected. Simultaneously, the arousal value is decreased with a negative-exponential function of time (like the voltage on a capacitor discharging through a resistor, if you like electronic circuit analogies), dropping the value again when the muscle contractions subside.
Once I had it working well, I stopped improving it and went on with half a dozen other projects. I don't use the length of the clenches, but it could very well be interesting. Indeed, being able to distinguish between voluntary and involuntary clenches would be awesome.