New release planning (3.2.4)/ 3.3.0

Discuss the future of Warzone 2100 with us.
OFelix
Rookie
Rookie
Posts: 25
Joined: 01 Jul 2015, 22:36

Re: New release planning (3.2.4)/ 3.3.0

Post by OFelix »

As an unsophisticated player, having the HQ be a requirement for map and designs (and nothing else) makes sense to me :-)

As an aside, does destroying an AI's HQ when playing in single player have any effect on them?
MIH-XTC
Trained
Trained
Posts: 368
Joined: 31 Jan 2014, 07:06

Re: New release planning (3.2.4)/ 3.3.0

Post by MIH-XTC »

OFelix wrote:As an unsophisticated player, having the HQ be a requirement for map and designs (and nothing else) makes sense to me :-)

As an aside, does destroying an AI's HQ when playing in single player have any effect on them?

Nexus does not check if it has a HQ before building combat units and it's possible that it can build tanks without a HQ but it has a hard coded starting strategy that ensures it has a HQ before it starts building units.
User avatar
NoQ
Special
Special
Posts: 6226
Joined: 24 Dec 2009, 11:35
Location: /var/zone

Re: New release planning (3.2.4)/ 3.3.0

Post by NoQ »

HQ being a requirement for design is there to nerf tank rushes (requires the rusher to pay $100 and around 30 seconds of two-truck work), and HQ being a requirement for mg tower nerfs truck/tower rushes in a similar manner. HQ requirement for the minimap (and designs) makes it a fun target for vtol diversions, but that's kinda minor.

Aesthetically, i guess it would make sense to require HQ for all defensive structure tech, because it is kinda similar to design, and would also be consistent with the mg tower research requirement, and it doesn't affect the balance strongly enough to care.
User avatar
Berg
Regular
Regular
Posts: 2204
Joined: 02 Sep 2007, 23:25
Location: Australia

Re: New release planning (3.2.4)/ 3.3.0

Post by Berg »

HQ are part of the game play and tactics and design.
User avatar
alfred007
Regular
Regular
Posts: 619
Joined: 31 Jul 2016, 06:25
Location: Stuttgart, Germany

Re: New release planning (3.2.4)/ 3.3.0

Post by alfred007 »

OFelix wrote:As an aside, does destroying an AI's HQ when playing in single player have any effect on them?
In campaign, there are several levels where enemy VTOL attacks stop after you destroyed the enemy HQ (Beta 01, Beta03, Beta 04, Beta 07, Gamma 01, Gamma 02 and Gamma 03).
Forgon
Code contributor
Code contributor
Posts: 298
Joined: 07 Dec 2016, 22:23

Re: New release planning (3.2.4)/ 3.3.0

Post by Forgon »

Zepherian has tested the buildbot release of January 14th and noticed that the behaviour of VTOLs is unacceptable:

All VTOLs try to land on the same pad, even if there are far more pads than VTOLs. I successfully reproduced this problem.
User avatar
Prot
Trained
Trained
Posts: 242
Joined: 29 Nov 2010, 12:41

Re: New release planning (3.2.4)

Post by Prot »

Per wrote:
vexed wrote:Wasn't it changed to 100% accuracy for some weapons?
That was rolled back, along with most other balance changes, before the release of 3.2.
If you talking about changes between 3.1.5 and 3.2, then well, how about cyborgs-repairers are built without having to research repair technology at all?
Plus jammers that came later..
Of rolled back I only remember the need to pay for the construction of the oil extractors.

As for HQ, strange, I recompiled my latest master and now do without HQ there is no design. Before that I won two computers without built HQ, maybe it was someone else's commit, or a bug caused by different versions of somewhere in rules.js I do not know.
User avatar
andrvaut
Trained
Trained
Posts: 200
Joined: 02 Jan 2016, 12:44

Re: New release planning (3.2.4)

Post by andrvaut »

Per wrote:That was rolled back, along with most other balance changes, before the release of 3.2.
I think you need to come back shortHit and shortRange.

Change led to a complete re-balance of branches and a change in roles.
The missiles can no longer defeat the guns at great distances.
Bullets do not kill all cyborgs near.
Fire often misses at point-blank range.
Guns do not need to converge to increase efficiency.

I believe that it is these changes that keep many players on version 3.1.5.
I'm ready to prepare a patch if it is accepted
vaut ΣΑ [GN], ru streamer.
Tournaments channel: https://www.youtube.com/channel/UCzusNa-54ydodtSz2TdHFww
Per
Warzone 2100 Team Member
Warzone 2100 Team Member
Posts: 3780
Joined: 03 Aug 2006, 19:39

Re: New release planning (3.2.4)

Post by Per »

andrvaut wrote:
Per wrote:I think you need to come back shortHit and shortRange.
...
I believe that it is these changes that keep many players on version 3.1.5.
Really? You believe it is the lack of short/long range accuracy that keeps players on 3.1? The narrative keeps shifting fast.

Let's recap a bit. This change is based on a couple of very long threads about accuracy back in 2012 (viewtopic.php?f=42&t=8964 and viewtopic.php?f=42&t=10794). This change was committed later that year, there was no controversy about it at the time (despite a lot of disagreements on pretty much everything else related to accuracy), and has been in the game for almost 6 years. You may argue that most players who really know the game played 3.1 in those 6 years, but it was still tested and played, and it got one balance adjustment in 2016 (see dc6ddae6953c7f8cb2b0bd073641f02ed348da8b) based on feedback. Nobody complained about the lack of short/long until now.

Let's also recap the reason for removing it, for those too lazy to read all those pages of forum posts above: It is a hidden game mechanic that is impossible for the player to figure out. There is no way of knowing when you cross the magic threshold and accuracy flips between short and long. Invisible game mechanics should be avoided.

So please be prepared to offer actual evidence that the lack of short/long range is leading to problems that are not possible to fix simply by adjusting existing accuracy values, rather than another round of reflexive "oh this is different than 3.1, let's change it back".
User avatar
Prot
Trained
Trained
Posts: 242
Joined: 29 Nov 2010, 12:41

Re: New release planning (3.2.4)/ 3.3.0

Post by Prot »

After raising topics about the accuracy I'm trying to understand but still don't understand why were originally long and short accuracy.
The flamethrower will never have long accuracy, as it is a close-range weapon.
On the other hand, there are rockets that always gets far more accurate than close range, this way of playing the rockets many players knew, and I realized got rid of it.
Including I learned about this feature with missile accuracy from this site, in the guide.
Some things, even though hidden - make the game unique, such as it is. Sometimes the obvious is too boring.
But new things are also needed(I'm talking about jammers, changes in the command tower, etc). I am opposed to return the balance of the game 1 to 1 as in 3.1.5.. But sad that the missiles now have the same accuracy in the near and far distance, i.e. the rockets lost zest to its mechanics.

BTW I do not remember what version, I found in stats that the body armor on different sides can have different characteristics of armor. Let's say if the tank is nose to the enemy, he receives less damage than if he retreated back, and the damage it would inflict more. If I am not mistaken, the mechanics of armor, but never used it and little about it who knows - this is a hidden feature I agree. But very funny and interesting in terms of moddings.
User avatar
Berserk Cyborg
Code contributor
Code contributor
Posts: 938
Joined: 26 Sep 2016, 19:56

Re: New release planning (3.2.4)

Post by Berserk Cyborg »

Prot wrote: If you talking about changes between 3.1.5 and 3.2, then well, how about cyborgs-repairers are built without having to research repair technology at all?
In rules.js for multiplayer there is:

Code: Select all

// enable cyborgs components that can't be enabled with research
makeComponentAvailable("CyborgSpade", playnum);
makeComponentAvailable("CyborgRepair", playnum);
That comment does not seem to hold any merit, at least with my observation, since removing those makes the template appear when the correct research is completed. I imagine some AI scripts would need to be accommodated for such a change.
User avatar
Prot
Trained
Trained
Posts: 242
Joined: 29 Nov 2010, 12:41

Re: New release planning (3.2.4)

Post by Prot »

Berserk Cyborg wrote:In rules.js for multiplayer there is:

Code: Select all

// enable cyborgs components that can't be enabled with research
makeComponentAvailable("CyborgSpade", playnum);
makeComponentAvailable("CyborgRepair", playnum);
That comment does not seem to hold any merit, at least with my observation, since removing those makes the template appear when the correct research is completed. I imagine some AI scripts would need to be accommodated for such a change.
Yeah, i find it in research.json on line 3231

Code: Select all

    "R-Sys-MobileRepairTurret01": {
        "iconID": "IMAGE_RES_SYSTEMTECH",
        "id": "R-Sys-MobileRepairTurret01",
        "msgName": "RES_REPTU1",
        "name": "Mobile Repair Turret",
        "requiredResearch": [
            "R-Sys-Engineering01"
        ],
        "researchPoints": 600,
        "researchPower": 18,
        "resultComponents": [
            "CyborgRepair",
            "LightRepair1"
        ],
        "statID": "LightRepair1"
    },
This already has resultComponents cyborgRepair. I don't know who and why added this in rules.js. Need to look at the commits before version 3.2.0, perhaps there will be a comment about it.

UPD: i found only this https://github.com/Warzone2100/warzone2 ... 960a4e9e21
I think before that the rules.js not exists.. Looks like someone just mistake while creating rules.js.

I think that need to fix and remove the line "makeComponentAvailable("CyborgRepair", playnum);" from rules.js
MIH-XTC
Trained
Trained
Posts: 368
Joined: 31 Jan 2014, 07:06

Re: New release planning (3.2.4)/ 3.3.0

Post by MIH-XTC »

I read the first 2 pages of that accuracy thread and then stopped.

Wouldn't the long/short hit values just be a linear function? So for example if a turret has the following stats:

short hit = 80
long hit = 30
min range = 2 tiles
max range = 12 tiles

then:
@ 2 tiles the accuracy is 80%
@ 3 tiles the accuracy is 75%
@ 4 tiles the accuracy is 70%
5 is 65
6 is 60
7 is 55
... 12 is 30

That just seems too obvious. What's wrong with this approach?

This is a rhetorical forum post, it's not asking for a response, I just merely want to throw out the idea of a linear function if it hasn't been considered.

The way reg312 explains accuracy for a weapon with max range 12 is that "long range" is defined something like tiles 6-12 and short range 0 - 6 but of course that's flawed to assign all tiles 6-12 the same accuracy%.
Per
Warzone 2100 Team Member
Warzone 2100 Team Member
Posts: 3780
Joined: 03 Aug 2006, 19:39

Re: New release planning (3.2.4)/ 3.3.0

Post by Per »

MIH-XTC wrote:Wouldn't the long/short hit values just be a linear function?
That is a good question. I have no idea why it was not implemented in this way. But the most important question is - does short range accuracy actually add anything useful to the game? Obviously, it does not do so immediately to the player, who is oblivious to it because it is a hidden game mechanic. I don't really see how it adds much to balancing, either. In fact, it just makes balancing a lot harder, as you have to consider a whole lot of extra variables, such as how players, the script AI, and the micro-AI all behave to optimize accuracy, and so on - and then you have to make balancing decisions based on the probability of players' knowledge of hidden game mechanics... It is just hairy all over.

Although, if instead of specifying accuracy at short range, you specify the shape of the graph going from 100% at point blank range (weapon's minimum range), to the accuracy at maximum range, then you've suddenly combined short and long range accuracy and my old suggestion of making direct fire weapons 100% accurate to eliminate the ridiculous close range misses..... Maybe not a good idea but fun to think about.

Keep in mind that in the original game you only had the dice roll accuracy. We also have hit box intersection, meaning you can literally dodge bullets by moving around and hit stuff that accidentally gets in the way.
User avatar
Lord_Kane
Trained
Trained
Posts: 104
Joined: 24 Nov 2016, 21:51

Re: New release planning (3.2.4)/ 3.3.0

Post by Lord_Kane »

No offense, but can we implement and test a new accuracy system in a new branch maybe after this release?
Post Reply