2300 FTL drive vs CT Jump Drives
- Thread starter Murph
- Start date
What always bothered me about Stutterwarp was that it ignored intrinsic velocity. If you go into stutterwarp in orbit at 5 km/sec and come out above a different planet, presumably you still have a 5 kps velocity in an inappropriate direction. When you figure the orbital velocity of the different planets and stellar systems, you can have 10-15 kps velocities and no drive capable of dealing with it.
Premumably you could also tunnel to a new KE state as well as position, but that would imply a ship's power supply of a different order of magnitude.
A jump drive at least required a maneuver drive and stopped working far enough out in space that killing intrinsic velocity was a trivial maneuver.
As for making sense, the jump drive can be considered a special case of the Alcubierre/Van den Broek drive. This makes as much sense as the synchronized quantum tunneling of the stutterwarp.
[This message has been edited by Uncle Bob (edited 09 May 2002).]
Premumably you could also tunnel to a new KE state as well as position, but that would imply a ship's power supply of a different order of magnitude.
A jump drive at least required a maneuver drive and stopped working far enough out in space that killing intrinsic velocity was a trivial maneuver.
As for making sense, the jump drive can be considered a special case of the Alcubierre/Van den Broek drive. This makes as much sense as the synchronized quantum tunneling of the stutterwarp.
[This message has been edited by Uncle Bob (edited 09 May 2002).]
Fulacin Highport
SOC-9
Except that the stutterwarp drive is based (somewhat loosely) on the very real tunnel diode effect: http://www.xrefer.com/entry.jsp?xrefid=493458
Don't get me wrong, the scale-up problems boggle the mind. But it *could* work; it does already, in many, many electronic devices.
Personally, I hope the Alcubierre/Van den Broek drive really works out, and is not another pipe-dream. But it would require absolutely Godzilla-sized amounts of energy.... and the only reasonable way to supply the required energy would be to develop working zero-point extraction generators. Not exactly off-the-shelf tech.
------------------
----------
Joe Brown
Don't get me wrong, the scale-up problems boggle the mind. But it *could* work; it does already, in many, many electronic devices.
Personally, I hope the Alcubierre/Van den Broek drive really works out, and is not another pipe-dream. But it would require absolutely Godzilla-sized amounts of energy.... and the only reasonable way to supply the required energy would be to develop working zero-point extraction generators. Not exactly off-the-shelf tech.
------------------
----------
Joe Brown
Actually, the van den Broek geometry requires only a few grams of negative energy, while applying quantum tunneling on that scale boggles the information theory side (10^30 variables?). I figure a small ZPE breakthrough is far more likely than a mind boggling info theory development. But your millage may vary.
You still haven't explained away the Kinetic Energy and conservation of momentum problems with stutterwarp.
You still haven't explained away the Kinetic Energy and conservation of momentum problems with stutterwarp.
Gallowglass
SOC-12
<BLOCKQUOTE>quote:</font><HR>Originally posted by Murph:
Argument gentlemen and ladies? I have to admit other than the really stupid 7.7 ly limit on the 2300 stutterwarp, I like it better than jump drive. Also it makes sense in a way.<HR></BLOCKQUOTE>
Stutterwarp is a nice idea, IMO. By being linked to a piece of real world physics "you can point at" it helps the hard SF feel, and whilst the 7.7 LY limit is a bit arbitrary, the basic principle is useful as it creates a genuine "astrography" in 3d space, the same way that limited jump distances do. Push the LY limit up by any significant degree and the whole arms of space etc structure that was integral to the 2300 setting disappears (we will gloss over the effect of switching to the more up to date Gliese 3.0 star data
).
I always treated the 7.7 ly limit as being like de-Gaussing old monitors: the field builds up and some time around the 7.7ly limit it becomes sufficiently strong that you either shut down the drive or expose the crew to a lethal dose of radiation.
In contrast I always treated the OTU jump drive as something of a black box and never dwelt to much on what underpinned its operation. The one thing I think TNE got absolutely right was the FF&S premise that the FTL drive is one of the key defining elements of your setting, and as I like both the OTU and the 2300 settings, I rather liked both Sutterwarp AND Jump
All I need now is a decent 2300 or Traveller compatible set of rules for CJ Cherryh's FTL drive...
[This message has been edited by Gallowglass (edited 10 May 2002).]
Argument gentlemen and ladies? I have to admit other than the really stupid 7.7 ly limit on the 2300 stutterwarp, I like it better than jump drive. Also it makes sense in a way.<HR></BLOCKQUOTE>
Stutterwarp is a nice idea, IMO. By being linked to a piece of real world physics "you can point at" it helps the hard SF feel, and whilst the 7.7 LY limit is a bit arbitrary, the basic principle is useful as it creates a genuine "astrography" in 3d space, the same way that limited jump distances do. Push the LY limit up by any significant degree and the whole arms of space etc structure that was integral to the 2300 setting disappears (we will gloss over the effect of switching to the more up to date Gliese 3.0 star data
I always treated the 7.7 ly limit as being like de-Gaussing old monitors: the field builds up and some time around the 7.7ly limit it becomes sufficiently strong that you either shut down the drive or expose the crew to a lethal dose of radiation.
In contrast I always treated the OTU jump drive as something of a black box and never dwelt to much on what underpinned its operation. The one thing I think TNE got absolutely right was the FF&S premise that the FTL drive is one of the key defining elements of your setting, and as I like both the OTU and the 2300 settings, I rather liked both Sutterwarp AND Jump
[This message has been edited by Gallowglass (edited 10 May 2002).]
<BLOCKQUOTE>quote:</font><HR>Originally posted by Uncle Bob:
What always bothered me about Stutterwarp was that it ignored intrinsic velocity. If you go into stutterwarp in orbit at 5 km/sec and come out above a different planet, presumably you still have a 5 kps velocity in an inappropriate direction. When you figure the orbital velocity of the different planets and stellar systems, you can have 10-15 kps velocities and no drive capable of dealing with it.
<HR></BLOCKQUOTE>
I've not got 2300 in front of me right now, but IIRC, stutterwarp isn't efficient enough to use within orbit. For that, you still need thrusters. I would assume that these are used to control relative velocity when entering system/orbit.
Thrusters are one thing 2300 glossed over in ship design.
------------------
Paul
What always bothered me about Stutterwarp was that it ignored intrinsic velocity. If you go into stutterwarp in orbit at 5 km/sec and come out above a different planet, presumably you still have a 5 kps velocity in an inappropriate direction. When you figure the orbital velocity of the different planets and stellar systems, you can have 10-15 kps velocities and no drive capable of dealing with it.
<HR></BLOCKQUOTE>
I've not got 2300 in front of me right now, but IIRC, stutterwarp isn't efficient enough to use within orbit. For that, you still need thrusters. I would assume that these are used to control relative velocity when entering system/orbit.
Thrusters are one thing 2300 glossed over in ship design.
------------------
Paul
TJP
SOC-12
The 7.7 ly limit never bothered me that much. I liked the "astrography" and limited options it gave. To me the 7.7 ly limit was no more arbitrary than the jump limits of the jumpdrive. Choosing which one's better is hard as it depends on so many things, but I like stutterwarp more. I can swallow the 7.7 ly limit much better than the "one week in jump regardless of distance" thing or temporal misjumps. But that's just me and I have no scientific (or technobabble) reasons for it.
IMHO there has to be some form of limiting FTL travel if you want to have a structured campaign or borders in space. If you can explain it in some reasonable way ("reasonable" used here in relation to the campaign feel/genre), then you have taken a big leap in the right direction of building a good sci-fi universe.
And a guestion (again) to the scientifically oriented people: What happens if you ram another ship with stutterwarp and its "pseudo-velocity," or for that matter, hit a micro-meteorite while in warp?
IMHO there has to be some form of limiting FTL travel if you want to have a structured campaign or borders in space. If you can explain it in some reasonable way ("reasonable" used here in relation to the campaign feel/genre), then you have taken a big leap in the right direction of building a good sci-fi universe.
And a guestion (again) to the scientifically oriented people: What happens if you ram another ship with stutterwarp and its "pseudo-velocity," or for that matter, hit a micro-meteorite while in warp?
Very dificult to ram another ship or "hit" a metior. This is because the ship tunnles from ons point in space to another, up to a hundres meters or so away, with no elaps time (as oposed to that in diodes where there is a n elaps time equal to time that would normaly be taken for a particle of the given energy to traverse the tunnled distance). This means that in order for two ships to ram they have to be stuttering with exactly the same frequency, and have to have real space velocity vectors that intersect during the very short interval that they are in real space together. The realy nasty outcome is for one ship to materialise inside the other momentarilt, as their paths cross, or for the ship to materialise around a metior of some kind, this would cause enormous damage to all concerned. Still difficult to achieve if you are trying to and your target is trying to stop you.
<BLOCKQUOTE>quote:</font><HR>Originally posted by Uncle Bob:
Actually, the van den Broek geometry requires only a few grams of negative energy, while applying quantum tunneling on that scale boggles the information theory side (10^30 variables?). I figure a small ZPE breakthrough is far more likely than a mind boggling info theory development. But your millage may vary.<HR></BLOCKQUOTE>
IIRC it would take more energy than we have in the universe to make a tunnel big enough for a starship to travel through by brute force. Since 2300AD starships can clearly do it (
) they must be scaling the effect but not the tunnel. Making the Jerone Effect a one (or less) dimensional tarnsformation makes things easier. Use particles' interrelation to compensate for the effects of a fractal/lossy algorithm for keeping track of the ship's atoms and you're just about there. But you still need a biiig computer and some very springy tantalum coils.
<BLOCKQUOTE>quote:</font><HR>You still haven't explained away the Kinetic Energy and conservation of momentum problems with stutterwarp.<HR></BLOCKQUOTE>
Either the tunnel is rotated slightly relative to the direction of travel so the ship's own momentum is used to slow itself several thousand times a second or captians make sure they're heading in roughly the right direction and use thrusters as another poster suggested.
[This message has been edited by robmyers (edited 18 May 2002).]
Actually, the van den Broek geometry requires only a few grams of negative energy, while applying quantum tunneling on that scale boggles the information theory side (10^30 variables?). I figure a small ZPE breakthrough is far more likely than a mind boggling info theory development. But your millage may vary.<HR></BLOCKQUOTE>
IIRC it would take more energy than we have in the universe to make a tunnel big enough for a starship to travel through by brute force. Since 2300AD starships can clearly do it (
<BLOCKQUOTE>quote:</font><HR>You still haven't explained away the Kinetic Energy and conservation of momentum problems with stutterwarp.<HR></BLOCKQUOTE>
Either the tunnel is rotated slightly relative to the direction of travel so the ship's own momentum is used to slow itself several thousand times a second or captians make sure they're heading in roughly the right direction and use thrusters as another poster suggested.
[This message has been edited by robmyers (edited 18 May 2002).]
<BLOCKQUOTE>quote:</font><HR>Originally posted by robmeyers:
IIRC it would take more energy than we have in the universe to make a tunnel big enough for a starship to travel through by brute force. <HR></BLOCKQUOTE>
You are not making a tunnel for the ship to pass through whole: that is called a wormhole and does not exceed the speed of light except in certain ways which may be prohibited by causality. In stutterwarp you are encouraging each particle to make a quantum state change simultaneously, viewed as a change in position. "Tunneling" is not a real hoole like Babylon V, but quantum mechanics slang, as Mssr Chadwick understood..
<BLOCKQUOTE>quote:</font><HR>Either the tunnel is rotated slightly relative to the direction of travel so the ship's own momentum is used to slow itself several thousand times a second or captians make sure they're heading in roughly the right direction and use thrusters as another poster suggested.
<HR></BLOCKQUOTE>
The thrusters and their power supplies are one, maybe two orders of magnitude too small for these changes.
As for the rest, it is possible that I misunderstand you, but it seems that you are answering with technobabble, and not very good technobabble. This is innapropriate when we are discussing the science behind T2300.
The essense of "hard" SF is eschewing the impossible while embracing the unlikely.
[This message has been edited by Uncle Bob (edited 18 May 2002).]
IIRC it would take more energy than we have in the universe to make a tunnel big enough for a starship to travel through by brute force. <HR></BLOCKQUOTE>
You are not making a tunnel for the ship to pass through whole: that is called a wormhole and does not exceed the speed of light except in certain ways which may be prohibited by causality. In stutterwarp you are encouraging each particle to make a quantum state change simultaneously, viewed as a change in position. "Tunneling" is not a real hoole like Babylon V, but quantum mechanics slang, as Mssr Chadwick understood..
<BLOCKQUOTE>quote:</font><HR>Either the tunnel is rotated slightly relative to the direction of travel so the ship's own momentum is used to slow itself several thousand times a second or captians make sure they're heading in roughly the right direction and use thrusters as another poster suggested.
<HR></BLOCKQUOTE>
The thrusters and their power supplies are one, maybe two orders of magnitude too small for these changes.
As for the rest, it is possible that I misunderstand you, but it seems that you are answering with technobabble, and not very good technobabble. This is innapropriate when we are discussing the science behind T2300.
The essense of "hard" SF is eschewing the impossible while embracing the unlikely.
[This message has been edited by Uncle Bob (edited 18 May 2002).]
<BLOCKQUOTE>quote:</font><HR>Originally posted by Uncle Bob:
In stutterwarp you are encouraging each particle to make a quantum state change simultaneously, viewed as a change in position. "Tunneling" is not a real hoole like Babylon V, but quantum mechanics slang, as Mssr Chadwick understood..
<HR></BLOCKQUOTE>
Of course, it isn't actually FTL, since tunneling is limited by the shape of the quantum wave, which propagates at the speed of light. It's also a statistical phenomenon that cannot be 'encouraged' without violating QM.
<BLOCKQUOTE>quote:</font><HR>
The thrusters and their power supplies are one, maybe two orders of magnitude too small for these changes.
As for the rest, it is possible that I misunderstand you, but it seems that you are answering with technobabble, and not very good technobabble. This is innapropriate when we are discussing the science behind T2300.<HR></BLOCKQUOTE>
Why? Stutterwarp is clearly technobabble.
In stutterwarp you are encouraging each particle to make a quantum state change simultaneously, viewed as a change in position. "Tunneling" is not a real hoole like Babylon V, but quantum mechanics slang, as Mssr Chadwick understood..
<HR></BLOCKQUOTE>
Of course, it isn't actually FTL, since tunneling is limited by the shape of the quantum wave, which propagates at the speed of light. It's also a statistical phenomenon that cannot be 'encouraged' without violating QM.
<BLOCKQUOTE>quote:</font><HR>
The thrusters and their power supplies are one, maybe two orders of magnitude too small for these changes.
As for the rest, it is possible that I misunderstand you, but it seems that you are answering with technobabble, and not very good technobabble. This is innapropriate when we are discussing the science behind T2300.<HR></BLOCKQUOTE>
Why? Stutterwarp is clearly technobabble.
<BLOCKQUOTE>quote:</font><HR>Originally posted by Anthony:
Of course, it isn't actually FTL, since tunneling is limited by the shape of the quantum wave, which propagates at the speed of light.<HR></BLOCKQUOTE>
Tunelling operates only on electrons. The Jerome Effect operates on other particles as well (and starships), so it is not just tunneling. Criticism of Stutterwarp based on contemporary knowledge of tunelling is therefore proceeding from a false premise.
<BLOCKQUOTE>quote:</font><HR> It's also a statistical phenomenon that cannot be 'encouraged' without violating QM.<HR></BLOCKQUOTE>
How does Quantum Mechanics know that the statistics are being violated? If it has to check every electron in the universe it's violating Relativity.
<BLOCKQUOTE>quote:</font><HR>Stutterwarp is clearly technobabble.<HR></BLOCKQUOTE>
That's a little harsh. It's a game device, designed to make an SF setting more fun. I'm sure a Quantum Mechanics expert could come up with amusing excuses for its operation.
Of course, it isn't actually FTL, since tunneling is limited by the shape of the quantum wave, which propagates at the speed of light.<HR></BLOCKQUOTE>
Tunelling operates only on electrons. The Jerome Effect operates on other particles as well (and starships), so it is not just tunneling. Criticism of Stutterwarp based on contemporary knowledge of tunelling is therefore proceeding from a false premise.
<BLOCKQUOTE>quote:</font><HR> It's also a statistical phenomenon that cannot be 'encouraged' without violating QM.<HR></BLOCKQUOTE>
How does Quantum Mechanics know that the statistics are being violated? If it has to check every electron in the universe it's violating Relativity.
<BLOCKQUOTE>quote:</font><HR>Stutterwarp is clearly technobabble.<HR></BLOCKQUOTE>
That's a little harsh. It's a game device, designed to make an SF setting more fun. I'm sure a Quantum Mechanics expert could come up with amusing excuses for its operation.
Hmm, I might say stutterwarp is good technobabble. IIRC, similar drives were used in the 1960s including Poul Anderson's Polesthotechnic League stories and in a wonderful novelette by Martinez called "Zero Sum." It is in the catagory of things that might work if we discovered a couple of new principles. After all, ten years ago warp drive was impossible, three years ago impossibly energetic, now just very difficult. The same thing could happen to stutterwarp.
What I object to is playing fast and loose with newtonian physics and celestial mechanics.
[This message has been edited by Uncle Bob (edited 18 May 2002).]
What I object to is playing fast and loose with newtonian physics and celestial mechanics.
[This message has been edited by Uncle Bob (edited 18 May 2002).]
<BLOCKQUOTE>quote:</font><HR>Originally posted by Uncle Bob:
After all, ten years ago warp drive was impossible, three years ago impossibly energetic, now just very difficult. The same thing could happen to stutterwarp.<HR></BLOCKQUOTE>
Amen.
<BLOCKQUOTE>quote:</font><HR>What I object to is playing fast and loose with newtonian physics and celestial mechanics.<HR></BLOCKQUOTE>
If you don't like the idea of warping at a slight angle to the direction of travel each cycle to compensate, just have ships leaving orbit in the direction of travel with thrusters and moving into orbit with them at the other end of the stutterwarp trip. The problem is then comparable to accelerating/decelerating for an interplanetary trip.
[This message has been edited by robmyers (edited 19 May 2002).]
After all, ten years ago warp drive was impossible, three years ago impossibly energetic, now just very difficult. The same thing could happen to stutterwarp.<HR></BLOCKQUOTE>
Amen.
<BLOCKQUOTE>quote:</font><HR>What I object to is playing fast and loose with newtonian physics and celestial mechanics.<HR></BLOCKQUOTE>
If you don't like the idea of warping at a slight angle to the direction of travel each cycle to compensate, just have ships leaving orbit in the direction of travel with thrusters and moving into orbit with them at the other end of the stutterwarp trip. The problem is then comparable to accelerating/decelerating for an interplanetary trip.
[This message has been edited by robmyers (edited 19 May 2002).]
<BLOCKQUOTE>quote:</font><HR>Originally posted by robmyers:
If you don't like the idea of warping at a slight angle to the direction of travel each cycle to compensate, just have ships leaving orbit in the direction of travel with thrusters and moving into orbit with them at the other end of the stutterwarp trip. The problem is then comparable to accelerating/decelerating for an interplanetary trip.<HR></BLOCKQUOTE>
"If you don't like the idea of warping at a slight angle to the direction of travel each cycle to compensate"
It is not a matter of liking, but what do you mean by this? Are you proposing that stutterwarp can change momentum and energy as well as position? If so, it implies additional technology that makes nanotechnology look boring.
It still requires an energy source in the gigawatt-terawatt range, not the 10-200 MW range available. Unles you are suggesdting a technology that can ignore conservation of energy, a god-like power.
As for using thrusters, we need a delta vee of 10-50 kps. Since 1/6 of the mass of a lander is needed to cancel the 10 kps orbital energy (an unreasonably low number anyway), a stutterwarp ship will need 1/6-9/10 of its mass in thruster fuel to compensate.
I have no problem with the current rules as game constructs, but they are neither scientifically plausible or particularly well thought out.
[This message has been edited by Uncle Bob (edited 19 May 2002).]
If you don't like the idea of warping at a slight angle to the direction of travel each cycle to compensate, just have ships leaving orbit in the direction of travel with thrusters and moving into orbit with them at the other end of the stutterwarp trip. The problem is then comparable to accelerating/decelerating for an interplanetary trip.<HR></BLOCKQUOTE>
"If you don't like the idea of warping at a slight angle to the direction of travel each cycle to compensate"
It is not a matter of liking, but what do you mean by this? Are you proposing that stutterwarp can change momentum and energy as well as position? If so, it implies additional technology that makes nanotechnology look boring.
It still requires an energy source in the gigawatt-terawatt range, not the 10-200 MW range available. Unles you are suggesdting a technology that can ignore conservation of energy, a god-like power.
As for using thrusters, we need a delta vee of 10-50 kps. Since 1/6 of the mass of a lander is needed to cancel the 10 kps orbital energy (an unreasonably low number anyway), a stutterwarp ship will need 1/6-9/10 of its mass in thruster fuel to compensate.
I have no problem with the current rules as game constructs, but they are neither scientifically plausible or particularly well thought out.
[This message has been edited by Uncle Bob (edited 19 May 2002).]
<BLOCKQUOTE>quote:</font><HR>Originally posted by Uncle Bob:
Are you proposing that stutterwarp can change momentum and energy as well as position?<HR></BLOCKQUOTE>
Well, I was implying rotation but if that's nonsense I'm sorry.
As for the amount of energy required, we're back to warp drive feasability.
Are you proposing that stutterwarp can change momentum and energy as well as position?<HR></BLOCKQUOTE>
Well, I was implying rotation but if that's nonsense I'm sorry.
As for the amount of energy required, we're back to warp drive feasability.
Mark Fletcher
SOC-1
Hi,
IMHO, 2300ad is supposed to be a hard science fiction game. My interpretation of this is that the science of 2300ad could be seen as being possible from a 20th/21st century point of view.
The principle of Stutterwarp is based on Quantum tunneling, but at macro scales. Hence in 2300ad's background, its a more believable means of propulsion than a jump drive.
Ok so, there are the problems that you mention that werent taken into account in 2300ad, and perhaps they should have. But you have to admit, that Stutterwarp qualifies as a technology that can be included into the background if you just give it a cursory glance.
For me 2300ad was playing in worlds and environments that were alien and believable. The technology behind the game was a means to an end, not the end itself.
Cheers,
Mark
IMHO, 2300ad is supposed to be a hard science fiction game. My interpretation of this is that the science of 2300ad could be seen as being possible from a 20th/21st century point of view.
The principle of Stutterwarp is based on Quantum tunneling, but at macro scales. Hence in 2300ad's background, its a more believable means of propulsion than a jump drive.
Ok so, there are the problems that you mention that werent taken into account in 2300ad, and perhaps they should have. But you have to admit, that Stutterwarp qualifies as a technology that can be included into the background if you just give it a cursory glance.
For me 2300ad was playing in worlds and environments that were alien and believable. The technology behind the game was a means to an end, not the end itself.
Cheers,
Mark
<BLOCKQUOTE>quote:</font><HR>Originally posted by robmyers:
That's a little harsh. It's a game device, designed to make an SF setting more fun. I'm sure a Quantum Mechanics expert could come up with amusing excuses for its operation.<HR></BLOCKQUOTE>
Sure. Just change the laws of QM. There's nothing wrong with technobabble, just don't try to believe it's anything more.
Incidentally, stutterwarp as written violates both conservation of energy and conservation of angular momentum. The conservation of energy problem can be dealt with (it's not a huge amount of energy), the angular momentum problem can't be dealt with without requiring reaction mass.
That's a little harsh. It's a game device, designed to make an SF setting more fun. I'm sure a Quantum Mechanics expert could come up with amusing excuses for its operation.<HR></BLOCKQUOTE>
Sure. Just change the laws of QM. There's nothing wrong with technobabble, just don't try to believe it's anything more.
Incidentally, stutterwarp as written violates both conservation of energy and conservation of angular momentum. The conservation of energy problem can be dealt with (it's not a huge amount of energy), the angular momentum problem can't be dealt with without requiring reaction mass.
<BLOCKQUOTE>quote:</font><HR>Originally posted by Anthony:
Sure. Just change the laws of QM. There's nothing wrong with technobabble, just don't try to believe it's anything more.
Incidentally, stutterwarp as written violates both conservation of energy and conservation of angular momentum. The conservation of energy problem can be dealt with (it's not a huge amount of energy), the angular momentum problem can't be dealt with without requiring reaction mass.<HR></BLOCKQUOTE>
How so? The velocity is retained, and turning is accomplised via thrusters.
Bryn
Sure. Just change the laws of QM. There's nothing wrong with technobabble, just don't try to believe it's anything more.
Incidentally, stutterwarp as written violates both conservation of energy and conservation of angular momentum. The conservation of energy problem can be dealt with (it's not a huge amount of energy), the angular momentum problem can't be dealt with without requiring reaction mass.<HR></BLOCKQUOTE>
How so? The velocity is retained, and turning is accomplised via thrusters.
Bryn
