Release 3.2.0 beta2

The projects speaking tube.
Add your two cents if you want to.
MIH-XTC
Trained
Trained
Posts: 368
Joined: 31 Jan 2014, 07:06

Re: Release 3.2.0 beta2

Post by MIH-XTC »

railgen wrote:True:
It will help out players, who don't put that much effort into games and it might be fine for ordinary warzone maps.
That makes it justified somehow as wz does not really support ntw type maps, I think.
I think you are under the impression this change is dumbing down the game and makes the unit handling require less strategy but I don't think that's true. This change is not removing functionality or strategy, it makes it easier to control units. Assigning artillery units to sensor might seem like it requires good unit handling skills but I think it makes unit handling unnecessarily complicated and convoluted for no good reason.

If I want to move my artillery units then they automatically become un-assigned to sensor when moving them and then I need to re-assign. Also I cannot assign to regular sensor and CB sensor at the same time. It's better to just have artillery droids use sensor without needing to be assigned to them.

The new change makes it easier for better players to more easily control their artillery units. It will make game play more enjoyable.
MIH-XTC wrote: I think this is very good as long as it applies only to sensor droids and not other structures such as an oil derrick. Including sensor towers?
No not including sensor towers. If all structures can cause artillery to fire simply because opponent is within visible range, this means the only difference between sensor towers and other structures is that sensor tower has more visible range but implies that all structures essentially act as a sensor tower. I think that's bad.
But mortars were never a problem? This update applies mostly on wide range weapons with wide explosions - Ripple, as example, is very rarely a good choice.
Mortars are hardly ever used on tanks and as a result they cannot be used as alternatives to mg, rocket and cannon but now they can be with auto-assignment to sensors. That's what makes this change good, we just need to recalibrate the build time.
MIH-XTC wrote: I know you are concerned about armies of mortars or artillery becoming too powerful or unstoppable in larger quantities as a result of this change but the solution for that is very easy :lol2: . We simply increase the build time of artillery turrets.

Under new sensor functionality, I would guess maybe a 25% increase in build times. I don't have time to test now to get exact numbers but it's no problem to rebalance artillery as a result of this improved functionality.
But what makes you think that would solve the problem? What you do is making it worse in low oil game but not that much in flat long high oil.
You can't put low, mid and high all in one balance mod. You will always outbalance yourself when you don't modify the actual game physics and AI.
This is a big misconception. It is very easy to change low oil balance without affecting high oil balance.

For example, if we increase build time for mortar from 500 to 5000 then mortar becomes absolutely worthless regardless of sensor behavior because it will take 2 minutes for each mortar tank. If we change it from 500 - 800 then it will take longer to get 20 - 25 mortar tanks and in that same amount of time opponent could have more rocket/cannon so now it's balanced again but with better unit control behavior.

Hopefully this makes sense.

We just need to calibrate the build timse so that 20 - 25 mortars under new behavior is balance with the amount of rockets/cannons needed to overcome mortar. The same goes for any other artillery unit.

If a unit is overpowered then adjusting build time simply means you reduce their effectiveness by reducing their quantity. Adjusting the build time and build power is the only rebalancing solution without affecting other balance variables.
The 25% won't change much after all, since there is no direct anti indirect weapon what will make it easy to stack up that units.
25% was just an estimate to illustrate the solution of increasing build times.

This is one of the reasons why engine upgrades need to make units progressively faster. The only reason groundshaker tanks win in current high oil balance is because units move too slow to get close enough, not because of splash damage. Rocket or flamer should be counter to mortar droids since they are supposed to move faster than mg and cannon

And don't forget that AI is always improvable. I could give you many ideas how you could ease a lot of things.
feel free to share your ideas. I know what it's like to have many ideas but just need to organize them for communication purposes.
Would be nice if you could show a multiplayer beta game so we can see a bit about the balance.
Currently you won't find regular games opened.
From a stats/balance perspective, not very much has seemingly changed. One change that I think needs to be reverted in 3.2 are the factory upgrade modifiers. Beforehand factory upgrades went 30, 120, 210, 270 but now they go 30, 60, 90, 120 meaning units produce slower in middle - late game. However, I don't think this was intentional. It turned out like this because the factory upgrades only use upgrades 1, 4, 7 and 9 but both 3.1 and 3.2 have the same incremental modifier (30).
railgen
Greenhorn
Posts: 9
Joined: 14 May 2016, 00:16

Re: Release 3.2.0 beta2

Post by railgen »

Making one balance change will cause who knows how much other exceptions, that were not included in calculations.
MIH-XTC wrote:Mortars are hardly ever used on tanks and as a result they cannot be used as alternatives to mg, rocket and cannon but now they can be with auto-assignment to sensors. That's what makes this change good, we just need to recalibrate the build time.
Mobile and static mortars are not good in the integrated warzone maps.
Long range actually is an bad option as well, since vtol is often superior there.
Many maps are designed for backdoors, so indirect weapons would be totaly useless in case you make them have even more disadvantages.
Are you referring to a certain map modification?
MIH-XTC wrote:If a unit is overpowered then adjusting build time simply means you reduce their effectiveness by reducing their quantity. Adjusting the build time and build power is the only rebalancing solution without affecting other balance variables.
Overpowered in very special map types. I would not care that much if it's not the map people are playing constantly.

MIH-XTC wrote:25% was just an estimate to illustrate the solution of increasing build times.

This is one of the reasons why engine upgrades need to make units progressively faster. The only reason groundshaker tanks win in current high oil balance is because units move too slow to get close enough, not because of splash damage. Rocket or flamer should be counter to mortar droids since they are supposed to move faster than mg and cannon
That's true, but it is a matter of opinion.
Fast units means faster games and the maps are very small, so why faster units?
Lowering explosion radius/damage would make the same results. Also it is unrealistic that very heavy tanks can travel the distance artillery can shoot that fast. You should have an idea about how far artillery in real can fire, so it's more realistic to lower explosions then making it seem like maps are ~2,5 km² (one neutral building = 1 tile = 10 m?) sized as max. where the most modern weapons can fire 1 km. Most wz weapons are the contrary to nuke weapons, already, so it would not matter much.
MIH-XTC wrote:From a stats/balance perspective, not very much has seemingly changed. One change that I think needs to be reverted in 3.2 are the factory upgrade modifiers. Beforehand factory upgrades went 30, 120, 210, 270 but now they go 30, 60, 90, 120 meaning units produce slower in middle - late game. However, I don't think this was intentional. It turned out like this because the factory upgrades only use upgrades 1, 4, 7 and 9 but both 3.1 and 3.2 have the same incremental modifier (30).
First you must test that on mp games on all possible modes but otherwise I have my doubts as balancing is everything else, but primitive. You can't see all the differences you make just by setting up stats - benchmarks. I guess you know that anyways.

That's also why I asked for a video. If you want to convince with your allegations then this is a must. Myself I just criticize an new update, that has not been tested well yet. Many players already dislike the strength of gs in mp. Some will like it now because it has became easier and others will hate it due to this reason. It simply removes a part of the know how and micro controlling.
MIH-XTC
Trained
Trained
Posts: 368
Joined: 31 Jan 2014, 07:06

Re: Release 3.2.0 beta2

Post by MIH-XTC »

Mobile and static mortars are not good in the integrated warzone maps.
Many maps are designed for backdoors, so indirect weapons would be totaly useless in case you make them have even more disadvantages.
Are you referring to a certain map modification?
no, no map modifications. The fact that mobile and static mortars are not good in the stock maps or low oil maps is the reason why I think this is a good change because now they will become good since they don't need to be assigned to sensors. Now we can play mortar tank + mg cyborg as an alternative to rocket, cannon and MG. I like this idea because as of right now, it's mostly just cannons vs rockets with MG being mandatory, flamer and mortar not really effective. This change allows for better possibilities.
That's true, but it is a matter of opinion.
Fast units means faster games and the maps are very small, so why faster units?
Lowering explosion radius/damage would make the same results. Also it is unrealistic that very heavy tanks can travel the distance artillery can shoot that fast. You should have an idea about how far artillery in real can fire, so it's more realistic to lower explosions then making it seem like maps are ~2,5 km² (one neutral building = 1 tile = 10 m?) sized as max. where the most modern weapons can fire 1 km. Most wz weapons are the contrary to nuke weapons, already, so it would not matter much.
Agreed, faster units mean faster games which may not be a good thing on smaller maps but the reason we have smaller maps is because units move too slow on big maps. The maps were designed around the stats, not the other way around. I think the splash radius for artillery is good, groundshaker is 2 tiles so I don't think that needs to be changed. Maybe splash damage can be slightly reduced, I did reduce GS from 350 to 300 (direct hit is 350). Current track propulsion speed is fine but I think that half track, wheel and hover should get progressively faster while track gets slightly more HP to compensate.
MIH-XTC wrote:From a stats/balance perspective, not very much has seemingly changed. One change that I think needs to be reverted in 3.2 are the factory upgrade modifiers. Beforehand factory upgrades went 30, 120, 210, 270 but now they go 30, 60, 90, 120 meaning units produce slower in middle - late game. However, I don't think this was intentional. It turned out like this because the factory upgrades only use upgrades 1, 4, 7 and 9 but both 3.1 and 3.2 have the same incremental modifier (30).
First you must test that on mp games on all possible modes but otherwise I have my doubts as balancing is everything else, but primitive. You can't see all the differences you make just by setting up stats - benchmarks. I guess you know that anyways.

That's also why I asked for a video. If you want to convince with your allegations then this is a must. Myself I just criticize an new update, that has not been tested well yet. Many players already dislike the strength of gs in mp. Some will like it now because it has became easier and others will hate it due to this reason. It simply removes a part of the know how and micro controlling.
I will make a video of changes that I propose once they are finalized but right now still making modifications. I don't have anything else to say on this topic. I think artillery being able to use sensors without assignment is good because controlling artillery tank right now is too cumbersome and not effective.
Per
Warzone 2100 Team Member
Warzone 2100 Team Member
Posts: 3780
Joined: 03 Aug 2006, 19:39

Re: Release 3.2.0 beta2

Post by Per »

MIH-XTC wrote:From a stats/balance perspective, not very much has seemingly changed. One change that I think needs to be reverted in 3.2 are the factory upgrade modifiers. Beforehand factory upgrades went 30, 120, 210, 270 but now they go 30, 60, 90, 120 meaning units produce slower in middle - late game. However, I don't think this was intentional. It turned out like this because the factory upgrades only use upgrades 1, 4, 7 and 9 but both 3.1 and 3.2 have the same incremental modifier (30).
Nice catch! Fixed.
MIH-XTC
Trained
Trained
Posts: 368
Joined: 31 Jan 2014, 07:06

Re: Release 3.2.0 beta2

Post by MIH-XTC »

Quick question to whoever converted 3.1.CSV to .3.2 JSON since someone obviously had to go through this, how did you programmatically create the nested .JSON arrays for certain parameters.

I'm having trouble creating the nested .JSON arrays from .CSV input.

Was that you Per?

Can we go back to .CSV pls lol?
railgen
Greenhorn
Posts: 9
Joined: 14 May 2016, 00:16

Re: Release 3.2.0 beta2

Post by railgen »

I'm also looking for something that can parse the .ini files quickly to .json.
Maybe add something to the toolset?

I hope 3.1.x converted stuff is compatible in 3.2..
KlassKill
Trained
Trained
Posts: 37
Joined: 04 Oct 2014, 21:35

Re: Release 3.2.0 beta2

Post by KlassKill »

Per
Warzone 2100 Team Member
Warzone 2100 Team Member
Posts: 3780
Joined: 03 Aug 2006, 19:39

Re: Release 3.2.0 beta2

Post by Per »

MIH-XTC wrote:Quick question to whoever converted 3.1.CSV to .3.2 JSON since someone obviously had to go through this, how did you programmatically create the nested .JSON arrays for certain parameters.
See tools/conversion/ini2json/ in source for the python tools. There are no tools to convert directly from CSV -> JSON, unfortunately, since we stopped by INI on the way.
User avatar
NoQ
Special
Special
Posts: 6226
Joined: 24 Dec 2009, 11:35
Location: /var/zone

Re: Release 3.2.0 beta2

Post by NoQ »

MIH-XTC wrote:Quick question to whoever converted 3.1.CSV to .3.2 JSON since someone obviously had to go through this, how did you programmatically create the nested .JSON arrays for certain parameters.

I'm having trouble creating the nested .JSON arrays from .CSV input.
You may also have a look at csv2ini.py at https://github.com/haoNoQ/wztools2100 so you might be able to chain those together (it converts most of the stuff, but it has bugs, which i haven't investigated yet, so just go ahead and convert and see if anything goes wrong and fix manually in the converted stats and/or fix the script and pull request).
User avatar
NoQ
Special
Special
Posts: 6226
Joined: 24 Dec 2009, 11:35
Location: /var/zone

Re: Release 3.2.0 beta2

Post by NoQ »

Guys, could you have a look on how Nullbot3 behaves here? Direct download link to the latest version: download/file.php?id=15422 (and pick the new AI in the list).

Because i plan to replace the current Nullbot with Nullbot3 in one of the next betas, since it should behave better in most cases, and lag up the game a lot less then the old one. Before i push, i plan to fix the build order (because balance changes were reverted) and add a Hover personality (not sure if it's worth it because the generic personality should already be good enough on hover maps) and probably a Flamer personality.
User avatar
montetank
Regular
Regular
Posts: 642
Joined: 14 Feb 2013, 00:05
Location: Montenegro

Re: Release 3.2.0 beta2

Post by montetank »

NoQ wrote:Guys, could you have a look on how Nullbot3 behaves here? Direct download link to the latest version: download/file.php?id=15422 (and pick the new AI in the list).

Because i plan to replace the current Nullbot with Nullbot3 in one of the next betas, since it should behave better in most cases, and lag up the game a lot less then the old one. Before i push, i plan to fix the build order (because balance changes were reverted) and add a Hover personality (not sure if it's worth it because the generic personality should already be good enough on hover maps) and probably a Flamer personality.
Not bad-not bad. The beginning is human like. I tested with your roughness map with scavs. The AI order was research fac-factory-power generator-then the 3 truck helped to build the oil-then command center. They worked together fine. But now the bad news: No combat units were build after 4 minutes. Instead the 4th and 5th truck were build and a second power generator and a 3rd research fac. Not very intelligent. The first combat unit after 6 minutes..... :roll:
Very good: A truck wanted to leave the base and noticed the scavs. :3 The truck returns to the base. no suicid :D
I wanted to made a screenshot, but the game crashed every time, when i try to make a screenshot. I dont know why.
Its my first impression.

PS The 4th and 5th truck were build after the 3rd truck noticed the scavs.
In case the WZ-game ends in a draw , the game winner will be determined by penalty shoot-out.
User avatar
NoQ
Special
Special
Posts: 6226
Joined: 24 Dec 2009, 11:35
Location: /var/zone

Re: Release 3.2.0 beta2

Post by NoQ »

Yeah, as i mentioned, early game will change.
Post Reply