Gameplay notes, suggestions and enhancements (VS 4.0.3)

[Shissui]

With a background in physics, this is likely a review for Peter. However, please allow me to more thoroughly kill this thread.

Relativity lesson #1 -- you are always stationary relative to yourself. Thus, if you are riding on a photon, then the destination is approaching you at the speed of light. It will take (time = distance/c) to get there.

Resolution of this apparent contradiction leads to the equations for special relativity. The equations for General Relativity derive from removing the assumption that all things move at a constant vector *without accelleration*.

***
The speeds in this game are very low relative to the speed of light. If anybody wanted to include relativistic effects, it would take a massive rewrite of the game system.

1) Rather than speed, it would make more sense to measure energy/mass. This is what would be used for determining how close to 'c' you are approaching. Note that your actual speed increases with the square root of your energy ratio.

2) Let us assume, for the moment, a constant vector of 0.99c [forward; star taken as frame reference]. What you observe in front of you is forshortened. A big space mine "spawns" directly in front of you, but far away. You will travel 99% of the way to that object before you discover that it has suddenly appeared. If it spawned behind you, it would take nearly twice the time for the knowledge of the spawn to reach you.
--> all information must be tagged with how old it is and this must be done in a way that is usefully displayed to the player.

3) Let us bring General Relativity back. I want to change direction -- the BigSpaceMine is actually still far enough away to do something about it. I now have a rank2 tensor (actually, a 4x4 matrix, not 3x3, as time must also be included in a GR tensor) to describe my current state and a second one to describe the effect that any thrust I apply to my ship could have on it.

4) If the mine is actually a torpedo (or any other self mobile object), then we have 2 more tensors to calculate for *its* position/vector & ability to change it. Then we need the product of these two calculations to know how to display that object in the frame reference of my ship. Finally, we must recalculate the time dilation effect to determine how long to wait before displaying these results.

5) If we add additional objects, the the problem becomes uncomputable on current home machines VERY FAST within the real time constraints of any realistic frame rate.

[peter]

With a background in physics, this is likely a review for Peter. However, please allow me to more thoroughly kill this thread.

I don't claim to be know the actual math behind GR. e.g. I've never studied tensors.

Relativity lesson #1 -- you are always stationary relative to yourself. Thus, if you are riding on a photon, then the destination is approaching you at the speed of light. It will take (time = distance/c) to get there.

Proper time=distance/c for an observer e.g. standing still at the destination. For the photon (and a hypothetical observer riding it), it takes _zero_ time.

qualitative argument: Photons have zero rest mass, so if they aren't moving at the speed of light (so their energy is 0/0) they can't exist.

BTW, a great book to give you a feel for relativistic effects is George Gamow's "Mr. Tompkins in Wonderland". http://en.wikipedia.org/wiki/Mr._Tompkins. The main character finds himself in a world where c = 30 miles/h, so relativistic effects are easily observed in everyday life, e.g. when looking at a distant clock tower, or passing cyclists.

Resolution of this apparent contradiction leads to the equations for special relativity. The equations for General Relativity derive from removing the assumption that all things move at a constant vector *without accelleration*.

***
The speeds in this game are very low relative to the speed of light. If anybody wanted to include relativistic effects, it would take a massive rewrite of the game system.

Yeah. Even getting sensors right wrt. SPECing ships would require consideration of the finite speed of light. The in-system map (shift-M) isn't limited by your sensor range, so it operates over distances of light-minutes. And then there's hails and wingman orders. It's much easier if we make like Star Trek and assume that all scanning and communications can happen at infinite speed if we want them to.

1) Rather than speed, it would make more sense to measure energy/mass. This is what would be used for determining how close to 'c' you are approaching. Note that your actual speed increases with the square root of your energy ratio.

2) Let us assume, for the moment, a constant vector of 0.99c [forward; star taken as frame reference]. What you observe in front of you is forshortened. A big space mine "spawns" directly in front of you, but far away. You will travel 99% of the way to that object before you discover that it has suddenly appeared. If it spawned behind you, it would take nearly twice the time for the knowledge of the spawn to reach you.
--> all information must be tagged with how old it is and this must be done in a way that is usefully displayed to the player.

I don't think you have the times right. I think you and photons from the mine would meet in the middle.

[Shissui]

Relativity lesson #1 -- you are always stationary relative to yourself. Thus, if you are riding on a photon, then the destination is approaching you at the speed of light. It will take (time = distance/c) to get there.

Proper time=distance/c for an observer e.g. standing still at the destination. For the photon (and a hypothetical observer riding it), it takes _zero_ time.

You, the observing rider, are not massless. Thus, you are always at rest relative to yourself. It takes the *photon* zero time, but it takes the rider proper time.

qualitative argument: Photons have zero rest mass, so if they aren't moving at the speed of light (so their energy is 0/0) they can't exist. :)

Now you are on the right track. :)

They cannot exist, yet there are measurable properties (energy = frequency = wavelength; phase; direction. The E & B field vectors are measurable, but also derivable from the energy + phase + direction).

Similarly, it is not possible to fully use Maxwells Equations (the physicist's 4 line description of all of electromagnetism) without integrating over a singularity (read: divide by zero).

***

I don't think you have the times right. I think you and photons from the mine would meet in the middle.

yes. (almost the middle)

***
While there are a variety of approximations & shortcuts available to simplify the task, my central point was that it is not computationally useful to implement a relativistic universe; especially when those computations must be made in real time and at the expense of the frame rate on the slowest computer deemed able to run the game.

There is a further argument that, even if implemented, it would be very difficult to display this universe in a way that is meaningful to the player. For example, it would be almost essential to use polar log coordinates centred on your own ship; and then apply a distortion to these coordinates proportional to the special relativity time compression. Long ago I drew a few of these to explain one of C.J.Cherryh's fleet battles --> I still needed 20 minutes to explain the resulting coordinates to a group of Physics grad students.

So, hopefully, I have finally killed this thread enough that it can continue in "off topic" ??

[peter]

Proper time=distance/c for an observer e.g. standing still at the destination. For the photon (and a hypothetical observer riding it), it takes _zero_ time.

You, the observing rider, are not massless. Thus, you are always at rest relative to yourself. It takes the *photon* zero time, but it takes the rider proper time.

I don't see how you can think that the photon experiences no time, but an observer at rest wrt. to it does experience time.

Anyway, the impossible thing here is getting anything with non-zero mass moving at c. Once you have that, it's the same as for a photon. From the photon's reference frame, everything ahead appears compressed, so the observer would measure any distance covered as 0m. Right? (I'm not considering any acceleration phase here.)

While there are a variety of approximations & shortcuts available to simplify the task, my central point was that it is not computationally useful to implement a relativistic universe ... in real time

Agreed. We're only debating the physics, not what should be in Vegastrike.

There is a further argument that, even if implemented, it would be very difficult to display this universe in a way that is meaningful to the player. For example, it would be almost essential to use polar log coordinates centred on your own ship; and then apply a distortion to these coordinates proportional to the special relativity time compression. Long ago I drew a few of these to explain one of C.J.Cherryh's fleet battles --> I still needed 20 minutes to explain the resulting coordinates to a group of Physics grad students.

cool.

So, hopefully, I have finally killed this thread enough that it can continue in "off topic" ??

I'm trying to stop myself from always having to have the last word, but there's always something I can find to reply to...

[Shissui]

This thread continues in the "off topic" forum thread=77885, as the discussion appears no longer to even contemplate the usefulness of the physics involved to any real implementation in VS.
[It would be nice if a moderator added the connecting link(s) actually.]

@Peter -- no need to kill the topic, I just think the forum should be changed.

[travis115]

i think it would help alot if you made things cheaper just a little bit, cuase one im always confused on how to make money and 2 unless im just doing a really crappy job on making money the prices still seem a little high, also make distances a little closer and dont make your ship slow down so fast when it gets close to planets object etc. (dont mean to criticize great job )

[peter]

i think it would help alot if you made things cheaper just a little bit, cuase one im always confused on how to make money and 2 unless im just doing a really crappy job on making money the prices still seem a little high, also make distances a little closer and dont make your ship slow down so fast when it gets close to planets object etc. (dont mean to criticize great job )

If you don't know what to trade where, you need vstrade. Check it out on http://vegastrike.sourceforge.net/wiki/OOG_Tools. I'm going to keep promoting it every time someone asks , because GoldenGnu and I are working on it, and pretty soon it's going to be even better than the latest alpha versions are already.