|
Post by dorn on Apr 29, 2019 22:42:34 GMT -6
This is amazing. Why didn't the RtW team not implement something similar already? Simulated battles are very much needed, especially having non-player nation vs nation engagements and wars simulated would be great. Because the principles of fight are different. You have quite complex creation of design, systém of fight and if you make auto-resolve of battle without taking this into considararion you have another game.
|
|
|
Post by oaktree on Apr 30, 2019 8:42:51 GMT -6
Given that Steam and Iron (SAI) was already around and RTW to a degree is a scenario generator for SAI it's not surprising that there was not a simple combat simulator in play. (I don't know what the auto resolve for raider interceptions does, but it by default is a simpler scenario than what Tortuga is trying to handle.)
|
|
|
Post by garrisonchisholm on Apr 30, 2019 17:30:53 GMT -6
We should also consider that "simming" a battle was in general the exact opposite of what NWS set out to do. They have an amazingly elegant physics system and great detail of ship-building, and just boiling things down to- essentially- a single dice throw is the opposite of those intentions.
That doesn't mean that after you've played it 200 times in 3 years you don't find situations where this would come in handy though. :]
|
|
|
Post by dorn on May 1, 2019 1:09:19 GMT -6
I agree with you. I do not want this but a lot of people like it and I just try to point out that it is probably not difficult to make it as there are missing 2 things: - all player force under AI control
- UI for auto-resolve using battle "in background"
|
|
|
Post by tortugapower on May 1, 2019 22:10:50 GMT -6
Hi dorn , I'll echo the sentiments here: this simulator is not meant to replace the existing combat entirely. It's intended use-case is when you are up 3x in victory points and do not wish to fight the little battles any more. Right now, the only alternative is declining every battle, but even then you get unexpected battles you must fight.
This is why my simulator intentionally returns conservative results. You can auto-sim without fear of really changing the war.
That said, there is a sizable group of people who would like to see an Auto-Resolve feature. I agree with garrisonchisholm and oaktree 's assessment that Fredrik did not envision the game played with auto-resolve. I believe RtW is supposed to be a strategic shell thrown around a tactical engine, but many people thoroughly enjoy the strategic management and prefer it as a strategic-level game with tactical combat included.
For those people, I don't see the harm in adding the auto-resolve feature: worst-case scenario it goes unused, and there is almost no dev-time wasted as I have done the work already.
|
|
|
Post by tortugapower on May 10, 2019 16:10:32 GMT -6
I just released a "big" update, essentially combining both types of my simulators (this one for RtW combat events, and the other for any custom user-made fleet composition). The backbone behind this merger is a new settings file that should be placed alongside the .exe in the same directory: rtw_sim.ini. This file allows input to the internals of the simulator. Here's the current settings file in its entirety: # Simulator Load Parameters Game=0 fleet1 = "./fleet1.txt" fleet2 = "./fleet2.txt"
# Fleet Parameters flee_if_smaller_than = 0.50 fleet_health_limit = 0.60
# Division Parameters div_health_limit = 0.65 move_failure_chance = 0.09 You can download it from the OP or you can copy/paste that text into a new file "rtw_sim.ini" (using Notepad). Instead of placing the simulator exe into each savegame folder (e.g. Game1 or Game4) -- how it previously worked -- the simulator now resides in the "Save" directory inside the RtW installation directory. The top definition, Game=0, tells the simulator what files should be loaded for simulation. - If "Game" is set to zero (Game=0), then it will load fleet1 and fleet2 text files, as defined in the next two lines. This allows for custom fleet combat).
- If the Game is set to a # 1 to 5, then it will load the RtWGame#.sac combat file from the appropriate directory, Game#.
Also note that the courage of fleets and retreat requirements are now accessible to the user. If you want fleets to fight to the death, set the health limits and flee_if_smaller value to zero! For a video tutorial, please reference first 4 minutes of this video: youtu.be/VRKyNhPAMKUGood hunting! Tortuga
|
|
|
Post by Adseria on May 11, 2019 12:42:04 GMT -6
Ok, I've attached a battle log that I found quite interesting, and it shows a minor flaw with the simulator.
So, the 2 fleets started 10kyds apart (poor weather made that the max sight range). Both fleets seemed to want to maintain 8kyds. However, they were both capable of moving 2kyds at a time. As a result, both sides close in to 6kyds apart. At this point, both sides see that they're too close, and try to move apart. The next round, they're 10kyds apart, and so they close in...
This repeats until round 6. Round 6 was different because one fleet (the british) were slowed by random factors. As a result, they only withdraw by 1kyds. The 2 fleets are now 9kyds apart. They both try to close by 1kyds... and find themselves at 7kyds. They try to withdraw by 1kyds... and they're back to 9kyds again. This once again repeats until round 10, when the American ships are also slowed by random factors. At this point, the British fleet withdraws by 1kyds, the Americans stay put, and the 2 fleets are now at 8kyds, where they wanted to be all along. They then maintain until one the British decide to increase the range by 1kyds, and do so. Then the Americans close back to 8kyds, then the British withdraw again...
My point it, in an actual battle, each side would be reacting to the movements of the other. Since both sides decide how they want to move at the same time, and then execute the movement, it leads to this oscillating around the point where they want to be. It would be better if one side (Britain, in this case) issued movement orders, and then the other side (America) reacted to those orders. Then, in the second round, Britain reacts to whatever America did in round 1. This way, they'd know where their opponent would be, and react to it, as would happen in a real battle.
Attachments:log.txt (34.19 KB)
|
|
|
Post by vonfriedman on May 12, 2019 3:35:01 GMT -6
I can agree to insert in RTW (and RTW2) an option that allows the player to resolve some minor clashes automatically, basically with the throwing of a dice. A great naval battle, however, cannot be reduced to a simple duel between parallel lines of ships. Not to mention Trafalgar, Jutland or Tsushima, even the most inexperienced player would move his fleet in a more complex way. A satisfactory simulation of a naval battle therefore would require - in my opinion - a considerable design effort. I would prefer the NWS team to devote such an effort to develop a new SAI2 game.
|
|
|
Post by tortugapower on May 13, 2019 11:53:12 GMT -6
Ok, I've attached a battle log that I found quite interesting, and it shows a minor flaw with the simulator.
<snip>
My point it, in an actual battle, each side would be reacting to the movements of the other. Since both sides decide how they want to move at the same time, and then execute the movement, it leads to this oscillating around the point where they want to be. It would be better if one side (Britain, in this case) issued movement orders, and then the other side (America) reacted to those orders. Then, in the second round, Britain reacts to whatever America did in round 1. This way, they'd know where their opponent would be, and react to it, as would happen in a real battle.
You are correct, this oscillation is possible, and it was the impetus behind adding a "slowed by random factors" mechanic -- this is essentially a damping parameter. I don't like that this can happen because it's artificial and ugly from a human perspective. However, I don't think that this has any impact on the simulation or its results. I could implement a system where one side acts first and then the other (IGO-YOUGO, the traditional tabletop turn format). However, I think that WEGO is more realistic and a better solution whenever it can be implemented. If these oscillations appear to be harmful to results, I could implement an IGO-YOUGO system, but it's possible that one side will gain an advantage.
|
|
|
Post by oldspec4 on May 23, 2019 2:40:15 GMT -6
Excellent..exactly what I needed to get back into the game.
Also, wanted to add that I enjoy your videos too.
|
|
|
Post by tortugapower on May 24, 2019 1:38:33 GMT -6
Excellent..exactly what I needed to get back into the game.
Also, wanted to add that I enjoy your videos too.
Thank you! If you have any problems getting it working, let me know.
(Huge update for this simulator coming up as well, after lots of feedback from the fleet design tournament.)
|
|
|
Post by cabalamat on Aug 13, 2019 12:10:25 GMT -6
I'm trying to run the fleet combat simulator (version 0.511) and failing. Here's a screenshot of me running it from the command line, with the .ini file open:
Any idea what the problem is?
|
|