I got the idea that a long stereo estim could partly be done by subtraction: lay down some signals and then do boosts or cuts according to a list of times. So, I have both the funscript JSON and this list of sections to work with. It might be a useful reference if anyone wants to skip to their favourite bits.
This is one of those cases where I wonder, "It's 3 hours... what was I thinking??" But, that being said, I've got stereo estim done from the beginning to the end of "Chocolate". A bit over 1 hour. Once I get to 1h 30, I'll post the draft file of the work so far.
My method may be be helpful to anyone wanting to do stereo audio for a PMV:
0. Obviously, phase shifts don't work for stereo. So there isn't a great way to use the funscript. In theory, you could turn the funscript into volume pulses but in the files I looked at, the times are way too close together. (millseconds not seconds).
1. On a spreadsheet, record the times of any audio transitions. Right now, I'm at the 1h 04m mark, and I have 144 different estim signals. So, usually I'm changing the estim signal about every 8 to 16 measures of the music... or whatever seems right. The spreadsheet also converts between hours/minutes/seconds and seconds... you need to do a lot of this!
2. As I make the spreadsheet, watching the video, I make note of what sort of signal would seem right (smooth, percussive, tease, etc).
3. I cut/paste all the times and comments to a script file (in emacs, my favourite editor). I'm making the commands in Easy2 audio generator but use whatever tool works for you.
4. I try to use a similar technique for segments in each chapter. The techniques are (so far):
a) Custom signal generation for each segment in some style. This makes the best signals.
b) "Wallpaper" a bunch of sections with one smooth signal, than use amplitude mods to make changes (or not, to keep it smooth) in each section. This saves a bit of time and usually eliminates transitions. If you were working in Audacity, this method would work well.
c) Load up the funscript and transform those times into a bunch of audio commands. I have a script for this... but in actual practice, this doesn't work so well and comes out overly percussive.
5. Next, load the audio up in Audacity and look for mistakes...
a) missing signals
b) quiet signals
c) harsh transitions, especially from silence
d) segments that don't "look" like what you thought you are creating.
When I use Easy2, these errors are usually when I've forgotten the "vol" command or missed a "fadein". And that list of signal segments in the spreadsheet is essential for locating the mistakes.
6. Go back to #1 and repeat for the next chapter.
Hey Fashion Techniques... Pink Endurance.... who is this cutie?
Fashion Technique wrote: ↑Thu Oct 20, 2022 9:36 pm
Some girls have their own rounds, here are their insta`s Don`t open before you play, it may brake your 'discoverer pleasure'
Thank you! My IG feed is quite pornographic. I wonder if my friends/children have noticed.
I'm almost at the 2 hour mark completed for stereo signals. It definitely needs corrections for signal leveling as some of the transitions are too harsh, but otherwise I don't spot many obvious bugs. Some of the signals turned out great - If anyone can make it much past the one hour mark, they are superhuman.
Thank you guys for doing this electro stuff (funscript, estim, etc.) for my video
I don`t have any toys, but maybe one day i will get them and play with your files
It will be amazing experience!
Can you give me "beginners guide", what device is better to buy? Thx
Fashion Technique wrote: ↑Wed Jan 24, 2024 12:53 am
Thank you guys for doing this electro stuff (funscript, estim, etc.) for my video
I don`t have any toys, but maybe one day i will get them and play with your files
It will be amazing experience!
Can you give me "beginners guide", what device is better to buy? Thx
Can you get a few oddball components? The oddball components are 2 specialized transformers, a few odd sizes/power transistors, some wire connectors, standoffs, and audio connections. A few switches and/or potentiometers? A amp board off Aliexpress... oh yeah, and a reasonable sized case to put it in. Prices and suppliers vary. It's usually about $120-$150 CAD per build for me. Much cheaper elsewhere.
The biggest challenge is the last question and I found getting all the parts was difficult (much easier in the US and Western Europe, I think). I blew a few months researching a few of the parts. Anyway, you can build your own Stereostim with a couple weeks of work. I've built 5 now and got better at it on each iteration. Join SocialStim or one of the other forums and you'll get lots of help. It's a pretty friendly community.
The reward is the best sex you can have by yourself.
The Easy2 commands are in there too. There's a lot of experimentation in the 250ish audio signals.
And a Torrent inside there that includes the 6GB version of the video.
Intended electrode setup:
L - head/shaft
R - base/balls or anal plug.
There is some intended variation in volume but should not be anything that forces you to jump too badly. It isn't a linear volume rise: there are some relaxed times. Start and finish wherever you feel like. Mess with your volume levels as you like... or strap down for the whole ride and martyr yourself to the electron gods.
OMG, now I just have to find a little time to revisit Intermarium !
Thank you Andrus for this huge e-stim file (in Audacity it seems really interesting ).
And of course a big thank you again, Fashion Technique, for this masterpiece
Andrus wrote: ↑Sat Feb 10, 2024 2:06 am
Thank you to Fashion Technique for creating this delightful video journey. Here's a stereo estim to go with the video:
The Easy2 commands are in there too. There's a lot of experimentation in the 250ish audio signals.
And a Torrent inside there that includes the 6GB version of the video.
Intended electrode setup:
L - head/shaft
R - base/balls or anal plug.
There is some intended variation in volume but should not be anything that forces you to jump too badly. It isn't a linear volume rise: there are some relaxed times. Start and finish wherever you feel like. Mess with your volume levels as you like... or strap down for the whole ride and martyr yourself to the electron gods.
I wanted to run this straight through without stopping, but within minutes I started needing to write down time sections to go back and extend some of these signals to 15min or more for my own use.
I posted this and hadn't checked back since because I've been busy coding.
@throwawayacct - Let me know what bits you liked.
If you like, it's usually pretty easy to cut-paste the script and generate it for you to whatever length. Most of this was experimentation and some of it was imitation. It's hard to know what will work.
I almost have include files working in the script and subroutines are next. The goal is to be able to assemble a estim file out of well-known segments from a script library.
Andrus wrote: ↑Mon Feb 19, 2024 10:03 pm
I posted this and hadn't checked back since because I've been busy coding.
@throwawayacct - Let me know what bits you liked.
If you like, it's usually pretty easy to cut-paste the script and generate it for you to whatever length. Most of this was experimentation and some of it was imitation. It's hard to know what will work.
I made a list of time-stamps and I have like 15 or so segments I did exactly that clip/paste/trim/effects -> repeat function with Audacity. I have a dozen more, but life calls and its not like the track is going anywhere.
When I saw the name "Easy2" I figured it'd be worth checking out the github...but yeah, I'm not a programmer or know computer languages, and running it is WAY over my head. So I'll happily wait for subsequent releases and play around with things I do know.
I made Easy V1 in Python and the idea was build a sound, modify a sound with stuff like phase shifts, then write the sound onto a timeline. It had a ton of commands but the problem was I couldn't remember all the commands. It was complicated. And keeping track of times was confusing. I got some nice sounds out of it but Python is also slow as fuck and hard to debug.
So Easy V2 is C++. I'm trying to imitate analog synths where the patch cables feed one thing into the next. Syntax is based on order-of-commands. I tried to keep the commands simple and make fallback defaults. The system keeps track of where you are on the timeline. I tried to make the syntax simpler so learn-by-example is possible.
Here's a few basics...
sound 10 vol .5
... makes a 10 second sound with 1000Hz (default) at 50% volume. Default is a sine wave.
sound 10 vol ramp .5 to 1
... makes a 10 second sound with volume rising from 50% to 100% The ramp command is a "NumberDriver"
sound 10 vol ramp .5 to 1 .2
... makes a 10 second sound with volume ramping from 50% to 100% repeating every 0.2 seconds
sound 10 vol osc .5 to 1 freq 2
... makes a 10 second sound with volume doing tremolo at 2Hz. The osc command is another "NumberDriver".
sound 10 vol osc .5 to 1 freq 2p
... makes a 10 second sound with volume doing tremolo at period of 2 seconds
sound 10 vol osc .5 to 1 freq 2p square
... makes a 10 second sound with volume doing tremolo at period of 2 seconds but make the volume modifier a square wave but the sound is the default sine wave
sound 10 vol 50% freq 600 to 800 freq 5p
... makes a 10 second sound with volume at 50% with frequency going up and down from 600Hz to 800Hz at 5 second period (p for period because I couldn't reuse s)
sound 10 vol .5 phase .4
... makes a 10 second sound with 1000Hz default at 50% volume with phase shifted by 50%
sound 10 vol .5 square
... makes a 10 second sound with 1000Hz default at 50% volume using square waveform
sound 10 vol .5 square duty ramp .2 to .9 10
... same as above by square wave duty cycle goes from 20% (sparse) to 90% (strong) - nice feeling!
sound 10 vol randseq .7 .9 .5
... 10 second sound at default 1000Hz with volume changing every 0.5s between 70% and 90%. The randseq command is another NumberDriver. This particular effect was unexpectedly delightful: like a massage.