Jump to content
Sign in to follow this  
You need to play a total of 10 battles to post in this section.
S4pp3R

[in-depth] How I would balance AA/CVs post 080X

29 comments in this topic

Recommended Posts

Member
575 posts
15,730 battles

My god. You should be replacing that fellow at wg doing balancing. At least you are making some convincing sense even though there are issues here and there. 

But the point remains that this current iteration was poorly planned and what we are now seeing is damagecon. (not using the premium version) It should have at least taken until mid year until the kinks were ironed out.

PS your basic theory for balancing can in reality be used for other ship classes.

Edited by dejiko_nyo

Share this post


Link to post
Share on other sites
261
[151ST]
Member
1,086 posts
5,239 battles
1 minute ago, dejiko_nyo said:

My god. You should be replacing that fellow at wg doing balancing. At least you are making some convincing sense even though there are issues here and there. 

But the point remains that this current iteration was poorly planned and what we are now seeing is damagecon. (not using the premium version)

Thankyou!

I even have proposed Commonwealth CL/DD lines I did up somewhere, and probs once a quarter I do something like the above with a full game change proposal.

So if yourself or others want to see some of these again, let me know!

The biggest issue I have on reddit is getting the posts seen, as memes and the like populate the feed unless you are a CC... 😑

Share this post


Link to post
Share on other sites
Member
575 posts
15,730 battles

Just another addendum: people keep saying PTS doesn't have enough players to generate data. Well, I would say, remove CVs from general play for the time being and let the STs test them out before releasing it public. You can even invite more established CV players as additional testers. If new ships can take months before release, why didn't a bigger project like this undergo more diligent testing?

Share this post


Link to post
Share on other sites
261
[151ST]
Member
1,086 posts
5,239 battles
2 minutes ago, dejiko_nyo said:

Just another addendum: people keep saying PTS doesn't have enough players to generate data. Well, I would say, remove CVs from general play for the time being and let the STs test them out before releasing it public. You can even invite more established CV players as additional testers. If new ships can take months before release, why didn't a bigger project like this undergo more diligent testing?

Only issue would be making sure there were two ST queued up

Share this post


Link to post
Share on other sites
Member
253 posts
1,474 battles
36 minutes ago, dejiko_nyo said:

people keep saying PTS doesn't have enough players to generate data. .....

What particular data? for balance? for load? player composition? Match making? and under what scenario?

For this not enough data thing, there were so many claims being made during the O.8.XX release in the last few weeks,  but those major balancing factors does not need huge user population to simulate and figure out. are they really made by the people who is doing the testing (PT or Internal test), by some PR people just don't know what their technical teams were actually doing?

Edited by tsuenwan

Share this post


Link to post
Share on other sites
261
[151ST]
Member
1,086 posts
5,239 battles
2 minutes ago, tsuenwan said:

What particular data? for balance? for load? player composition? Match making? and under what scenario?

For this not enough data thing, there were so many claims being made during the O.8.XX release in the last few weeks,  but those major balancing factors does not need hugh user population to simulate and figure out. are they really made by the people who is doing the testing (PT or Internal test), by some PR people just don't know what their technical teams were actually doing?

When we play the PTS, 80%+ of players are bots, therefore the data is 'tainted' by a lack of humans so to speak.

WG gets 100 times more date from a day of the live server...

Share this post


Link to post
Share on other sites
Member
575 posts
15,730 battles

I have been on PTS a few times, it's poorly populated and the lag times are soso. Matchmaking there is not reflective of the live environment, in terms of player coordination and regional server "playstyle". Hence the need for live server STs.

Obviously the bots did not abuse the F key either. As I mentioned before, the testing process was done poorly for something more major than introducing Alaska or Azuma which haven't even come out despite them being announced way before this "cv rework".

@S4pp3RI have to strongly agree with the low tier CVs are utter garbage. Just playing ops where previously in a Ryujo I can actually contribute. Now... I rather use my crappy New Mex which can perform better.

I cannot understand how people will be encouraged to play CV line when their initial experience is utterly crap.

Share this post


Link to post
Share on other sites
Member
253 posts
1,474 battles
13 hours ago, S4pp3R said:

When we play the PTS, 80%+ of players are bots, therefore the data is 'tainted' by a lack of humans so to speak.

WG gets 100 times more date from a day of the live server...

1. I have used automated test tool in my work also. Different test scenarios can be composted to look at different factors, In WOWS case, I just can not believe tester population is so low that PT MM can not composite some full human test match.

2. Balance problem appeared in 0.8.0 and later, many of them do not required huge number of professional gamer-testers and scenarios to discover. Factors affecting the last few percent of Win/Lost rate would need large population to confirm. but it does not seemed to be the current situation

Automated test bots are good for regression test to reflect what was working still works. meaning using known user action and behavior to confirm known function and outcome of the software

Edited by tsuenwan

Share this post


Link to post
Share on other sites
Beta Tester
2,280 posts
6,849 battles

as for DD vs CV aside from vision they should make it that the only effective weapon for dd are the rocket

for now any of 3  weapon system are effective against dd. unlike other ship class that has to change ammo to deal respectable damage to dd.

torpedo bomber, increase their arming distance. thus increase the difficulty of hitting maneuverable ships. torpedo bombers are used against capital ship. but in this game it can bomb any type of ship without problem.
diver bomber: usn and rn dive bombers are nasty against dd, they could take half hp of dd. they are far too accurate. need to increase dropping circle.
rocket planes: too nasty. if it didnt kill you, it will destroy torpedo tubes permanently. far too annoying. can also punish heavier armored ship. need some nerf.

idc if they need to buff the HP of planes but right now their damage should be toned down. right now CV is the only ship that can top both spotting damage and damage, and they accumulate spotting damage just by attacking. they are not expossing themselves to harm too. unlike surface ships.

reduce plane LoS. dive bombers should have highest LoS among the squadron since it fly higher, while torpedo and rockets planes have lower LoS. also while performing an attack run. rockets and torpedo bomber should receive LoS penalty reduction, since they are flying lower to attack.

Share this post


Link to post
Share on other sites
Member
253 posts
1,474 battles
5 hours ago, dejiko_nyo said:

I have been on PTS a few times, it's poorly populated and the lag times are soso. Matchmaking there is not reflective of the live environment, in terms of player coordination and regional server "playstyle". Hence the need for live server STs.

Obviously the bots did not abuse the F key either. As I mentioned before, the testing process was done poorly for something more major than introducing Alaska or.....

Bot did not discover F key nor the Hakuryu torpedo, but some human testers did discovered them and reported.

ReportedProblemWOWS.jpg.7a02ea5927d6dfc77cb59efdc56c054d.jpg

Share this post


Link to post
Share on other sites
Member
1,607 posts
10,070 battles
9 hours ago, S4pp3R said:

Shamelessly plugging this post,

Any comments from the forum regulars that are avoiding Reddit?

Great write up. It is nice to see someone actually putting thought and effort into fixing CVs instead of just complaining about them like so many others on Reddit/forums/etc. Have an upvote.

I like most of your suggestions, though I do disagree with some portions.

AA

I am not sure if AA should be unified across gun mounts. Imo they should be balanced individually so as to avoid having situations where a ship has far too powerful AA for its tier because it is using the same AA guns as its later tier brethren. The Scharnhorst vs Bismarck jump in AA is one of the bugs that the 0.8.0.2 hotfix is supposedly going to fix. Perhaps having the same AA guns can be made to have a small variation in damage between tiers and ships, but they shouldn't do all the same damage imo.

DD spotting

I think your suggestion is a bit too complicated for WG to implement. Like how does the DD how much spotted time is remaining? How about the CV? What happens when a plane leaves the spotting radius and comes back? Does the timer reset?

To fix the spotting problem, one idea I have seen others suggest is to simply make ships which are spotted by the CV not render in-game, so they only show up on the minimap. Essentially the CV can only for spot for itself.

Another idea I have is to for planes to have a cone of vision, so they can only spot whatever is in front of the plane and not behind. This means that planes can only spot ships on approach, but not after they fly over or turn away.

A third idea would be to make it such that DD aerial detection does not bloom when it fires its AA. This may make things interesting as DDs would be the only class that can stealth AA.

 

Apart from those issues, I like the rest of your ideas. I hope WG at least takes note of it. :Smile_medal:

Share this post


Link to post
Share on other sites
Member
575 posts
15,730 battles

@tsuenwanThat is the message I hoped that I am wrong about. It's basically reaffirming my fears that wg is playing lipservice to the community instead of listening to the feedback and adjusting the release of stuff.

Share this post


Link to post
Share on other sites
Member
222 posts
368 battles
6 hours ago, dieselhead said:

CV's cant be balanced in this game, its that simple. They will always be OP no matter what you do due to the nature of what they are.

they can..... just more tweak

 

17 hours ago, S4pp3R said:

S4pp3R’s CV/AA 0801/2 Rebalance

Reddit:

 
Please comment, vote on reddit as well if able, I really want to get this idea out there!

they blocked reddit in my country :fish_panic:

you should copy paste it in here

Share this post


Link to post
Share on other sites
Member
253 posts
1,474 battles
1 hour ago, dejiko_nyo said:

@tsuenwanThat is the message I hoped that I am wrong about. It's basically reaffirming my fears that wg is playing lipservice to the community instead of listening to the feedback and adjusting the release of stuff.

On the TDB blog, I remembered reading something written by YUZORAH, WG actually inform him that they could actually gathered enough data to proof Radar Line of Sight did not work,  by their own internal test team.

RadarLOS.jpg.32cb16d9f74a557567d771a5a3ebf2b4.jpg

In fact for the Bigworld engine Wows running on. I understood the effort to implement LOS is quite a tedious job compare with adding the timer.

Share this post


Link to post
Share on other sites
Member
575 posts
15,730 battles
31 minutes ago, tsuenwan said:

RadarLOS.jpg.32cb16d9f74a557567d771a5a3ebf2b4.jpg

In fact for the Bigworld engine Wows running on. I understood the effort to implement LOS is quite a tedious job compare with adding the timer.

This one I remember reading. The issue I have is they know implementation of a mechanic is tedious, they know of the problems and concerns , yet continue to spam radar ships.

It is as if they are creating more problems for themselves. Right now as I have pointed out, they are more concerned with keeping with their timeline despite major issues cropping up and making problems worse.

Share this post


Link to post
Share on other sites
Member
569 posts
8,469 battles
3 hours ago, Akyamarukh said:

they can..... just more tweak

They can't, as if you have 2 cv players in one match and 1 is way better than the other then the influence on the game is too much.

Unlike any other class where 1 OP player wont have the same impact on the match, sure he will impact the match but not to the same extent that an OP cv player will.

Basically have a look at the update thread, there are 9 pages of mostly hate for CV's. What does that tell you? It's not rocket science, most player hate CV's.

Share this post


Link to post
Share on other sites
Member
575 posts
15,730 battles

If Starcraft can match player ability, the CV players should be matched for skill. Yes you will have to wait but it will be the downside of playing a limited player base class. Most of the hate comes from players in specialized lines. From the matches on this server that I've seen, the main problem with CVs in matches is that people play selfishly and do not work as a team. Then the opposing CV proceeds to rip the team to shreds.

The only way I see to improve team play is to reward passive actions like being target practice and spotting on an equal level with damage/kills. It will piss people off but at least then the unsung players who work behind the scenes to victory are rewarded for their sacrifice.

Also, crazy CV idea: Pilot CV like regular ship but instead of guns, shoot planes at target (autofly/bomb/torpedo drop). 1-2 minute reload between "shots" Maximum number of "shots" in air. That will bring CV into line with other ships. :P

Share this post


Link to post
Share on other sites
261
[151ST]
Member
1,086 posts
5,239 battles
5 hours ago, Akyamarukh said:

they can..... just more tweak

 

they blocked reddit in my country :fish_panic:

you should copy paste it in here

I apologise to those unable to read it... and Apologise for the long post... here goes...

S4pp3R’s CV/AA 0801/2 Rebalance

Edit: To those struggling to read the long post and not interested in deep analysis, read the tldr and the CV line ideas (bottom of the post), you may find them intriguing...

tldr

Nerf AA across the board to the point where AA difference between tiers is far less, then nerf planes accordingly; bottom tier AA ships and CVs should find being uptiered far easier. Introduce a maximum spotted time for DDs by CVs per minute, still have DDs appear on minimap, however specific CV can still see DD per normal.

NB

Keep in mind that this is a broad spectrum proposal so don't get too tied down in any of my numbers, obviously balance from that point accordingly. Also as you read I bring up ideas and points that I then quickly turn on and trash, these are musings, please actually read the post before jumping to conclusions - I've tried to look at the whole situation hollistically so something may make more sense further down the post.

I also wouldn't implement my proposed AA nerf one patch and then CV fix in another, both would need to be done in conjunction to avoid CVOP.

 

OK so I’ll list the main issues I see so you understand where I’m working from.

- AA inconsistent across tiers (even prior to 080)

- Zone AA control is boring and silly. Using it can often expose you rather than aid you.

- Bottom tier CVs horrible to play

- Bottom tier AA is horrible to play

- 3 CVs per side shouldn’t break the game

- DDs being perma-spotted isn’t good

- CVs need to have some actual build options so that play isn’t stale.

- CVs need some sort of DPM drop-off to match AA dropoff

- Alternative CV lines need viability and individuality

OK so this is based on the game as of 0801/2 HotFixes, however should still work no matter how many hotfixes they apply as it's essentially an overhaul of numbers.

 

So how do you fix CVs and AA? Where do you start? That’s easy, start with AA.

 

Ship Anti-Aircraft

Ship AA is defined (and has been for a while) as being inconsistent across tiers and being bottom tier against a CV just being painful. The old adage prior to 080 was ‘if high tier CV wants to kill you, they’ll get through your AA’. Since the 080 change we’ve had the Zone AA reinforcement over ctrl-clicking, which basically dumbed down the AA play (IMO and I will explain how).

Let’s start with the Zone AA control. Currently by pressing the ‘O’ key you can increase either port or starboard AA strength, at the expense of the other side. There is usually around a 10s window (depending on your ship AFAIK) while the AA focus switches. There are two reasons why this system is stupid.

  1. Very little risk-reward, and

  2. Over-simplification meaning very little skill involved.

Simply put, the old system should be reintroduced. The immediate response from people is ‘but there’s only one squadron in the sky’ or something of the like. So enable enemy teams to see enemy sqns flying off when they’ve dropped, you now have a risk-reward system. Target the planes that have dropped, trying to hurt the CVs reload times more, or target the planes that are yet to drop in the hope of stopping the next drop. I’ve tested the new system and honestly, it’s more effort to manage than the old ctrl-click system, which is sad.

DFAA as it is now, simply increases the DPM.

AA inconsistency and how horrible it is being bottom tier AA are basically the same thing, AA balance.

AA balance should be blatantly obvious to WG but sadly never has been. Change the whole system, reducing overall DPS across tiers, unify DPM of turret mount types (No more Scharnhorst 105s 300, Bismarck 105s 1700- HotFix 0801, not sure 0802) it’s silly. By reducing the overall DPM (I’m not saying get rid of the Flak system, I actually think it’s a great idea) you reduce the difference tier-to-tier. While you’re at it fix any of the absurd AA ships and while keeping them ahead of the pack, bring them a bit ahead of the pack, not DM/Mino AA build ridiculousness from pre 080.

Now that AA doesn’t have such absurd variations, being bottom tier isn’t as big of a deal anymore. This balances T8 AA v T10 CVs and T10 AA v T8 CVs… If you want to have AA powerhouses, give respective ships ‘better’ DFAA, be it longer or bigger buff or both. This will balance out the issues which will arise from the above system, ship types.

BBs should have the most amazing AA, that is a fact, however if you balance the AA well enough, planes should still get through. CL/As should have situationally better AA through the effective use of DFAA.

If there are issues with certain higher tier ships having worse AA than their predecessor, so be it. Who said the ship up 1 tier in the line had to have better AA? Maybe it’s got other things going for it? The other way to address this discrepancy is to abandon the idea that the ships are historically accurate. They aren’t, Warships is an arcade game, get over it. Once you accept that, you simply add a hypothetical Hull(C) to any ships that are suffering in this department. If a ship has absurd AA for tier, sweet, remove that final AA refit and give it a Hull with ½ a refit. Balance should always trump historical accuracy, if there are too many issues in this area then maybe have a look at what tier the ship sits in?

 

CV Balance

Now that you have a balanced AA system, you address CVs. The fundamental perk from the way the AA system has been rebalanced above is that the differences in tier are less, meaning you can keep +/-2 MM, so now we’ve fixed bottom tier CV/AA. Personally I don’t think there should be +/-2 anyways, but that’s that.

CV Balance will always be a hard thing. However some basic concepts based around the current system should be implemented:

- A full 3 passes of any type should result in an average of 1 DOT (Fire, Flood)

- A top tier ship out on its own should be able to be punished by a bottom tier CV

- CV damage should be on par with BB at tier.

 

By reducing the amount of DOTs that can be applied, CV needs to watch for enemies who have used DAMCON to get some ticks of it, not too dissimilar to normal ships. The reason the chance needs to be so low is the ability of the CV to target just about anyone on the map. To balance out reduced DOTs, damage might need to be buffed a little.

In buffing/nerfing planes HP v AA a simple standard should be applied, roughly how many passes before destroyed. WG could easily work out the optimum number for but for example:

- AA CA DFAA active top tier v bottom tier CV. 1 pass.

- AA CA no DFAA top tier v bottom tier CV. 2 passes.

- AA CA DFAA active same tier. 2 passes.

- AA CA no DFAA same tier. 3 passes.

- AA CA bottom tier DFAA active. 3 passes

- AA CA bottom tier no DFAA. 3 passes, half or a third of sqn survives.

Now I’m not saying the above is balanced, it’s just to illustrate how you go about balancing CVs. You set a standard for standard CV, when you’ve achieved that balance then start adding flavour to the BB by buffing and nerfing certain features, all while keeping it to the above standard.

I honestly think that WG doesn’t actually design properly, like above. All of their ship releases never feel quite right and there’s always this back-and-forth on them after they’ve been released. They’ve already got balanced content in the game, in other words examples of what’s balanced at what tier. This should make it so easy to balance ship releases but I digress…

If you balance CVs well, allowing them to strike with regularity but without huge Alpha, instead of having their planes shredded at the first sign of an enemy, CVs will be fun and they will work. The whole trick is to have AA kill a few planes but planes can still drop stuff consistently. In the end if CV v AA was balanced, 3 CVs per side wouldn’t be an issue.

 

CV damage drop-off

The reason CVs need some sort of damage drop-off is simple, AA mounts degrade. The other reason that no one else ever thinks of is that it allows you to give them a little bit more damage, to keep their damage numbers in the right area. Damage drop-off could be done many different ways, however my fave is replenish rate. The only issue with this is that it indirectly buffs any CVs with long range attacks (see Haku). So the solution is simple, have a standard replenish rate per bomber type per CV. Doing an attack run, you should have to switch bomber types to wait for them to replenish. If that means that as a Haku player you’re essentially doing APDB and TB runs alternating constantly, so be it. If it needs it, you can even make it get longer and longer the more you use a type of Bomber, meaning that CVs will actually have to think about which plane type to launch.

 

DDs shouldn’t be Perma-spotted

There are multiple ways to do this, however my fave is thus. CV sees what his/her planes see all the time, however if you are spotted by CV aircraft, you cannot be spotted for longer than 20s every minute. Doesn’t matter if there are 3 CV SQNs above you or 1, just set it as a hard cap. Normal BB/CA spotter/cata planes would obviously work as they do now. Obviously adjust the 20s window as needed for balance.

 

CV Variety in Builds and Lines

This is literally the easiest thing to do and something it seems WG just hasn’t put any effort into.

Each CV should have options for all three groups, minimum.

Each CV should have loadout options that influence aircraft quality, service times, flight speed or concealment. Let’s call them Engineering Focus.

So how do we do this? There are a multitude of different metrics that can be played with, drop circle, damage, flood/fire chance, DW v normal torps, AP v HE DBs, AP v small AP DBs, AP v HE rockets.

Balancing it may sound like a nightmare but is actually quite simple. When you play any online game the best builds are all about min-maxing. So stack up each area as heavy as you can, see if it breaks the balance. Eg. Haku flood chance. Balance accordingly.

The way I might envision the current two CV lines is as follows (keeping in mind the replenish cool-down system I mentioned above):

 

IJN (Haku line)

Average rockets, poor fire chance. Choice between a bit more damage for almost 0 fire chance and a bit more fire chance but poorer damage or maybe accuracy.

Long range, quick torps, can ‘stealth’ torp at 8 and 10 and almost at 6. Option between standard torps and DW torps (can’t hit DDs).

AP v small AP DBs. AP would be like currently, perhaps with a bit more damage, intended for capital ships. Small AP DBs, for DD-CA.

The overall flavour for the IJN line would be almost no fire options however with slightly better alphas strike potential. TBs would need to be a little squishy so you don’t want to get too close to enemies.

 

USN (Midway line)

Rockets as is now, small or large, refine them a bit further so definite benefits and choice involved.

TBs: more torps per drop, close quarters TB or less per drop but more range.

HE v AP DBs

USN overall would have a bit more survivability and would be more effective with Rockets, while their shorter TB would be very effective against solo targets and HE or AP DBs would offer a choice to have a bit more alpha strike or push fire chance.

 

IJN Second Line (possible idea)

Similar to Haku line, however with some differences.

Rockets, same as Haku line but more per salvo.

Less torps per drop, shorter range, slightly higher alpha, TBs are quicker. Not really a ‘stealth drop’ TBs. Same DW/non options

Same DBs, however less strikes per salvo, faster aircraft, more accurate.

The idea of this CV line is a 'lighter', 'quicker' style of CV play allowing more of a shift in target focus aka more flexibility

 

USN Second Line

Similar to Midway line with some perks.

AP or HE Rockets, AP would be effective against CLs

TBs, 1 less per drop compared to Midway line, far more accurate.

HE v big HE bombs with more pen and fire chance, less number of bombs.

Idea of the second USN line is more about precision strikes and choices around that with a bit less possible DPM.

 

Conclusion

Well the above is how I’d be doing the CV/AA over the next few months if I was a WG dev. Don’t get too tied down in my specifics, it’s just a reference point to see how it could be done. Little balancing issues; WG is perfectly able to do themselves but there are still some gaping issues and ‘beta-ness’ about the CV rework, the AA and just general tier balance.

 

Thoughts from the floor?

I'm very open to in-depth discussion and will reply as able (with respect of course)

Share this post


Link to post
Share on other sites
Member
253 posts
1,474 battles

Unit to Unit engagement dynamic shall be uniformly inline in a game. Otherwise one type of engagement will disrupt the others.  

In WOWS cases,  Air vs Surface unit engagement is very different to surface to surface engagement. This create a situation making the player like playing 2 type of game at the same time. unless the Air-Surface model change enough to be inline with the surface-surface model. This is clearly shown in cases where someone go in to a game with AA build and found no CV.

Edited by tsuenwan

Share this post


Link to post
Share on other sites
261
[151ST]
Member
1,086 posts
5,239 battles
33 minutes ago, tsuenwan said:

Unit to Unit engagement dynamic shall be uniformly inline in a game. Otherwise one type of engagement will disrupt the others.  

In WOWS cases,  Air vs Surface unit engagement is very different to surface to surface engagement. This create a situation making the player like playing 2 type of game at the same time. unless the Air-Surface model change enough to be inline with the surface-surface model. This is clearly shown in cases where someone go in to a game with AA build and found no CV.

I agree, that disparity was what I was trying to address with my proposal

Share this post


Link to post
Share on other sites
Member
1,607 posts
10,070 battles
4 hours ago, dejiko_nyo said:

Also, crazy CV idea: Pilot CV like regular ship but instead of guns, shoot planes at target (autofly/bomb/torpedo drop). 1-2 minute reload between "shots" Maximum number of "shots" in air. That will bring CV into line with other ships. 😛

Tbh, when I first heard about the rework, I thought it would have been something along those lines.

Share this post


Link to post
Share on other sites
Sign in to follow this  

×