Post Reply 
 
Thread Rating:
  • 3 Votes - 3.67 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Buff APS accuracy! (or at least correct it!)
2017-12-15, 11:11 PM
Post: #1
Buff APS accuracy! (or at least correct it!)
Following on from my rifled barrel post which started to go off on a tangent, I've been using my mathematically inclined mind to work out how FtD APS fairs against real guns. Hold onto your horses *cough* ships, this is gonna be long and interesting!

Using trigonometry, we can work out the dispersion of guns in FtD. I must also note that FtD dispersion IS A LIE! Well kind of. It's misleading. The inaccuracy in degrees seems to be the radius and not the diameter of the 'dispersion cone'. Your actually accuracy seems to be twice the angle given.


Take this pretty standard cannon I'm designing. Ignore the aesthetics as it's only for testing. Including the mantlet the barrel is 10 metres long or, because it's a 293mm gun, approximately 34 calibres long (10,000mm/293mm = 34.13) which is a bit lower than the standard 38 to 40 calibre guns used across both world wars.

[Image: 7JdfBuS.png]

It has quite a low muzzle velocity. More gunpowder only increases the accuracy by small amounts and I lose a lot of explosive power and of course cool-down times are longer.
It has an inaccuracy of 0.746°.
To work out the dispersion we need trigonometry and for this purpose, Tangent is our best friend.

Using these calculations we get these results at various ranges (0.746°):
500m = 6.5m [Image: lJ7R5td.png]
1000m = 13.0m [Image: uOUFCby.png]
2500m = 32.6m [Image: PQ44I2m.png]
5000m = 65.1m [Image: htPFfed.png]

Now these don't look too bad... but they're wrong! After testing all the distances in FTD, I was hitting around an area twice the size of the circles drawn! So I remade the formula, not halving the angle to overall get twice the dispersion!


500m = 13.0m [Image: uOUFCby.png]
2500m = 65.1m [Image: htPFfed.png]
5000m = 130.2m [Image: 4Gorcjc.png]

NO WONDER I CAN'T HIT ANYTHING!

So yeah... the accuracy reading needs to be more... accurate.

Now...

Lets talk about REAL guns. Specifically the American 16"/50 calibre gun, used on the IOWA class, info here; http://www.navweaps.com/Weapons/WNUS_16-50_mk7.php
In the 1987, tests were conducted to access the accuracy of the gun. 15 shells were fired to a range of about 32,000m with a dispersion of just 200m. Using MORE Trigonometry, we get an accuracy of just 0.179°, or by the games incorrect accuracy reading, it would read just 0.09°(but would get a 0.179 dispersion) in FtD. Lets have a look at the inaccuracy at 5000m if FtD was true to physics:
[Image: nGKZHqA.png]7.9m at 5000m.
That's basically what my gun can do at 250m. Real life guns are 20 times more accurate guys n gals...

And finally, I leave you with this. If I want the accuracy of the 16" gun, how long does my barrel need to be?

Well for 50 calibres, I need a 15m barrel, and I get an inaccuracy of 0.487°. For the accuracy of the 16" I need a 75m barrel, or a 256 calibre gun!
[Image: RSjBA2R.png]
But at least now I hit every time!!!

So cmon guys, should we buff the APS accuracy? Maybe not by 20 times, but 4/5 times?
Find all posts by this user
Quote this message in a reply
2017-12-15, 11:40 PM
Post: #2
RE: Buff APS accuracy! (or at least correct it!)
I think people seem to forget that accuracies and velocities are reduced in order to compensate that the max combat range in FTD is 5km, instead of the 32km you used from real guns.
Find all posts by this user
Quote this message in a reply
2017-12-16, 12:16 AM
Post: #3
RE: Buff APS accuracy! (or at least correct it!)
(2017-12-15 11:40 PM)Liondrome Wrote:  I think people seem to forget that accuracies and velocities are reduced in order to compensate that the max combat range in FTD is 5km, instead of the 32km you used from real guns.

Velocity yes. Accuracy? Not that I know of... perhaps you're correct.
Find all posts by this user
Quote this message in a reply
2017-12-16, 01:18 AM
Post: #4
RE: Buff APS accuracy! (or at least correct it!)
When you stop trying to play Theme Hospital it's entirely possible to have accurate guns.
Decreasing inaccuracy to <25% would be just silly, real life comparisons are useless.
Find all posts by this user
Quote this message in a reply
2017-12-16, 04:10 AM (This post was last modified: 2017-12-16 04:16 AM by Richard Dastardly.)
Post: #5
RE: Buff APS accuracy! (or at least correct it!)
* Vastly reduce barrel length required for whatever the ammo controller deems good accuracy
* Have any accuracy gains from extra barrel length fall off hard. Really if you've burned all your propellant you're not going to gain anything from more barrel.

No more noodle barrels, no 0.001% inaccuracy either. I built a fairly nice - or at least reasonably balanced - looking turret with 4m shells, and then rebuilt the insides with 6m loaders. The ammo controller is telling me to use 23m of barrel, this is just stupid. With some fairly intrusive rail assist it's around 0.3-0.4% inaccuracy with 17m of barrel, but even that is idiotically long.

Old old argument though. I guess one more go at it wouldn't hurt.

Poke my boat! mostly pre-2.0 learning & catalogue thread - Update: Heavy & light tanks 12/01/18 for 2.1. 6 ships made 2.0 aware. If it's not AotE I can't do it regularily - sea -> post processing -> eyestrain.
Find all posts by this user
Quote this message in a reply
2017-12-16, 06:04 AM
Post: #6
RE: Buff APS accuracy! (or at least correct it!)
(2017-12-16 04:10 AM)Richard Dastardly Wrote:  * Vastly reduce barrel length required for whatever the ammo controller deems good accuracy
* Have any accuracy gains from extra barrel length fall off hard. Really if you've burned all your propellant you're not going to gain anything from more barrel.

No more noodle barrels, no 0.001% inaccuracy either. I built a fairly nice - or at least reasonably balanced - looking turret with 4m shells, and then rebuilt the insides with 6m loaders. The ammo controller is telling me to use 23m of barrel, this is just stupid. With some fairly intrusive rail assist it's around 0.3-0.4% inaccuracy with 17m of barrel, but even that is idiotically long.

Old old argument though. I guess one more go at it wouldn't hurt.

Another suggestion is to have shell velocity begin to decrease if your barrel is too long compared to the number of GP/Rail casings, as after a point friction begins to take its toll.
Find all posts by this user
Quote this message in a reply
2017-12-16, 08:53 AM (This post was last modified: 2017-12-16 08:55 AM by Richard Dastardly.)
Post: #7
RE: Buff APS accuracy! (or at least correct it!)
(2017-12-16 06:04 AM)Liondrome Wrote:  Another suggestion is to have shell velocity begin to decrease if your barrel is too long compared to the number of GP/Rail casings, as after a point friction begins to take its toll.

I would actually think compressing the air in front of the shell is a bit of problem, you can make the shell as aerodynamic as you like but if the air has nowhere to go, it's not going to matter what shape it is. As long as the propellant is expanding behind the shell it'll still try and accelerate it, but obviously with fixed propellant amounts the pressure behind the shell will be dropping the entire time the shell moves. Meanwhile there's air in front being moved and some barrel friction.

Really though we're just looking for excuses for game balance at this point.

Poke my boat! mostly pre-2.0 learning & catalogue thread - Update: Heavy & light tanks 12/01/18 for 2.1. 6 ships made 2.0 aware. If it's not AotE I can't do it regularily - sea -> post processing -> eyestrain.
Find all posts by this user
Quote this message in a reply
2017-12-16, 09:09 AM
Post: #8
RE: Buff APS accuracy! (or at least correct it!)
I would actually agree that APS accuracy could be increased by ~20-50% which would help greatly to stop having unboiled spaghetti as barrels...

There is always a weak-spot if you search Hard enough.

If you fire enough AP at that shield, at some point you're going to come through.

There is no "best" I wouldn't even say there is anything universally good, Good is subjective, I find everything bad even if it's in theory good against this or that.
Find all posts by this user
Quote this message in a reply
2017-12-16, 12:16 PM
Post: #9
RE: Buff APS accuracy! (or at least correct it!)
(2017-12-16 09:09 AM)Skyer Wrote:  I would actually agree that APS accuracy could be increased by ~20-50% which would help greatly to stop having unboiled spaghetti as barrels...

Yep, that was mentioned a few times when the propellant requirements were halved.
Didn't see anything reasonable in this suggestion though.
Find all posts by this user
Quote this message in a reply
2017-12-16, 04:46 PM (This post was last modified: 2017-12-19 05:24 AM by Richard Dastardly.)
Post: #10
RE: Buff APS accuracy! (or at least correct it!)
(2017-12-16 12:16 PM)draba Wrote:  
(2017-12-16 09:09 AM)Skyer Wrote:  I would actually agree that APS accuracy could be increased by ~20-50% which would help greatly to stop having unboiled spaghetti as barrels...

Yep, that was mentioned a few times when the propellant requirements were halved.
Didn't see anything reasonable in this suggestion though.

Yeah, that isn't the way to go about it - in fact you should get penalised after a certain barrel length & start losing accuracy again, because the thing will start bending - in addition to possibly losing some muzzle velocity unless it's a railgun. However, having the optimal lengths the same for propellant burn & accuracy & then having anything more have quite severe diminishing returns seems a way to get sensible barrel lengths because you'd never get any more accurate than you are as it is right now.

Having the behaviour a little different for railguns would help differentiate railguns too - in their case you're accelerating the projectile at the same force for the length of the magnets rather than having the force accelerating the projectile diminish as the propellant gas expands, and if we accept they're spinning the shell up for accuracy reasons then that is also somewhat detached from barrel length, because it can do that even if the shell isn't in the barrel anymore. In this case, *magnet* length should matter.

Edit: sort-of when it comes to railgun acceleration force - that's ignoring any induction effects of a lump of metal moving through a magnetic field. Some diminishing returns in magnet length would be sane anyway.

Poke my boat! mostly pre-2.0 learning & catalogue thread - Update: Heavy & light tanks 12/01/18 for 2.1. 6 ships made 2.0 aware. If it's not AotE I can't do it regularily - sea -> post processing -> eyestrain.
Find all posts by this user
Quote this message in a reply
Post Reply 


Forum Jump:


User(s) browsing this thread: 1 Guest(s)