Devs: Please explain Ion Railgun vs. Ion Missile discrepancies

[Verain]

The ion cannon really is the best weapon for stripping shields.

 

I don't think this conclusion follows from your given.

 

The dps of ion cannon versus shields is the highest of anything in the game- 2k is a good approximation. But ion cannon suffers from an EXTREMELY close range. Also note that the "1 to switch weapon" thing is rather disruptive to your firing pattern- much more so than on a gunship or Type 2 Strike, as it interrupts the flow of the shots (on a gunship you have a reload time, and swapping during that guarantees that it disrupts nothing- on the type 1 strike, you need to reclick the button to get your shots going again, and there seems to be some kind of delay value at work, likely to prevent it from somehow increasing rate of fire).

 

Again, I don't think this is very important for talking about ion railgun. But I will say that in my experience, ion railgun strips the most shields in any game I gunship. If I run my type 1 strike fighters (both of which run mastered ion cannons), that is a niche ability- if I find myself close to someone who has shields, then I use them (and they are very powerful), but often they are incorrect versus a full shield opponent for other reasons. Even in comparison to total damage dealt and discarding aoe, I definitely deal more shield damage with ion railgun than with ion cannon. Opening with a generous ion railgun shot eliminates a lot of shields, and a full charge would normally clear it. Then swap to slug. But versus many enemies on the strike fighter, I will often be fighting an enemy out of ion cannon range, and their shields get stripped by concussion missile, cluster missile, or heavy laser. Only my "Whiskey Gladiator" build (ion/rapid/cluster) features a decent amount of ion damage, and even then, some of the shields are devoured by clusters- and only because this ship is fully committed to close combat.

Edited March 25, 2014 by Verain

[Kuciwalker]

Yes yes there are a bunch of fundamentally unquantifiable differences between cannon and railgun. Nevertheless, I feel like the cannon is superior to the railgun qua shield-destroying by the following measure: if the status debuffs were removed from the railgun, I would prefer a strike fighter (with ion+heavy) to a gunship for bomber-killing.

 

The short range of the ion cannon is also compensated for by the wide firing arc, lowish tracking penalty, and high rate of fire (which makes even low-accuracy shots good).

[Verain]

Strikes are good against bombers, but it's certainly not due to their short range armaments, but to their long. My Starguard is almost never on ions versus a bomber- heavies are a much better choice. Versus a minelayer, it's very important not to get in melee- he's a bear. Versus a dronecarrier, sure- she's a mother hen who's egg has a railgun, her defense isn't melee.

 

 

Remember that bombers are exemplary for their hull. Their base shield value is the same as a strike fighter, though they are very likely to have cooldowns to increase shields.

 

 

I just don't think that ion cannon is that good against bombers.

[Nemarus]

 

Remember that bombers are exemplary for their hull. Their base shield value is the same as a strike fighter, though they are very likely to have cooldowns to increase shields.

 

 

Actually Bomber base shield capacity/regen is worse than everything but Scout.

 

Base Shield values are:

 

Strike: 1800 capacity / 95 regen

Gunship: 1700 / 85 regen

Bomber: 1500 / 75 regen

Scout: 1300 / 65 regen

 

The trick is that Bombers have several components which significantly boost Shield Capacity: Overcharged Shields, Charged Plating, Engine-to-Shield Converter, Large Reactor. Plus E2S Converter let them recharge shields quickly, even if they have inherently slow regen. The only Bombers that don't take E2S Converter are Minelayers who opt for a Hyperspace Beacon.

Edited March 25, 2014 by Nemarus

[Zoom_VI]

The ion cannon has a huge damage per shot and an even higher rate of fire. If one of your shots lands the second is likely to also just because of the ROF. The ion cannon really is the best weapon for stripping shields.

 

The problem is that both weapons strip so much shields that it doesn't really matter. If you are in a situation where your purpose is to kill a opponent as fast as possible, and ignoring outside variables, your never going to ion him twice since the first shot strips so much of his shields its better to just go to slug for the second hit.

 

The other issue with ion cannons is that the cooldown when switching weapons completely murders the advantage. The time it takes to strip their shields with ion cannon added on top of the cooldown adds up to about as much time as it takes to strip their shields with a standard weapon. Plus since Ion cannon doesn't wipe out half of the target's power reserves, there sin't exactly a huge advantage to using it over the other primaries.

[Kuciwalker]

I don't agree. I find myself frequently using ions on my strike and being thankful for them. They are especially valuable when killing satellite orbiters.

[Zoom_VI]

I don't agree. I find myself frequently using ions on my strike and being thankful for them. They are especially valuable when killing satellite orbiters.

 

But are they as effective as your railgun?

[Altheran]

But are they as effective as your railgun?

 

For killing purpose, more than Railgun. For anything else no.

Why ? Because it has no meaningful side effect, so Railgun has purposes that Cannon doesn't. But for killing, Ion Railgun doesn't improve my ability to kill much. In the other hand, Ion Cannon, allow me to kill way faster with them than without them.

[Ramalina]

Ok this is the TL;DR version because the long analysis got inadvertently deleted. Yeah, I know, it's still pretty long.

 

All of the ion weapons are secondary in role if not in slot. You wouldn't equip any of them as an acceptable sole weapon for a starfighter.

 

If you do the math, the vs Shield DPS is roughly equal for all three ion weapons at full upgrades, though ion cannons suffer a bit from drop off at range.

 

The DPS vs Hull is crap for all three at present, though if Dulfy's 2.7 build calculator is correct in anticipating changes then Ion Missiles, well they'll still have crap DPS, but their damage per missile will almost be halfway decent.

Until such changes though, ion weapon dps is so low that it's not worth worrying about. Unless you're just complaining for the sake of complaining.

 

None of them are excessively difficult to use at appropriate ranges. Ease of use is greatest for the railgun, for missiles and cannons relative difficulty depends quite a bit on the particular opponent and situation.

 

Sustained rate of fire favors the Ion railgun. Flying to range with ion cannons and missile reload times decrease the effective rate of fire more often than having to reposition affects the railgun's rate of fire. In a swarm of closely packed fighters the ion cannon pulls ahead, but those are much less common than they were pre-2.6.

**Edit: The difference though, is probably not big enough to be a problem.

 

Engine power drain per second. This is the special ability that causes the most aggravation. The ion railgun drains 18 to 29 engine power per second depending on how charged the shots are. It allows two shots (or three if the target has an engine power cooldown) to render the target almost completely helpless for 6 seconds using the normal upgrade choices. The ion missile, based on estimates that it's drain is of similar size to a railgun's, manages a measly 4 energy drain per second, but it does all come in one fairly large lump. With max upgrades it can potentially disable a target as long as the target had no more than 60 to 70% engine power, and no engine power cooldowns. The ion cannon manages to drain about 11 engine power per second and doesn't interfere with regeneration. It's a bit irrelevant though because firing ion cannons at the target for long enough to drain the target's engine pool is usually an inferior choice to switching to a high hull dps weapon. In which case the engine drain of the ion cannons is negligible (12 to 24 engine power drained if you switch when shields go down).

 

For the most part I think the differences are acceptable ship class based tradeoffs. To bring Ion cannons in line with other disabling/special effect weapons it's really just the engine drain rate that needs a beating from the nerf bat. At full charge, the engine drain is somewhere between extremely strong but acceptable and slightly overdone. At minimum charge, it's wildly overpowered. By maximum rate it's 7 times better than ion missiles, and almost 3 times better than ion cannons. In terms of mechanics, it's about twice as good as ion missiles, and almost infinitely better than ion cannons.

 

The underlying design theory is that all the ion weapons should strip shields and make it more difficult for the target to escape being hit by follow up shots. The potential balance issue, is that while the ion missile and ion cannon can make it moderately or slightly more difficult to escape, the ion railgun can make it virtually impossible to escape. Unspammable big drain is ok, spammable small drain is ok, spammable big drain is not ok.

 

My suggestion would be to look at keeping the maximum theoretical rate of engine drain below 20/second or maybe even 15/second and not allowing a single hit drain to exceed 60 engine power (though I imagine most strike and scout pilots would prefer a per-hit cap in the 30-40 engine power range, enough to force them onto defense but not enough to make defense impossible). I think my first guess would be to look at a 40-45 power drain/shot cap for railguns with max average drain rate set to about 20/s. Then test that internally or on a PTS build to see how it plays and evaluate for futher adjustments. The goal being that after two shots, a target at full would be able to boost a short distance, but probably wouldn't have enough power to use an engine component evasive maneuver.

 

I'll also note that if engine drain on ion railgun isn't brought a bit more into line, the effect and the complaints will likely grow after the increases to evasive maneuver cooldowns and costs go live.

Edited March 25, 2014 by Ramalina

[Nemarus]

Ok this is the TL;DR version because the long analysis got inadvertently deleted. Yeah, I know, it's still pretty long.

 

<snip>

 

Nice breakdown!

[Gavin_Kelvar]

My suggestion would be to look at keeping the maximum theoretical rate of engine drain below 20/second or maybe even 15/second and not allowing a single hit drain to exceed 60 engine power (though I imagine most strike and scout pilots would prefer a per-hit cap in the 30-40 engine power range, enough to force them onto defense but not enough to make defense impossible).

 

I'd also add not completely blocking regen (or if it's going to be a large drain that can have relatively quick follow up shots have no block to regen). If you retained regen abilities the large drain would be powerful (perhaps still a bit too powerful) but not completely broken. The ability to prevent any regen is also a problem.

 

Otherwise I agree with your analysis.

[Ramalina]

Yeah, well the logical follow on from my conclusion it that the likely balancing option for ion railguns would be to reduce damage and drain stats but slightly increase the rate of fire, because that's probably the easy way to smooth things out a bit if you don't go with a linear (or close stepwise) scaling of engine drain with railgun charge.

 

Scaling would be the best solution, but given that they introduced a minimum charge instead of scaling last time they adjusted, I sort of suspect that their existing code makes implementing scaling a real pain in the ***.

 

Scaling would also probably generate a lot less complaints than other options, but given that it didn't happen in the last adjustment to ion railguns it makes me suspect that there are barriers to doing it. Though come to think of it, I wonder why they can't just recycle the damage with distance scaling code that's used for blaster damage and apply it to ion railgun engine drain and charge level at time of firing.

 

If there aren't huge barriers I think the devs should argue for doing it, because then you have a nice block of scaling code that can probably be recycled elsewhere.

 

Still, not optimistic that that will happen.

 

Gavin, as far as regen afterwards goes, I'm not sure that's a big issue. If you get hit by an ion weapon, the goal is to break contact before the next shot hits. Regen at 'recently consumed' rates doesn't help with that very much. It's more a matter of, do you have enough to use an engine maneuver, and if not can you at least boost to try to break LOS if there's cover within a few km or the possibility of briefly making it out of the attacker's weapons' ranges. For non-cooldown based 'recent' regen, by the time it would make a difference you've probably already either escaped or died.

 

Where the regen comes into play is a bit later, if you've escaped the initial attack, how prepared are you to continue if the attacker pursues to press the attack.

Edited March 25, 2014 by Ramalina

[Nemarus]

Yeah, well the logical follow on from my conclusion it that the likely balancing option for ion railguns would be to reduce damage and drain stats but slightly increase the rate of fire, because that's probably the easy way to smooth things out a bit if you don't go with a linear (or close stepwise) scaling of engine drain with railgun charge.

 

Scaling would be the best solution, but given that they introduced a minimum charge instead of scaling last time they adjusted, I sort of suspect that their existing code makes implementing scaling a real pain in the ***.

 

Scaling would also probably generate a lot less complaints than other options, but given that it didn't happen in the last adjustment to ion railguns it makes me suspect that there are barriers to doing it. Though come to think of it, I wonder why they can't just recycle the damage with distance scaling code that's used for blaster damage and apply it to ion railgun engine drain and charge level at time of firing.

 

If there aren't huge barriers I think the devs should argue for doing it, because then you have a nice block of scaling code that can probably be recycled elsewhere.

 

Still, not optimistic that that will happen.

 

Gavin, as far as regen afterwards goes, I'm not sure that's a big issue. If you get hit by an ion weapon, the goal is to break contact before the next shot hits. Regen at 'recently consumed' rates doesn't help with that very much. It's more a matter of, do you have enough to use an engine maneuver, and if not can you at least boost to try to break LOS if there's cover within a few km or the possibility of briefly making it out of the attacker's weapons' ranges. For non-cooldown based 'recent' regen, by the time it would make a difference you've probably already either escaped or died.

 

Where the regen comes into play is a bit later, if you've escaped the initial attack, how prepared are you to continue if the attacker pursues to press the attack.

 

They obviously already scale Railgun damage with charge, so I don't think the scaling logic is the problem.

 

Rather, I think having effects with variable magnitude or duration is the difficulty, as I can't think of any other buffs or debuffs that work that way.

[Altheran]

Rather, I think having effects with variable magnitude or duration is the difficulty, as I can't think of any other buffs or debuffs that work that way.

 

May be they can make it work that way : the more time spent charging, the more it builds stacks of a buff (not shown on UI). Stacks are consumed when firing. Effect amplitude would depend on number of stacks.

It would not scale smoothly, but at least would give the feel of a scaling.

Edited March 25, 2014 by Altheran

[Ramalina]

Engine power isn't a buff though, it's a finte resource pool, the same as health. In terms of algorithms, once you get past the variable name, it shouldn't be any different than damage and health pools. So unless some strange things were done, applying the weapons damage to distance code to scaling engine power drain based on charge percentage should just be a matter of changing some coefficients and parameters, and a few variable name substitutions.

 

I was going to write an example, but it would take longer than I feel like for something that's not going to be actual functional code. But assuming you know the code reasonably well, and that the code is well documented through commenting, and whatever change tracking system they use, I'd think that it's something that one person might be able to do in an afternoon or less. Of course, there may be a lot of extraneous things that hook into it somehow, and if that's the case I could see making sure that everything is connected, and connected properly, could be a real pain even if the actual algorithm for calculating results is fairly simple change.

 

Of course if they don't have well documented and commented code, well, then I feel quite sorry for them. They'd have to be crazy not to have good documentation though, given their update cycle.

 

**Edit: The above is an unclear mess.

Restated: The engine drain is just subtraction from the engine pool in the same way that hull damage is subtraction from the hull health pool. It's a straight up subtraction, buffs, debuffs, upgrades, etc would just be coefficients or operators in the algorithm that calculates how much to subtract. So I don't see why you'd try doing novel or complicated things under the hood. They already have a way to do engine pool subtractions, such as the cost of activating an engine ability. It's just that for scaled ion weapon damage you might call a subroutine instead of having a fixed value. Though really, they might already do that to account for things like efficiency modifications.

 

The hardest part would probably be not accidentally messing something else up in the process of making the change. More a matter of tedious attention to detail than difficulty.

Edited March 25, 2014 by Ramalina

[Gavin_Kelvar]

Gavin, as far as regen afterwards goes, I'm not sure that's a big issue. If you get hit by an ion weapon, the goal is to break contact before the next shot hits. Regen at 'recently consumed' rates doesn't help with that very much. It's more a matter of, do you have enough to use an engine maneuver, and if not can you at least boost to try to break LOS if there's cover within a few km or the possibility of briefly making it out of the attacker's weapons' ranges. For non-cooldown based 'recent' regen, by the time it would make a difference you've probably already either escaped or died.

 

Where the regen comes into play is a bit later, if you've escaped the initial attack, how prepared are you to continue if the attacker pursues to press the attack.

 

If the drain scaled or was only applied at a full charge that's all true that the goal becomes to try to break LOS before the next shot and in that regard it wouldn't be as bad. The problem as I see it is that you have a 1 second or so window to break LOS right now before the follow up shot that drains you completely hits. Then you're pretty much dead in the water for 6 seconds. That's where I saw one of the major problems: you have such a small window of opportunity to escape before getting stunlocked it may be too small for you to either reach cover or get out of range.

 

It just overall seems cheap to be able to fully drain an enemy in 2 hits and have them helpless for 6 seconds (especially since the 2nd hit doesn't have to be fully charged). At least with ion cannons and missiles you always have enough power to fight back when fleeing isn't an option. I grant that the regen after use isn't great but at least you can regen power to have some means of defending yourself or can hope that you'll regen enough to flee. Sometimes regening enough for 1 shot or 1 very quick boost makes the difference so the impact of completely stopping regen shouldn't be underestimated.

Edited March 25, 2014 by Gavin_Kelvar

[Ramalina]

They obviously already scale Railgun damage with charge, so I don't think the scaling logic is the problem.

 

Rather, I think having effects with variable magnitude or duration is the difficulty, as I can't think of any other buffs or debuffs that work that way.

 

Engine drain isn't a buff or debuff, it's subtraction from a resource pool. The basic internal coding should be almost the same as for other resource pool mechanics like health, hull, shields, force, ect.

 

Ha! There. Only took three tries and half a page worth of wasted writing to finally formulate a clear response. I must need to get some sleep.

[Kuciwalker]

Engine power isn't a buff though, it's a finte resource pool, the same as health. In terms of algorithms, once you get past the variable name, it shouldn't be any different than damage and health pools.

 

It's not a question of algorithms; it's a question of what behavior the engine supports. The engine might handle "damage" the same way as "energy drain", or it might (probably does) handle damage in a very special, privileged way, while handling energy drain through the same engine that handles all other attack effects (including in the ground game).

 

The code that resolves attack effects may simply not be able to access the "duration charging" variable in the current engine.

 

In fact, I regard the above as particularly plausible and even likely, since there are a lot of hints that the GSF engine is actually just a very radically modified version of the ground game engine.

 

edit: replying to your edit/clarification:

 

Engine drain isn't a buff or debuff, it's subtraction from a resource pool. The basic internal coding should be almost the same as for other resource pool mechanics like health, hull, shields, force, ect.

 

No, I very much doubt that it's handled the same way. Consider that regular damage involves a distance calculation and a critical roll.

Edited March 25, 2014 by Kuciwalker

[Ramalina]

It's not a question of algorithms; it's a question of what behavior the engine supports. The engine might handle "damage" the same way as "energy drain", or it might (probably does) handle damage in a very special, privileged way, while handling energy drain through the same engine that handles all other attack effects (including in the ground game).

 

The code that resolves attack effects may simply not be able to access the "duration charging" variable in the current engine.

 

In fact, I regard the above as particularly plausible and even likely, since there are a lot of hints that the GSF engine is actually just a very radically modified version of the ground game engine.

 

edit: replying to your edit/clarification:

 

 

 

No, I very much doubt that it's handled the same way. Consider that regular damage involves a distance calculation and a critical roll.

 

If you're radically modifying the engine, then the ability to do generalized code for resource pools really shouldn't be a problem.

 

If it's just different artwork and scripting rather than actual work on the engine then I could see pre-existing constraints being a big problem. But that's a matter of content creation vs programming.

 

So, if they're programming for GSF I don't think it should be a problem unless they've used a lot of really bad coding practice up to this point. If it's content creation for the ground game that just happens to look a bit different, then yeah, it could be hard to impossible if the engine wasn't particularly well designed.

[Zoom_VI]

No, I very much doubt that it's handled the same way. Consider that regular damage involves a distance calculation and a critical roll.

 

distance calculation and does it crit are likely functions of the damage class. if engine drain is handled similarly it would like be a separate class from damage but perhaps it inherits from the same base class, thus not having all the functions.

[Armonddd]

So, if they're programming for GSF I don't think it should be a problem unless they've used a lot of really bad coding practice up to this point.

 

swtor.exe uses two processes to facilitate an engine that won an award in 2006 but wasn't used in any released games until 2011.

 

The chances that things aren't held together by black magic and nightly sacrifices are essentially nil.

[Zoom_VI]

swtor.exe uses two processes to facilitate an engine that won an award in 2006 but wasn't used in any released games until 2011.

 

The chances that things aren't held together by black magic and nightly sacrifices are essentially nil.

 

Honestly armonddd, I think most of the people posting on how the "code" works are probably just pulling bullsh*t out of their ears and going with it.

[Kuciwalker]

If you're radically modifying the engine, then the ability to do generalized code for resource pools really shouldn't be a problem.

 

No one is suggesting in the least that designing the engine to do what you want from the beginning is hard. It's trivially easy! The suggestion is that they did not anticipate your use case and so did not design the engine to support it.

 

You're free to mouth off over how incompetent their programmers are or whatever but "did not anticipate every possible future use case" is the rule, not the exception.

Edited March 26, 2014 by Kuciwalker

[Kuciwalker]

distance calculation and does it crit are likely functions of the damage class. if engine drain is handled similarly it would like be a separate class from damage but perhaps it inherits from the same base class, thus not having all the functions.

 

You are making detailed and wildly unjustified assumptions about their code.

[Ramalina]

Well all my training in programming has been with FORTRAN in an engineering (and I mean real engineering not computer 'engineering')/scientific context. It's an area where code that produces good output may be recycled for 40 years or more, and my professors went to great lengths to show us why ALWAYS using good design practice matters in terms of saving yourself giant heaps of self-inflicted future misery.

 

Of course, the department felt strongly enough about that, that instead of shuffling us off to the Comp Sci folks the engineering computational classes were all done in house. There were a couple of upper division classes that we'd sometimes share with the math department. Mostly on how to deal with using computers to do kinds of math that computers aren't good at, without getting garbage for results.

 

Basically the idea of shackling your future options by writing bad code, it seems really abnormal to me.

 

Eh, I guess Brad Finney did a good job of brainwashing me.

 

If it is an old engine done by someone else, I hope they asked the most important question before taking that sort of thing on. . . .

"How, much are they paying?"

 

When it comes to fixing other peoples code any answer other than, "A lot," is the wrong answer.

 

**Edit looked up but did not buy the engine.

 

Also, at a guess, Bioware went with the Starting at $75,000 + revenue share option, rather than the $99/yr option. Meaning that in the event that you couldn't script charge based engine drain scaling using the stock engine tools, they should still be able to monkey around with the source code if they want to.

 

If it were $20/yr I'd download it and try using the default tools to build a GSF resources system mockup, but I'm not $100 worth of curious about it.

 

Also, if I'd paid $75k + for the engine, I'd be pretty pissed if it couldn't do a flexible interacting multi-resource system, given how much design flexibility is talked about in the engine's advertising material. It's hard to evaluate going just off of sales material, but it really looks like something the default engine should support through scripting. Otherwise the bits about create whatever kind of character system and mechanics you want, wouldn't make much sense.

Edited March 26, 2014 by Ramalina