Traveller Starship combat sucks!
Captain Mike Rogers
broadside of a scout ship when in the same hex? Can you say Imperial
Marksmanship Academy?
Starship weapons have too much range, and when 1 hex = 10,000 miles an
Traveller Imperial Scout ship could attack any planet from orbit with
their puny turret mounted weapon.
Starting to see a problem here?
The most common starship in the Spinward Marches, the 200-ton Beowulf
Class Free Trader, is only about 65-70 meters long, and at 1 hex (10,000
miles) away that makes it about the same size as a Honda Civic at 1 mile
away, think of how big that'd be at 3 hexes away?
Does anybody have a good, simplified system that can be used?
Mike Rogers
PS- Writing this post has ticked me off something fierce and I will
design a simplified starship combat system.
G. W. Roberts
Captain Mike Rogers <mtro...@airmail.net> wrote in article
<B662AE347EA3E1B1.AA2F58F3...@library-proxy.airnews.ne
t>...
Must be you, bud. I've played and enjoyed several times. I don't expect a
killing shot each and every time, and neither do the players I game with.
The only thing I can think of to make the system simpler would be to just
roll against a ship captain's skill and decide if the other guy dies then
and there.
Have fun.
Captain Mike Rogers
And I don't expect a killing shot everytime, only an idiot would, but
battles do not take place on battlefields that are 10,000 miles across.
Rolling a quick contest of captains skill is not what I was looking for,
but thanks for the
Mike idea.
Jack Tatum
Just how do you hit, when your chance to hit is less than zero?
Jack Tatum
Andrew Batishko
>Can someone tell me why the average starship gunner can't hit the
>broadside of a scout ship when in the same hex? Can you say Imperial
>Marksmanship Academy?
Hit probabilities seem a little low, but not unreasonable. Let's step
through things... (or skip it all for the solution presented at the end).
Start with a starship gunner. Assume that this is what he does for a
living. Figure that he probably has 8 points in the skill, a DX of 12, and
an IQ of 11 (he had to be at least a little smart to get through the academy
training). That gives a skill of 15 (counting the +1 bonus for the IQ).
Next double this value due to the accuracy bonus. This gives a 30.
Next comes the RoF bonus of +4 for standard Traveller weapons, for a 34.
Next is the targeting program. The rules suggest giving your ship the best
targeting program it can handle. For a TL 12 command bridge (which is what
most military ships will be), this is would be about a complexity of 9
(since the ship only has 3 complexity 10 computers available to run
software). This brings the skill level up to 43.
Next add in the target size, which would be +8 for a Beowulf class. Brings
us to 51.
Next add +2, because of course we've got the active sensors locked onto
it... 53.
Now, you take a look at the ranges... You have an effective skill of 10 at
2 hexes (-43 to skill). Even a skill of 8 (which isn't too horrid) out to 4
hexes.
Now, I will admit, drop the targeting program down to a 7, since you are in
a Beowulf yourself, and it only has 3 complexity 7 computers, but less
weaponry, so you can afford to run a higher complexity targeting system.
But on the other hand, your average gunner character may have a DX of 13
instead of 12, so your skill is up to 16 and don't forget the extra accuracy
bonus since it effectively doubles your skill, so you are brought back up to
exactly the same skill level.
Like I said, I'd prefer ranges of perhaps 8-10 hexes being reasonable, but
there IS a solution. Merely declare that targeting software is particularly
advanced. Rather than added a value equal to complexity to the skill level,
just add complexity plus some amount (such as 4 or 5). This will give you a
pretty good extension on the range.
Andrew
Tomansky
> Hmm... let me do a little reality check here. A 70 meter long craft at
> 10,000 miles (ooh, bad boy, mixing units!) covers an arc of roughly
>
> (70 meters * 3 feet/meter) / (10,000 miles * 5280 feet/mile)
>
> 4e-6 radians
>
> 2e-4 degrees
>
> Hey, Iceman, what was the accuracy of a good sniper rifle again?
> Something like a 3" group at 1000 yards? That would be an arc of 8e-5
> radians, an order of magnitude better.
>
> Maybe that has something to do with it.
Contemporary radar can target a half meter object at over 30,000
km. Modern US gunnery computers can consistently hit a three meter
moving target over five km away. Advanced tech. would considerably
improve things, although stealth is bound to brow also.
David Levi
Scott M
modified to GURPS. I always thought High Guard was much more elegent than
either the CT or Mayday system. However, some don't like it because it is
more abstract. But hey, a lot of my characters don't want to figure out
vectors every turn.
This assumes you have damage/penetration tables from High Guard.
The advantage to this system is that I get to keep all my ship plans and
ship designs from Classic Traveller. I have too many ship plans to just
forget about them.
SPACECRAFT SKILLS
Some skills will be useful for spaceship operation and combat:
Gunnery/TL skill gives plus to hit in starship combat, as long as a Gunnery
Interact program is running. Gunnery-12 +1, Gunnery-13 +2, Gunnery-15
+3, Gunnery-16 +4, Gunnery-18 +5, Gunnery-19 +6, Gunnery-21 +7. This
skill covers all offensive weapons; there is no need to take a specialized
gunnery skill as indicated in GURPS (such as missiles or beam weapons).
Tactics(Space) helps with initiative rolls. Tactics-12 +1, Tactics-13 +2,
Tactics-15 +3, Tactics-16 +4, Tactics-18 +5, Tactics-19 +6, Tactics-21
+7.
Piloting (Large Spacecraft) is necessary to run a starship, and for various
rolls of maneuvering and getting out of tight spots. In addition, Pilot-15
gives the ship a +1 Agility, Pilot-18 gives at +2 Agility, and Pilot-21 +3
Agility.
Electronics Operations/TL (Sensors) is necessary to use ship’s sensors,
which are used to detect other ships and various other space features.
Without this skill, the ship’s computer will notify you only of emergency
situations (i.e. – there is a ship 500 meters away firing at us, or
collision warnings), provided your computer is working. Other specialties
which may be useful are (Communications), (Computers), (Security Systems).
Note that just plain Electronics lets you design and build electrical
equipment.
Mechanic/TL will allow you to fix problems aboard a starship. Useful in a
fight, where things always seem to get damaged. You must specialize in one
of three areas: (Jump drives), (Maneuver Drives), or (Power Plants). It
defaults to IQ-5, to any Engineer/TL specialty-4, or other Mechanic/TL
specialty-4.
Engineer/TL will allow you to design, build, identify, or fix complex
machinery on spacecraft. . You must specialize in one of three areas:
(Jump drives), (Maneuver Drives), or (Power Plants). It defaults to
appropriate Mechanic specialty-6, or , or other Engineer/TL specialty-4.
This skill is harder to learn than Mechanic, but is more wide-ranging.
Astrogation is necessary to create flight plans, and for making sure you
don’t misjump too often.
SPACE COMBAT
Batteries: Turret Weapon batteries are ranked from USP Code 1 (the least
effective) to Code 9 (most effective). For ships 1000 tons and under, mixed
turrets (weapons of different types in a turret) are allowed. In these
cases, each weapon is a battery. Ships with more than one weapon mount of a
type may group them into batteries. Batteries are determined when ship
built, not spur of the moment. See weapons table in High Guard to see how
many weapons are required for each code ranking.
First, determine each ship’s:
-- Agility, which is agility of ship plus pilot modifiers. Pilot-15 gives
the ship a +1 Agility, Pilot-18 gives a +2 Agility, and Pilot-21 a +3
Agility.
-- Emergency Agility, which is equal to maneuver drive or power plant
number, whichever is less, plus the pilot modifier above.
Other information needed will be on each ship’s Universal Ship Profile.
Steps in Combat
1) Battle Formation Both players form their ships into two lines each: line
of battle and reserve line. Ships in line of battle may fire and be fired
upon. Ships in reserve cannot fire or be fired upon unless line of battle
is broken.
Launch and recover ships. Launch facility may launch or recover one
ship/turn, launch tube 40/turn. Ship with dispersed structure all per
turn. Put ships into line of battle or reserve line.
2) Initiative Roll 2d6 each for initiative, with modifiers:
Faster fleet (one whose least agile ship has highest agility) +1
Most ships in line of battle (capable of fire and maneuver) +1
Tactics (Space) skill:
Tactics-12 +1
Tactics-13 +2
Tactics-15 +3
Tactics-16 +4
Tactics-18 +5
Tactics-19 +6
Tactics-21 +7
3) Determine Range Two ranges: long and short. Range always long on first
round. Subsequent rounds player who wins initiative determines range.
(Lasers, particle accelerators, meson guns better at short range. Missiles
most effective at long range. Plasma and Fusion guns only operate at short
range.)
4) Pre-Combat Decision Step Each player determines:
--Break off or not. To break, any ship may either jump or accelerate.
To jump, expend energy points (of power plant) equal to two turns output of
jump drive. For example, if a ship with Power Plant 8 (which produces 8
energy points) attempts jump-5, it takes two turns; if it attempts jump-4
(or less) it takes one turn. Can’t use energy for the jump for other
purposes.
To accelerate, need to start at long range. Ship(s) automatically escapes
if not pursued. Else, see pursuit section below.
-- Use emergency agility or not. If so, ship may not fire any energy
consuming weapons (all but missiles and sandcasters), and ship agility
becomes equal to maneuver drive or power plant number, whichever is less.
Can still add pilot’s modifier, as above. Can still use computers and
screens.
-- Determine if black globe is on, and flicker rate. Each 10% flicker adds
two levels armor. This armor also effects meson guns. Agility -10% for
every 10% flicker rate. See High Guard for capacitor rules if using black
globes.
5) Combat Arrange battle lines by size. Player with initiative determines
goes first, by putting out largest ship as a target. Other player may fire
with any batteries of any ship. Then other player puts forth largest ship.
Continue alternating with all other ships. For each ship:
-- Indicate which batteries fire. Indicate if missiles are High Explosive
(HE) or Nuclear.
-- Determine which batteries hit. Role 2d6. Modifiers to hit role:
+/- target size:
under 100 tons (size code 0) -2
100 to 1000 tons (size code 1 to A) -1
2000 to 10,000 tons (size code B to K) no modifier
20,000 to 50,000 tons (size code L to P) +1
75,000 tons and over (size code Q+) +2
+ computer size (bis counts as one higher)
+ Pilot skill:
Pilot-15 +1
Pilot-18 +2
Pilot-21 +3
+ gunnery (if have gunnery interact running):
Gunnery-12 +1
Gunnery-13 +2
Gunnery-15 +3
Gunnery-16 +4
Gunnery-18 +5
Gunnery-19 +6
Gunnery-21 +7
lasers at long range -1
missiles at short range -1
meson guns at short range +2
if meson guns -target agility rating
(can’t fire Plasma and Fusion guns at long range)
One gunner may operate one turret or bay / weapons system.
-- Determine penetration. Roll 2d6. Modifiers:
+ computer size
plasma and fusion guns +2
(HE missiles ignore dampers)
-- Determine damage (but not applied yet). Roll 2d6 on each applicable
damage table. Modifiers to roll:
+ armor of defending ship on surface explosion and all but meson guns on
radiation damage table
if nuclear missile, -6 on surface explosion table
if pulse laser, -2
All spinal mount weapons that hit past size 9 gets one extra damage roll on
each appropriate table. The number of extra rolls for spinal particle
accelerators is reduced by one for each factor of armor on defending ship,
but not less than one.
All batteries whose weapon code exceeds size code of target ship will
inflict (if they hit and penetrate) automatic critical hits equal to size
difference. These are reduced by one for each two factors, rounding down.
Meson guns ignore armor, except for black globes.
6) Damage Damage inflicted above takes effect. Each critical hit reduces
target armor by one. If “Crew-1” is obtained, and ship is at USP level 1
already (1 to 9 people), then one person is out of action on the player’s
ship(s).
7) Breakthrough If all of one player’s ships in line of battle cannot fire
offensive weapons, line is broke. Other player can then fire their line of
battle ships at other’s reserve line. The reserve line ships cannot fire
back, but can fire defensively. In next turn, do new line of battle.
8) Pursuit May accelerate alone or in groups. If in group, agility is that
of slowest ship. Modifiers to agility:
If broke off from reserve, agility +2
(remember if emergency agility has been used)
Any enemy ship can pursue if agility is at least equal to group breaking
off.
9) Terminal step
--Can revive frozen watch to restore one crew factor.
--Damage control. Can roll on Mechanic or Engineer or Electronics to repair
to effect of one hit. Can’t repair same system more than once in a turn.
If all offensive weapons, black globe, and maneuver drive disabled, can then
board.
Gregory Loren Hansen
Captain Mike Rogers <mtro...@airmail.net> wrote:
>Can someone tell me why the average starship gunner can't hit the
>broadside of a scout ship when in the same hex? Can you say Imperial
>Marksmanship Academy?
>
>Traveller Imperial Scout ship could attack any planet from orbit with
>their puny turret mounted weapon.
>
>Starting to see a problem here?
>
>The most common starship in the Spinward Marches, the 200-ton Beowulf
>Class Free Trader, is only about 65-70 meters long, and at 1 hex (10,000
>miles) away that makes it about the same size as a Honda Civic at 1 mile
>away, think of how big that'd be at 3 hexes away?
Hmm... let me do a little reality check here. A 70 meter long craft at
10,000 miles (ooh, bad boy, mixing units!) covers an arc of roughly
(70 meters * 3 feet/meter) / (10,000 miles * 5280 feet/mile)
4e-6 radians
2e-4 degrees
Hey, Iceman, what was the accuracy of a good sniper rifle again?
Something like a 3" group at 1000 yards? That would be an arc of 8e-5
radians, an order of magnitude better.
Maybe that has something to do with it.
>Does anybody have a good, simplified system that can be used?
>
>Mike Rogers
>
>PS- Writing this post has ticked me off something fierce and I will
>design a simplified starship combat system.
You might want to look at Cyberpunks near-Earth orbit supplement. I
forget exactly how the rules go, but you have a good chance of hitting the
target unless the target is actively defending itself. The pilot presses
the fire button, but aiming is entirely out of his hands, so I think pilot
skill only comes into initiative, not aim. And they assume that any
weapons a ship will use is vastly overpowered, so any target is destroyed.
That means smaller ships are automatically destroyed, and a targeted
section of something bigger, like a space station, is automatically
destroyed. And that makes perfect sense to me. A railgun might be able
to put a stream of bullets through 2" thick steel, but the target sure
won't be protected by 2" worth of steel or else it could never get off the
ground or move. And historically, how many Sidewinders does it take to
knock down a typical fighter plane?
--
"Besides, it doesn't take much creativity or courage to figure out that
something which reads 'Danger: Flammable' on the label might be fun to
fool about with." -- Joris van Dorp
RPGMeister
In article
<D06DA375F998F26F.1343AB92...@library-proxy.airnews.net>,
Jack Tatum <Jack_...@panthercity.net> writes:
>
>
>Just how do you hit, when your chance to hit is less than zero?
"Use the Force, Luke..."
I tried to resist, but I was weak...
>
>Jack Tatum
>
Dylan Knight
Every time I try to sell out, no one is buying...
HAL
Skill 14 gunner +14
Accuracy bonus +14
RoF bonus + 4 (I have designed point defense lasers with +9)
Active Sensor lock + 2
Target computer + 9 (TL12 computer max is +12)
Size bonus + 8 minimum, +9 or +10 more likely.
Totals: 51
Minus 43 for range of 3 hexes, and you now have an 8 or less on 3d6.
Range 1 is a 10 or less. Range zero is a 12 or less.
With a point defense laser granting a rate of fire of 1 shot every other
minute - and having two such lasers - the to hit at range 1 is now +11
instead of +4. Granted, the damage is not hot, but then again, you are
likely to get more hits per round. Has anyone noticed that with the
exception of the Rampart and the Dradgon SDB, that even low level damage
such as 96d6 will average 286 points of damage per hit? By the by, I
didn't chose 96d6 at random. That would be the stats of a 14.93-Mj point
defense laser.
In short gentlemen, I don't think that firing upon a ship at out to 3
hexes is a terrible problem. With a skill of 16, the previous example
goes from a base 51 to a base 55. Increasing the Targeting computer from
a cheap TL12 Complexity 8 targeting computer to a roughly 3 Mcr Targeting
complexity 11 computer is not a major stretch. Then the odds go from a
base 55 to a base 58. Slave enough lasers together, and the odds go up
even more due to the RoF bonus.
Hal
PS Check out http://www.buffnet.net/~hal
In there, you will find revised Hull charts for different types of
streamlined hulls along with some other modules plus... (best part of all
<grin>) stats on lasers for TL12 of different types. In addition, if you
wait a week, I will also have new stats on Meson Spinal mounts that
simulate the old HIGH GUARD stats on 1,000 ton, 2,000 ton, 5,000 ton, and
8,000 ton Meson guns based upon TL's 9,10, 11, and 12 GURPS VEHICLES stats
(yes, I know it states that Meson Guns are TL10 devices, but Canon
TRAVELLER states that Meson guns show up at what amounts to GURPS TL9 for
GURPS TRAVELLER).
HAL
only one thing - act as targeting computers. They are CHEAP!!!! There is
no reason why you can't get one for every weapon you get at really cheap
prices. The only thing you really pay for is the cost of the program
itself. Overall, the best targeting computer is a Microframe dedicated
computer with the Genius option. At TL12, this will cost $35,000
unhardened, or $175,000 hardened. It will be a complexity 9 computer able
to run Targeting 9 (+10 bonus). If you go one level cheaper, you can get
a dedicated hardened targeting computer for only $26,250 that will run a
computer program granting you a +9 bonus to hit. Oh, almost forgot:
The cost of a program is equal to $1,000 x 2^complexity. Thus, a
complexity 10 program will run you 1.024 Mcr. A complexity 8 program will
run you an additional .256 Mcr.
: Now, I will admit, drop the targeting program down to a 7, since you are in
JDP
impossible to hit anything. However, GT assumes that you'll have a
Targeting program of at lest 7+
If all else fails look for the "Role Playing Space Combat System." It
assumes you're using ships from T4 but can easily be modified (it's
available on the Web, but I'm not sure where right now.)
On Sun, 06 Dec 1998 17:04:41 -0600, Captain Mike Rogers
<mtro...@airmail.net> wrote:
>Can someone tell me why the average starship gunner can't hit the
>broadside of a scout ship when in the same hex? Can you say Imperial
>Marksmanship Academy?
>
>Starship weapons have too much range, and when 1 hex = 10,000 miles an
>Traveller Imperial Scout ship could attack any planet from orbit with
>their puny turret mounted weapon.
>
>Starting to see a problem here?
>
>The most common starship in the Spinward Marches, the 200-ton Beowulf
>Class Free Trader, is only about 65-70 meters long, and at 1 hex (10,000
>miles) away that makes it about the same size as a Honda Civic at 1 mile
>away, think of how big that'd be at 3 hexes away?
>
Peter Janik
pretty nice recap of High Guard but there is a small omission in it (
either in
High Guard or your description )
Scott M wrote:
> Snip <
> Mechanic/TL will allow you to fix problems aboard a starship. Useful in a
> fight, where things always seem to get damaged. You must specialize in one
> of three areas: (Jump drives), (Maneuver Drives), or (Power Plants). It
> defaults to IQ-5, to any Engineer/TL specialty-4, or other Mechanic/TL
> specialty-4.
Hmm... How about Mechanic/TL for the Computer? Targeting Systems?
And THE most important LIFE SUPPORT?
> Snip <
>
> 9) Terminal step
> --Can revive frozen watch to restore one crew factor.
> --Damage control. Can roll on Mechanic or Engineer or Electronics to repair
> to effect of one hit. Can’t repair same system more than once in a turn.
> If all offensive weapons, black globe, and maneuver drive disabled, can then
> board.
Kinda useless to repair the ship if you die of asphyxiation or
atmosphere poisoning.
You might have some time but you still need the skills don´t you?
Just my 2 cents worth
Peter Janik
Pauli Hakala
On 7 Dec 1998, RPGMeister wrote:
> In article
> <D06DA375F998F26F.1343AB92...@library-proxy.airnews.net>,
> Jack Tatum <Jack_...@panthercity.net> writes:
>
> >
> >
> >Just how do you hit, when your chance to hit is less than zero?
>
> "Use the Force, Luke..."
>
> I tried to resist, but I was weak...
Lets see how GURPS Handles it;
Gunner Skill: 14
Weapon Accuracy: 30 (Or something like this, don't have the books with me
right now. This all is from memory.)
Firing Distance+Relative Velocity: 16,000,000 yards, -35.
Target Size: 70 yards, +9.
Targeting Computer Bonus: +6
Active Sensor Lock Bonus: +2
Now, with the *10 minute* aim, one can ignore skill restrictions on
accuracy.. So the chance to hit will be: 14+30-35+9+6+2 = *26*.
That looks like a damn good chance to hit for me, at least.
Of course, if one is impatient enought to fire off after only 4 seconds of
aim, chances to hit drop somewhat;
14+14+3-35+9+6+2 = *13*
Not exactly a zero chance to hit either.
(In GURPS Traveller, Luke will use targeting computers and sensors, not to
mention patience in taking aim.)
-Pauli
Gregory Loren Hansen
I can't argue with your analysis or understanding of the game rule. But I
really, really don't think operator skill should be involved in the hit
probability. I know it's more dramatic and everything, but the computer
is going to put up the targets on a display with a little box around each
one. The operator will tell the computer which one to shoot at, maybe by
entering a number at a keyboard. When you're talking about distances of
tens of thousands of miles, a human operator wouldn't even be able to see
the target, much less hit it! Maybe he'd see a shiny dot, like a star.
For that matter, the targeting computer shouldn't make that much of a
difference, either. Why? Whatever you get is going to have way
more processing power than you actually need, especially with future
computing technology, but it's true even today. And it's going to be a
dinky part of the total cost.
To put it a little bit in today's terms, and pulling some numbers out of
my butt, say you spend $100,000 on a 5-inch gun for your fishing boat,
another $20,000 for servos and other gear to automate the operation of
your gun, $20,000 on a full load of ammunition, and a good $500,000 on
radar. So you've sunk maybe $700,000 so far into a system that can shoot
down an airplane that you can't see with the naked eye. Now it's time to
buy a computer. Do you choose an Alpha-based system for $10,000 (1.4% of
system cost)? Or do you try to save some money by getting a PII system
for $2000 (0.28% of system cost)? It doesn't matter, they both have a
buttload more processing power than you need. The Phalanx has been
shooting at fragments of destroyed missiles long before Pentium-class
processors were available in the consumer market. You want the most
reliable computer. Choose the one that runs a real-time mission-critical
operating system -- if you're running NT you might wind up dead in the
water after a divide-by-zero error (e.g. the USS Yorktown). Whatever
computer you get, the cost is too dinky compared to the hardware to make
a difference.
If you want to improve your accuracy, you don't get a better computer.
It's not that easy! You get a better mount and servos for the gun, with
better stabilization and vibration dampening, and more repeatable
positioning. You get a better gun, maybe one with bigger bullets, or
improved ammunition (match-grade 5-inch shells?). You get a better radar
with higher resolving power. You get a bigger boat that doesn't toss
around in the water so much. Or you go to guided missiles. Unless you're
using a Commodore 64, and I know you're not, you could try replacing just
about anything *but* the computer to get better accuracy. And it's going
to cost a lot more than a new computer would, and it could turn into a
major refitting job.
And from what I've seen, that's just the way it goes. There was only so
much that computers could do to clean up the space telescope's images, it
really needed new optics. There's only so much that post-processing can
do to correct for wall effects and other spurious results from an ion
chamber, it's better to improve the design of the chamber and get cleaner
raw data. And there's only so much a computer can do with a radar image
and a gun that's vibrating around, it really needs a better mount or
better sensors.
qbra...@csc.uvic.ca
Gregory Loren Hansen <glha...@copper.ucs.indiana.edu> wrote:
[snip...]
>
>For that matter, the targeting computer shouldn't make that much of a
>difference, either. Why? Whatever you get is going to have way
>more processing power than you actually need, especially with future
>computing technology, but it's true even today. And it's going to be a
>dinky part of the total cost.
>
[snip...]
Maybe, but consider this. Shooting at a missile fragment isn't that hard,
because you have a nearly 100% accurate model of its future movement.
It has its existing velocity and gravity will pull it down, air resistance,
etc..
Now you are shooting at an enemy ship in space. They know you are there
and you know they are there. Their piloting program is adding little
random movements, or big ones, so that by the time your weapon reaches
them they are moving somewhere else. Now your targetting computer
is tracking their past movements and building a model to predict their
future positions. Perhaps it is even building a 3 dimensional space
which probability densities for all possible positions based on knowledge
of the capabilities of the ship, or the past behavior of the ship. The
piloting program knows that the targetting computer is building this model
so it switches between different evasion algorithms in some kind of way
that it hopes the targetting program will have a problem predicting.
Anyway, I have complete confidence that targetting computers make
perfect sense. Having said that, I'd give anything other than a
critical failure a hit on a target which is not evading, but just
moving in a straight line with constant acceleration, as long as you
bother to use your targetting computer at all.
I don't know what this says about human gunner's skill.
--
Quetzalcoatl Bradley
qbra...@csc.uvic.ca
Gregory Loren Hansen
<qbra...@csc.UVic.CA> wrote:
>In article <74jdis$8k1$1...@flotsam.uits.indiana.edu>,
>Gregory Loren Hansen <glha...@copper.ucs.indiana.edu> wrote:
>
>[snip...]
>
>>
>>difference, either. Why? Whatever you get is going to have way
>>more processing power than you actually need, especially with future
>>computing technology, but it's true even today. And it's going to be a
>>dinky part of the total cost.
>>
>
>
>Maybe, but consider this. Shooting at a missile fragment isn't that hard,
>because you have a nearly 100% accurate model of its future movement.
>It has its existing velocity and gravity will pull it down, air resistance,
>etc..
>
>Now you are shooting at an enemy ship in space. They know you are there
>and you know they are there. Their piloting program is adding little
>random movements, or big ones, so that by the time your weapon reaches
>them they are moving somewhere else. Now your targetting computer
>is tracking their past movements and building a model to predict their
>future positions. Perhaps it is even building a 3 dimensional space
>which probability densities for all possible positions based on knowledge
>of the capabilities of the ship, or the past behavior of the ship. The
>piloting program knows that the targetting computer is building this model
>so it switches between different evasion algorithms in some kind of way
>that it hopes the targetting program will have a problem predicting.
>
>Anyway, I have complete confidence that targetting computers make
>perfect sense. Having said that, I'd give anything other than a
>critical failure a hit on a target which is not evading, but just
>moving in a straight line with constant acceleration, as long as you
>bother to use your targetting computer at all.
It depends on what you're shooting. If you're shooting bullets that might
only be going 20 miles per second, it would take eight minutes to hit a
target 10,000 miles away. Don't bother, nothing you can do will hit the
target if it can see the bullets coming and dodge. If you're shooting
lasers or particle beams, the flight time is essentially zero (.06
seconds, the acceleration required to move the ship away from a known shot
becomes hundreds to thousands of Gs). The problem reduces from predicting
target behavior to making small corrections to account for essentially
random movement.
Besides, if computers continue to double in power every eighteen months,
by that time a computer that can do what you described will be available
at grocery store checkouts for $15. And that's mostly for the plastic and
the keyboard and display, not the little chip inside.
If it can be done without complicating the game, I'm all for tying sensor
quality and mounts into accuracy, and selling weapons and mounts
seperately. It's not hard to imagine both of those costing more than the
gun itself. And I'd just say the targeting computer is an integral part
of the weapon mount, it and the software will always be adequate, and the
cost of both is rolled into the total price of the mount. For game
purposes there's not a lot of difference, except you'd read "mount"
instead of "computer", and installation would be harder.
Makes sense, doesn't it? Your car is controlled by a computer, but you
don't buy the computer seperately, and it's not a big part of the cost of
the car. The space shuttle has five computers on board, but they're not
very expensive or interesting compared to the rest of the hardware.
Twenty years ago, a computer was a big deal. Nowadays they're cheap and
small and capable enough that they're built into consumer products without
a second's thought, and taken for granted. Twenty years ago, a Game Boy
would have filled a room. Okay, I'm making that up. But it is true that
a six year old Sun computer has the power of a Cray 1. And today you can
get something from Best Buy for $900 that will out-perform that Sun. At a
higher tech level it will be even more true.
Tomansky
> Maybe, but consider this. Shooting at a missile fragment isn't that hard,
> because you have a nearly 100% accurate model of its future movement.
> It has its existing velocity and gravity will pull it down, air resistance,
> etc..
>
> Now you are shooting at an enemy ship in space. They know you are there
> and you know they are there. Their piloting program is adding little
> random movements, or big ones, so that by the time your weapon reaches
> them they are moving somewhere else. Now your targetting computer
> is tracking their past movements and building a model to predict their
> future positions. Perhaps it is even building a 3 dimensional space
> which probability densities for all possible positions based on knowledge
> of the capabilities of the ship, or the past behavior of the ship. The
> piloting program knows that the targetting computer is building this model
> so it switches between different evasion algorithms in some kind of way
> that it hopes the targetting program will have a problem predicting.
>
> Anyway, I have complete confidence that targetting computers make
> perfect sense. Having said that, I'd give anything other than a
> critical failure a hit on a target which is not evading, but just
> moving in a straight line with constant acceleration, as long as you
> bother to use your targetting computer at all.
That's why smart captains use trans warp battle maneuvers. (Not
that drivel in Star Trek.) Moving in and out of a warp field as
rapidly as possible for combat. Less time in "normal" space is less
time to get hurt.
David Levi
David P. Summers
glha...@copper.ucs.indiana.edu (Gregory Loren Hansen) wrote:
> I can't argue with your analysis or understanding of the game rule. But I
> really, really don't think operator skill should be involved in the hit
> probability. I know it's more dramatic and everything, but the computer
> is going to put up the targets on a display with a little box around each
> one. The operator will tell the computer which one to shoot at, maybe by
> entering a number at a keyboard. When you're talking about distances of
> tens of thousands of miles, a human operator wouldn't even be able to see
> the target, much less hit it! Maybe he'd see a shiny dot, like a star.
Well, I think character involvement is important for roleplaying
purposes. My vision is that the character spends his time refining
the accuracy of the computer lock (like a sonar operator in a submarine).
However, if you don't agree you can simply assign the computer that locks
on, fires, etc. a skill and use that.
> For that matter, the targeting computer shouldn't make that much of a
> difference, either. Why? Whatever you get is going to have way
> more processing power than you actually need, especially with future
> computing technology, but it's true even today. And it's going to be a
> dinky part of the total cost.
A lot of what the targetting computer does is look at the movement and
make predictions about where the ship will be when the shot hits a
fraction of a second later.
thew...@earthlink.net
>That's why smart captains use trans warp battle maneuvers. (Not
>that drivel in Star Trek.) Moving in and out of a warp field as
>rapidly as possible for combat. Less time in "normal" space is less
>time to get hurt.
>
>David Levi
Um, David, have another look at the subject line and give that one
another go, yeh?
David P. Summers
> hard for the machines of a technologically more advanced civilization.
Except that the foe is presumably making evasive manuevers.
Gregory Loren Hansen
David P. Summers <summers@remove_alum.mit.edu> wrote:
>In article <74jdis$8k1$1...@flotsam.uits.indiana.edu>,
>> really, really don't think operator skill should be involved in the hit
>> probability. I know it's more dramatic and everything, but the computer
>> is going to put up the targets on a display with a little box around each
>> one. The operator will tell the computer which one to shoot at, maybe by
>> entering a number at a keyboard. When you're talking about distances of
>> tens of thousands of miles, a human operator wouldn't even be able to see
>> the target, much less hit it! Maybe he'd see a shiny dot, like a star.
>
>purposes. My vision is that the character spends his time refining
>the accuracy of the computer lock (like a sonar operator in a submarine).
>
>However, if you don't agree you can simply assign the computer that locks
>on, fires, etc. a skill and use that.
I've come to think that spaceship combat is fun and exciting when you see
it on TV, but it's not a lot of fun to role-play. Role-playing is fun
when you have options to exercise, when your character's personality makes
a difference in the game. The person who sends the ship into a danger
zone, who decides to engage an enemy, he's probably having fun. But the
player on the weapons and the player on the sensors and the player in the
engineering room are probably getting distracted. That's why all the main
characters on Star Trek were on the away teams.
>> For that matter, the targeting computer shouldn't make that much of a
>> difference, either. Why? Whatever you get is going to have way
>> more processing power than you actually need, especially with future
>> computing technology, but it's true even today. And it's going to be a
>> dinky part of the total cost.
>
>A lot of what the targetting computer does is look at the movement and
>make predictions about where the ship will be when the shot hits a
>fraction of a second later.
It's not that hard. It's not hard for today's machines, and it won't be
hard for the machines of a technologically more advanced civilization.
Gregory Loren Hansen
David P. Summers <summers@remove_alum.mit.edu> wrote:
>glha...@copper.ucs.indiana.edu (Gregory Loren Hansen) wrote:
>> >A lot of what the targetting computer does is look at the movement and
>> >make predictions about where the ship will be when the shot hits a
>> >fraction of a second later.
>
>> It's not that hard. It's not hard for today's machines, and it won't be
>> hard for the machines of a technologically more advanced civilization.
>
>
Say you're aiming for the center of a 70 meter long ship that's 50,000,000
meters away, with a laser or particle beam. The target has 0.17 seconds
to pull away. Assuming you had a naive firing program that doesn't try to
predict anything, the target must accelerate (or change acceleration) by
a = 2d/t^2
= 2*(35m)/(.17s)^2
= 2422 m/s^2
= 242 g
at the moment the shot is fired in order to avoid the shot. It will pulp
the crew. Having said that, I'm pretty sure the computers of today have
the needed calculational power, and for a lot less than $120,000. But the
computers of two hundred years from now? If Moore's Law holds, they'll
double in power more than a hundred times by then, making them about 10^47
times more powerful than they are today. I don't expect Moore's Law to
hold that well over that much time, but it gives me some weight to swing
around. I think we can just assume any computer that comes with the
system will be up to the task, and the main cost of the computer is in the
hardening and shielding, not the computing power.
We do, after all, have radar-guided, computer-controlled anti-aircraft
guns today that shoot at evading targets. And they work.
Tomansky
I did indeed read the subject as well as all posts to the thread.
If you think the subject line and the thread have deviated, well . .
. welcome to usenet. OTOH, if you dislike my approach, don't use
it.
David Levi
Steve McDonald
>David P. Summers <summers@remove_alum.mit.edu> wrote:
>Say you're aiming for the center of a 70 meter long ship that's 50,000,000
>meters away, with a laser or particle beam. The target has 0.17 seconds
>to pull away.
You have made a small error here. You are correct that the laser beam
or (presumably nearly light-speed) particle beam would take 0.17 seconds
to reach the target. However, you have neglected an important point.
You do not have FTL sensors--at any given time you only know where the
target was 0.17 seconds *ago*. This means that the target actually has
at least 0.34 seconds to dodge between the time you last detected it
and the time that your beam reaches it. This delay could actually be
larger if you factor in things like the amount of time to process the
sensor data, the time to make a targeting prediction, and the time to
actually fire the weapon. Granted this will be fairly small but could
add up, particularly the firing time since there are mechanical operations
involved as well as electronic. Laser beams and particle beams do not
fire instantaneously.
This does not refute your argument, since the acceleration necessary
for the target to successfully dodge is still high, but not nearly as
high as you had calculated.
-----steve m
mcdo...@mit.edu
Andrew Batishko
>You have made a small error here. You are correct that the laser beam
>or (presumably nearly light-speed) particle beam would take 0.17 seconds
>to reach the target. However, you have neglected an important point.
>You do not have FTL sensors--at any given time you only know where the
>target was 0.17 seconds *ago*. This means that the target actually has
>at least 0.34 seconds to dodge between the time you last detected it
>and the time that your beam reaches it. This delay could actually be
>larger if you factor in things like the amount of time to process the
>sensor data, the time to make a targeting prediction, and the time to
>actually fire the weapon. Granted this will be fairly small but could
>add up, particularly the firing time since there are mechanical operations
>involved as well as electronic. Laser beams and particle beams do not
>fire instantaneously.
Actually, if you reverse the calculation to determine the amount of time
required in order to make a dodge possible at 3 Gs, you discover the 1.54
seconds is needed. Subtract your 0.34 seconds of "travel" time and you're
left with 1.2 seconds for processing and firing orders. Seems completely
sufficient to me. ...unfortunately so, as I rather like the way the system
is right now. Makes it more interesting for the characters to have things
be more "personal" (eg having your skill make a difference in being able to
hit). But then I guess an increased lack of "personal" combat is
characteristic of higher tech combat anyway (clubs -> bows -> guns ->
missiles -> complete computer control). Just doesn't make for as
interesting an RPG, that's all...
Andrew
pulver
On 8 Dec 1998, Gregory Loren Hansen wrote:
> If it can be done without complicating the game, I'm all for tying sensor
> quality and mounts into accuracy, and selling weapons and mounts
> seperately. It's not hard to imagine both of those costing more than the
> gun itself. And I'd just say the targeting computer is an integral part
It's a reasonable argument, but it has no real bearing on Traveller *per
se*, because Traveller is, basically, a space opera, not a hard sf game.
(It's 5,000+ years in the future: does anyone actually believe that the
way developments in biotechnology, computing, etc. are projected for the
next *century* that we'll have a society that looks so much like our own?)
More seriously, the Trav combat system was designed to replicate the feel
of Classic Traveller. That is, long distance engagements, slow 10-20
minute turns, 1-6 G accelerations, missile and beam weapons both being
used, vector movement -- and character skills and computer programs being
both equally important. SOme differences pop in (the current system is
somewhat based on the classic trav missile supplement, with more punch
given to kinetic kill missiles, but blame this on the missing missile
rules in the Traveller book...).
Sure, it isn't realistic. It can't really *be* realistic and still capture
the feel of Traveller. In fact, the current version actually ups the
potency of character skill levels *deliberately* to capture the primitive
"man in the loop" feel of Traveller combat. This was a deliberate design
decision made involving myself and Loren. Compare to the space combat
system in Vehicles, where you add the weapon's full Acc bonus.
The rules do assume you are firing at maneuvering targets. If they aren't
maneuvering, assume you hit automatically if you've detected them. I'd
also say that the system is a direct extroplation of the GURPS combat
system.
For a modified version, ignore Acc, Targeting programs and skill. Roll
against 55 - range modifier + size modifier. An attack can only take place
if you have a sensor detection. Ignore Dodges but apply an additional -1
per G of acceleration of a maneuvering target. Add the +10 for point
defense fire. Add a +10 bonus if the target *isn't* maneuvering
defensively -- this includes targets that haven't detected any opponents.
If I was writing a *serious* space combat system:
* the game would be split into strategic turns for long-distance
maneuvering at the edges of sensor range, with much easier long range
detection, and tactical turns lasting only 1-3 seconds each for whenever
a battle group got within combat range
* firing ranges would be a lot shorter (mostly taking place within a
thousand miles or less), due to limits on laser and beam weapon technology
* standard weapons would be more serious. At Traveller TLs, I'd want
anti-matter pumped gamma-ray laser weapons; if I was doing a more
realistic near-future game, lasers and particle beams and such.
* missiles would accelerate at 50 G or more for short periods and have
nuclear warheads; if the technology allowed, x-ray laser bomb-pumped
* the actual character interaction would involve steering during the
strategic turns, and picking from a list of offensive/defensive options
prior to a combat turn... crews would probably suit up into acceleration
tanks prior to battle for the occasional very high gee burst.
* things like "fighters" would be completely out the window.
* non-nuke missiles would probably still be remotely-controlled (but by
the ships AI, with human strategy where necessary) but just before they
entered target engagement range, they would seperate into dozens of small
submunitions for intercept, each with its own AI system aboard.
* ships would have rapid-pulse point defense lasers for intercepting the
above, all computer controlled, of course.
* meson guns would go out the window.
* so would the reactionless drives...
-David
pulver
On 9 Dec 1998, Gregory Loren Hansen wrote:
>
> Say you're aiming for the center of a 70 meter long ship that's 50,000,000
> meters away, with a laser or particle beam. The target has 0.17 seconds
> to pull away. Assuming you had a naive firing program that doesn't try to
> predict anything, the target must accelerate (or change acceleration) by
Does that assume a 70 meter sphere? What if the ship was needle shaped and
made a vertical acceleration?
> = 242 g
Of course, that would only bring it down to about 50 G with a 10:1 ratio
of length width thickness, but... Also, are you allowing for any nessary
time to spin the ship to bring turrets to bear, and these rather hefty
100+ metric ton turrets to move a few degrees to track targets? (Of
course, once you've got it pointed in the right general direction...)
> the needed calculational power, and for a lot less than $120,000. But the
> computers of two hundred years from now? If Moore's Law holds, they'll
I've read that Moore's Law should freeze by about 2025-2050 as quantum
limits are encountered. Quantum computers may be possible after that, but
seem to require very different engineering techniques and have certain
problems...
> We do, after all, have radar-guided, computer-controlled anti-aircraft
> guns today that shoot at evading targets. And they work.
Sergeant York, anyone?
(Of course, tracking a target with ground clutter is harder, but it was a
fun cheap shot).
More seriously, most radar-controlled anti-aircraft guns that I've read
about track the bullet or cannon shell stream and track the target and try
to get both to impact with one another. You'd need to use a slightly
different technique with lasers (probably low power aiming beam, high
power pulse once it is aimed) but the analogy doesn't seem quite the same.
I do recall reading in discussions of SDI problems that the computer code
to get things to work right was one of the main difficulties (next to
laser power) that was expected with the system. Of course, a lot of
garbage was written about SDI ... so I can't judge that.
JefWilson
In article <Pine.BSF.4.02A.98120...@dillinger.io.com>, pulver
<pul...@io.com> writes:
Any chance of getting someone to write _GURPS Honor Harrington_ (based on the
books by David Weber)?
>* so would the reactionless drives...
except for this last item.
Jeff Wilson
http://members.aol.com/JefWilson/rpg/
ajac...@iii.com
> books by David Weber)?
And this is considered _serious_ starship combat? David Weber throws around
impressive-sounding verbiage, but if you look closely a lot of it doesn't
make a whole lot of sense....
I suspect licensing for GURPS HH would cost rather more than SJGames would
want to pay ;) Write your own unofficial version, David Weber has a tendency
to put appendices with all kinds of useful technical information in his
books.
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
David P. Summers
glha...@copper.ucs.indiana.edu (Gregory Loren Hansen) wrote:
> >Except that the foe is presumably making evasive manuevers.
> Say you're aiming for the center of a 70 meter long ship that's 50,000,000
> meters away, with a laser or particle beam. The target has 0.17 seconds
> to pull away.
Actually, it has the time it's sensor signal travels to the sensor,
gets aquire, processed and and a firesolution is computed, and the
time the return shot takes to get to the ship. A ship only has
to move the distance equal to half its shortest distance to avoid a
shot that is dead on. It has to move a fraction of that distance
to avoid a grazing shot.
Now we don't know how fast TL 12 sensors can aquire signals, etc. bit
if we assume that the whole "aquire, process, decide" routine is 100
millseconds and the ship is 20 meters wide. It can avoid a shot
by a change in acceleration of 10 G (acheivable by accelerating
5 G in one direction after accelerating in the other). If it is
avoid grazing shots, then the requirement is less.
David L. Pulver
> Any chance of getting someone to write _GURPS Honor Harrington_ (based on the
> books by David Weber)?
Hey, I said a *serious* system. Weber's system is basically an artificial
set up designed (initially -- systems change in later books) to justify
age of sail tactics in a far future science fiction universe. This isn't
to say the books aren't fun, of course...
Personally, I suspect that Weber has a good idea how he would want to game
out his own system, since he's worked on the STARFIRE tactical space
combat system for Task Force Games. I doubt he'd find working on a game
supplement financially rewarding compared to novels, though. Perhaps you
could hire Steve White to do it? He's Weber's occasional writing partner
(e.g., on the recent IN DEATH GROUND) and is familiar with GURPS...
Uxi10801
>
<snip>
If you ever do write this "serious" space combat system and it's printed
(GURPS, presumably), please lemme know, so I can at least browse it in the shop
(if not buy it outright). ; )
Eberhard Schulz
> Say you're aiming for the center of a 70 meter long ship that's 50,000,000
> meters away, with a laser or particle beam. The target has 0.17 seconds
> to pull away.
No, it has not. Since the laser beam travels at the speed of light,
it cannot be observed in advance. You don't know what hits you until
it does hit you.
All you can do while you are being it is move out of the danger zone,
or maybe rotate your ship to distribute the energy over a larger
surface so that it doesn't penetrate.
NEil Phillips
share the following wisdom:
>Besides, if computers continue to double in power every eighteen months,
>by that time a computer that can do what you described will be available
>at grocery store checkouts for $15. And that's mostly for the plastic and
>the keyboard and display, not the little chip inside.
Recent indications seem to be showing that that simply is not the
case.. (IE. computers will level out eventually..)
The problem's all to do with how many transistors you can fit on a
chip, and how hot the chip can run.. there will probably be ways
around it, but there will likely be a slowing untill that solution is
found.
Possiiblities are room temp superconductors, bio-computers (why
bother, seems to be enough of them around..), optical computers - a,ll
will need a little effort. Then again, there could be another
breakthrough just as earth-shattering as the silicon chip was, and
it'll be like moving from valves to transistors all over again..
desktop-sived computer on your watch, interfacing through the nerves
in your wrist directly with your brain - but thats just speculation
basic point is, they'e gonna reach a limit with what the current style
of computers can do, eventually.. (you can, f course, get more speed
and memory, big it's gonna reach a point where they start getting
bigger again..)
--
NEil (phil...@gwbbs.net.au)
a.k.a. N-ster...
The opinions expressed in this message are not my own,
but rather are those of Microsoft Corporation.
NEil Phillips
wisdom:
>You have made a small error here. You are correct that the laser beam
>or (presumably nearly light-speed) particle beam would take 0.17 seconds
>to reach the target. However, you have neglected an important point.
>You do not have FTL sensors--at any given time you only know where the
>target was 0.17 seconds *ago*. This means that the target actually has
>at least 0.34 seconds to dodge between the time you last detected it
>and the time that your beam reaches it. This delay could actually be
>larger if you factor in things like the amount of time to process the
>sensor data, the time to make a targeting prediction, and the time to
>actually fire the weapon. Granted this will be fairly small but could
>add up, particularly the firing time since there are mechanical operations
>involved as well as electronic. Laser beams and particle beams do not
>fire instantaneously.
>
>This does not refute your argument, since the acceleration necessary
>for the target to successfully dodge is still high, but not nearly as
>high as you had calculated.
plus, if the ENEMY does not have these "ftl" sensors, by the time they
see teh laser firing at them, it has allready hit them..
pulver
Heh. Actually, I've drafted a prototype, but not for GURPS (oddly enough,
it was written for the Legions of Steel universe, but that's another
story). Since I severed my relationship with Global Games, that one isn't
likely to see the light of day. Maybe in a few years...
Gregory Loren Hansen
NEil Phillips <phil...@gwbbs.net.au> wrote:
>mcdo...@mit.edu (Steve McDonald) saw fit to share the following
>wisdom:
>
>>You have made a small error here. You are correct that the laser beam
>>or (presumably nearly light-speed) particle beam would take 0.17 seconds
>>to reach the target. However, you have neglected an important point.
>>You do not have FTL sensors--at any given time you only know where the
>>target was 0.17 seconds *ago*. This means that the target actually has
>>at least 0.34 seconds to dodge between the time you last detected it
>>and the time that your beam reaches it. This delay could actually be
>>larger if you factor in things like the amount of time to process the
>>sensor data, the time to make a targeting prediction, and the time to
>>actually fire the weapon. Granted this will be fairly small but could
>>add up, particularly the firing time since there are mechanical operations
>>involved as well as electronic. Laser beams and particle beams do not
>>fire instantaneously.
>>
>>This does not refute your argument, since the acceleration necessary
>>for the target to successfully dodge is still high, but not nearly as
>>high as you had calculated.
>
>plus, if the ENEMY does not have these "ftl" sensors, by the time they
>see teh laser firing at them, it has allready hit them..
I feel better about it after David ???'s explanation that they were
deliberately trying to capture the "man in the loop" feel. (Does hit
chance depend on the target pilot's skill? If not, maybe it should.)
Knowing that it's not simply an oversight removes most of the reason for
arguing.
But with all the talk about evasion and trying to predict the target's
next location, I had assumed the ship was continually bouncing around like
a high-speed chase on a bad dirt road. There's a lot of reasons not to do
that unless you have good Star Trek-style inertial dampers, but that's all
that made sense from the arguments I was getting.
I still think improved stabilizer mounts are more useful than a faster
targeting computer. Besides, replacing a small box to get better accuracy
is too easy, too Star Trekkie. The job should involve lifting and cursing
and spinning of wrenches, grease on the work suits, machining of clamps or
other odd peices of hardware, and three hours spent zeroing it.
David P. Summers
> deliberately trying to capture the "man in the loop" feel.
That was the primary reason, but there are physical rationales
behind them to.
> But with all the talk about evasion and trying to predict the target's
> next location, I had assumed the ship was continually bouncing around like
> a high-speed chase on a bad dirt road. There's a lot of reasons not to do
> that unless you have good Star Trek-style inertial dampers, but that's all
> that made sense from the arguments I was getting.
Actually, Traveller does. They are called Grav Compensators in Taveller.