Oh, that'd be it then. I must have overlooked the bit where you mentioned the Silesia dataset and skipped straight to enwik9. With that corrected, I get similar results to yours.
One issue I did notice during testing is that lzav's compression ratio is fairly unimpressive (~42:1) on all-zero data. Surely it should be able to do better than that?
A speaker can be used as a microphone. It's not a good microphone, but it's enough to save your voice and ambient sound, but the quality is bad.
So in the best case, the energy will go from the power source of the generator to the speakers, travel through air, and be absorbed by the other speaker acting like a microphone.
If the other microphone has a signal, the signal will try to move the internal parts of the microphone in one direction and the air in the other directions (or the same direction, according to the distance and frequency). This will cause a weird movement, and the amount of current and voltage drop in the speaker will not be the usual one. Anyway, there will not be something too strange unless you have a lot of oscilloscopes recording the process.
In a more real case, some of the energy will be transformed in heat in the wires, and some energy will escape because the sound does not hit the other microphone, and you will not recover the 100% of the energy.
Energy Conservation is one of the most fundamental laws of Physics. It may be wrong, but I don't expect that this device can prove it.
Anyway, Physics is an experimental science, so you can prove that the Conservation of Energy is wrong with an experiment. But you must:
* Make a video showing it. People will claim it's fake. I'll claim it's fake! Use a signal generator that has a battery so the maximal energy in the initial system is clear. Use a lamp to show the unexpected additional energy.
* Make the instruction 100% clear. Better 200% clear. Better 1000! clear. Check with a friend that s/he can build the device and reproduce the experiment without your help. Rewrite the instructions and check with another friend until they are 1000000% clear.
* If the instructions are clear and your device work, someone will not believe the video, and not believe the instructions, and then build the device as an amusement and get a huge surprise.
As you can see, you are making a lot of shaky assumptions about physics of the process. That's good, you can't tackle it! The problem is, video is not a proof, you can find a lot of videos of "free" energy generation that are fake. Idiots created a situation where nobody believes this. It's not a violation of conservation of energy, it's a crack in physics understanding itself. I have not seen anybody experimenting with forcing physics "collapse" itself. I'm probably the first one to do it. You can't find a better system than these simple loudspeakers to try oppose physics to physics.
I'll say more about this system. As such "opposition", however small it is (no reason to talk about how much energy is lost in system), creates a "paradox" between consuming and generating, the "physics" as a system of laws experiences "stress" and tries to solve the problem by expending energy. Same happens with a human when in tries to solve an untackable, but utterly important, problem-he experiences a mental breakdown. The space may have infinite energy at its base. It just strives to keep it in balance, hence "unbreakable" law of conservation of energy exists. I do not argue with "newtonian age" laws of physics, I just propose a way to put them out of balance. I succeeded at that in my opinion. Videos can be faked, so absolutely no reason to record it. You can see a similar speaker arrangement in any "surround" consumer sound system. Nothing new, except the way to arrange and operate the sound system.
Which problems do you see in the explanation and graph? It's as easy as it can be, totally nothing fancy, you can create the system out of scrap basically.
Note that I'm not an electrical engineer and can't write diagrams. I do not have "friends" in this area.
Anything far from C. Anything related to AI-driven mechanism operation: you'll be put in jail if it kills a human. Start your own software business, learn marketing and design, invent a solution to some problem, and become best in that. The other way is to becom a "dev" perpetually fulfilling some capitalist's dream of becoming rich and famous.
Oh yea, to me Lisp looks like Python, that's far from C, and usually it's a "interpreted" language, far from machine. Java and PHP is an example of interpreted language that's close to C, you can easily switch your mind from Java/PHP to "big boys" C/C++ if needed, and back. C/C++ is best paid "for hire" language, for a reason.
> Lisp looks like Python, that's far from C, and usually it's a "interpreted" language, far from machine
the currently most popular Common Lisp implementation is based around an optimizing native code compiler. That compiler has its roots in the early 80s.
Yes, I understand. Python can also be compiled to native code, if its developers decide to go that way. PHP8 is JIT compiled by now. It's more about being "closer to the machine". You can solve much more practical coding problems being "closer to the machine". Just check which class of specialists use Lisp/Python, compare to C/C++. I guess the initial question is "how to be relevant for a longer time on this market". Native code is a very big market, considering "IOT", much like website dev market. Can't say the same about Lisp and Python.
> Python can also be compiled to native code, if its developers decide to go that way
With Lisp this is usually done. The first compiler has been written early 60s. Since then zillions have been implemented and used.
> It's more about being "closer to the machine".
If one looks at a single CPU Lisp can compile itself to fast code, even in the range of 50% slower than C - compared to standard Python which can be much slower, unless computation stays in the Python libs which are written in C.
Where Lisp loses is the whole infrastructure to more advanced parts of the hardware and the systems. Linux for example is largely defined with C level interfaces. Lisp uses them, as foreign code and data - it can't use its GC managed memory.
> Native code is a very big market
Still Lisp compiles to native code, usually. There must be other reasons than native code.
Oh no, my question is not about money at all. My question is about being mighty scientist having some programming skills as core ability. Also my question is about being a good man, for example, I do not want to develop (or to help to develop) any malware, weapons, fake news providers and surveillance stuff.
For science, Python probably, it gained much tracktion in scientific community in the recent couple of years. But it's only a subset of "programming". C-like languages usually offer to become full-time programmer. In science you may need to read books and papers more often than to code I think.
You can only create "bad" code in programming, there's nothing bad in programming I know of. Ah, don't forget that as per GitHub statistics, programmers usually have psycho-neurological problems, it's a hard thing for the mind to stay in balance.
Do you really have that psycho-neurological problems? If yes, what do you think has caused them? Another question is how to predict which technology will have a long live, like Facebook's React, and which one will die shortly after birth, like Google's Angular.
Yes, to some degree, I have problems, but not too severe. The problem is you have to think logically, this puts your mind into "black or white" mode, and you have to find strength in yourself not to apply it to all surrounding life. There's no way to predict which technology will last longer. I think C will last indefinitely, because it's an "optimal" close-to-machine language. C++ is much better at coding large systems, but you have to be very smart at handling it, it has a lot of "hype-driven" innovations, not necessarily needed. Web is a very "interpreter" driven market, so "interpreters" will born and fade indefinitely, you can't predict it. Too many ways to do the same thing, basically.
PRVHASH is a hash function that generates a uniform pseudo-random number sequence derived from the message. PRVHASH is conceptually similar (in the sense of using a pseudo-random number sequence as a hash) to keccak and RadioGatun schemes, but is a completely different implementation of such concept. PRVHASH is both a "randomness extractor" and an "extendable-output function" (XOF), however the resulting hashes have security level that corresponds to the hash length specification: the collision resistance is equal to 2^(n/2) while the preimage resistance is equal to 2^n, where n is the resulting hash length in bits.
PRVHASH can be also used as a very efficient general-purpose PRNG with an external entropy source injections (like how the /dev/urandom works on Unix): the 64-bit hash value can be used as a pseudo-random number, spliced into 8 output bytes each round: this was tested, and works well when 8-bit true entropy injections are done inbetween 8 to 2048 generated random bytes (delay is also obtained via entropy source). An example generator is implemented in the prvrng.h file: simply call the prvrng_test64() function. The prvrng_test32() implements the same technique, but with 32-bit hashes, for comparison purposes.
prvrng_gen64()-based generator passes PractRand 8 TB threshold, without or with only a few "unusual" evaluations. Which suggests it's the first working universal TRNG in the world. This claim requires a lot more evaluations from independent researchers.
