I see your point but I suspect its only a matter of time before someone learns how to exploit the no mid-move interception feature. I've already discovered its possible to charge down enemy CV's if you time it right.Originally posted by Paul Vebber
The problem with "reaction" at more than a single hex or so is the "chain reaction" problem, especially around the Slot... If TF A reacts to B and moves out of range of C reacting to B and A then reacts to C you can get some nasty 'infinite loops' or other logic bombs that can lead to really bizzare cases of multiple TFs chasing each others tail, or the "but the stupid thing went after THAT TF, I wanted it go after THE OTHER one... or the "I'll put a gauntlet of week TFS' to casue the enmy to stop and never get where they want to go". It really measn a LOT of new code and LOT of testing to put the required guards in against "gamey" behavior.
And the possibility of all sots of gamey tactics of sending out small single ship task forces to try to "soak off" reaction, stream raids to "clear the way of reactors"...it can readily get way out of hand...
Surface Combat
Moderators: Joel Billings, Tankerace, siRkid
Didz
Fortis balore et armis
Fortis balore et armis
Paul wrote:
"If a TF "Acts like minefield" and sits in a hex the entire turn, its easier
to quickly figure out, that is the line along which possible future tweaks
may be made, so you can patrol a choke point by begining the turn there
with no movement orders (patrolling)."
This would be a very big improvement over what we have now.
And it would avoid the raised issue of TF's chasing each other
all over the place in some sort of goofy chain reaction.
As long as the player can send a surface combat TF to any hex
he wants with orders to "react to enemy", so that if the TF spots
enemy ships passing through that hex, the commander will make
his best effort to intercept and engage those ships.
P.S. I want to further clarify what I mean, since it might have
come across wrong above.
Surface combat TF's need the ability to intercept enemy TF's whether
either of the TF's are in mid course or at a destination hex.
Making TF's like mines once they get somewhere and don't
move doesn't help, because I believe it's been pointed out
that TF's can already be ordered to go to a destination hex
and have the TF react once they are there sitting at the
destination hex. So that wouldn't change anything.
If the player sends a surface combat TF up the slot and a
jap CA TF is coming down the slot, there should be a
chance that they will spot each other once they both enter
the same hex as they travel along their routes. Once one TF
spots the other, and if the TF has previously been
given orders to engage enemy ships, then the commander
would try to intercept and fire on the enemy TF.
If the other TF has orders to retire (or if the commander decides
he had better run because he's not a very aggressive CO or
because he sees he's outclassed), and if it's speed is close to, equal
to or better than the attacking TF, then there is a very good
chance that only a few shots would be traded, if any. But if the
evading TF is significantly slower than the attacking one (or if the
other TF commander also chooses to engage), then there should
be a real chance that they can force the fight on the fleeing TF.
The exact speed difference between each TF, commander
experience, weather, time of day, etc. would obviously play a
big role in all this.
And this would only apply to TF's that pass through the same hex,
not other nearby hexes. This will keep potential encounters from
becoming common, and yet provide for realistic battle opportunities
between surface combat TF's that want to fight it out
(or slow TF's that can't get away from the attacking TF).
"If a TF "Acts like minefield" and sits in a hex the entire turn, its easier
to quickly figure out, that is the line along which possible future tweaks
may be made, so you can patrol a choke point by begining the turn there
with no movement orders (patrolling)."
This would be a very big improvement over what we have now.
And it would avoid the raised issue of TF's chasing each other
all over the place in some sort of goofy chain reaction.
As long as the player can send a surface combat TF to any hex
he wants with orders to "react to enemy", so that if the TF spots
enemy ships passing through that hex, the commander will make
his best effort to intercept and engage those ships.
P.S. I want to further clarify what I mean, since it might have
come across wrong above.
Surface combat TF's need the ability to intercept enemy TF's whether
either of the TF's are in mid course or at a destination hex.
Making TF's like mines once they get somewhere and don't
move doesn't help, because I believe it's been pointed out
that TF's can already be ordered to go to a destination hex
and have the TF react once they are there sitting at the
destination hex. So that wouldn't change anything.
If the player sends a surface combat TF up the slot and a
jap CA TF is coming down the slot, there should be a
chance that they will spot each other once they both enter
the same hex as they travel along their routes. Once one TF
spots the other, and if the TF has previously been
given orders to engage enemy ships, then the commander
would try to intercept and fire on the enemy TF.
If the other TF has orders to retire (or if the commander decides
he had better run because he's not a very aggressive CO or
because he sees he's outclassed), and if it's speed is close to, equal
to or better than the attacking TF, then there is a very good
chance that only a few shots would be traded, if any. But if the
evading TF is significantly slower than the attacking one (or if the
other TF commander also chooses to engage), then there should
be a real chance that they can force the fight on the fleeing TF.
The exact speed difference between each TF, commander
experience, weather, time of day, etc. would obviously play a
big role in all this.
And this would only apply to TF's that pass through the same hex,
not other nearby hexes. This will keep potential encounters from
becoming common, and yet provide for realistic battle opportunities
between surface combat TF's that want to fight it out
(or slow TF's that can't get away from the attacking TF).
-
juliet7bravo
- Posts: 893
- Joined: Wed May 30, 2001 8:00 am
"looping"; OP points are already part of the game. What if a TF that reacts expended OP points...checking OP points could be part of the pre-reaction move check. TF's refueling, expended all their movement points for the turn, or that have just completed a reaction move would flunk this test. All newly created TF's could be automatically defaulted to "do not react" to help reduce the number of TF's needing to be "checked" on any given turn.
I think eveyone agree's that in an ideal world TF's would be capable of interception at any point along their movement path.Originally posted by Dunedain
Surface combat TF's need the ability to intercept enemy TF's whether
either of the TF's are in mid course or at a destination hex.
Making TF's like mines once they get somewhere and don't
move doesn't help, because I believe it's been pointed out
that TF's can already be ordered to go to a destination hex
and have the TF react once they are there sitting at the
destination hex. So that wouldn't change anything.
However, as Paul was quick to point out UV has not been designed with the capability to do that and building it in after the event is going to be impractical.
My idea to use the MINE sub-routine to test for the existence of a TF along the route of the subject TF was a compromise solution that gave a limited opportunity the intercept a moving TF with a carefully placed Surface force.
Thus in your example placing a strong Cruiser TF on overwatch in the slot would produce a good chance that any enemy convoy coming down it would trigger a surface intercept and a battle.
Personally, I am prepared to accept that two opposing TF's both en-route somewhere else are more likely to slip past each other than if one of them was specifically assigned to patrol and interception duties.
Again! at the risk of flogging this dead horse even more I must just add that if the turn length was reduced to 12 hours then this problem would largely disappear simply because the TF's would probably end their moves within reaction distance of each other.
Didz
Fortis balore et armis
Fortis balore et armis
juliet7bravo: Those are some good ideas on how to realistically limit
when surface combat TF's could attempt to intercept enemy TF's.
A surface combat TF that just finished trying to engage an enemy
TF but the enemy TF got away, is likely not going to be in a good
position to attempt another interception for awhile. On the next impulse,
they could try again, if an enemy TF was present in their current hex
during that impulse. Whether either TF was moving through the hex
or stopped in that hex shouldn't matter, though.
when surface combat TF's could attempt to intercept enemy TF's.
A surface combat TF that just finished trying to engage an enemy
TF but the enemy TF got away, is likely not going to be in a good
position to attempt another interception for awhile. On the next impulse,
they could try again, if an enemy TF was present in their current hex
during that impulse. Whether either TF was moving through the hex
or stopped in that hex shouldn't matter, though.
-
Paul Vebber
- Posts: 5342
- Joined: Wed Mar 29, 2000 4:00 pm
- Location: Portsmouth RI
- Contact:
I think that would quickly get some folks to attemt to "soak off" op points form a large task force by getting it chase after smaller fast TFs sent in front of the "main body" while there is som potential historic value in such, I think for teh most part it would get players thinking up ways to 'game the system' The current system is bassically a big randomizer, and while it has its problems in choke points, I think in most games reported, you are seeing fairly historical numbers of "blue water" surface actions and an historically difficult time chaseing down damaged survivors.
-
juliet7bravo
- Posts: 893
- Joined: Wed May 30, 2001 8:00 am
Yes, but generally you have to expend a serious amount of time, effort and tedium trying to make those engagements happen. Plotting intercepts on a specific hex by hex basis is tactical. It's either an operational level game or a tactical level game, one or the other, and right now it's really neither one.
"Soaking off" reaction/guard forces is strictly historical, and that is precisely why they are reactive forces. The solution is very easy, in game terms you have a main body set "do not react" squatting on the OBJ. The enemy can fritter away small elements trying to suck off your reactive elements and weaken your forces. Alternatively he has the option of sending in heavier forces in an effort to chop up your light reactive forces in detail. As the Operational Commander you would have the option of making the reactive forces just as powerful or weak as you like in responce, with orders set as you desire, or to sortie with the main body...this is "operational control", and is a vast improvement in gaming pleasure and realism over "I'll set a mother huge TF on top the base so that anything he sends in has to fight it". Besides, at sea intercepts shouldn't happen each time, every time for the reasons put forth here, and all the factors leading to a react/don't react decision should come into play during the reaction check.
"Soaking off" reaction/guard forces is strictly historical, and that is precisely why they are reactive forces. The solution is very easy, in game terms you have a main body set "do not react" squatting on the OBJ. The enemy can fritter away small elements trying to suck off your reactive elements and weaken your forces. Alternatively he has the option of sending in heavier forces in an effort to chop up your light reactive forces in detail. As the Operational Commander you would have the option of making the reactive forces just as powerful or weak as you like in responce, with orders set as you desire, or to sortie with the main body...this is "operational control", and is a vast improvement in gaming pleasure and realism over "I'll set a mother huge TF on top the base so that anything he sends in has to fight it". Besides, at sea intercepts shouldn't happen each time, every time for the reasons put forth here, and all the factors leading to a react/don't react decision should come into play during the reaction check.
-
Paul Vebber
- Posts: 5342
- Joined: Wed Mar 29, 2000 4:00 pm
- Location: Portsmouth RI
- Contact:
The solution is very easy, in game terms you have a main body set "do not react" squatting on the OBJ.
Thats basically how it works now?
I thought the problem was with units not intercepting "while en-route" somewhere? I sounds like your talking more about choke-point barriers?
If you want to block a choke point, then the "Stationary TF as minefield" works pretty well, with the ability to try to 'sucker it" into going after fodder while your main force slipsby, unless it makes short shrift ot he fodder and turns back.
Once you get out of a chocke point situation the difficulty bringing an enemy TF to battle is daunting and I'm not sure expending a lot of CPU cycles to check for situations that might occur 3 or 4 times in a game is really worth it?
I'm not saying its a bad idea, just looking at "bang for programming buck" value.
What do you feel "all teh factors are" that need to be considered in the reaction check?
-
juliet7bravo
- Posts: 893
- Joined: Wed May 30, 2001 8:00 am
"The solution is very easy, in game terms you have a main body set "do not react" squatting on the OBJ."
That is how it works, period. That isn't very realistic or "fun", to be basically shuttling TF's from port to port.
From what you're saying, interceptions "en route" would be prohibitive in CPU time...I can understand and live with that.
By a limited reaction move, I (and most people here I think) are talking about being able to station a TF in hex X/Y, with "react to" and "patrol" orders and having some hope of it making a limited reaction move to close with and engage the enemy instead of putzing around trying to guesstimate where they're going and getting your TF into the same hex manually. This could take place as part of the end of turn sequence...no moving parts. Currently only Air TF's will do so against enemy air TF's, and they usually do it and end up deep into enemy AZOC's.
I can think of scores of opportunities for surface TF reaction moves occuring every game. I've had surface TF's end up in hexes adjoining or closely adjoining enemy TF's numerous times, even Carrier TF's socked in by weather...they might as well be on Mars.
Checks...OP check, Commanders aggression level check, sighting reports and how "solid" they are, weather, radar, fuel state, enemy AZOC. That would do for starters.
I know there's movement checks going on already...you have OP checks, "fleeing from enemy carrier" checks, carrier reaction checks...
Two "operational" fixes (1) a limited EOT reaction moves, (2) being able to set limits to carrier reaction moves, would, IMO add a GREAT deal to the game from the operational end.
That is how it works, period. That isn't very realistic or "fun", to be basically shuttling TF's from port to port.
From what you're saying, interceptions "en route" would be prohibitive in CPU time...I can understand and live with that.
By a limited reaction move, I (and most people here I think) are talking about being able to station a TF in hex X/Y, with "react to" and "patrol" orders and having some hope of it making a limited reaction move to close with and engage the enemy instead of putzing around trying to guesstimate where they're going and getting your TF into the same hex manually. This could take place as part of the end of turn sequence...no moving parts. Currently only Air TF's will do so against enemy air TF's, and they usually do it and end up deep into enemy AZOC's.
I can think of scores of opportunities for surface TF reaction moves occuring every game. I've had surface TF's end up in hexes adjoining or closely adjoining enemy TF's numerous times, even Carrier TF's socked in by weather...they might as well be on Mars.
Checks...OP check, Commanders aggression level check, sighting reports and how "solid" they are, weather, radar, fuel state, enemy AZOC. That would do for starters.
I know there's movement checks going on already...you have OP checks, "fleeing from enemy carrier" checks, carrier reaction checks...
Two "operational" fixes (1) a limited EOT reaction moves, (2) being able to set limits to carrier reaction moves, would, IMO add a GREAT deal to the game from the operational end.
-
Paul Vebber
- Posts: 5342
- Joined: Wed Mar 29, 2000 4:00 pm
- Location: Portsmouth RI
- Contact:
J7B - You make good points and I sense we are arguing in circles a bit. There won't be any thing related to this in this patch I don;t think, it would probably take a week or two of coding and at least a month of testing to do even a minor change of this sort and be sure it ws fixing more than it was breaking. IT has implications of the AI too tht would be hard.
But a variety of things are under consideration for the future.
But a variety of things are under consideration for the future.
I think that reacting to surface forces would be a nice addition to the game.
However, I've been pretty successful in forcing interceptions of crippled ships. You have a pretty good idea where the enemy is going and in restrict waters they have limited choices.
I generally pick a point 4 hexs from the current location back the enemies main base. If the TF is slightly damaged it will be going speed 4 and you'll catch at night. If it is severely damaged it will be going speed 2 and you'll have a day surface battle. Yes sometimes the go speed 3 and you miss them completely or they use a different route but... That's war plans for you.
However, I've been pretty successful in forcing interceptions of crippled ships. You have a pretty good idea where the enemy is going and in restrict waters they have limited choices.
I generally pick a point 4 hexs from the current location back the enemies main base. If the TF is slightly damaged it will be going speed 4 and you'll catch at night. If it is severely damaged it will be going speed 2 and you'll have a day surface battle. Yes sometimes the go speed 3 and you miss them completely or they use a different route but... That's war plans for you.
-
juliet7bravo
- Posts: 893
- Joined: Wed May 30, 2001 8:00 am
Paul...understood. Rome wasn't built in a day, and the next patch sounds like it's going to tweak a variety of issues. I do think these "Command and Control" issues are the 2 biggest issues that need to be discussed and sorted out in the future.
strollen; yeah, it works sometimes, but that's taking tactical command. You're probably always have instances were you need to do so, but in an "operational level" game the ships/TF Commanders oughta do more than sit there like lumps waiting for you to hand carry them around. If you can get them in approximately the right location and with the right orders, your TF commanders ought to have some limited autonomous ability to have/take the initiative to close that last 1-3 hex gap on their own.
strollen; yeah, it works sometimes, but that's taking tactical command. You're probably always have instances were you need to do so, but in an "operational level" game the ships/TF Commanders oughta do more than sit there like lumps waiting for you to hand carry them around. If you can get them in approximately the right location and with the right orders, your TF commanders ought to have some limited autonomous ability to have/take the initiative to close that last 1-3 hex gap on their own.
Paul
Please make any significant change to the current system an optional rule. I'm sure there are others like me who are happy with the current method of initiating surface combat and are worried about the potential for "gaming the system" that allowing picket line tactics and mid-turn interceptions will create. Thanks.
elmo3
Please make any significant change to the current system an optional rule. I'm sure there are others like me who are happy with the current method of initiating surface combat and are worried about the potential for "gaming the system" that allowing picket line tactics and mid-turn interceptions will create. Thanks.
elmo3
We don't stop playing because we grow old, we grow old because we stop playing. - George Bernard Shaw
WitE alpha/beta tester
Sanctus Reach beta tester
Desert War 1940-42 beta tester
WitE alpha/beta tester
Sanctus Reach beta tester
Desert War 1940-42 beta tester
-
juliet7bravo
- Posts: 893
- Joined: Wed May 30, 2001 8:00 am
The current system is totally gamey and already allows itself to be "gamed" in a variety of ways. One key way that strikes me offhand is the current thread about attacking major bases...just move to an empty beach next/close to the port and unload. Bammo...now you've created what is effectively a new base side by side with the base you're attacking and can unload fuel/supplies (at the same rate you could inside the enemy base during a frontal assault), base ships out of it, your screening ships can "anchor" in it for long periods, shuttle next door to bombard the enemy base without taking system damage from being "at sea", repair damage, plus resupply fuel/ammo each turn. You've did it totally without risk of interception by enemy surface units, and your transport TF's can move in/out of it unescorted by anything other than an ASW element at will without even the slightest risk of interception by enemy surface units. Now THAT is "gaming the system".
It's an "operational level" game, let's focus on operations and let our TF commanders earn their pay by making some tactical decisions.
It's an "operational level" game, let's focus on operations and let our TF commanders earn their pay by making some tactical decisions.
I'm no expert on the Solomons Campaign. So can anyone give a couple of examples, even one, where an intentional interception and subsequent surface combat occurred more than 30 miles from a base (i.e. not in a base hex) during the campaign? I'm not talking about hunting down cripples. Even if there were several then we're still only talking about such an event happening once each 100 days of the time period covered by the campaign.
If that is a good rough estimate then that would make implementing such a feature a very low priority IMO. It's all well and good if the game allows people to explore "what-if" type surface encounters as an option, but it should not allow an unrealistically high number of them as a matter of course.
If that is a good rough estimate then that would make implementing such a feature a very low priority IMO. It's all well and good if the game allows people to explore "what-if" type surface encounters as an option, but it should not allow an unrealistically high number of them as a matter of course.
We don't stop playing because we grow old, we grow old because we stop playing. - George Bernard Shaw
WitE alpha/beta tester
Sanctus Reach beta tester
Desert War 1940-42 beta tester
WitE alpha/beta tester
Sanctus Reach beta tester
Desert War 1940-42 beta tester
-
Wilhammer
- Posts: 401
- Joined: Fri May 24, 2002 4:00 pm
- Location: Out in the Sticks of Rockingham County, North Caro
- Contact:
I can't think of any surface engagement that was fougt more than 20 miles from land in this campaign.
An example of what did happen in the war, that CANNOT happen in this game, due to the end point find only feature, First Savo Island.
Look at the map.
Lunga and Tulagi and the hex next to Lunga towrds Shortland.
That hex is Savo Island and Tassoforanga (?) Point.
In that hex, the Allied cruiser/DD force screened the Transports in Lunga hex, over 10 miles away.
The goal of the IJN was to get through that screen and hit those transports.
The screen in that hex in UV will not interfere if the plotted end point is Lunga.
The IJN will shoot throught Savo and hit Lunga.
The End Point Design Decision (EPDD) is most glaring at this zero to one hex passing in the night situation.
Actually, the "excitement" of the slot is fairly mitigated by this design.
Ships won't go "Bump" in the night, so to speak.
An example of what did happen in the war, that CANNOT happen in this game, due to the end point find only feature, First Savo Island.
Look at the map.
Lunga and Tulagi and the hex next to Lunga towrds Shortland.
That hex is Savo Island and Tassoforanga (?) Point.
In that hex, the Allied cruiser/DD force screened the Transports in Lunga hex, over 10 miles away.
The goal of the IJN was to get through that screen and hit those transports.
The screen in that hex in UV will not interfere if the plotted end point is Lunga.
The IJN will shoot throught Savo and hit Lunga.
The End Point Design Decision (EPDD) is most glaring at this zero to one hex passing in the night situation.
Actually, the "excitement" of the slot is fairly mitigated by this design.
Ships won't go "Bump" in the night, so to speak.
Unfortunately, as has been documented elsewhere the TF Commanders can't be trusted to make sensible decisions. So, most players are switching 'React to enemy' and 'Retirement Allowed' off.Originally posted by juliet7bravo
It's an "operational level" game, let's focus on operations and let our TF commanders earn their pay by making some tactical decisions.
Didz
Fortis balore et armis
Fortis balore et armis
-
HMSWarspite
- Posts: 1405
- Joined: Fri Apr 12, 2002 10:38 pm
- Location: Bristol, UK
Dont follow you. You cannot repair ships at most beaches - they will be port 1 or 0. You can only refuel/rearm if you ship in supplies/fuel. This supply hex can then be attacked by the enemy (air/naval). TF in a hex can still be attacked, even if docked - in fact that is the only time most level bombers (except naval specialists like Betty's) will hit them. Walking 30 miles to attack the next hex will really take the edge off most troops, hence isn't a cake walk either. TF can be intercepted in either hex, since they will start in one, and finish in the other (unless the unload time is very short, in which case they would do better to be well out at sea heading for home at the end of the turn). Finally, the defender can of course, defend the beach in question by shipping supplies, garrisoning and fortifying it. What would be gamey is to force the attacker always to assault frontally in a 30mile hex!Originally posted by juliet7bravo
The current system is totally gamey and already allows itself to be "gamed" in a variety of ways. One key way that strikes me offhand is the current thread about attacking major bases...just move to an empty beach next/close to the port and unload. Bammo...now you've created what is effectively a new base side by side with the base you're attacking and can unload fuel/supplies (at the same rate you could inside the enemy base during a frontal assault), base ships out of it, your screening ships can "anchor" in it for long periods, shuttle next door to bombard the enemy base without taking system damage from being "at sea", repair damage, plus resupply fuel/ammo each turn. You've did it totally without risk of interception by enemy surface units, and your transport TF's can move in/out of it unescorted by anything other than an ASW element at will without even the slightest risk of interception by enemy surface units. Now THAT is "gaming the system".
.
Why is landing next door to the objective 'gamey'?
I have a cunning plan, My Lord
-
HMSWarspite
- Posts: 1405
- Joined: Fri Apr 12, 2002 10:38 pm
- Location: Bristol, UK
You're thinking too precisely for the scale of the game. Savo Island can (in effect) happen just as reality. There is a slight abstraction in that (if you are correct as to the placement of Savo vs Lunga, and the hex boundary), the battle takes place 'slightly nearer the landing' (one hex over). The effect is just the same - hence no problem with the simulation. BTW a good naval battle at 1+hr and an average speed of close to 30 kts will actually take place in 2 or more hexes, and the game doesn't take account of that either. No big deal, the effect is captured.Originally posted by Wilhammer
I can't think of any surface engagement that was fougt more than 20 miles from land in this campaign.
An example of what did happen in the war, that CANNOT happen in this game, due to the end point find only feature, First Savo Island.
Look at the map.
Lunga and Tulagi and the hex next to Lunga towrds Shortland.
That hex is Savo Island and Tassoforanga (?) Point.
In that hex, the Allied cruiser/DD force screened the Transports in Lunga hex, over 10 miles away.
The goal of the IJN was to get through that screen and hit those transports.
The screen in that hex in UV will not interfere if the plotted end point is Lunga.
The IJN will shoot throught Savo and hit Lunga.
The End Point Design Decision (EPDD) is most glaring at this zero to one hex passing in the night situation.
Actually, the "excitement" of the slot is fairly mitigated by this design.
Ships won't go "Bump" in the night, so to speak.
You could, of course, set up the 'perfect' simulation. Just get yourself a network or really high spec PCs, and about 50 people a side, wait for someone to do about a zillion man years of programming, and than take about as long as the real thing to play it......
I have a cunning plan, My Lord
-
Wilhammer
- Posts: 401
- Joined: Fri May 24, 2002 4:00 pm
- Location: Out in the Sticks of Rockingham County, North Caro
- Contact:
Refueling = Rearming
The rules state that you reload ammo in two ways:
1. Disband a TF in level 3 or greater port.
2. Refuel a TF.
Concerning 2, the rules are not clear on this.
Will a ship get reammoed if refueled at sea?
If that is the case, then we have a problem.
I can see getting some new ammo from a Replenish convoy, but not from refueling from say a BB or CV.
And, if you can via a Replenish Convoy, you should not get much ammo.
BTW, how many Supply Points are used when you get ship ammo?
Any at all?
1. Disband a TF in level 3 or greater port.
2. Refuel a TF.
Concerning 2, the rules are not clear on this.
Will a ship get reammoed if refueled at sea?
If that is the case, then we have a problem.
I can see getting some new ammo from a Replenish convoy, but not from refueling from say a BB or CV.
And, if you can via a Replenish Convoy, you should not get much ammo.
BTW, how many Supply Points are used when you get ship ammo?
Any at all?