How do ships see?

[Anthony]

Originally posted by atpollard:

First. Do you know what the range is that the sunlight reflected off a 300 degree K scout ship averaged with the background will be magnitude 44 per arc second? (the scout ship is twice as bright as the average sky).

Twice as bright as the background is magnitude 20.5, not 44, and it depends on the color. It's also not necessary to be twice as bright as the background; equal will do, since ships don't block the background out. Assuming a fairly dark hull, and ignoring heat output, probably on the order of 20 light-seconds.

Second. Have you considered that this computer/sensor problem needs to be solved in real time?

You don't have to. It's much easier to track a known target than to find an unnoticed target. You can use real-time imaging to track known targets, and a modest scan time to find new targets.

If my ship is being shot at by another ship at a range of 100,000 kilometers, I may want to target the incoming missiles with my laser. I cannot abort a 10 minute scan because my ship performed evasive maneuvers and start a second scan with the sensors if I want to target the missile.

The example I gave involves a signal integration time of about ten seconds; it takes longer than that because each frame covers a fairly small part of the sky, but if you already know what part of the sky you care about (the one with the enemy ship in it), that's not a factor. You can fairly easily reduce the time to a second or less, at some loss in sensitivity.

Third. If we use hardware to detect the moving object by comparing two images taken 1/10 of a second apart (an impressive exposure time) and only report the changes to the ship computer (to filter out stationary background objects) – then how fast will a ship need to travel (perpendicular to the field of view) at a distance of 900,000 km (maximum tracking range) to travel 1 arcsecond and be detected as “moving” by this 500 gigapixel detector.

About 45 kps. However, you wouldn't do that. It's not like you can't just compare the images to ones you took ten seconds ago, even if you're taking multiple images per second.

I would personally like to know what could be built TODAY using non-military technology and an unlimited budget. Mounted to a moving space shuttle, how far away could you detect another shuttle on a transfer orbit (no engines firing, but life support at normal levels). Is the range closer to kilometers or AUs or hundreds of AUs? Are CT sensors weaker than real life or “handwavuim” more advanced than real life?

Depends on the time frame and on whether we knew where to look. Assuming we knew where to look, it's probably on the edge of detectable at an AU. If we didn't know where to look, it's not really realistic to figure on detection beyond a few million kilometers.

 

[ElHombre]

Good points all around.

Assuming a fairly dark hull, and ignoring heat output, probably on the order of 20 light-seconds.

Does anyone have any ideas on how a ship's heat signature would be handled? You can't hide it and I'm asuuming any starship will have some sort of IR sensing array given how useful it can be. The power plants in Traveller have large outputs and that doesn't make a corsair's job any easier.

 

[sgbrown]

I don't think detection is nearly as easy as folks make it out. There is a tremendous amount of background noise - Some of which is also moving. The issue is how do you distinguish a ship from an asteroid at long distance? Not saying it "can't" be done - just that the problem is far from trivial. So yes you can hide in space - particularly if you want to follow a non powered orbit that is reasonably similar to asteroids. Even easier if you know which direction the enemy lies in.

 

[Scott Martin]

Perhaps a bit of overanalysis is being done.

Let's look at this from the point of view of a Merchant, who wants to spend as little as possible on his sensor array.

What a merchant cares about is detecting a "bandit" far enough out that they can get to the 100D limit before the (potential) hostile gets within weapons range.

I don't think that anyone here is saying that firing ranges are greater than detection ranges.

I see 2 primary alternatives:
1) The "Bandit" comes in at high accelleration from a long ways out and tries to hit the merchant before they can jump.
2) the bandit "lies low" and tries to ambush the merchant far enough out that the merchant cannot run for part, but close enough in that the merchant can't reach 100D.

Personally I'd opt for the second if I'ma privateer: land on the planet (where ground clutter will make it almost impossible to see me) wait for the merchant to hit about 3 Diameters and then go after them at 4+ Gravities.

Option 1 leaves you a lot less potential places to hide (the surface of a local moon maybe?) and because Traveller has a firly "low" thrust range, 6G isn't a lot of juice to catch anyone on a "closing" intercept, especially given that local traffic control is pretty much guarenteed to have better sensor arrays than said merchant (of course someone may have paid traffic control to look the other way, but an Imperial (or Megacorp backing the financing bank) )Investigation following that up would mean that whoever was looking the other way better make enough to retire...

You will note that neither of these has anything to do with ships "skulking" through space: in both cases the privateer is landed (or attached to) another body and has something to hide on / in to spoof long-range detectors. Trying an "ambush" with a force that needs to jump in-system is probably a complete waste of time, and all of these types of "ambush" situations will be close-in to a planetary body.

If I were a deep space fleet, I'd pointedly park my capital ships a light second or three away from any hostile planet and use them for fire support while I sent in the Marines in their (expendible) landers...

Scott Martin

 

[Anthony]

Originally posted by SGB - Steve B:
The issue is how do you distinguish a ship from an asteroid at long distance?

Generally speaking, you distinguish ships from asteroids by the fact that they're in an interesting position, or headed towards somewhere interesting. As a defender, you can also distinguish them by the fact they aren't in your asteroid database.

There may be lots of asteroids out there, but the vast majority are nowhere near anything of interest. Objects become interesting when they're going to pass near to a target. It's perfectly realistic for a ship to hide out in the asteroid belt, but even at 6 Gs it's at least two days to get from the asteroid belt to anywhere interesting, which is typically plenty of time to react.

The issue is not whether stealth is possible in space at all. The issue is whether stealth is possible at ranges that are tactically relevant, which is generally ten light-seconds or so.

 

[far-trader]

A small tangent if you all don't mind, I was just curious what the opinions are of active stealth measures. What I mean are tactics like dazzle where rather than trying to not be noticed you go the other extreme and try to overload the oppositions sensors such that they can't tell you from all the noise. Anybody got some reality based ideas on making that work? Or is it pretty much a no go? About the only one I recall being offered previously when this discussion comes up is to keep blowing up a bunch of nukes all over to temporarily blind sensors. Seems to me there might be an easier option, and one that doesn't involve blinding yourself in the process.

Maybe something like just bouncing some strong signals off other sky objects to radiate multiple sensor pings from many different points in the sky. That would only be of limited use though since you'd need those objects so it'd have to be in a crowded sky scenario. Like an asteroid belt or maybe a gas giant orbit with lots of moons and/or rings. Of course those are just the places that a pirate is gonna jump you so maybe it would work.

Or would it be just as effective to blast a single strong(er?) signal from your own location?

How much dazzle noise would it take to cause a processing overload on a little old model/1 and effectively blind that Merchie?

What would it take to produce said noise signals?

Opinions is all I'm asking, but hey if you can back it up with science all the better

 

[Anthony]

Well, the unit responsible for the dazzle is automatically detected, like any active jammer. Other than that, it depends on your sensor assumptions. With real-world sensors and SF-level energy weapons, blinding is trivial (in fact, it's so trivial that it makes spaceship combat unplayable).

 

[ElHombre]

The issue is how do you distinguish a ship from an asteroid at long distance?

The ship has an infrared signature from its power plant. The ship will also be changing its course.

Let's look at this from the point of view of a Merchant, who wants to spend as little as possible on his sensor array.

Except that's not the case. Even a small sailboat on a lake has to keep its eyes out for other traffic. Larger ships will have radars and lookouts.

What I mean are tactics like dazzle where rather than trying to not be noticed you go the other extreme and try to overload the oppositions sensors such that they can't tell you from all the noise.

Dazzling techniques wouldn't keep a ship from being noticed. Nukes would have to be launched from a missle, which itself would be noticed. The flash would also not last long in space; there's no atmosphere to keep the effects going.

The biggest problem for the ship trying to hide isn't getting pinged with active sensors, it's the fact that its crew doesn't respond well to having the heat shut off. To quote from 'Apollo 13': Power is Everything. Generating and using that power creates heat which is all-but-impossible to hide when you're in space.

What dazzlers could do however is affect the targets targeting ability. Most merchies aren't going to have much in the way of an ECCM suite. In fact, that could be used to the Authorities advantage ('And exactly what are you doing with a piece of advanced military hardware on your ship, Captain Solo?').

I could envision a scenario like this: A pirate hides on the ice-bound moon of a gas giant (this would have to be a pre-prepared position. maybe the pirates use a seeker to set up these places all over its operating territory. it would explain why seekers are found as pirate ships). As a merchant comes in for frontier refueling, the pirate emerges and uses its ECM gear to keep the merchant from hitting with its own weapons. The pirate has the High Guard position and the acceleration advantage combined with the fact that the 100D limit for a gasball is far larger than for a planet.

 

[atpollard]

Originally posted by Anthony:
</font><blockquote>quote:</font><hr />Originally posted by atpollard:

First. Do you know what the range is that the sunlight reflected off a 300 degree K scout ship averaged with the background will be magnitude 44 per arc second? (the scout ship is twice as bright as the average sky).

Twice as bright as the background is magnitude 20.5, not 44, and it depends on the color. It's also not necessary to be twice as bright as the background; equal will do, since ships don't block the background out. Assuming a fairly dark hull, and ignoring heat output, probably on the order of 20 light-seconds.</font>[/QUOTE]Thank you for the correction - my bad (I deal with light in foot-candles which are a linear scale and incorrectly assumed that magnitudes were also linear).

I am genuinely confused about something in your post. Given that each sensor "dot" reads the average magnitude of 1 square arc-second of space and the background is magnitude 22. If the sensor dot detects a ship and reads its magnitude as 22 ("equal will do, since ships don't block the background out.") How will my sensor know that there is a ship and not just a background. I am misunderstanding one of your basic assumptions. Please clarify this for me.

 

[TheEngineer]

Well, I guess the frequency distribution of the reflected stuff is fairly different from the background...
If the sensor could just get a full frequency range intensity, he would not be able to detect any differences.
Thing is to analyse the spectrum. The spectrum can tell a story

 

[atpollard]

Originally posted by atpollard:
If my ship is being shot at by another ship at a range of 100,000 kilometers, I may want to target the incoming missiles with my laser. I cannot abort a 10 minute scan because my ship performed evasive maneuvers and start a second scan with the sensors if I want to target the missile.

Originally posted by Anthony:
The example I gave involves a signal integration time of about ten seconds; it takes longer than that because each frame covers a fairly small part of the sky, but if you already know what part of the sky you care about (the one with the enemy ship in it), that's not a factor. You can fairly easily reduce the time to a second or less, at some loss in sensitivity.

I like your posts, this is an enjoyable discussion. One second or even 10 seconds sounds reasonable for most things (I would question the accuracy for targeting a missile with a laser – they are both small objects that might need greater precision to intersect). I was mostly trying to rule out any design that involved a 30 minute scan of the sky with a 10 minute exposure time as the basis for combat in Traveller.

Originally posted by atpollard:
Third. If we use hardware to detect the moving object by comparing two images taken 1/10 of a second apart (an impressive exposure time) and only report the changes to the ship computer (to filter out stationary background objects) – then how fast will a ship need to travel (perpendicular to the field of view) at a distance of 900,000 km (maximum tracking range) to travel 1 arcsecond and be detected as “moving” by this 500 gigapixel detector.

Originally posted by Anthony:
About 45 kps. However, you wouldn't do that. It's not like you can't just compare the images to ones you took ten seconds ago, even if you're taking multiple images per second.

The two ways to process this data are hardware or software. I proposed a hardware based system to shift the burden off the main computer. The advantages of hard wiring this at the sensor is increased speed (I can identify changes at the exposure speed of the scan) and faster scans require less compensation for movement by my ship. The disadvantage is the need for greater precision (I must be able detect only a fractional second of movement by an object).

The advantages of your software approach is lower precision (you can check for movement in minutes between exposures) and more complex analysis of the data is possible (apply filters and compare the data again). The disadvantages of the software solution is the need to store huge amounts of data (500 gigapixels per scan times the bit depth x the number of scans per second x how long you want to keep this data for) and process the analysis using software (comparing multiple 500 gigapixel images in 1 second requires Teraflop processing power or greater). The other disadvantage to software analysis is the greater correction needed to compensate for the motion of the scanning ship. Nothing in the background will remain fixed if the ship is moving and the amount of change will vary with the distance the object is from the ship). Compensating for motion could bump the required processing power from Teraflops to Pentaflops. All of this needs to be done with a Model 1 computer if a Free Trader is to be able to shoot at incoming missiles with his laser turret.

 

[sgbrown]

Originally posted by Anthony:
</font><blockquote>quote:</font><hr />Originally posted by SGB - Steve B:
The issue is how do you distinguish a ship from an asteroid at long distance?

Generally speaking, you distinguish ships from asteroids by the fact that they're in an interesting position, or headed towards somewhere interesting. As a defender, you can also distinguish them by the fact they aren't in your asteroid database.

There may be lots of asteroids out there, but the vast majority are nowhere near anything of interest. Objects become interesting when they're going to pass near to a target. It's perfectly realistic for a ship to hide out in the asteroid belt, but even at 6 Gs it's at least two days to get from the asteroid belt to anywhere interesting, which is typically plenty of time to react.

The issue is not whether stealth is possible in space at all. The issue is whether stealth is possible at ranges that are tactically relevant, which is generally ten light-seconds or so. </font>[/QUOTE]For one on one ship combat I concur that it would be difficult to sneak up on some one in space. But I don't concur with the idea that detection of someone actively hiding (as opposed to trying to get close) is a given.

If I'm trying to hide and maybe even observe and report, tactically relevant may be a lot larger than ten light-seconds. The reporting part may be dicey, but the hiding and observing part shouldn't be too difficult if I'm already in position.

It's not like finding a lit candle in a dark stadium. It's like finding a particular moving lit candle in a stadium filled with millions of lit candles, tens of thousands of which are moving (and actually only about half of these are in the main belt - and some of those in the belt move in and out of it from time to time.) Oh yeah, and there's a several of lanterns(planets)in the stadium with a big spot light shining down from a blimp on the center of the field. Picking the candle out gets a lot easier if the person moving the particular candle is doing things a lot different than the other candles or trying to get within a few feet of you - but if they're actively trying to blend in and know where you are, detection gets a lot harder.

That's why I tend to figure you can "hide" in space. Given just the right set of conditions - you might even be able to ambush somebody. Granted the ambush is a lot harder than just hiding and nearly, if not completely, impossible without special circumstances - but still possible under the right circumstances.

 

[atpollard]

Originally posted by far-trader:
A small tangent if you all don't mind, I was just curious what the opinions are of active stealth measures. What I mean are tactics like dazzle where rather than trying to not be noticed you go the other extreme and try to overload the oppositions sensors such that they can't tell you from all the noise. Anybody got some reality based ideas on making that work? Or is it pretty much a no go? About the only one I recall being offered previously when this discussion comes up is to keep blowing up a bunch of nukes all over to temporarily blind sensors. Seems to me there might be an easier option, and one that doesn't involve blinding yourself in the process.

Maybe something like just bouncing some strong signals off other sky objects to radiate multiple sensor pings from many different points in the sky. That would only be of limited use though since you'd need those objects so it'd have to be in a crowded sky scenario. Like an asteroid belt or maybe a gas giant orbit with lots of moons and/or rings. Of course those are just the places that a pirate is gonna jump you so maybe it would work.

Or would it be just as effective to blast a single strong(er?) signal from your own location?

How much dazzle noise would it take to cause a processing overload on a little old model/1 and effectively blind that Merchie?

What would it take to produce said noise signals?

Opinions is all I'm asking, but hey if you can back it up with science all the better

Underlying any answer is the even more basic Science vs Fiction in the particular Traveller Universe. The rules are general enough that almost anything could be possible or impossible in the area of stealth and ECM.

As an example, Photoelectric cells already convert Light into Electricity, so with a little wave of the hands, the Traveller Power Plant incorporates a “Photoelectric” shell that converts heat to electricity. Suddenly the heat signature of a ship is drastically reduced (in most Traveller rules, ships are not thermal propulsion rocket ships, so heat is not essential for maneuvering).

Example 2: Gravity bends space, and all reactionless Maneuver Drives bend space around the ship rendering it invisible to detection by EM detectors (light, heat, x-ray, etc.). Sensors detect gravity distortions and so detect ships only indirectly.

Since there is no way to quantify and analyze “incomprehensible future technologies”, no appeal to real science is possible.

 

[atpollard]

Originally posted by TheEngineer:
Well, I guess the frequency distribution of the reflected stuff is fairly different from the background...
If the sensor could just get a full frequency range intensity, he would not be able to detect any differences.
Thing is to analyse the spectrum. The spectrum can tell a story

That sounds reasonable, but our sensor has gotten much more complex than just a simple Radio Shack Photoelectric cell detecting a bright/hot ship against a cold/dark background.

 

[ElHombre]

Originally posted by atpollard:
That sounds reasonable, but our sensor has gotten much more complex than just a simple Radio Shack Photoelectric cell detecting a bright/hot ship against a cold/dark background.

But is it something a civvie ship would carry? I'd argue yes for the simple reason that many Free/Far Traders carry weaponry. It wouldn't do any good to have them and not have the sensors needed to find a target.

 

[TheEngineer]

Well, You could use a single cell but simply filter out the well known background frequencies ...
That should keep construction simple, too

But don't you think sensors are awfully cheap compared to the rest of a spaceship ?

Anyway, the answer to the question "How do ships see (in the TU)?" is just given in the rules.
At least MT describes types of sensors quite detailed, and theit usage and efficiency is covered in the space combat rules. Well, the rules are focussed on combat purpose and sadly more general sensoring is not really covered here. Grand Survey was a great help here ....

Now, a VERY crude (no concern of sensor types or targets cloaking activities) task for general scanning is given in the MT RefManual:

To scan the area for unidentified starships:
Routine, Sensor Ops, Edu, 10 sec (uncertain)

Regards,

TE

 

[mike wightman]

Originally posted by atpollard:

Example 2: Gravity bends space, and all reactionless Maneuver Drives bend space around the ship rendering it invisible to detection by EM detectors (light, heat, x-ray, etc.). Sensors detect gravity distortions and so detect ships only indirectly.

This would also shield the ship from weapons fire

 

[TheEngineer]

Ho, ho, if bending is done well, the ship could be effectively invisible

 

[Valarian]

Originally posted by TheEngineer:
Ho, ho, if bending is done well, the ship could be effectively invisible

Nah ... Star Trek III - the Klingon ship is detected due to the distortion caused by the cloaking device. This'd cause the same sort of effect.

You'd have to have a sensor tech who's alert and on the ball though.

 

[atpollard]

Originally posted by ElHombre:
</font><blockquote>quote:</font><hr />Originally posted by atpollard:
That sounds reasonable, but our sensor has gotten much more complex than just a simple Radio Shack Photoelectric cell detecting a bright/hot ship against a cold/dark background.

But is it something a civvie ship would carry? I'd argue yes for the simple reason that many Free/Far Traders carry weaponry. It wouldn't do any good to have them and not have the sensors needed to find a target. </font>[/QUOTE]Perhaps the "weapon grade" sensor is part of the Book 2 hard point cost, Book 2 turret cost, and 1 dTon of fire control. Without lasers, a ship could make do with a less precise targeting system (missiles and sand are proximity weapons, lasers require perfect accuracy). Without any weapons, a ship just needs to detect navigation hazards and other ships on an intercept/collision course.