Alright, this is something I have pondered for some time and never really followed up on. I know many of you have much higher educations than I and can perhaps shine some light on the thought.

We send things out into space, we communicate with distant probes and devices. We constantly monitor the heavens for signals from distant lifeforms or space anamolies. However, sound (of any frequency) travels by wave correct? (some even argue light travels in waves?, thats not the point)

I always assumed that sound was the traveling vibration or 'wave' that travels through matter. We then detect this and enterpret(gah cant spell?) it accordingly. However, isn't space a vacuum? Void of any matter? How does sound travel? Light would be more understandable since it's (considered by most) a beam or ray.

Am I wrong in this somewhere? I have to be wrong... somehow? I just need that explained.

There is no sound in space.

sound DOESN'T travel in space, because, as you said, it is a pressure wave through matter. However, when we send signals, it's not sound that we send it with. Otherwise you'd hear a godawful noise coming from your computer's wireless card all the time, and people in between your radio and the station would all get blasted by the music.

We send it as radio waves, which are electromagnetic waves, just like light. Excluding fancy quantum theories and all kinds of new ideas which I don't know and probably wouldn't understand, we'll simplify it and say that electromagnetic radiation (in the form of waves) travels through a vacuum. That signal gets interpreted and translated mechanically into sound.

Someone help me out though, what's the current state of science explaining electromagnetic waves? It's a probability wave of the existence of a photon or something, right? I can't really remember...

I always thought it was just sound waves on frequencies we couldnt hear... IE 900mhz, 1.7ghz, 2.4ghz ect ect.

No, it's light waves in frequencies we can't see.

WAVE-PARTICLE DUALITY TO THE RESCUE!!!

So basically, light/radio waves/x-rays, all types of electromagnetic radiation can be simultaneously though of as both a wave and a particle.

So when we are detecting radio signals from space, or even when we are seeing light from the sun, we can think of it a stream of particles (or photons to be particular), and of course there is no reason why a stream of particles can't travel through a vacuum.

BUT, we can also think of the same radio signal/light/whatever as being a waveform traveling without a medium (when sound waves travel through the air, we call the air the medium. As you pointer out there is no "traditional" medium in a vacuum, but yet photon waveforms can still be transmitted). We generally consider this waveform to be a probability function, but this gets pretty hard to explain (and I don't feel up to explaining it).


Or, you can just subscribe to the camp that believes there to be an all permeating Luminiferous Æther... but that went out of fashion along with hoop-skirts and frock coats...

they've also demonstrated wave-particle duality with larger particles - neutrons, atoms, even buckyballs.

In space, No-one can hear you scream.

if a tree falls down in space and no ones around, does it make a noise?

*cough* ahem...

Okay so the 'waves' we send are actually energy waves in the form of light or magnetic waves? Wouldn't bursting a signal like that into space create a major disturbance of most electrical devices in the area of the transmitter?

I always wondered about those FCC labels saying 'this device must accept any interference' or something to that effect.

Sound also travels MUCH slower than light does, so telecommunications would be a heck of a lot slower if they relied on sound to transmit information. Think about lightning: the flash is a form of information, and so is the resulting thunder. If the lightning is about 1 mile away, then you can expect to wait about 5 seconds between seeing the bolt and hearing the clap of thunder. Thats just one mile of separation between the sender (bolt) and the receiver (you). If this were the case for radio waves etc. then something like a GPS would never be practical for sending signals between earth's surface and space.
Now sound does travel faster through other materials like metal and water, but it still wouldn't begin to approach the speed needed say for networked gaming.

Actually, GPS is so precise that they take relativity into effect when calculating satellite-receiver transmission times.

Different electronic devices are designed to detect specific frequencies. Thats why you can tune in to one single radio station at a time. The FCC regulates who gets to use which frequencies in order to keep the spectrum from getting too cluttered. Now certain devices can interfere with each other, which is why you have to turn off your electronics on planes and in some hospitals.

Have you got a link for any info on that? It's not that I don't believe you or anything, it just sounds frikkin' cool!

hmm, well thanks for clearing this up. I guess my biggest mistake was thinking 'radio' meant sound waves.

Im really amazed that the community was so helpful here.

http://en.wikipedia.org/wiki/Wave-particle_duality, scroll down.

wow btw.

That one quantom thingy where 2 objects can be in one place is awesomer.
And so is.. That thing where 2 atoms no matter how far apart can react to eachother.

yes.

our universe is messed up.

I told my EE major roommate about it and he was unimpressed. He said you actually needed it to prove alpha decay.

But then again this is also the guy that thinks hand written asm would let you run everything on 2k of RAM and that LCD displays have liquid in them.

Why does the L stand for liquid if they don't? I'm pretty sure they do.. I've broken calculators and black liquid spills out everywhere.

Whoever thought up that abbreviation is a retard, apparently.

Umm... Liquid Crystal Displays really do have liquid in them....
In modern High resolution Displays the liquid is usually kept in such small cells that even if you break the LCD in half, only a little bit will leak out (which is good, because the stuff is toxic anyway). The Liquid aspect is actually what makes the whole LCD work, because the crystals are in a fluid state, they can re-align themselves when exposed to an electric field.

As for the ASM bit, well, hand optimization can make a world of difference, especially for critical sections of your code. But it is simply unreasonable to try and hand optimizing even the most minimal of modern programs. Plus, it's just silly to put absolute limits on resources. Sure I could write virtually ever program in the world such that it never needs more than 2k of memory, but runtime and disk usage would both skyrocket... anyway... Meh...

solid state transistors (like the millions and millions and millions in your cpu and gpu) are based off of wave/particle duality eg. quantum physics

Sorry I was drunk when I posted that.

No see I was disassembling an somewhat broken LCD and he thought this entire, say 3/4" wide area was filled with liquid. IIRC it was just area for backlight diffusion. The panel itself was on the surface and very thin. He went on to talk about how even modern LCDs have several millimeters of liquid behind them. :S


He actually said using 2K of RAM and doing it at speed which is impossible since disks are way slow.

hand optimization is impractical in almost every situation because it severely limits the number of platforms it can run on. The UNIX/FOSS world favors AT&T syntax for assembly, while Windows and ICC on all platforms use the Intel syntax. Compliler support for inline assembly on non-x86 processors is spotty (often not existing at all, even for x86-64).

this also only really applies to C and C++ (which are basically higher-level, platform-neutral macro assembly languages), or other languages which compile to native code, have a relatively basic runtime and simple calling conventions. and even then, it's practically impossible to get significant results hand-optimizing for a wide range of processors like what's available in the x86 world.

by the time you finished hand-optimizing your code for the entire range of processors it would be running on, the total amount of extra time you spent on the project would be greater than the amount of CPU time saved.

I think we already had this argument a while ago... might be wrong... But, once more into the breach...

I beg to differ. Hand optimization still has it's place. Functions that perform fixed manipulations on an isolated set of data, especially when frequently called, can be optimized for a huge increase in throughput. We are talking about things Computational, and Rendering Libraries (though the latter have been getting a little less common).

Replacing 5-10 lines of good old C code that happens to be at the crux point of your critical path, with 20 or so lines of processor specific optimized assembly really isn't that much of a times sink, and as I've said before, can make a world of difference.

Your right of course, you loose cross-platform compatibility with that. But these tender little golden nuggets of assembly can just be re-written for the handful of platforms you really care about. And if thats still not good enough, just make a halfway intelligent makefile that detects the hardware platform, utilizes optimized assembly if available, and falls back to the original implementation if it can't.

We are talking about AT MAX, a man-weeks worth of work, total, for four different assembly "nuggets", and a smart build system. And in exchange you can probably achieve at least 50% higher throughput, and MUCH higher is possible. Seems worth it to me.

Also, I'm not entirely certain what you mean by "A Wide range of Processors like what's available in the x86 world". X86 is an architecture and the instruction set is defined by IA32. Every single processor in the x86 family, from Pentium to Pentium 4 including all AMD interludes, complies to the same fundamental instruction set, and can execute the same assembly (they may add more instructions to the top, and they did). This is actually one of the weaknesses of the x86 Arch, as it leaves a metric-****load of legacy requirements on new processors.

I have to agree with Jon'C. Unless you are coding embedded systems, operating systems, or you need to guarantee an operation is atomic on your architecture, hand optimized assembly is usually a waste of time.

See, this is all new to me. I always thought that scientists use one of those audio amplifiers that they use in playgrounds and point them up, and the astronauts put their hand in a cup over their ears to hear them from space.

the instruction set is the only thing that different x86 processors have in common.

did you know that, for instance, Pentium 4 processors don't have a barrel shifter? yet on every other x86 processor it's immensely faster to use shifts instead of multiplying or dividing by a power of two. so you write some fancy-pants bit manipulation code to squeeze out an extra few nanoseconds on your Athlon 64 gaming rig and surprise! it's actually slower on my laptop. A lot slower.

Your argument essentially ignores the fact that we have optimizing compilers now, a lot of which (such as ICC) support CPU-specific optimizations and branching. This happens automatically. And the only way you can write better optimization code is if you had a hand in designing the processor (ICC = Intel C Compiler)

also, I'm not sure if you intended this or not, but 'critical section' means something quite different when you're talking about computer programming. hand-optimized assembly wouldn't do anything special in a critical section (presumably your hand-written synchronization primitives would come into play before and after execution rather than during)

I think he just meant frequently used portions of code.

I have a physics question:

If I rolled 3 bouncy balls in a box that had no air resistance or friction, would the balls maintain an equal net force?

are you sure you meant net force? are you sure you didn't mean momentum?

You need to make sure that any collisions are perfectly elastic.

I'm confused also.. you mean will the balls keep constant speed? or will they bounce consistently? Because even without air resistance and friction, bouncing is not perfectly elastic.

hmmm. What I want to know is if they're constantly exchanging energy, there should be a consistent energy among them all.

The bouncing dissipates energy through heat and sound, so even if all three entered the box exactly the same, I think 'd have to ensure they all bounce exactly the same for that to happen.

alright. well, see, force is the change in momentum so

Ignoring the fact that you posed the question in a profoundly poor way, and assuming that the system is perfect and isolated, and that the collisions are perfectly elastic, the balls will always have the same total amount of energy.

Sorry, and thank you.

I was going to say something like that followed by WOW BUCKYBALLS.

Cheers for that heads up Joncy - I felt like melting my brain some more than usual this morning!