Not a member yet? Why not Sign up today
Create an account  

Poll: What would you put in your ship?
You do not have permission to vote in this poll.
Beltfed
66.67%
2 66.67%
Standard
33.33%
1 33.33%
Total 3 vote(s) 100%
* You voted for this item. [Show Results]

  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
 
Beltfed vs 1M Loaders with two 5x5 Optimal Turrets

#1
[Image: sxEDyTD.png]
These fire 142 mm sabot shells

These are two 5x5 turrets, one beltffed with 12 beltfeeders, 1 clip each and 5 inputs per clip. It has a burst RPM of 308 and sustained ~ 160. it required a constant and high ammunition input, for small ships requireing many ammo processors. It is not very explodey. It costs 2455 res.

The other turret is standard loader fed, with 19 loaders with 3 clips per loader and one input feeder. This turret can sustain fire for 5 mins at 150 RPM then  at 122 RPM for 9 mins 40 seconds then 91.5 RPM indefinitley. It holds so many rounds that it does not require much ammunition supply during battle, eliminating the need for ammunition processors making it free to fire. It is very explodey. it cost 1980 res.

What one would you chose? Why?
Reply

#2
"free to fire" is misleading, that ammunition is still going to come from somewhere. 

I would choose the belt-fed.  Loading Sabot for ammunition my biggest advantages are DPS, and range.  I'll assume that both fire a shell with the same range/accuracy at which point, for combat effectiveness I want the most turret for the space I spend.  What you've told me is that in a head-to-head match the belt-fed gun is worth 2 of the other cannons, initially.  This gives me a far greater alpha-strike and a better chance of dealing critical or crippling damage first.  as the battle drags on, the belt-fed gun never drops below the DPS of the normal cannon, and it's advantage, after the initial drop to 160 rpm will continue to grow until the belt-fed gun is once again worth almost 2 of the normal cannons, since they are steadily slowing down.

From a cost perspective, considering total fleet logistics you will also tend to find that the belt-fed gun is cheaper to field.  Yes, the gun itself appears, on paper, to cost more to operate but a fleet armed with the belt-fed gun will need to field less total ships to match the combat power of the other weapon and in a head-to-head match with even numbers of guns the belt-fed equipped ships have a much higher chance of escaping with minimum damage because they hold a firepower advantage.  That advantage translates directly into more chances to knock out enemy weapons, blind enemy ships, or expose critical systems first, thus minimizing damage to friendly forces.  Minimum damage, leads to minimum replacement costs and trust me, you would rather spend 2k more on a gun and then spend 5k (random numbers) more operating it to save yourself the loss of a 30k vessel.

I'll point out another aspect of cost.  Larger ships cost more, which means larger systems, that require a larger ship to carry also cost more.  processors are small and material storage is also small.  I have generally found that ships large enough to get away with using ammo box generation instead of processors would be much better off replacing that block of ammo with another engine or similarly useful system.  while most small ships are simply too small to sustain any useful weapon without processors.
-Do not bring forth an argument as fact that can be disproven with a 10 minute Google search.
Reply

#3
You should place at least 2 input feeders for each clip section to make "time to clip" time equal to "time to loader", so your rate of fire will not drop that drastically when clips are empty. It should improve performance of the second cannon.

These are 4 "best" pure kinetic shells for a 1m loader:
D=166mm [apcap, solid, solid, gunpowder, gunpowder, gunpowder], V=423m/s, kinetic=5054, DPS=1150
D=166mm [apcap, sabot body, solid, gunpowder, gunpowder, gunpowder], V=468m/s, kinetic=4944, DPS=1125
D=142mm [apcap, solid, solid, gunpowder, gunpowder, gunpowder, gunpowder], V=477m/s, kinetic=4217, DPS=1120
D=200mm [apcap, solid, solid, gunpowder], V = 344, kinetic=5800, DPS=1113

These compositions are taken from top-4 generated by optimal solver, so it is unlikely there are better pure kinetic shells (or my optimizer is broken).

Generally, you should set largest diameter for a shell to achieve maximum DPS for gunpowder-based cannons. In most cases it could be picked according to formula: D=min(2*loader_length/shell_modules, 500).
Reply

#4
(2019-06-14, 11:49 AM)TinGoose Wrote: You should place at least 2 input feeders for each clip section to make "time to clip" time equal to "time to loader", so your rate of fire will not drop that drastically when clips are empty. It should improve performance of the second cannon.

These are 4 "best" pure kinetic shells for a 1m loader:
D=166mm [apcap, solid, solid, gunpowder, gunpowder, gunpowder], V=423m/s, kinetic=5054, DPS=1150
D=166mm [apcap, sabot body, solid, gunpowder, gunpowder, gunpowder], V=468m/s, kinetic=4944, DPS=1125
D=142mm [apcap, solid, solid, gunpowder, gunpowder, gunpowder, gunpowder], V=477m/s, kinetic=4217, DPS=1120
D=200mm [apcap, solid, solid, gunpowder], V = 344, kinetic=5800, DPS=1113

These compositions are taken from top-4 generated by optimal solver, so it is unlikely there are better pure kinetic shells (or my optimizer is broken).

Generally, you should set largest diameter for a shell to achieve maximum DPS for gunpowder-based cannons. In most cases it could be picked according to formula: D=min(2*loader_length/shell_modules, 500).

I would question that optimizer.  From an engineering perspective the system itself can be 'perfect' but if the inputs are bad the results will be too.  The fact that you don't include any high AP options and all the shells are slow for what is typically the fastest shell type are both red flags to me.  As far as optimizers go, their best use is finding the specific shell diameters that offer the most shell modules at a given length, most of these are known values by this point.  Using them to pick components is suspect as that decision should really be made based on the anticipated target profile, and the role of the ship.
-Do not bring forth an argument as fact that can be disproven with a 10 minute Google search.
Reply

#5
Beltfed, smaller volume for same firepower means it's easier to defend(on top of being less volatile).
Hard to quantify but being compact in general is very important, I think it's well worth the extra materials even if you reload the other gun exclusively through passive regen.

(2019-06-14, 11:49 AM)TinGoose Wrote: D=200mm [apcap, solid, solid, gunpowder], V = 344, kinetic=5800, DPS=1113

These compositions are taken from top-4 generated by optimal solver, so it is unlikely there are better pure kinetic shells (or my optimizer is broken).

Your thing ignores AP, so it gives wrong results.
A quick example:
  • 200mm AP/2X solid/2X GP is roughly 345 m/s, 5865 damage, 9.1 AP(only ~65% damage against 1m stone), ~53.4K kinetic potential
  • 200mm AP/2X sabot/2X GP is 408 m/s, 5348 damage, 14.5 AP, ~77.5K kinetic potential for a ~45% increase
That's ignoring that misses do nothing and low caliber shells lose a lot of damage in the outermost layer of armor on angled shots.
Reply

#6
Good point. I have forgotten about armor part. Fixed that. Now I can pass armor value to optimizer.

Here are variants against armor=50, 1m length:
DPS     DAMAGE                    D      VEL      T      BLOCKS  SHELL
244.44 kinetic=3845:20.1   142   565.60  3.17   34         ['apcap', 'bsabot', 'bsabot', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder']
245.14 kinetic=2275:23.9   100   670.98  2.22   38         ['apcap', 'bsabot', 'bsabot', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder']
250.46 kinetic=2687:22.9   111   645.11  2.46   36         ['apcap', 'bsabot', 'bsabot', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder']
251.47 kinetic=3203:21.7   125   611.24  2.77   35         ['apcap', 'bsabot', 'bsabot', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder']

Variants vs armor=100, 1m length. Max modules=10
DPS        DAMAGE                 D      VEL       T    BLOCKS  SHELL
122.22 kinetic=3845:20.1   142   565.60   3.17    34    ['apcap', 'bsabot', 'bsabot', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder']
122.57 kinetic=2275:23.9   100   670.98   2.22    38    ['apcap', 'bsabot', 'bsabot', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder']
125.23 kinetic=2687:22.9   111   645.11   2.46    36    ['apcap', 'bsabot', 'bsabot', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder']
125.73 kinetic=3203:21.7   125   611.24   2.77    35    ['apcap', 'bsabot', 'bsabot', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder', 'gunpowder']

At first I was surprised results are the same. But since relevant AP values are lower than 2*armor, all generated variants have their DPS reduced in absolutely the same proportion.

*Since there are so many GP blocks, base bleeder would be nice addition to these shells.  But my optimize slightly underestimates velocity for such shells. I should dig deeper into the code

By the way, is there any way to make a proper table formatting, like bbcode/html? Manual formatting is so painfull.
Reply

#7
probably easier to just post a snap of the output of the optimizer itself.

and again, even with AP added, I still have to question the optimizer.  the core question becomes what are you optimizing *for*?  An optimization program will give you *exactly* what you ask for, but it doesn't validate the question itself.  It's easy to guild a system that will give you the best answer to the worst question rather than asking the best question and then answering that.
-Do not bring forth an argument as fact that can be disproven with a 10 minute Google search.
Reply

#8
(7 hours ago)Lincrono Wrote: probably easier to just post a snap of the output of the optimizer itself.

and again, even with AP added, I still have to question the optimizer.  the core question becomes what are you optimizing *for*?  An optimization program will give you *exactly* what you ask for, but it doesn't validate the question itself.  It's easy to guild a system that will give you the best answer to the worst question rather than asking the best question and then answering that.

Right now it generates a shell design with largest effective damage per second against specified armor. For kinetic shells it is just about measuring direct damage, AP and reload time from shell volume. It sorts results by the damage delivered. It grinds through all possible shell configurations. Current setup generates pure kinetic shells, though other combinations are possible (HE, frag...).
Reply



Forum Jump:


Users browsing this thread:
1 Guest(s)