More power than your body has room for!

I recently stumbled across a post on a forum where one of the people posting in the thread stated that they were using subliminals by a producer who had "what may be the most powerful subliminals on earth" because he uses 1 septillion different voices at a time, and they're all adjusted by pitch to allow the subconscious to differentiate and understand them.

For reference in the rest of this post, I quote: "The affirmations are repeated 1.8×e^4240 times per second. and he says " I made enough different pitches of the voices to attempt to replicate 1,000,000,000,000,000,000,000,000 (1 septillion) different voices" such that the subconscious can tell apart the different layers."

I'm going to give it the benefit of the doubt and say they must have been joking, because it would be so far beyond ludicrous to believe this to be true that you'd have to completely ignore even the smallest hints of logic, reason and common sense.  But in case they weren't, and someone actually believes claims like this, let's look at this with some logic and show why it's so ridiculous that there are no words in the English language that adequately describe how ridiculous, preposterous and flat out insane it actually is... 

1.8×e^4240 (1.8 to the 4240th power) translates to "Overflow: the result couldn’t be calculated" in the scientific calculator running on my production system.  Each one of those 4,240 represents the original number (1.8) being multiplied by itself as many times as the exponent; that's what a "power" means.  So in this case, 1.8 multiplied by itself again and again, a total of 4,240 times. 

The number that results is so stupendously large that it is absolutely meaningless outside of exponential format, and no human could possibly ever even hope to comprehend it. I suspect that outside of certain niche scientific fields, it is far beyond absurdly useless in it's size.

In fact it's so large that no computer or calculator I can find will work with it for me to show how preposterous it is.  They all just throw errors when I enter that value.  That's how ridiculously, insanely preposterous it is.  We can't even work with it.

Let's look at the claim of "a septillion voices all at once", separated by pitch such that all those layers could be told apart by the subconscious.  At least a calculator can represent and work with that number to some degree before it dies of exhaustion!

According to the experiments I have done, the subconscious mind can usually detect audio between a pitch of about 5-15 cycles per second on the low end and around 25,000 cycles per second on the high end.  These appear to be the limits of audio input through the ears, and they far exceed what the conscious mind can detect through the ears, which is typically 20 cycles per second  on the low end to around 14,500 cycles per second on the high end for most people.  Let's presume this gives us 24,995 different whole number values for cycles per second pitches to use.

But if we are trying to divide 1 septillion by 24,995, that means that there must be how many different pitches per whole number of the pitches that a human body can perceive at the subconscious level?  Well, my calculator responds to this calculation with the error message, "Malformed expression".  In other words, this calculation is so preposterous that even a desktop scientific calculator app can't handle it.

According to Google, a septillion divided by 25,000 is 4e+19.  If I recall correctly, that translates to 4 with 19 zeroes after it.  So 40,000,000,000,000,000,000, or 40 quintillion.  That's how many different voices would have to be spread across every hertz (one cycle per second) of difference in pitch frequency, if there were a septillion voices at once.  There would be 40 quintillion different voices, all at different pitches that were 1/40 quintillionth of a cycle per second different, just going from a pitch of 24,995 cycles per second to 24,994 cycles per second.

Now my question to you is, what digital audio workstation (DAW) can even create two sounds that have that much resolution to their pitch values?  Not a one, because no such thing exists.  Two sounds that are 1/40 quinitillionth of a second different in pitch simply cannot be created by any modern computer.  No audio recording format has that level of pitch resolution designed into them.  And even if they could, you wouldn't be able to tell the difference, because your ears simply cannot register pitch differences that minute.  Not even close! 

But let's play around with this for a minute, presuming that your ears and subconscious and physical brain, all of which are different beasts entirely, could all register pitch differences down to 1/40 qunitillionth of a second.  What then?  Then you'd have to be able to differentiate that many different sounds at once.  Once again, your ears and brain simply cannot do that; they just don't have the resolution to even separate or process 10,000 voices at once, never mind 40 quintillion, and they sure as hell cannot differentiate a septillion.

But wait, there's more!  For you to have a track composed of a septillion voices, you would have to make a mixdown of 1 septillion tracks.  There is no digital audio workstattion software in the world that can handle 1 septillion, or even 40 quintillion, tracks at a time. Sorry.

But if there was, how long would it take just to create the empty tracks, without even loading any audio into them?  If we estimate that a fast DAW on a fast computer can create 10,000 empty tracks per second - I don't know how many it can create, but this seems like at least a plausible maximum number - then it would take 1e+20 seconds - one hundred quintillion seconds - which is 3.17 BILLION years.  Just to create that many empty tracks to hold the audio we want to work with.  If we say it can create 100,000 empty tracks per second, we still have to wait a whopping 31.7 million years!

Then, there is also no disk file system in the world that can keep track of one septillion files at once, although Google's AI seems to think NTFS could "potentially" do so, as it can handle up to 8 petabytes of disk space.  But, if each "track" were just one byte (which is one character, like a single X) you'd have a septillion bytes, which translates to one yottabyte of disk space, which is VASTLY bigger than any hard drive or any other storage format on earth. Never heard of a yottabyte?  Neither had I, and I used to work in computers for a decade.  This is what Google tells me in response to the question, "What comes after a terabyte?"

"Petabytes, exabytes, zettabytes, yottabytes, and brontobytes are all units of measurement that come after terabytes. Each unit is 1,000 times larger than the previous one."  Some people may argue that Google is wrong, because computers work in powers of 2, meaning every one of them is not 1,000 times bigger than the last, but 1,024 times.  Let's use 1024, since that's what the computer will use.

So to help you comprehend to some small degree how insanely large a yottabyte of disk space is, let's break it down.

A byte is one character, like X.

A kilobyte is 1,024 bytes. (Thousands of bytes.)

A megabyte is 1,024 kilobytes. (Millions.)

A gigabyte is 1,024 megabytes. (Billions.)

A terabyte is 1,024 gigabytes. (Trillions.)

A petabyte is 1,024 terabytes. (Quadrillions.)

An exabyte is 1,024 petabytes. (Quintillions.)

A zettabyte is 1,024 exabytes (Sextillions).

A yottabyte is 1,024 zettabytes. (Septillions.)

And a brontobyte is 1,024 yottabytes. (Octillions.)

Even if each of 1 septillion "tracks" were just one byte, the smallest size of a sector you can store data to in a file system is 512 bytes.  Which means that if we store each "track" as just one byte, that 1 septillion tracks track would not need 1 septillion bytes of storage space to store, it would need 512 times that, or 512 yottabytes!  Again, vastly more than anything on earth.  But according to Google, modern file systems use a minimum sector size of 4,096 bytes, which means at one byte per "track", you would need 4,096 yottabytes, or 4 brontobytes!

But wait!  It gets even more preposterous!! If we have just 1 second of audio data per track, and we use the current bare minimum audio resolution for audio production, which is 44,100 samples per second, and we say that it requires 1 processor operation to process one of those samples, then how long would it take to mix down a septillion tracks using the fastest supercomputer we currently have?

According to Google, the fastest supercomputer in the world right now is called Frontier, and it can do more than 1 quintillion calculations per second.  Let's round this to an even 1 quintillion to make it easier to work with.  A septillion tracks divided by 1 quintillion operations per second gives us 1 million seconds of processing time required.  That's "only" 11.5740741 days to theoretically mix that many tracks down, if we ignore a lot of factors that would make it take longer.

1 septillion tracks of 1 second each at 44,100 sample rate with that processing speed at one operation per sample mixed down is 1 septillion times 44,100 divided by the speed of processing.  1 septillion times the sample rate of 44,100 is 4.41e+28 calculations, and this can be performed at 1 quintillion calculations per second, resulting in a time to process the mixdown of just one second of audio on 1 septillion tacks of 44,100,000,000 seconds, which is 1,397.473368 years. 

And this is if we completely ignore all the processing power needed at the same time to run the operating system and do all of everything necessary to enable the computer to actually perform this mixdown operation, AND we assume that each sample only needs 1 operation of processing time in the processing of the mixdown. (It would almost certainly need more.)

Are you beginning to see just how far beyond insanely, ridiculously preposterous it is for anyone to say that they made a single track that consists of a septillion different voices at once?  Or say that any supercomputer or software could create or process it?  Or any hard drive could store the data necessary to make it?  Or that anyone has access to a computer that could do this in a single human lifetime?  We didn't even take into consideration how long it would take to save that mixdown, or how much RAM would be required to perform the calculations necessary to do the mixdown!

I'm really hoping the original poster of that statement was joking. 

Oh, and just for your information... I do NOT use 1 septillion voices at a time in my subliminals, and I do NOT use 1.8^4240 repetitions per second... Not even close.  Why?  Because that would take longer than 60 million years to make if I had the most powerful supercomputer in the world, more disk space than humankind has ever created, more electricity than is available in the entire world and somehow could mix down that many tracks at once. And the result would just be... white noise.  Which you can generate for yourself, in seconds, for free.

For comparison, I use a very specific number of voices (which doesn't end in a 0 or a 5!) that has been carefully calculated to work with the limits of what your ears, your brain and your subconscious can actually input, decode, decipher and execute.  It took me more than a decade to work out what that value was based on real research and real live experimentation.  And as for repetitions per second, there is no realistic way for me to calculate that.  That's the truth, and those are the facts.