flat assembler
Message board for the users of flat assembler.
 flat assembler > Heap > sleepsleep's vitally important things Goto page Previous  1, 2, 3 ... 102, 103, 104 ... 125, 126, 127  Next
Author
Furs

Joined: 04 Mar 2016
Posts: 899
Ok, let me try a different way, since revolution doesn't seem to make you understand. Please do not nitpick the example for being too cheesy or "simple" since that's the whole point of it (just for you to understand, not to be "full").

I ask you first, how do you know something has length of exactly 0.1 anything (i.e. any unit)? Did you measure it? How do you know it's 0.1 and not, say, 0.100000000000000000000000000000000000000000000000000000000000000000000000000125??? Or something else representable by a 512-bit float? hmm? Or maybe 0.0999999999999999999999999999999999999999999999999999999999999999999999999999999. Which is the same thing.

In natural sciences, unlike math, we don't construct anything. All the math we do is to please the data/measurements, not the other way around. Math is a tool, not the truth. In fact, it's not even correct, but it's "good enough" for most tasks (note that is engineering, not physics). You seem hell bent on the idea that, if a math equation says something, that's exactly how it works, but that is not true and it's even known.

Saying stuff like "the equation says that" is not proof of how the Universe works. Physics is based on data, not math. Math is just a tool to put the data to use. If math can't reasonably (within a specific error tolerance) predict data, it is discarded, because data is king.

But if you can't measure data exactly then how can you say we know 0.1 exists? That's nonsense.

Nobody ever measured something with higher precision than Planck Length (if even that).

Listen, even our own algorithms use math. And yet, we can describe our algorithms with math. But they don't function with 0.1, even if "the math" does.

Math is ideal, the reality/Universe is not. Hence why 0.1 is not "real".
09 Aug 2017, 16:48
sleepsleep

Joined: 05 Oct 2006
Posts: 6961
Location: ˛　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣ Posts: 6699
all defined human units is to define some sort of measurement, they aint real, but the logic of maths is real, i suppose,

math could be truth because it doesn't matter what kind of unit any civilization invented or defined, but the logic of such symbol would always remain the same,
09 Aug 2017, 22:19
sleepsleep

Joined: 05 Oct 2006
Posts: 6961
Location: ˛　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣⁣ Posts: 6699
btw, do you guys/gals feel, such threading way of interaction doesn't really help in advancing human understanding / discussion about something?

problem: how to allow 5 and more people to talk and discuss?
09 Aug 2017, 22:23
YONG

Joined: 16 Mar 2005
Posts: 8000
Location: 22° 15' N | 114° 10' E
 sleepsleep wrote: btw, do you guys/gals feel, such threading way of interaction doesn't really help in advancing human understanding / discussion about something?
In fact, lots of the research on education methods suggest that online discussions tend to give better outcomes than do face-to-face discussions. Take a look at the link below:

Online vs. face-to-face discussions in a web-based research methods course for postgraduate nursing students : A quasi-experimental study
http://eprints.ioe.ac.uk/3291/

10 Aug 2017, 04:31
YONG

Joined: 16 Mar 2005
Posts: 8000
Location: 22° 15' N | 114° 10' E
 Furs wrote: let me try a different way, since revolution doesn't seem to make you understand.
Haha ... You actually sound like you are a physics guru.

10 Aug 2017, 04:35
YONG

Joined: 16 Mar 2005
Posts: 8000
Location: 22° 15' N | 114° 10' E
 Furs wrote: how do you know something has length of exactly 0.1 anything (i.e. any unit)?
I don't. I have never said so.

I pointed out a simple truth in computer science -- the exact value of 0.1 cannot be represented by floating-point numbers regardless of their bit-lengths.

And then you just shifted the focus to something else.

10 Aug 2017, 04:40
YONG

Joined: 16 Mar 2005
Posts: 8000
Location: 22° 15' N | 114° 10' E
 Furs wrote: In natural sciences, unlike math, we don't construct anything.
Really?

For example, in life sciences, there are lots of growth models specifically-constructed to study population dynamics. Refer to:

Population dynamics
https://en.wikipedia.org/wiki/Population_dynamics

Another example: In chemistry, the collision theory was constructed to explain how chemical reactions occur and why reaction rates differ for different reactions. Refer to:

Collision theory
https://en.wikipedia.org/wiki/Collision_theory

One more example: In physics, the BCS theory was constructed to explain how superconductivity arises in certain substances. Refer to:

BCS theory
https://en.wikipedia.org/wiki/BCS_theory

Enough?

10 Aug 2017, 04:56
YONG

Joined: 16 Mar 2005
Posts: 8000
Location: 22° 15' N | 114° 10' E
 Furs wrote: Physics is based on data, not math.
Really?

For example, have we collected any data from the event horizon of a black hole?

Still, we have lots of theories that describe black holes. In particular, some of those theories might be able to solve the information paradox. Refer to:

10 Aug 2017, 05:03
YONG

Joined: 16 Mar 2005
Posts: 8000
Location: 22° 15' N | 114° 10' E
 Furs wrote: Math is ideal, the reality/Universe is not.
I have never said that realistic simulations are not possible. In fact, I would say that even with our latest supercomputers, it is more than possible to create simulations that are realistic enough to "trick" the average human mind into believing that he/she is living in a "reality", at least for a short period of time.

My mention of 0.1 was meant to indicate the limitation of computing. To create an exact copy of the reality we are facing -- including all the microscopic phenomena like interactions between particles and quantum fluctuations, the "simulation" would have to be the reality itself.

10 Aug 2017, 05:21
Furs

Joined: 04 Mar 2016
Posts: 899
 YONG wrote: I pointed out a simple truth in computer science -- the exact value of 0.1 cannot be represented by floating-point numbers regardless of their bit-lengths. And then you just shifted the focus to something else.
In that case, you are obviously right, but what has that to do with what the context was? I assumed it wasn't just a side remark.

 YONG wrote: For example, in life sciences, there are lots of growth models specifically-constructed to study population dynamics.
They are theories though, and besides I don't know if it's the correct term, but I don't think those qualify as "Natural" science that is to deal with analysis of the nature around us (i.e. physics, chemistry, biology). If that's the wrong term then I used it wrong.

When I said construct I mean we don't get to define how the Universe works. We analyze it and make theories for it, which are just ideal approximations.

 YONG wrote: My mention of 0.1 was meant to indicate the limitation of computing. To create an exact copy of the reality we are facing -- including all the microscopic phenomena like interactions between particles and quantum fluctuations, the "simulation" would have to be the reality itself.
That's not true though. Microscopic phenomena (I know you meant it not literally, but I'm just saying) is only on the order of 10^-6.

We can perfectly calculate such things with our current tech and floats (no 512-bit floats needed), in fact we already do, how else would we create things as small as transistors are in the CPU these days? (which are between 10^-9 and 10^-8 )

The "quantum fluctuations" don't have to be stored, they can be calculated on the fly. Since they're random it is not possible to KNOW if it was calculated on the fly (procedurally) or not. Randomness doesn't make it harder to simulate, it makes it easier. Randomness hides the precision loss inherent from calculations. In fact that is one reason why I strongly believe we're in a simulation, because it is random.

Let me give you an example I'm familiar with. Quantization Distortion results when you try to store an audio PCM file with a limited precision, such as 16-bits. I'm sure everyone is aware of the 16-bit "CD Quality" PCM format. What fewer are aware of is that the quantization distortion is masked on purpose by a process called dithering. (note that the distortion is extremely unlikely to be noticed, but people want to be "perfect").

Since 16-bits have a extremely faint quantization distortion, they decided to mask it with white noise dithering, which is noise that gets added on the last bit to produce a noise floor. Noise is random, by definition. The noise will drown the quantization distortion completely.

i.e. it drowns the inherent imprecision of a 16-bit PCM number into random noise on the last bit. This makes the 16-bit format with no calculation / imprecision distortion AT ALL. The only thing this will have is a soothing "white noise" in extremely low volume. Like extremely quiet TV static. If you want actual numbers and dB levels see something like this article. (start with "Sampling fallacies and misconceptions" if you want the interesting parts)

So, randomness in quantum mechanics could very well intentionally mask some imprecise calculations in the simulation. We'll never see them because of it; all we see is a small amount of randomness (dithering).

The "fields" in QFT don't need to be stored whatsoever. They only "carry particles", and as such, they are implemented as "ideal math". Of course, even our own algorithms are implemented with ideal math. But the end result is that the calculations are imprecise.

Thus, it's not a very alien concept to realize that the designers of the simulation made it such that such imprecision loss is dithered away. While the calculations do have a particle end up in either one slit or the other (in the double slit experiment) based on ideal math conditions, the resulting particle seems "random" to us because it is dithered after the calculations.

(dither is always applied last -- after all processing has been done on the audio -- in this case, on the particle)

 sleepsleep wrote: btw, do you guys/gals feel, such threading way of interaction doesn't really help in advancing human understanding / discussion about something? problem: how to allow 5 and more people to talk and discuss?
That's a good question. I'm still thinking about it. But if you have a different viewpoint (than me or YONG) you're very welcome to express it ofc Most people tend to have a bias with one or the other because they share someone's viewpoint, so they just read/lurk. This is perfetly normal.
10 Aug 2017, 11:39
YONG

Joined: 16 Mar 2005
Posts: 8000
Location: 22° 15' N | 114° 10' E
 Furs wrote: ... I don't know if it's the correct term, but I don't think those qualify as "Natural" science that is to deal with analysis of the nature around us (i.e. physics, chemistry, biology).
The study of population dynamics is an important topic in ecology, which is a branch of biology.

10 Aug 2017, 12:50
YONG

Joined: 16 Mar 2005
Posts: 8000
Location: 22° 15' N | 114° 10' E
 Furs wrote: Microscopic phenomena (I know you meant it not literally, but I'm just saying) is only on the order of 10^-6.
I did give a couple of examples of microscopic phenomena. By "interactions of particles", I mean "interactions of any known particles in physics". Take the radius of an electron as an example. We are talking about something like 9.1 x 10^(-17) m. Refer to:

Determining the Electron Structure
http://www.alternativephysics.org/book/ElectronStructure.htm

A quark's radius is even smaller. Refer to:

10 Aug 2017, 13:08
YONG

Joined: 16 Mar 2005
Posts: 8000
Location: 22° 15' N | 114° 10' E
 Quote: To simulate the universe perfectly, you need a computer the size of the universe.
Refer to:

Are We Living in an Ancestor Simulation? ft. Neil deGrasse Tyson | Space Time

10 Aug 2017, 13:12
YONG

Joined: 16 Mar 2005
Posts: 8000
Location: 22° 15' N | 114° 10' E

How much computing power would be needed to simulate a virtual world that is indistinguishable from the real world?
https://www.quora.com/How-much-computing-power-would-be-needed-to-simulate-a-virtual-world-that-is-indistinguishable-from-the-real-world

How much processing power would we need to simulate a perfect replica of the world?

10 Aug 2017, 13:20
Furs

Joined: 04 Mar 2016
Posts: 899
Those links make a lot of assumptions and are outright wrong though, because there is no answer. It is not possible to know it, because we do not know the implementation. You can even calculate much slower than "realtime". That's because the simulation itself controls the time within it. Plus, video games employ tons of tricks to not calculate redundant stuff. I mean, our video games, not anything imaginary. So if a mere stupid game can employ such tricks, there's no way an assumption that not using any of them is "reasonable".

Lastly, it assumes that the outside Universe has similar performance bottlenecks, which probably isn't the case and they have vastly different implementations.

EDIT: What I mean to say is... everyone who assumes we need a computer the size of the Universe to store all that data is assuming that the implementation is brute force and raw and simplest possible. For example, it cannot possibly use procedural calculations (we use them e.g.: https://en.wikipedia.org/wiki/Procedural_texture), seriously?
10 Aug 2017, 13:33
YONG

Joined: 16 Mar 2005
Posts: 8000
Location: 22° 15' N | 114° 10' E
 Furs wrote: it assumes that the outside Universe has similar performance bottlenecks, which probably isn't the case and they have vastly different implementations.
We agree that aliens should not be very different from humans, right?

If so, it is reasonable to assume that the aliens create the simulation based on their daily-life experiences. Thus, the problems faced by the beings in the simulation should be very similar to those faced by the aliens in their real life.

Just my "silly" thought.

11 Aug 2017, 02:55
revolution
When all else fails, read the source

Joined: 24 Aug 2004
Posts: 15324
Location: Bigweld Industries
 YONG wrote: My mention of 0.1 was meant to indicate the limitation of computing. To create an exact copy of the reality we are facing -- including all the microscopic phenomena like interactions between particles and quantum fluctuations, the "simulation" would have to be the reality itself.
And this where the 0.1 thing has no relevance to whether or not we are simulated. We don't have 0.1 in our world, anywhere. We have the idea of 0.1, but not the implementation. This is where the form/function confusion comes about. BTW: We can replace 0.1 with 2, or pi, or any other number we like, and it still holds. We don't have 2 in our world, only the idea of 2, the concept of 2, not an actual 2.
11 Aug 2017, 04:10
YONG

Joined: 16 Mar 2005
Posts: 8000
Location: 22° 15' N | 114° 10' E
 revolution wrote: And this where the 0.1 thing has no relevance to whether or not we are simulated.
First, I am glad that you are back -- I thought that you would hide forever.

Second, how come you did not have the guts to admit that you were wrong about your "any such simulation" statement?

Third, why would "pointing out the limitation of computing" be irrelevant to the discussions of simulation?

No more "0.1 is not real" nonsense!

11 Aug 2017, 04:27
YONG

Joined: 16 Mar 2005
Posts: 8000
Location: 22° 15' N | 114° 10' E
Should we trust revolution given that he/she constantly tries to trick other forum members?

It is not just my opinion. Both guignol and T.G. have expressed similar opinions before.

Refer to:
https://board.flatassembler.net/topic.php?p=198425#198425

11 Aug 2017, 04:35
revolution
When all else fails, read the source

Joined: 24 Aug 2004
Posts: 15324
Location: Bigweld Industries
Hmm, firstly, why would I hide?

Secondly, I don't think I'm "wrong". There appears to be a disconnect between my explanation and your interpretation.

Thirdly, a limitation of computing doesn't invalidate that we are simulated, or not. And the specific mention of a value 0.1 has no relevance, but you appear to apply a representation of something as evidence of an actual thing. This is the form/function thing that you still seem to have trouble understanding. How would you comment on 2 instead of 0.1? 2 can be represented in a binary float format, but 2 is not an actual thing, it's just a concept. So whether or not 2 is representable in binary float format gives us no insight to whether or not we are simulated.
11 Aug 2017, 04:36
 Display posts from previous: All Posts1 Day7 Days2 Weeks1 Month3 Months6 Months1 Year Oldest FirstNewest First

 Jump to: Select a forum Official----------------Blog General----------------MainDOSWindowsLinuxUnixMenuetOS Specific----------------MacroinstructionsCompiler InternalsIDE DevelopmentOS ConstructionNon-x86 architecturesHigh Level LanguagesProgramming Language DesignProjects and IdeasExamples and Tutorials Other----------------FeedbackHeapTest Area
Goto page Previous  1, 2, 3 ... 102, 103, 104 ... 125, 126, 127  Next

Forum Rules:
 You cannot post new topics in this forumYou cannot reply to topics in this forumYou cannot edit your posts in this forumYou cannot delete your posts in this forumYou cannot vote in polls in this forumYou can attach files in this forumYou can download files in this forum