Weapons System, An MVP Approach
Slydev's picture
Posted on:
Tuesday, November 6, 2018 - 12:54

So this is a post I have wanted to do for a while, basically for the MVP we have to simplify the weapons system to see what works and what does not. We can build up from there in each iteration.

I have boiled down my weapons system into core points:

  • Torpedoes should be customizable so that we can pick a loadout on the fly to deal with the situation we are facing.
  • There should be a strong risk VS reward on payloads (e.g. more damaging payload would be slower and thus could be detected easier).

In the real version I think we would all like something where payloads are individual objects that travel in 3d space. For the MVP though that will not be the case, the firing system will be simple:

  • Choose payload
  • Choose target
  • Fire payload

At the time payload is fired the system will calculate two values: 

  • The time to impact
  • If the payload will hit (For MVP it will most likely be hit or miss but later there can be other values that can vary damage.)
  • If the payload is detected/destroyed at what time in the flight path it is destroyed

Time to impact is calculated fairly simply. This will be the distance the target is from us divided by the speed the payload will travel at. This will give us a time (in seconds most likely) to impact.

After the payload has been fired there will be a countdown to impact. This will continue to count down until either the destroyed time is reached (in which case the interface will let us know the payload has been destroyed) or it reaches 0 which means it hit its target.

So even though the system knows if will hit or not the crew will not.

Detecting a miss

The logic I am using here is that a payload can be detected by an enemy ship and destroyed (most likely shot at by a countermeasure, that isn't so important right now I think). To work this out we will need a couple values:

  • Detecting strength of enemy ship (How 'strong' their detecting radar or skills are)
  • Detecting range of enemy ship (What the max range they can detect is, I think post MVP this will be logarithmic, basically the further out adds a larger penalty to their strength but for MVP it can just be a drop off)

So what we need to calculate is how many seconds a payload will be inside the detection range of a ship before impact.

For example we have a ship that has a detection range of 2kms. We fire a payload travelling at 600km/hr from 3kms away. This would give the payload a travel speed of 0.166km/second which means it will be in the detection range for about 12 seconds.

So to get the seconds will be

floor([Detection Range]/[Payload KM/sec])-1

We will then do a roll for every second it is in detection range until we either run out of rolls or we get a detection (in which the payload is destroyed)

We will NOT do a roll for the last 1 (or 2) seconds, at that point the payload is too close.

So if an enemy ship has a detection rate of 5% there is a 5% chance that it will detect 11 times. This is NOT the same as multiplying the detecting rate by the seconds and there are a few reasons for doing it this way:

  • This gives us the time that the payload is destroyed (add a random fraction of a second to the time
  • This gives us an advantage if we are within the detection range of a ship as they will get less time to detect the payload
  • This gives us guaranteed hits if we are very close to their ship (which in the future will be risky to do)

What makes up a torpedo

From my other posts I discussed an automated process for creating the torpedoes (the AFP) which gives customisation so that the crew can use a payload that will work for the situations that they might find themselves in.

For MVP these values will be very simple:

  • There will be a slider that will let us choose between speed and damage. Faster torpedoes will spend less time in the detection zone and thus have more chance to hit, but they wont do as much damage and vice versa.
  • I would also like the option to add shielding to the torpedo, this will either give a speed penalty or make longer to create but will add a penalty to the detection strength of the enemy ship (so it might make it 30% less we should tweak this to find what suits best)

Damage for now will be boring, enemy ships will have a single HP value that we will be witling down.

We also need some values about the Endeavour to be tweakable:

  • Amount of torpedo tubes
  • Cooldown time on firing a payload out of a tube
  • Time to build a payload (Maybe can use this instead of the cooldown above)

We wont know till we start but will have to play around with some scenarios to work out what will be the best numbers here. We want to limit the amount we can fire so that the crew will have to pick and choose but not make it too limited that it takes an age to destroy anything.

I feel that this system will let us play with a range of payload types as well as a few interesting enemy types (High detection strength but low range. High detection range but low strength. Ships with low HP but high detection range and skill. Ships with high HP but low detection range and skill). These will all require the crew to use their intuition to work out what the best payload and approach is to destroy. We could also say have a ship with an extreme detection range and skill that we would essentially have to broadside up close to hit. There are a lot of options and as the system gets more dynamic, they will grow.


I have tried to think about a simple approach to a weapons system while still achieving my primary goals with the final system. This system is quite different to what the final will be however it is the bare minimum to let us know if the full system will work and be engaging while letting us tweak values on enemy ships and payload types to make for interesting dynamic combat. I look forward to getting feedback from anyone on what they think. Please keep in mind again that this is to be as simple as possible, I have a tonne of ideas that will make it a lot more complex but that isn't what the MVP is for at all! The tests that we do with this system will give us a lot more direction on where to go with the weapons system from here on out so its pretty exciting!

This system doesn't yet include payloads being fired at us. I think that should be a bit more advanced and be using the EMDAR plus navigation controls, we should see how this system works first and then push into the next step there. If anyone has any questions just let me know :)



mattm's picture

So to summarise what I think you see as included in the MVP:

  • Choice of multiple ordnance types: with tradeoff between range/speed, countermeasure evasion and damage infliction
  • Simple trajectory: unguided torpedoes with simple probability calculation(s) of successful prosecution based on distance, speed and effectiveness of detection/countermeasures
  • Automated countermeasures: Simple calculation of effectiveness based on limited criteria
  • Simple damage: Recorded as "structural integrity points" rather than attributed to specific location or ship systems

More complexity/realism can be added as the system develops.

NeerajA's picture

Picking up on your point about a "cool down" time between launches, the railgun launchers will be magnetic which is therefore frictionless, but the torps will be launched explosively to give initial velocity which then gets amplified by the railgun's magnetic field. This will have an impact on launcher condition and will need to be monitored - regular maintenance will be needed to prevent launcher failure.

But the biggest impact to to be turnaround times will be the process of charging the capacitor used to dump the huge amount of power needed at the instant of launch.

I went into more detail on this in a separate post here.

mattm's picture

I'm wondering if we should look at including more around trajectory elements in the MVP, as this is pretty fundamental to tactical management and are vital skills and experience to ground early, not just for the tactical team but for helm and command as well.

This would cover three post-launch phases:

  • Wire-Guided: The torp is being guided by the vessel's own sensors via a tight-beam data link. This link is easily broken if either vessel or torp maneuver too far out of line-of-sight.
  • Search: Once the wire is lost, the torp uses uses its own less powerful passive sensors and follows pre-configured search patterns to track the target.
  • Prosecution: Once close enough, the torp uses active sensors (making it detectable) to close and kill.

To give us the level of realism needed to make it a useful training experience we'd probably need to look at managing the torp track on the TOE grid, rather than just probability calculations.

As this is a whole phase in itself in the service delivery forecast and compared to the size of previous phases, including this seems about right?