DSubs – abstract multiplayer subsim


Pdf manual is bundled in the zip, or can be found in the repo:


TLDR: 2D cold-war-techlevel DW-inspired sensor-focused subsim, multiplayer infinite quick battle scenario with bots and other players.

Hello everyone!

I’m Boris-Barboris and I would like to make the progress of my hobby project public knowledge. I’m steadily moving towards the technical demo stage and expect to be able to let the general audience test it till the end of 2019.


I was a great fan on Sonalyst’s work on Dangerous Waters, as well as of DCS/LockOn simulator franchise. I have spent weeks in those titles, but the past is now long gone. Youth was gone as well, and so, gone was my blind respect for realism. Time grew much in value, and it is the time that was rarely respected by the simulator genre: realism dictates large intervals of wasted time and no gameplay just for the sake of realism. Time acceleration helps, but it is unsuitable for multiplayer – the prime thing I was enjoying in those games. I just want a fun PvP/Coop submarine-related game to play with real people for an hour or so. It appeared to me that the gameplay must be designed from the ground to account for higher desired combat intensity and lower time intervals.

Exactly 2 years before this post, March 2017, I have started my small project: DSubs (D for distributed), which I hope you will find in your favor.

Player controls one submarine. Everything that exists is submerged as well.
The only thing I enjoyed in DW was sub warfare. I never cared about surface of the water, nor did I care for air. The fact that half the time I open some Cold Waters video and see some annoying helicopter spamming torps on the player’s head tells me that I’ve made a right choice.
Not a simulator.
No real submarines, no real weapons, no real geography. I am inspired by the real world, but have no desire to replicate it.
Literally no graphics
I have no time for this anyways. Gameplay first. Whistles are easy to add later, core systems are not.
World is 2D.
If there was a depth, there would be surface, and surface is of no interest to me. Either way, I cannot implement 3D world mechanics in so little time that I have. But I would like to: sound speed profile is an interesting mechanic of great gameplay value.
Ranges, speeds and timings are tuned for action.
I expect the game to suit a working man with as little as 30 minutes of free time. Complete game loop cycle of spawn and death should be short enough for everyone.
Online only. Centralized server as an umpire.
I like to play with people, so I focus on the multiplayer from the start. I will make single-player tutorials and a campaign system for loners but it will still require internet.
Focus on sensors and imperfection of knowledge.
You only see what your submarine hull sees: the fact that you are alive, your ammo leftovers and the output of your sensors. Kills will not be reported, and the kill-feed/player stats will be significantly delayed.
Simple yet fun infinite scenario.
I have a nice idea about the rule set, which involves restricted circle-shaped arena that sometimes moves, forcing the players to rapidly relocate on near-flank speed in order to shuffle the cards and prevent stasis. I want the players to be able to spawn and reload ammo continuously, so it will not be round-based.
Indiscriminate solo and cooperative play.
The thing was written to be coop-capable from the start. Players can join the server solo, or join the other player and help him to control his submarine. I don’t like “multi-station” concept, coop players share equally full control of the sub.
What is done (September 2020)
Submarine navigation control: target course and throttle.
Underwater rigid body dynamics.
Interesting enough sound propagation model, tuned for shorter detection range and faster action.
Completely server-side kinematics and hydroacoustics calculations: client is lightweight and ignorant.
“Fine by me” propulsor sound emission model: server-side real-time DSP to generate hydrophone signal and stream it to client.
OK active sonar model, picture slices streamed to client from server.
Client-side CIC (Combat information center) COOP server, that manages contacts, data samples, solutions, hydrophone trackers…
UI for hydrophone waterfall and active sonar sensors.
UI for contact management and TMA.
Towed hydrophone arrays.
Active sonar decoys.
Simple military bots.
Simple tutorials that cover most of the mechanics.
Coop control for all of the above.

AI Bot running SUBSIM, what could go wrong?!