Thread: You have been waiting for this! Join the revolution

That looks awesome!. Good job!, is it going to be available to the IPhone 3 by any chance?.

$20? lol... Fail.. There areapps that cost $2.99 which work good.. but who buys apps anymore?

The app is 4$, the armband is somewhere around 12-16. And you'd be surprised about apps

It might run on iPhone 3 GS, but I don't have a device to test with, so iPhone 3GS might happen only after the release.

Shout out for EV, this is definitely something to be excited for!

Finally, a definitive proof of the wristband's effectiveness.

The algorithm fires before I wake up from the dream, just in the right time frame for me to be certain that I'm dreaming. Combining all 6 algorithms = success at EILD

This kind of accuracy is impossible with any mattress based app. My iPhone has 40% battery life left after 8 hours of wrist operation.

Spread the word
Pre-order armband! » Lucid Dreaming App

This looks incredible! How did you get the data onto Excel by the way?

Two collumns; Activity, Time
Paste em' in, then highlight and make a graph.

Awesome stuff Ev.

thats some pro research i will never understand..'

Thanks guys! Today marks the day that *may* change the world... Lucid dreaming aid for under 20$! 6 years of research into lucid dream induction, and I'm finally looking at something effective, affordable and widely available. Yet the day keeps unfolding just like any other day

I would greatly appreciate if you indicate your interest by either voting in the poll on the pre-order wristband page or sign up for the waiting list. If you got any non-confidential questions about the product, I'd be happy to answer them!

The wristband is comfortable to sleep in, it is not hot, does not seem to get sweaty, does not compress or restrict the blood flow while sitting comfortably in one position over the whole night.

Originally Posted by Ev

Thanks guys! Today marks the day that *may* change the world... Lucid dreaming aid for under 20$! 6 years of research into lucid dream induction, and I'm finally looking at something effective, affordable and widely available. Yet the day keeps unfolding just like any other day

I would greatly appreciate if you indicate your interest by either voting in the poll on the pre-order wristband page or sign up for the waiting list. If you got any non-confidential questions about the product, I'd be happy to answer them!

The wristband is comfortable to sleep in, it is not hot, does not seem to get sweaty, does not compress or restrict the blood flow while sitting comfortably in one position over the whole night.

Hey there! Really appreciate the research you've been doing, but I have to ask -- will this update be released for Android? Will the wristband support my phone? I have an Inspire 4G, the dimensions are 4.84in x 2.68in x 0.46in -- the iPhone 4G dimensions are 4.5in x 2.31in x .37in.

Maybe it'd fit tight? I'd definitely purchase this armband if it does! Please don't drop android development -- it's truly appreciated.

This update will be released for Android, however I'll need a new development device for that. The detection algorithm 6 requires a gyroscope and the device that I use for development does not have that. Additionally, android may have battery issues. iPhone can survive for 8 hours running the app and still have 40% battery left. My android device discharges really quickly because it runs on top of Windows Mobile.

There may be a wristband version for oversized devices. Or you can just ask you local tailor to re-fit an existing armand for you. It cost me 40$ to make one from my components.

Do you think the android version will be any less accurate than the gyroscope-powered iphone?

Also battery life will be so heavily dependent upon everybody's ROM, kernel, radio and device -- it would be no problem for me with my current setup but might be for somebody else; simply keeping it plugged in would be a quick fix.

This sounds awesome and I will buy and try it (Love the futuristic idea!) =) But can you explain how this would work exactly?
Sorry if you already have done that and if I have missed it..

Hi Choi,
I cannot go into the details of how/why the algorithm works, I'd like to keep it a trade secret for a while. Let's say that it works precisely because of the gyroscope, and this is why I must've missed it when developing the original android app. Because actigraphs are very expensive, and actigraphs with gyroscopes are even MORE expensive, I suspect this algorithm has been missed by sleep researchers for quite some time. I keep hearing that noone ever detected REM with actigraphy from the clinical research community.

Berryman, the android app would suffer from not having a gyroscope AND not having the same motion fusion algorithms that Apple provides to me as a developer. In Android, I was working with the raw acceleration, measured in Gs. In iPhone, I'm working with the gyroscope and *user* acceleration. This kind of filtering removes the effects of gravity and only leaves out the acceleration produced by the user. This kind of acceleration is a little bit indicative of the sleep onset, although the ratio

Yesterday Arch performed a quick test, and his results were very similar. About 4 minutes of sharp rise in Algorithm 6's output, followed by a certain plateau. His test indicated significant spike at Time in Bed +1:30 and Time in Bed + 2:05, then the makeshift armband became too uncomfortable causing an awakening.

Very interesting Ev, had a quick look on your site and joined the mailing list.
Been looking for a decent app for iphone and this could be it!
I wish you best of luck with it let us know how it goes

Ev, does that gyroscope make such an astronomical difference? Lucid app 0.7.3 seems to have perfectly fine accelerometer movement detection for me.

Originally Posted by Ev

Hi Choi,
...

Aha yeah of course you can't say exactly how it works I ment the precise function like: Keeps track on when you sleep and when you are awake and so on. But thank you =) But I understand now.

I would also love to add that some android phones do in fact have a 3-axis (I believe?) gyroscope; such as the samsung infuse 4g.

edit: infuse has a 3-axis gyroscope and a 3-axis accelerometer.

I went to school, enduring 4.5 years of grueling Electrical Engineering training. I took Calculus1,2,3, Physics 1,2,3. In a bunch of my classes I had to deal with differential equations. If only I was developing this project back then.

Lucid Dreaming App 0.7.3 is my attempt at an app with computer science and logic. This is something that I taught myself. Lucid Dreaming App was my first major programming project.
While I put a lot of effort into that, and it has been a bit better than all the other apps on the market, there are some fundamental errors in the lucid dreaming app beta 0.7.3 . It uses old actigraphy algorithms, dating as far back as 1992. Those are used by *extremely* omgwtfbbq kind of slow actigraphs. They sampled data very slowly, and did not have the gyroscope. These devices couldn't even imagine having user acceleration or gyroscope in them!

The kind of data processing that I'm running right now, in excel 2010 are killing the 4 cores of my CPU and 8 gigs of memory.

As a result, they tell you if you are asleep or awake. That's it. 1 or 0. Lucid dreaming app 0.7.3 is kinda like that. It's more accurate than some other apps out there, but it still uses technology that is almost 15-20 years old. There's only so much I could do with that. I should pull the Beta down or put a disclaimer saying that "this does not work for lucid dreaming". It is an ok sleep quality app.


I don't know the state of the current actigraphs, but what I'm seeing on my excel documents is nothing short of a major breakthrough.

I'm seeing the sleep cycles of a person clear as day.
It's almost like looking at the 4 stages of REM sleep, based solely on the wrist actigraphy. This is my own's body motion, and I'm just looking at it the correct resolution, with the right window size and looking for the right things.

The discovery of the detection algorithm 6 has made me look deeper into the math of the whole thing. I went to school for engineering and now recall just enough to say what a first deriviative is. I have pages and pages of equations, all of which point out to one thing, which I appear to confirm as I do more calculations:

MA = mental activity.

Mental Activity = Sum of functions. (function of the body(t) + function of the mind (t))

Awareness = Rate of change of mental activity. = MA'
Consciousness = Rate of change of awareness = MA"

I'm trying to prove this. If this is true, it would allow me to not only to *detect* when the dream onset is happening, but also *control* your mental activity level by observing how it changes to my input (the audio/vibration/light cue).

Together, I would be able to correct the guess of my equation and predict when you are going to fall asleep, see when you ARE falling asleep, see when you dreaming, see how you respond to the stimuli, and as a result, produce a certain change in your mental activity. The rate of change of MA would either increase or diminish the awareness. Think bineural beats on some serious steroids.

If I'm correct, this would be able to produce lucidity with only minor training, by observing how the body responds to the stimuli.

I present to you, detection algorithm 5:

Here are the 3 day "forecasts" for the sleep cycle changes, as well as the lengths of each sleep cycle.

In the image above :

There was a lucid dream at 5:44, followed by some dream journaling, creating an anomaly in the output. Complete awakening has been reported at 6:07AM. The Detection algorithm 5 predicted that I would arrive at the end of the sleep cycle at 6:07. A short lucid dreaming sequence has been reported at 5:48, followed by a near-dream state. <strong>This is the direct result of interfering with the sleep pattern via lucid dreaming. </strong>

TIB row stands for the time in bed. This is when I went to sleep at those days. By running the detection algorithm 5, I can predict <strong>the end times of all sleep cycles that the body has went through. </strong>For example on Day 1, I went to <strong>sleep at 00:41 </strong>and had the first sleep cycle at 1:53 The sleep cycle length column shows sleep cycle lengths for each cycle. These align closely with the 90-110 minute figure that is often quoted for sleep cycle length. This algorithm in real time predicts when the next end of a sleep cycle would be.

You're saying this (can't speak for anybody else) to someone who's scraping by Calculus 1 ಠ_ಠ

Looks absolutely nothing short of astounding, though. Sounds like the gyroscope is as useful as it seems! Might I ask though, is the 3-axis gyroscope found in some newer android devices different from that of the iPhone 4? If this application is as much of a breakthrough as it seems, I'd be willing to pay a lot more than $4 for it -- and I'm sure others would too.

By the way -- don't be discouraged at those "I'll Pirate It!" votes on your blog... I'm sure this could be very profitable for you if you present it as it really is. I wouldn't worry so much on developing an intricate licensing system and such; it'll get cracked one way or another and people will pirate it. Sad fact for such an enthusiastic developer, but know there will be many people who would happily purchase it as well.

this is really a breakthrough I commend you for your efforts and hope to try this app in the near future

I was stunned when i saw the waveform below


This is the biological clock sine wave from 3 days,

It is roughly aligned on the time in bed. This is a 90 minute period Sum of sinusoid waves, which might just correspond to the functions of the mind and the body that I was gonna plug into the equations of the awareness and concsciousness :

awareness = rate of change of the functions of the mind and body.

When the wave is rising, the awareness is increasing. when the wave is falling, the awareness is decreasing. When the wave is peaked, the awareness is either at the highest or at the lowest. When it is at the highest, it's easy to wake up and remember dreams, this is when dreams happen.
When it is at the lowest, the awareness reaches another turning point, which may be difficult to recall due to the sedating effect this may have on the body.




You will notice that the top of the sine wave is jagged, cratered. This is either because my detection criteria is incorrect at the moment, or there's another type of activity is started. the mental activity wave, as described above is heavily oscillating, and if awareness is the rate of change in the function of the mind and the body, this is when the awareness is increased and dreaming occurs.

Interesting...what is the difference between detection algorithm 5 and 6?

And are you proposing that the sleep cycle can be, although individualized, degraded down to simply a mathematical equation?