Is it ideal to have one set of stats?

The place to discuss balance changes for future versions of the game.
(Master releases & 3.X)

Is it ideal to have one set of stats?

Postby MIH-XTC » 04 Oct 2017, 04:22

Do you think we should strive to have the same set of stats for multiplayer and campaign?

I think so, I think it's best practice to have campaign emulate skirmish to reduce the learning curve.

The tech trees need to be the same too?

Any interest in unifying the two?

The campaign would need to add/remove droids as necessary to adjust the scenarios.

Just a random discussion.
MIH-XTC
Trained
Trained
 
Posts: 195
Joined: 31 Jan 2014, 07:06

Re: Is it ideal to have one set of stats?

Postby Berg » 04 Oct 2017, 05:08

I think campaign skirmish and multi player should be the same as you learn and migrate to online you will not have to change your tactics too much.
User avatar
Berg
Regular
Regular
 
Posts: 1882
Joined: 02 Sep 2007, 23:25
Location: Australia

Re: Is it ideal to have one set of stats?

Postby -Philosopher- » 07 Oct 2017, 13:10

Is there somewhere I could see and compare the respective stats? Tech trees?

I'm aware there are some technologies (and structures) available in campaign and not skirmish, and vice versa. How else do they differ?

How was it done in 1.10?

Could it be achieved without changing the campaign?

On a related question, I believe stats have been changed in different releases. Does anyone know where I could find a good source of information to understand how these have changed over time?
-Philosopher-
Trained
Trained
 
Posts: 115
Joined: 08 Oct 2014, 11:34

Re: Is it ideal to have one set of stats?

Postby Bethrezen » 07 Oct 2017, 14:43

The tech trees need to be the same too?


I can see issues with doing that because there are a bunch of extra research topics in skirmish/online play that are not available in campaign mode, for example the dragon body the full spectrum scanner to name a few but there are numerous others, so to be honest I'm not sure this is achievable, without significant alterations to campaign mode, and since the intention seems to be to try and keep the campaign close to the original while at the same time ironing out any glaring issues or things that where not well though out, well designed or well implemented in the original, then I'm not sure that this would be a good idea.

If someone wanted to take up this challenge and do some experiments to see if it would be feasible then I would encourage that but it should probably be done as a mod.

Do you think we should strive to have the same set of stats for multiplayer and campaign?


With regards to balance it would be possible to have units balanced the same in campaign and skirmish / online so that they have the same rate of fire, same damage per shot, move at the same speed etc across the different modes, having said that however if that was to be done then again it would probably be best to do that as a mod to begin with because I can see this causing serious issues if it's not handled correctly because campaign is a very different environment to skirmish or online.

On a related question, I believe stats have been changed in different releases. Does anyone know where I could find a good source of information to understand how these have changed over time?


Can't say i know of any, that being the case what I'd be tempted to do is to just look at the original source because presumably there is going to be a file in there somewhere that contains a table that lists the stats for all the units, although i couldn't tell you where or which file but it has to be in there somewhere, I'd then look at the current source find the equivalent file and at that point its simply a case of doing a comparison between them.
Bethrezen
Trained
Trained
 
Posts: 346
Joined: 25 Sep 2009, 02:05

Re: Is it ideal to have one set of stats?

Postby MIH-XTC » 07 Oct 2017, 15:42

Bethrezen wrote:
The tech trees need to be the same too?


I can see issues with doing that because there are a bunch of extra research topics in skirmish/online play that are not available in campaign mode, for example the dragon body the full spectrum scanner to name a few but there are numerous others, so to be honest I'm not sure this is achievable, without significant alterations to campaign mode, and since the intention seems to be to try and keep the campaign close to the original while at the same time ironing out any glaring issues or things that where not well though out, well designed or well implemented in the original, then I'm not sure that this would be a good idea.

If someone wanted to take up this challenge and do some experiments to see if it would be feasible then I would encourage that but it should probably be done as a mod.


Using spreadsheets I just did a quick string comparison to quickly identify the discrepancies. They are below.

Spoiler:


I’m familiar enough with the stats such that adding them into campaign can be done in a couple of hours. It actually isn’t much work at all. What’s more time consuming is understanding the missions and balancing the placement of the research items.

There is too strong of adherence to preserving the original campaign when it was made by people with no WZ experience over the course of one year. We have 18 years of gameplay data now and much more advanced programming capabilities. I understand there is a sense of attribution and respect for original authors but that doesn’t mean things shouldn’t be changed. If we can make the campaign missions more interesting then do it… improve the game and make it better. Maybe the original authors will come play newer versions and appreciate it equally. We act like this is heresy.



Do you think we should strive to have the same set of stats for multiplayer and campaign?


With regards to balance it would be possible to have units balanced the same in campaign and skirmish / online so that they have the same rate of fire, same damage per shot, move at the same speed etc across the different modes, having said that however if that was to be done then again it would probably be best to do that as a mod to begin with because I can see this causing serious issues if it's not handled correctly because campaign is a very different environment to skirmish or online.


Right, the scenarios would need to be reworked so that missions are not too easy or too hard but that should always be done several times. We should constantly be trying to improve the mission balances as opposed to avoiding changes. The way I see it is that the original authors didn’t have time to unify the two so they made separate versions for the sake of pushing out a product. It was only supposed to be a temporary fix. I think skirmish should serve as the foundation because that's where the game replay resides. Making stats for specific scenarios that are only good one time around isn't worthwhile.
MIH-XTC
Trained
Trained
 
Posts: 195
Joined: 31 Jan 2014, 07:06

Re: Is it ideal to have one set of stats?

Postby Berserk Cyborg » 07 Oct 2017, 17:42

Campaign has always had a weird, and evolving, balance issue. Lancers destroy cyborgs easily, flamers are useless, assault gun bullets are like being hit with a mini plasma cannon blast. Only required difference is how enemy howitzer/rocket/missile artillery emplacements are much more powerful to the point of destroying player droids with ease (Seraph is much different as well and probably should stay that way for campaign). Eventually it needs a solution and unification is probably inevitable.

MIH-XTC wrote:Using spreadsheets I just did a quick string comparison to quickly identify the discrepancies.
There is only one research file now for campaign (if it matters). I squashed the three files into one and removed useless/repeated research like CAM2... and anything like that and linked similar research together. So it might give a clearer understanding of the current campaign tech tree. Cyborg research grants the weapon and not some specialized body component for each and every cyborg template. Stat wise, I replaced producible cyborg bodies (all those grd/ground ones) with the skirmish cyborgLightBody and Nexus uses type II propulsion.

The tech tree can still contain hidden research for potentially more campaigns (dragon and EMP and whatever). "keyTopic" is used to suppress research and that alone would require two research files. I highly suggest keeping map placements the same or you will see outrage like never before.

Anyone who dare disturb campaign will become a super-villain. Though, if you think you could do it and do it well, then I would encourage you to try reworking the stats/balance regardless of what anyone says. After all, you are the master of stats and balance. :)
User avatar
Berserk Cyborg
Code contributor
Code contributor
 
Posts: 340
Joined: 26 Sep 2016, 19:56

Re: Is it ideal to have one set of stats?

Postby MIH-XTC » 08 Oct 2017, 03:52

Berserk Cyborg wrote:
MIH-XTC wrote:Using spreadsheets I just did a quick string comparison to quickly identify the discrepancies.
There is only one research file now for campaign (if it matters). I squashed the three files into one and removed useless/repeated research like CAM2... and anything like that and linked similar research together. So it might give a clearer understanding of the current campaign tech tree. Cyborg research grants the weapon and not some specialized body component for each and every cyborg template. Stat wise, I replaced producible cyborg bodies (all those grd/ground ones) with the skirmish cyborgLightBody and Nexus uses type II propulsion.



I think that's a good thing to merge the 3 research files into one, I prefer having fewer but larger files. I did the same thing with the research messages. I also try to avoid references to specific cams if possible, the terminology can be confusing for newcomers.

Berserk Cyborg wrote:I highly suggest keeping map placements the same or you will see outrage like never before.


Yea I wouldn't even try, I don't have enough time to do unit placements because I'm sure the coordinate positions are tied to logical conditions in the mission plot so it's really easy to have unintended consequences. I haven't done any testing in campaign either so there's more overhead.

Berserk Cyborg wrote:Anyone who dare disturb campaign will become a super-villain. Though, if you think you could do it and do it well, then I would encourage you to try reworking the stats/balance regardless of what anyone says. After all, you are the master of stats and balance. :)


I would approach it by importing what I have from mp into campaign by chunks but I don't have time to do this either. I probably won't work on WZ again until christmas break time. I just became fluent with using Git at work so I need to decompose what I changed into commits or pull requests instead of using forum attachments as a code repo.

One thing for sure I know can be ported from mp to campaign are the propulsion speed modifiers, engine outputs/upgrades and turret/body weights. Basically the components that effect the droid speeds. The idea is to make droids progressively faster with engine upgrades but right now most droid designs are capped by the max speed of their propulsion by default so there's no room for speed improvements. I have a good scheme setup for this.
MIH-XTC
Trained
Trained
 
Posts: 195
Joined: 31 Jan 2014, 07:06

Re: Is it ideal to have one set of stats?

Postby -Philosopher- » 08 Oct 2017, 20:56

So... I asked a bunch of questions earlier that could be mistaken for expressing an opinion...

I'm new(ish) here, so I doubt my view carries much weight, but under the assumption I might not be the only one thinking it, I should probably explain -

My questions didn't come from some dogmatic desire to avoid change to the original, but instead an attempt to form a view as to whether the change is likely to be good or bad - a cost/benefit/risk assessment, if you will. For that I needed data.

MIH-XTC wrote:I understand there is a sense of attribution and respect for original authors but that doesn’t mean things shouldn’t be changed.

Is that really the primary motivation behind objection to change, or is it more often something more real-world than that? I'm sure purists with such abstract motivations exist, but I'm sure a lot of objections to change come from elsewhere. For example - me - a number of recent changes have made the game worse, not better, thus it's about past experience rather than lofty goals or attachments. The point is, we shouldn't make assumptions.

Also - speculating - the current exercise to convert (or fix the conversion of) the campaign to jscript might be clouding the issue. Its stated objective is to convert the wzscript campaign essentially without change, but that doesn't mean to say there shouldn't be any change, ever (just not for that exercise), but as it's got focus at the moment it might seem that way (?).

MIH-XTC wrote:If we can make the campaign missions more interesting then do it… improve the game and make it better.

I agree. But only if it is genuinely better. The problem is "better" is subjective. e.g. 3.2.x was, presumably, released because someone or group of people thought it was better than what came before, but many people still play on the 3.1.x series (as I understand it) and so presumably don't agree. So either you need a way of getting input on it, at a detailed level, from a large enough subsection of the user base (a lot of work) or you end up carrying a lot of risk (and some of the "villainy" and "outrage" may be end up being with good reason).

MIH-XTC wrote:We act like this is heresy.

Maybe for some, but I doubt true for all. It could be just objections to individual changes, or a desire to avoid opening the pandora's box of what constitutes "better" (i.e. just not go there in the first place), or something else entirely. The danger of making assumptions about motivation is it risks dismissing counter arguments as heresy too, which would be just as bad.

Anyway, it's certainly not where my thoughts are coming from, which are, after some reflection:

My concern is this seems complicated to get right - that pandora's box, if you will, with a high risk of unintended consequences making the end result worse than what we have today, despite best intentions. Which, in itself, wouldn't be enough of a reason to not attempt it, but -
- it should be optional - a mod, as suggested above - that way you'd avoid the worst of the risk and let Darwin decide instead
- what would be the objective anyway? How valuable is that objective? I agree with @Berg's point - if that's the answer - but it feels like a "nice-to-have" rather than a "critical", which leads me to wonder...
- is it the right priority? Because of the aforementioned complexity, I imagine it'd take a considerable time input to get right and I constantly hear about how limited everyone's time is (it's certainly true of me). I think there are more important things that need to be looked into ahead of this.
- might be going off-topic here but if there is a strong desire to spend time on this sort of thing, what would be involved in creating an entirely new campaign (to complement the current one)? No pandora's box with something entirely new, and even a small, modest set of scenarios might be of more interest to most than time spent tinkering with the old. Just a thought... don't shoot me for having it...
-Philosopher-
Trained
Trained
 
Posts: 115
Joined: 08 Oct 2014, 11:34

Re: Is it ideal to have one set of stats?

Postby cybersphinx » 08 Oct 2017, 22:29

Some random thoughts:

A single player campaign (a lot of single player games really) is some kind of power fantasy, you start weak and get stronger. The experience of your opponent doesn't matter, since it's a computer. But this (lack of) balance might not make for a good multi player experience.

So you'd probably want to balance multi player occasionally, to make it more interesting that just "choose the one thing that's op." But, if single and multi player use the same stats, every time you change multi player balance you'd need to check if the whole campaign still feels good. That seems pretty silly (not to say too much work) to me.

If you take all the multi player tech and put it into the campaign, would that even fit in? From what I remember, I'd not even use all options currently in the campaign, so adding more might just increase the feeling of "I have all this (potentially) interesting stuff and don't do anything with it."
We want information... information... information.
cybersphinx
Inactive
Inactive
 
Posts: 1671
Joined: 01 Sep 2006, 19:17

Re: Is it ideal to have one set of stats?

Postby MIH-XTC » 11 Oct 2017, 02:32

cybersphinx wrote:Some random thoughts:

A single player campaign (a lot of single player games really) is some kind of power fantasy, you start weak and get stronger. The experience of your opponent doesn't matter, since it's a computer. But this (lack of) balance might not make for a good multi player experience.



Right, I played starcraft and stronghold and the campaigns were progressive in nature. However they had decent AI's though or at least challenging scenarios. I feel like WZ campaign doesn't have enough power/units/action or doesn't progress as fast. I always got the impression that it wasn't really challenging or not enough units.

cybersphinx wrote:So you'd probably want to balance multi player occasionally, to make it more interesting that just "choose the one thing that's op." But, if single and multi player use the same stats, every time you change multi player balance you'd need to check if the whole campaign still feels good. That seems pretty silly (not to say too much work) to me.


That's true, I didn't consider future divergences. I was thinking more like set it and forget it. So probably more like a one time sync-up to have both aspects of the game more reflective of each other

cybersphinx wrote:If you take all the multi player tech and put it into the campaign, would that even fit in? From what I remember, I'd not even use all options currently in the campaign, so adding more might just increase the feeling of "I have all this (potentially) interesting stuff and don't do anything with it."


It would fit in the sense that we can properly assign extra research items to the proper cam scenarios but not in the sense that a player would have any use for it.

I guess that is kind of what I meant above when I said there isn't enough action to make the campaign interesting.

There needs to be more units and structures and stuff to do. It seems like something is missing. I think an ideal campaign would end with some super scenario where the AI has max allowable hardpoint structures that takes several hours to beat. I think the campaign needs more AI droids and higher oil rates. Higher rate of "Activity". Higher oil rates is just like pouring gasoline on a smothering fire :)

Disclaimer: I'm hardly familiar with the specifics of the campaign. I never play it. I think too many of the missions are only with 10 units?
MIH-XTC
Trained
Trained
 
Posts: 195
Joined: 31 Jan 2014, 07:06

Re: Is it ideal to have one set of stats?

Postby Bethrezen » 10 Dec 2017, 05:47

It would fit in the sense that we can properly assign extra research items to the proper cam scenarios but not in the sense that a player would have any use for it.

I guess that is kind of what I meant above when I said there isn't enough action to make the campaign interesting.

There needs to be more units and structures and stuff to do. It seems like something is missing. I think an ideal campaign would end with some super scenario where the AI has max allowable hardpoint structures that takes several hours to beat. I think the campaign needs more AI droids and higher oil rates. Higher rate of "Activity". Higher oil rates is just like pouring gasoline on a smothering fire :)

Disclaimer: I'm hardly familiar with the specifics of the campaign. I never play it. I think too many of the missions are only with 10 units?


As i see it the issue with the campaign is that the levels are to scripted, at least by comparison to games like C&C or DoW, one of the other problems is that you are very limited by the timer and that really restricts dynamic play, don't get me wrong timed missions have there place but for WZ to have a truly dynamic campaign you would need to redesign elements of the game, the resource system would need to be made more like it is in C&C/DoW, the timer would need to go unit limits would need to go, the AI would need some work and a lot of the underpinnings would need to be redesigned to work more like other RTS games
Bethrezen
Trained
Trained
 
Posts: 346
Joined: 25 Sep 2009, 02:05


Return to Balance