My personal take on a very old dispute...

[Zarax]

If we need smaller slots, it's even easier. As I said, we are remaking the bodies in AR, so making small slots on medium and heavy bodies won't affect the overall design too much. I already thought of a way to do exactly that.

If they are made not to look wrong when with a single turret, I don't see why not using the same models. But toggling the availability of secondary turrets would be interesting, if possible.

In a perfect world we would have a unit design interface where you could place a turret at your discrection in one of several available weapon slots followed by the secondary weapons.

Since that is not feasible in the near future, what can be made is a system where we can have a main turret plus a secondary slot for small (and/or medium) bodies and 2 for heavy (3 for super heavy?).

Secondary weapons could require research like super cyborgs do, but with the single prereq of having researched the standard version (unlike super HPV borgs, that require super heavy gunners to be available). Those could be called "<weapon name> miniaturization" or something like that.

Research would be mostly a balancing device imho. Theorethically as soon as a cyborg can carry a certain weapon it should also be available as support too but let's leave that open for the balancing stage.

What we need to know right now is if it's possible, so that we don't waste our times discussing wether it would be good or not, how should it be called, what's the best design options, etc.

We need to know if the following is possible, before going any further with discussions, IMO:
Is it possible to have a separate turret category (secondary weapons) for certain connectors assigned to the bodies?

That requires source modifications, Zarel or some other dev can answer best on how hard it would be.

We can have that along with weight limitations (preferrably relative to the body's engine output), so that the player may choose only balanced combos.

Rather than weight limitations we could have different slot limits per body. If we make the additional weapons give very low (or even 0) extra HP the player will have to choose his favourite trade off between speed, price and firepower.

Is it possible to limit the firing angle to the frontal 180° (maybe even 210°, covering a bit of the sides as well) for all those secondary weapons or for specific connector's weapons?

I think that's already supported, at least there is code in weapon.txt that discriminates between fixed and rotating weapons.

Can we make it so that the secondary weapons only fire automatically (will fire at nearest targets, won't necessarily fire at the targeted object)? That would be useful to control the main weapon to perform its main role and have the secondary weapons automatic, so that they can fire at other threats while the main gun fires at your designated target.

From my multi turret tests it seems that this is the case already at least partially. A shorter range turret will fire indipendently from the main weapon.

I thank anybody who can answer those questions insightfully, and also thanks those who understand that knowing this is essential to get into further details. I think Zarax agrees on that.

~Olrox

Indeed we agree on most points.
The only thing I would like to add is that it would be also good to leave some flexibility within the starting specs (for example a greater number of slots than trunk balance would deem ideal) in order to give mods the possibility to further expand on it later.

[Zarel]

Since that is not feasible in the near future, what can be made is a system where we can have a main turret plus a secondary slot for small (and/or medium) bodies and 2 for heavy (3 for super heavy?).

That's way too many. 2 turret slots for light/medium, 3 for heavy, and 4 for super heavy? Seriously? That is way ridiculously too much. If you want four different turrets, just build four different tanks...

This is just... no! Why would you ever want that? Give me one good reason! One!

Research would be mostly a balancing device imho. Theorethically as soon as a cyborg can carry a certain weapon it should also be available as support too but let's leave that open for the balancing stage.

i.e. as soon as a weapon is researched, it will be available as a secondary turret? :rolleyes:

Why are we overcomplicating these things so much? Anything that can be a primary turret can be a secondary turret. All problems solved!

That requires source modifications, Zarel or some other dev can answer best on how hard it would be.

It's possible. May or may not require breaking mod format.

But I still don't see why we need specialized "secondary turret" turrets.

I think that's already supported, at least there is code in weapon.txt that discriminates between fixed and rotating weapons.

Yes, between fixed and rotating, not between fixed and 180°-restricted and rotating.

It's possible to implement, but it's not easy, and it certainly isn't "already supported". :/

From my multi turret tests it seems that this is the case already at least partially. A shorter range turret will fire indipendently from the main weapon.

To be exact, a turret will fire independently only if it cannot fire on what the primary turret is firing at.

[Olrox]

Since that is not feasible in the near future, what can be made is a system where we can have a main turret plus a secondary slot for small (and/or medium) bodies and 2 for heavy (3 for super heavy?).

That's way too many. 2 turret slots for light/medium, 3 for heavy, and 4 for super heavy? Seriously? That is way ridiculously too much. If you want four different turrets, just build four different tanks...

This is just... no! Why would you ever want that? Give me one good reason! One!

I must agree, that's too many slots, even if the weapons are going to be weak. Light bodies - never. Medium bodies - hardly, if there's and exceptiopnal reason, 1. Heavy bodies, 1. Super heavies - two normal turrets or 1 normal and 2 secondary.

Why are we overcomplicating these things so much? Anything that can be a primary turret can be a secondary turret. All problems solved!

Well, I was just thinking about yet another side to balance them (making them somewhat more costly - speaking about power and time - to have). But thinking it over, it sounds over-complicating simple stuff.

But I still don't see why we need specialized "secondary turret" turrets.

Directly - because it's nice to have multi-role units, as they aren't helpless if something they aren't primarily supposed to defend from appears in their way. This also reduces the ammount of units you'd need to have - instead of producing 2 machinegunners for each HCannon Tiger Tracks you have, and having to micro-manage their movement because speed limitation was removed due to bugs, you simply have a secondary machinegun attached to your tank.

This frees space in the battlefield, but on the other side, it's all or nothing - if you loose one unit, you'll loose all of their weapons.

Overall, concentrating firepower in a single strong unit helps to prevent swarming tactics. The unit would be more power-costly to produce, so it wouldn't be a good idea to make a hell of a lot of them to spam. On the other hand, it might be a good idea because then the light weapons would take advantage from the unit's high defense. Along with our large possibilities of balancing, I'm sure we can make it so that it'll be a mighty good tradeoff.

Another direct reason is that tanks with secondary weapons are coherent looking at RL tanks - if you have a heavy weapons platform, why not mount additional light weapons to eradicate light threats? Usually tanks that have space to mount a 120mm cannon have enough space to mount an additional m60, for example.

Heavy units with secondary weapons also do look cool if designed properly

Now, a somewhat indirect reason would be that, if firepower is slightly more concentrated in a little lower number of units, it'll be easier to gather experience - helping to minimize another controversial "issue" with MP games. Stronger, a bit more expensive units, mean more durable, a bit less numbersome units, and that makes the number of kills to be divided into fewer units.

Multi-role units would be nice for patrols, as another kind of weapons can be assigned to a whole group resulting on units with the exact same speed (unless some unit has somehow higher rank than another, but that's unlikely because they're the same units, set on patrol to the same areas, fighting the same groups). This would make patrols more efficient, and yet again make the units stick together without needing group movement speed restrictions.

I'm sure there are minimal drawbacks, as I'm similarly sure that there are other reasons I couldn't come up with. But for me, those are already interesting enough to make the cost/benefit worthy. I think that most of the community would like that, and am sure that the developers can make it in a way that doesn't screw up balance and raises the "fun factor", nevertheless - if it is possible.

I think that's already supported, at least there is code in weapon.txt that discriminates between fixed and rotating weapons.

Yes, between fixed and rotating, not between fixed and 180°-restricted and rotating.

It's possible to implement, but it's not easy, and it certainly isn't "already supported". :/

That's for artillery I think - those are the only units who rotate around their own axis to fire (behavior that should be extended to retreating units, but that's a completely different story, and the devs already accepted to take a deeper look into movement codes, methinks)

From my multi turret tests it seems that this is the case already at least partially. A shorter range turret will fire indipendently from the main weapon.

To be exact, a turret will fire independently only if it cannot fire on what the primary turret is firing at.

That wouldn't be enough, I think. They should be completely automated, like sentry guns, IMHO - otherwise they would loose a great deal of their utility as secondary weapons (not requiring the main weapon to change targets to fire at a weak enemy that's harassing the unit).

Man, it'd add so many new viable design opportunities!
Seriously

~Olrox

[Zarax]

This is just... no! Why would you ever want that? Give me one good reason! One!

Flexibility. We're probably looking at the issue from the same POV Zarel but with different concerns in mind.
Rest assured that my aim is not to create super units but rather to have a viable solution.

I always start from the assumption that secondary slots would carry weapons somewhat weaker than their main turret counterparts so that it may be acceptable to mount one or multiple ones depending on the body.

It does NOT have to end up in the trunk (which I think is where your issue is in terms of balancing) that way but it would be nice to leave the door open to further modding (which is my approach).

The way I see it it's better to design the system and code it to be flexible rather than have it fitting strictly trunk balance.
It's easier to do balance changes from the txt files in a open ended system rather than the code to adapt it each time.
With a balanced trunk and the possibility for modders to go beyond that I think we would make everybody happy.

i.e. as soon as a weapon is researched, it will be available as a secondary turret? :rolleyes:

Why are we overcomplicating these things so much? Anything that can be a primary turret can be a secondary turret. All problems solved!

I think we're saying the same things in different ways here. Again, I prefer secondary turrets to not require extra research as well but I wouldn't be against an extra layer of research to obtain them either.

That requires source modifications, Zarel or some other dev can answer best on how hard it would be.

It's possible. May or may not require breaking mod format.

But I still don't see why we need specialized "secondary turret" turrets.

Yes, between fixed and rotating, not between fixed and 180°-restricted and rotating.

It's possible to implement, but it's not easy, and it certainly isn't "already supported". :/

I thought VTOL turrets already had limited rotation? Having it fixed at least in the initial phases wouldn't be too much of a problem either imho.

To be exact, a turret will fire independently only if it cannot fire on what the primary turret is firing at.

Which sounds reasonable enough to me.

Lastly: Zarel, please try avoid assuming that I want to make this feature OP in game.

Our differences are mostly because I feel you're thinking from a balancing POV while I prefer to leave a wider array of choices available leaving the balance side to "soft" coding (the txt files).

Being a modder for years made me appreciate every bit of flexibility available even if I didn't use it myself.
If you own rome: total war I would be glad to link you to my work on another platform and you can see my take at game balance (My modding for coolness period is long past).

[Zarel]

Flexibility. We're probably looking at the issue from the same POV Zarel but with different concerns in mind.
Rest assured that my aim is not to create super units but rather to have a viable solution.

I always start from the assumption that secondary slots would carry weapons somewhat weaker than their main turret counterparts so that it may be acceptable to mount one or multiple ones depending on the body.

It's not so much a matter of overpoweredness, but a matter of complexity. You have to learn the stats of a whole new set of weapons, and you have to design tanks with a whole new set of weapons.

One primary turret and three secondary turrets? Why not just two regular primary turrets?

The only game I know of with units tons of secondary turrets is Supreme Commander, which has no idea of what "playability" means, which is why Supreme Commander 2 doesn't do that anymore.

It does NOT have to end up in the trunk (which I think is where your issue is in terms of balancing) that way but it would be nice to leave the door open to further modding (which is my approach).

Oh, in that case, I have significantly fewer objections.

I think we're saying the same things in different ways here. Again, I prefer secondary turrets to not require extra research as well but I wouldn't be against an extra layer of research to obtain them either.

Fair enough. For the record, I would prefer to simply have bodies with two primary turrets, than use "secondary turrets" for anything.

I thought VTOL turrets already had limited rotation? Having it fixed at least in the initial phases wouldn't be too much of a problem either imho.

VTOL turrets are either full rotating, or fixed... I think it depends on the turret. VTOL turrets don't have limited rotation, though.

Our differences are mostly because I feel you're thinking from a balancing POV while I prefer to leave a wider array of choices available leaving the balance side to "soft" coding (the txt files).

Being a modder for years made me appreciate every bit of flexibility available even if I didn't use it myself.
If you own rome: total war I would be glad to link you to my work on another platform and you can see my take at game balance (My modding for coolness period is long past).

Our differences are mostly because we are discussing this in Ideas/Suggestions. Mod suggestions go in Mapping/Modding, where I usually refrain from yelling at people about balance.

If you just want a game engine change, though... I suppose I would be fine with that, although not enough to implement it myself.

[Zarax]

Zarel: First, I think we have to establish a common vocabulary as I'm getting the feeling that while we're looking for similar solutions we have misunderstandings on how they are worded.

For reference: harcoded = C code that needs to be compiled, softcoded = txt files in mp.wz, trunk = how it would be playing in the standard releases.

Second, I have the utmost respect for all the work you did, especially in balancing.
I know how complex it is to balance hundreds of units and what a pain is when someone asks to alter the often delicate relationship between strength, cost and tech level.

Third, I'm going on the assumption that we all want to see a viable multi-turret system in game and want to find an optimal compromise between differing points of view so that one day it could end up in trunk as a well thought and balanced new feature replacing the current unsatisfactory situation.

Let me know if we are in the same track, otherwise it means that we'll have to better clarify what expectations are and how they differ.

That said, I'll try to be clearer on my vision:

I'm pushing for harcoded dedicated secondary weapon slots because I feel that what has been proposed before while simpler to implement has many drawbacks that would make it much harder to balance satisfactorily. I made a list of those in the first post so I'm not going in detail here.

What may have not transpired is that I may have a peculiar development philosophy and this might have transpired in an unclear message.

What I'm calling here is to have the hardcoded part of this system to be as flexible as possible (as in capability to support N "primary" turrets and N "secondary" ones) while trunk balance can be reached by fine tuned soft coding.

This means that I am NOT advocating to see 3 secondary turrets in trunk but rather to leave that window open in the hardcode in case for example somebody wants to make a "warzone 1900" where a mark VIII tank can be designed with 2 "primary" cannon turrets and 1 "secondary" mg turret.

To sum it up, I would like to see this feature implemented in trunk softcoding with strict balancing so that it can add to the game without becoming overpowering while being more aesthetically pleasing than previous solutions.
The hardcoded part instead can have the potential to support an higher number of slots to give modders a new features for their own work.

Do you think we're getting in the same boat now?

[Zarel]

Zarel: First, I think we have to establish a common vocabulary as I'm getting the feeling that while we're looking for similar solutions we have misunderstandings on how they are worded.

For reference: harcoded = C code that needs to be compiled, softcoded = txt files in mp.wz, trunk = how it would be playing in the standard releases.

Mmkay.

Let me know if we are in the same track, otherwise it means that we'll have to better clarify what expectations are and how they differ.

Yes, that's all fine.

For reference, I do indeed understand that you are not proposing changing trunk balance, but rather simply to open up this functionality for modders. What I am doing now is simply trying to persuade you why modders should not use this functionality, either.

I'm pushing for harcoded dedicated secondary weapon slots because I feel that what has been proposed before while simpler to implement has many drawbacks that would make it much harder to balance satisfactorily. I made a list of those in the first post so I'm not going in detail here.

I disagree.

I'm going to reply to your points in the first post in detail:

- Very few weapon configurations looks good

A new body model would fix that, and is planned (your own proposal would require a new body model, as well).

- It's indeed easy to make overpowered units

Having rebalanced/new "multi turret bodies" would prevent that, and is planned (your own proposal would require rebalancing, as well).

- Clipping kills most of the "ok looking" combinations

Angle restriction would prevent that, and is planned (your own proposal would require angle restriction as well).

In other words, 1. the drawbacks can be fixed, and 2. your own proposal has the same drawbacks, leaving your proposal with no benefits, which is the reason I objected a bit vehemently (and slightly rudely in some parts - for that, I apologize).

What I'm calling here is to have the hardcoded part of this system to be as flexible as possible (as in capability to support N "primary" turrets and N "secondary" ones) while trunk balance can be reached by fine tuned soft coding.

I am fine with adding this functionality. If you write a patch to add support for this (and it is good code and all that, and it doesn't break savegames, which I believe is possible as long as you restrict it to fewer than 3 total turrets), I would be happy to commit it (although other developers may object - if they did, I would not apply it - if you do plan on writing the patch yourself, I will ask the other devs if they want to commit such a thing).

However, because my own balancing plans will not be using this system, and because I have limited free time, I am unable to write this feature for you, so you will have to either write the patch yourself, or see if you can convince another developer to write it.

I hope this is fair enough?

Do you think we're getting in the same boat now?

I think I already understood that this was what you were saying, but I suppose I have been a bit unclear, myself. Have I explained my position more clearly, now?

[Zarax]

For reference, I do indeed understand that you are not proposing changing trunk balance, but rather simply to open up this functionality for modders. What I am doing now is simply trying to persuade you why modders should not use this functionality, either.

I guess we could agree to disagree on this but I'm afraid that I don't get why even modders wouldn't benefit from it.

Having rebalanced/new "multi turret bodies" would prevent that, and is planned (your own proposal would require rebalancing, as well).

I suspect this is the heart of our disagreement.

We both want to make a multi turret system but we're proposing different solutions.

While I think that creating specific weapons and slots for them on the existing bodies would be the way to go you'd rather create purpose-made multi-turret units.

Is that right?

- Clipping kills most of the "ok looking" combinations

Angle restriction would prevent that, and is planned (your own proposal would require angle restriction as well).

Guess it was an obvious one for any kind of solution. Glad we agree on this one.

In other words, 1. the drawbacks can be fixed, and 2. your own proposal has the same drawbacks, leaving your proposal with no benefits, which is the reason I objected a bit vehemently (and slightly rudely in some parts - for that, I apologize).

I still think my proposal has better potential for fine tuning and creative output but if I got your intentions right I'm extremely fine with that as well.
Your solution is most likely easier to code and I'm sure you'll do an excellent job at balancing it.

I work in customer support so I've got pretty thick skin for criticism. Clearing up any misunderstandings and understand each other's reasons is the important thing imho.

I am fine with adding this functionality. If you write a patch to add support for this (and it is good code and all that, and it doesn't break savegames, which I believe is possible as long as you restrict it to fewer than 3 total turrets), I would be happy to commit it (although other developers may object - if they did, I would not apply it - if you do plan on writing the patch yourself, I will ask the other devs if they want to commit such a thing).

However, because my own balancing plans will not be using this system, and because I have limited free time, I am unable to write this feature for you, so you will have to either write the patch yourself, or see if you can convince another developer to write it.

I hope this is fair enough?

As I'm understanding you've got an alternative solution to that it's far more efficient to see how that will play rather than trying to code two different perspectives for the same issue, especially as my C knowledge is extremely basic.

I think I already understood that this was what you were saying, but I suppose I have been a bit unclear, myself. Have I explained my position more clearly, now?

Much clearer, thank you.
The only doubt left is if I'm missing anything about your own solution to the multiple turret problem.

[Zarel]

I guess we could agree to disagree on this but I'm afraid that I don't get why even modders wouldn't benefit from it.

It's partly because I believe a mod with dedicated secondary turrets would be made better if they simply used primary turrets. I guess it's mostly because I'm too lazy to add such a feature. As an open-source project, we don't really have the time to add features not used in the core game (unless it happens to also improve the code, which is really the only reason I'm planning to add unit transport functionality for skirmish modders).

I suspect this is the heart of our disagreement.

We both want to make a multi turret system but we're proposing different solutions.

While I think that creating specific weapons and slots for them on the existing bodies would be the way to go you'd rather create purpose-made multi-turret units.

Is that right?

Yes, pretty much.

The problem with "specific weapons and slots for them" is "added complexity".

And, of course, the standard balance argument. Either the secondaries are so strong as to be overpowered (cf. the Aeon Restorer in SupCom), or so weak as to be useless (cf. the AA weapon of every other T3 gunship in SupCom).

(Eheh, ever since SupCom2 came out, and took out a lot of the bad features in SupCom, I've been frequently using SupCom as an example of what not to do...)

I still think my proposal has better potential for fine tuning and creative output but if I got your intentions right I'm extremely fine with that as well.

"Potential for fine tuning" hardly sounds like a benefit to me... It sounds to me like a euphemism for "extremely complex and will take a long time to balance and frustrate users by giving them even more weapons whose statistics they have to memorize"

As for "creative output"... I suppose that's true. But see above "complexity".

The only doubt left is if I'm missing anything about your own solution to the multiple turret problem.

Not that I know of.

[Zarax]

Yes, pretty much.

The problem with "specific weapons and slots for them" is "added complexity".

And, of course, the standard balance argument. Either the secondaries are so strong as to be overpowered (cf. the Aeon Restorer in SupCom), or so weak as to be useless (cf. the AA weapon of every other T3 gunship in SupCom).

(Eheh, ever since SupCom2 came out, and took out a lot of the bad features in SupCom, I've been frequently using SupCom as an example of what not to do...)

Guess we differ on our view about complexity. You see it as an issue where I see potential.

"Potential for fine tuning" hardly sounds like a benefit to me... It sounds to me like a euphemism for "extremely complex and will take a long time to balance and frustrate users by giving them even more weapons whose statistics they have to memorize"

As for "creative output"... I suppose that's true. But see above "complexity".

Yet another side of our differing view on complexity.
On the opposite side, have you ever thought about making the unit design interface more informative?

Imho it would be nice to have more stats displayed or at least more intuitive ones (I.E. unit speed instead of engine output).
If this (unlikely) never surfaced I would like to open a discussion on a dedicated thread.

Finally, I want to apologize if I sounded stubborn, most of my arguing was generated by the misguided assumption that there were no plans on the feature.

All in all I think we had a nice warm exchange of ideas

[Zarel]

On the opposite side, have you ever thought about making the unit design interface more informative?

Imho it would be nice to have more stats displayed or at least more intuitive ones (I.E. unit speed instead of engine output).
If this (unlikely) never surfaced I would like to open a discussion on a dedicated thread.

What kind of stats do you want? A body can't display unit speed because unit speed depends on propulsion. And propulsions already display unit speed...

[DuKe2112]

Well you seem to have some misunderstanding about complexity, because complexity is about restrictions. "You can do that with this, but not that, you rather need something else but it can't do this other thing". It also confuses the hell out'a people.

I still say size/weight restrictions on turrets is the simplest method to implement with the most options, you can even have your secondary weapons by saying everything above size X is main and everything below is secondary, but it also offers a dynamic approach.

I might take a look at it, but it will take time anyway as I don't now WZ codebase at all yet.


And yeah GUI information doesn't belong in a thread about turrets.

[Olrox]

Ok, I've already posted my own points about that, and add that it'd be nice to have weaker (than any Dragon body unit) multi-purpose units, before going through all the body researches. I don't know why I even added that anyway, it'll probably have no effect.

But I'm really getting tired of all this discussion, I won't take part on it any further - they get me bored easily. I can already see the outcome of this very clearly anyway, so I'll just get back to my original place and remake the bodies for AR in the standard way. If the forementioned code modifications are committed, please let me know if anyone wants models that support secondary turrets, I'll be glad to make those adaptations.

I'm -not- getting out of here because anyones post, I didn't even read most of today's ones. I just don't want to augment the discussion, or keep trying to achieve something that I just can't

~Olrox

[Zarax]

Olrox, the point reached by the whole conversation is that Zarel wants to have custom built specialized multi-turret bodies and he's writing the code to support them optimally.

This means that we will have multi-turret units, just in a different way than I envisioned.

[Zarel]

Yeah, I think I mentioned in one of my threads already that what we need is a good version of the Dragon body that multi-turrets look nice on.