Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Sean Mirrsen]

Therefore, it's spoilered. It doesn't really warrant an own thread, we're only discussing it because lightspeed rocks came up.

[Lost Requiem]

I missed the relativity discussion! 

Since I actually am a theoretical physicist, I live for this stuff.

Just so that nobody argues about it again, I explain it here:

Spoiler (click to show/hide)

The key to understanding special relativity is to use constant-velocity frames of reference (observers), and to realize that light in a vacuum travels at the same speed (c) in all of these reference frames.

If you move slower than light relative to an observer, they will see you flatten, increase in mass, travel slower than light and your clock slow down.  You will see the distance between yourself and your destination decrease (the universe flattens parallel to your velocity), everything else's mass increase and all other clocks slow down.  If you going fast enough, it will seem to you that you traveled faster than light.

If you go at the speed of light, an observers sees you as flat, unchanging, infinite mass and traveling at the speed of light.  (Photons don't seem to have infinite mass, because they have no rest mass.  Zero times infinity gives you the mass of the photon, which is why photons can have differing masses.)  You will seem to reach your destination instantly.  (No time to brake, though.)

If you go faster than light, stuff gets really weird overall.  It will seem to you as if you arrive before you depart, if you watch the clock, but your memories might not agree, because you might have turned into anti-matter and perceived time backwards.  It has been quite a while since I did the calculation, so I cannot quite remember the details.

If you have accelerating frames of reference, you can handle any acceleration as a type of gravity, or vice versa.  This is Einstein's principle of equivalence.

Hopefully we won't have any more relativity discussions in this thread.

I wrote a research paper about this last term.

Spoiler (click to show/hide)

The definition of light-speed is, strangely enough, hard to pinpoint because we're already traveling at X-speed around the sun, and the sun is traveling X-speed around the galaxy, and hell if I know what velocities exist past that.

The most difficult part of traveling faster than light is the fact that light exists as an electromagnetic field, which is the same thing that holds the atomic structure of everything together. By this logic, traveling faster than light will make you travel faster than what holds you together.

[quinnr]

Therefore, it's spoilered. It doesn't really warrant an own thread, we're only discussing it because lightspeed rocks came up.

Oh, I don't really care either way...it's pretty interesting..

[Vicomt]

very big flamewar starting spoiler

Spoiler (click to show/hide)

matter /anti-schmatter. it's all waves anyway, oh and dark matter and dark energy = charged plasma

sorry, couldn't resist.

[smjjames]

very big flamewar starting spoiler

Spoiler (click to show/hide)

matter /anti-schmatter. it's all waves anyway, oh and dark matter and dark energy = charged plasma

sorry, couldn't resist.

You really love derailing this thread don't you? lol.

[Shoku]

@shoku re: water

Basically, the game needs to be aware of the difference between a static body of water and a dynamic one.  As it is, it makes no distinction.  Once this sort of thing happens, then FPS would only get hurt when water starts to flow.  Furthermore, I don't see any other way around modelling beyond the tiles immediately adjacent to a tile of water.  Somehow the game needs to realize:  "Hey, I'm a 7/7 deep thingy of water next to a completely vacant thingy of space.  I better change that fast!  So let's see, where should I move to..." and kill the FPS if need be, at least it will only be killed for moment if the flow isn't into a chasm or some other endless extreme. 

It does that just fine. You can expect a large body of water to very quickly fill up a room you've accidentally carved into it and piercing an aquifer floods your fort at a suitable rate.

The slow water issue only shows up when you've got 7/7 next to a long stretch of 6/7 next to a long stretch of 5/7 and so on.

No, no it doesn't do it just fine.  Something that only works under certain circumstances and not under others is not working fine, in my opinion.

Read carefully. I'm saying that water quickly moves into a space when you've got 7/7 next to 0/7. THAT works in a way that makes perfect sense.

We're not disagreeing on the issues, of course.  Just apparently on whether we should be satisfied with the way it works (ie whether it is a problem or not).  A 7/7 tile should be able to "look ahead" a bit (beyond its immediately adjacent tiles) and more aggressively dump its content.

It's going to have to look an awfully long way for the point when the system ends up slow.
But yes, having some kind of additional nudging to spread out would definitely give more pleasing results.

It would be interesting to experiment by surrounding a space of 7/7 water with floodgates and then open them simulateously to see if the 7/7 immediately becomes 7 1/7 tiles in an appropriately large room.  I am willing to bet 50 dwarfbucks it doesn't.

It will be comparably fast in relation to water trying to travel down a single tile wide tunnel. The perimeter of the water area with grow quite quickly as the water spreads meaning that instead of water needing to go one very specific direction to fill in space it will instead accomplish this by going in any direction.

You may end up with a corner or two that are dry while the other two have become wet but it wouldn't be all that unsatisfactory a result. It would look well enough like a wave of water for a few tiles and then the water would just slosh around till low.

Yeah, a small room will flood *pretty* quickly, but a long channel simply will not.  Anyway, another problem is the way time is very relative in dwarf fortress.  But even given that, I say it should not take a full year for water to flow from one end of a channel to the other given uninterrupted input.  That's the basis of that argument.

I agree. Propose a cheap method of giving it directionality and we may see a rewrite after the upcoming release.

If we can't agree on that, then there's no point arguing about flow and this other esoteric stuff.

Last time I checked U-bends didn't work perfectly, they only fill to one level under the source level.
To ilustrate this situation:

Code: [Select]

Source
   |
   V      
OOOOOO  OOOOOOOO
77777O  O.......  <-Safe area
OOOO7O  O7OOOOOO
   O7O  O7O
   O7OOOO7O
   O777777O
   OOOOOOOO

I tested with an underground pool, there were probably some strange things going on there...

Code: [Select]

OO77777777777777777OO
OOO7777777777777OOOOO
OO7777777777777777OOO
OOO7777777777777OO __←mined out with open (channeled floor)
OO77777777777777777OO
Anyway, got no love from that.  Unfortunately, I don't have the save anymore to go back and check.  So I might be wrong.

edit: mispelled "tiles"

Ya, that's not working as intended. If you dump a single unit of water on top of it I bet it would all start rushing in though.

Spoiler: Quantum physics issues (click to show/hide)

If you go faster than the speed of light relative to an observer, you cannot be perceived by or affect that observer, but, if you could, you would seem to have negative mass, thus you would be antimatter.  (Antimatter seems to have opposite charges, but that is actually an illusion caused by having negative mass.)

That's not true. If you go faster than the speed of light, your effective mass becomes imaginary, not negative (due to the negative inside the square root in the gamma function). You can still affect an outside observer by sending radio signals, etc. (but only to observers lying in certain angles such that the projection of your velocity onto the vector from you to them has a magnitude less than the speed of light).
Further, antimatter does not have negative mass. The mass of antimatter (positrons and anti-protons) has been measured, found to be positive, and in agreement with theory. Negative mass implies gravitic repulsion. There are no known materials that have this property (except for "dark energy" which is simply some unknown force pushing the universe to expand faster than current theories suggest). Further, negative mass would imply negative energy, which seems absurd to me (though I guess it may yet exist as part of the bookkeeping the universe needs to do for hawking radiation, etc.).

Spoiler (click to show/hide)

Can't we do away with calling dark energy an anti-gravity by saying that all the junk in the universe is mostly stationary but space itself is growing? Or is that just one of those lies told to everyone that didn't dig themselves all that deep into the physics mountain?

Spoiler (click to show/hide)

The definition of light-speed is, strangely enough, hard to pinpoint because we're already traveling at X-speed around the sun, and the sun is traveling X-speed around the galaxy, and hell if I know what velocities exist past that.

The most difficult part of traveling faster than light is the fact that light exists as an electromagnetic field, which is the same thing that holds the atomic structure of everything together. By this logic, traveling faster than light will make you travel faster than what holds you together.

Spoiler (click to show/hide)

Not really. Relativity makes light speed to same to you no matter how fast you're already moving.

[decius]

Spoiler (click to show/hide)

Negative mass would not only imply repulsion by gravity, it would imply a negative magnitude of momentum. That is to say, if you were hit in the face by a negative mass, it would pull you. Which, in a weird way, has a certain internal consistency, if you don't require conservation of charge in positron and negatron (antiproton) emission.

Special and general relativity rely on the premise that c is constant with respect to every observer. They allow for distance and time to be different for different observers, however. I can't reconcile those conclusions with the definition of meter being based on wavelength, so I don't try.

I'd love to see water work in some realistic fashion, rather than taking months or years to flow through a channel. I just don't see a cheap way to model it. A way of determining a 'steady state' condition would help FPS, but I don't think the cycles required to determine/confirm steady state water would be fewer than those required to model current behavior. Oh, and someone get a cheap pathfinding algorithm too. I'm tired of single-digit FPS on my laptop.

[Reese]

Spoiler (click to show/hide)

I'm also not a scientist, but every discussion of astrophysics I've been witness to involving he big bang always describes how the universe is expanding.  This is one the reasons, as it was explained to me, that we see things that appear to exceed the speed of light; galaxies that are at the far reaches of what we can see appear to be moving faster than light, but in reality they are moving at sub luminal velocities and the apparent speed is caused by the distance between those objects and us increasing at a rate that is faster than light as space expands.

I think that any water movement improvements that don't kill FPS are going to be reliant on a pathing rewrite.  I imagine that what is needed is a system that dynamically defines contiguous spaces as rooms or bodies of water that are used as nodes, then a large body of water would not need to check each square, just if there are any water pathable links from the one node to another node with a lower level of water, and then flow would only need to be calculated along the path between bodies of water and you can avoid silly situations like all the water in a cistern draining out the bottom in a short period of time and instantly flooding the area below without actually causing any flow thanks to the water teleportation behavior.

[Shoku]

Spoiler (click to show/hide)

I'm also not a scientist, but every discussion of astrophysics I've been witness to involving he big bang always describes how the universe is expanding.  This is one the reasons, as it was explained to me, that we see things that appear to exceed the speed of light; galaxies that are at the far reaches of what we can see appear to be moving faster than light, but in reality they are moving at sub luminal velocities and the apparent speed is caused by the distance between those objects and us increasing at a rate that is faster than light as space expands.

I think that any water movement improvements that don't kill FPS are going to be reliant on a pathing rewrite.  I imagine that what is needed is a system that dynamically defines contiguous spaces as rooms or bodies of water that are used as nodes, then a large body of water would not need to check each square, just if there are any water pathable links from the one node to another node with a lower level of water, and then flow would only need to be calculated along the path between bodies of water and you can avoid silly situations like all the water in a cistern draining out the bottom in a short period of time and instantly flooding the area below without actually causing any flow thanks to the water teleportation behavior.

So really huge rewrite of the system. Ok but you'll have to wait quite awhile for it.

In the meantime we can just make use of pumps and other pressure thing to quickly get water where we want it.

[dreiche2]

Yeah hopefully Toady will be better soon and... stuff.

Spoiler (click to show/hide)

Hello theoretical physicist...

If you going fast enough, it will seem to you that you traveled faster than light.

Why would it seem that way?? Relative speed is always symmetrical. If you travel almost at c from an outside observer, the outside observer will travel with almost c from your viewpoint...

If you go at the speed of light, an observers sees you as flat, unchanging, infinite mass and traveling at the speed of light.  [...] You will seem to reach your destination instantly.  (No time to brake, though.)

Just for the record, the question what it would be like to travel at the speed of light or beyond is, according to current physics, meaningless.

(Photons don't seem to have infinite mass, because they have no rest mass.  Zero times infinity gives you the mass of the photon, which is why photons can have differing masses.)

Uhm what?

I wrote a research paper about this last term.

The definition of light-speed is, strangely enough, hard to pinpoint because we're already traveling at X-speed around the sun, and the sun is traveling X-speed around the galaxy, and hell if I know what velocities exist past that.

The whole point of the relativity is that it doesn't matter who travels around whom how. The speed of light is always c, and you can just measure it. Or am I misunderstanding you?

[Flaede]

To hopefully move away from the theoretical physics*, I have a question about the water-flow diagrams that were getting tossed around.

Spoiler (click to show/hide)

I tested with an underground pool, there were probably some strange things going on there...

Code: [Select]

OO77777777777777777OO
OOO7777777777777OOOOO
OO7777777777777777OOO
OOO7777777777777OO __←mined out with open (channeled floor)
OO77777777777777777OO
Anyway, got no love from that.  Unfortunately, I don't have the save anymore to go back and check.  So I might be wrong.

edit: mispelled "tiles"

Ya, that's not working as intended. If you dump a single unit of water on top of it I bet it would all start rushing in though.

I'm not sure I follow this diagram, but I am fascinated by funny quirks in simulations. They entertain me just as much as the meticulous details represented properly. How does this water quirk work, can someone explain? I am even having trouble telling if that diagram was intended as a horizantal or vertical 'slice'.

*not that I have a problem with it... in it's own thread! I'd like it to have freedom to expand into a science-off free-for-all.

[Sysice]

Um... 0*∞ is 0. That's why photons actually can travel at the speed of light, because they have no mass.
Hm. How do I make this post on-topic?

Anyone want to place bets on when Toady''s going to recover? 

[zchris13]

Most likely vertical.  That SHOULD rush in, unless he has a diagonal below the walls next to the tunnel.

[CobaltKobold]

Vertical, and I agree with the others on the hypothesis- ponding another unit on top should start the whole mess going, since it thinks it's a static body of water right now.

[zchris13]

Unless he forgot to unpause the game.