Help needed testing 3.2.x Campaign games!

The projects speaking tube.
Add your two cents if you want to.
User avatar
vexed
Inactive
Inactive
Posts: 2461
Joined: 27 Jul 2010, 02:07

Re: Help needed testing 3.2.x Campaign games!

Post by vexed » 13 Jan 2018, 05:48

Sorry for my absence, been traveling (work related) and haven't been here much, if at all. :(

Really, REALLY glad this thread is still active, and people are still testing!
It really does help everyone out to get these things ironed out before the next release!

Thanks to everyone who has contributed so far, if you haven't done so, now is the perfect chance to start from the beginning, and chime in on how things went for you!
/facepalm ...Grinch stole Warzone🙈🙉🙊 contra principia negantem non est disputandum
Super busy, don't expect a timely reply back.

Bethrezen
Regular
Regular
Posts: 584
Joined: 25 Sep 2009, 02:05

Re: Help needed testing 3.2.x Campaign games!

Post by Bethrezen » 13 Jan 2018, 17:32

I loaded up your interlude save, "let me win" all the way to Beta 3 and watched a VTOL hit a python tracked hurricane unit (professional rank). I admit I now see that it does count the second rocket, but it is far from a "one shot"
Not if you only get hit by one vtol no, the issue is that all 4 normally focus fire a single target and when all of them focus fire a single target then it is an instant one shot because what ever value it says in the design screen for lancers you have to double that to get the true damage that a single volley from a single lancer will do before any modifiers because as Alfred already pointed out each shot is counted individually.

So with a 30% damage bonus from research for lancers and a damage value of 600 in the design screen then that would mean that each lancer vtol should inflict 1200 damage before any modifiers are applied and that is indeed what I saw when I tested that yesterday.

[edit]

With the 20% modifier the damage value in the design screen is 480 for lancer vtols so we double that since lancers fire 2 shots and then we multiply that by 4 vtols so

480 x 2 = 960
960 x 4 = 3840

so assuming that all 8 rockets hit a single target of professional rank then that target will still be destroyed we work that out like so

First we find 25% of 3840 why 25% ? because yesterday when i worked out what the damage difference between an unattached rookie python tracked lancer and unattached professional python tracked lancer the damage difference came out at around 25% give or take so

3840 /100 x 25% = 960

3840 - 960 = 2880

AAA only have about 2232 hp and lancers only have 1909 hp so even at professional rank they will be destroyed unless some of the rockets miss.
My guess is that the research change is only seen on saves that have not already granted those rocket damage upgrades yet.
You sure about that because I can load up beta 03 and the damage of the ground based lancer has dropped from 400 to 320 so the damage reduction is defiantly being applied even when loading an existing save.

Having said that however I'm wondering if the reduction to the research bonus should be reverted back to 30% the reason I'm wondering about that is because I didn’t realise that altering the research bonus would also affect ground based lancers as well I had assumed that they used separate values so that you could reset 1 without affecting the other.

See my intention in all of this is simply to bring the damage of lancer vtols down to a point where its survivable and if a single target gets hit by all 4 vtols and all 8 rockets contact its not going to be instantly destroyed, It's why I reckon that when you go in to the design screen lancer vtols shouldn't have a damage value of more then about 235 / 240 and I know that is less then the ground based equivalents and i know at first glance that doesn't make a lot of sense and you are probably wondering why lancer vtols need to have there damage value set so much lower that that of there ground based equivalents, in a nut shell this has to do with the way the damage is applied.

When facing off against lancer tanks or lancer turrets it's rare then a single units is going to take 4 consecutive hits in the space of a couple of seconds which would destroy them if each of those 4 hits where inflicting a total of 1200 damage each before modifiers and assuming that all 8 rockets connected, ground based lancer fire is also easier to avoid because you can just pull back a bit, use the terrain etc.

The issue with vtol lancers and why they can still cause problems even with a damage value lower than that of there ground based equivalents is that the damage is all but impossible to avoid unless you are relying on exploits and all 4 units will focus fire a single target which means that what ever unit is being targeted has no chance of retreating because the damage is applied so quickly in essence its almost like you are taking a single shot that is inflicting somewhere between 3000 and 4000 damage which will just waste most of your units as most only have around 2000 hp give or take and even when you take into account damage reductions from armour and rank it still doesn't lower the damage enough that a single units can take all 4 hits and not be instantly destroyed and It's this that I have a problem with because in my opinion that's not fair, because you can't defend against it.

I mean if you are getting destroyed because you are doing something wrong then that's one thing but if you are being destroyed simply because to much damage is being inflicted to quickly then there is nothing that you can do about that which makes it an unfair advantage and it's why i reckon that the damage of lancer vtols should be lower than there ground based equivalents.

alfred007
Trained
Trained
Posts: 409
Joined: 31 Jul 2016, 06:25
Location: Stuttgart, Germany

Re: Help needed testing 3.2.x Campaign games!

Post by alfred007 » 13 Jan 2018, 18:10

Bethrezen wrote:I mean if you are getting destroyed because you are doing something wrong then that's one thing but if you are being destroyed simply because to much damage is being inflicted to quickly then there is nothing that you can do about that which makes it an unfair advantage and it's why i reckon that the damage of lancer vtols should be lower than there ground based equivalents.
I think the reason why VTOL lancer having a bigger base damage than ground-based lancer is that they need to be rearmed at a rearming pad. A ground-based lancer is standing in front of his target and can shoot one salvo after the other without moving away. A VTOL lancer shoots his salvo and must be rearmed after that. That means that ground-based lancer can do more shoots in a defined time than VTOL lancer. And this disadvantage is compensated by a higher base damage.
At the moment the ratio between the base damages is that the VTOL lancer is doing 50% more damage per shot than a ground-based lancer. But this is no dogma. If you say a base damage for a VTOL lancer of for example 150 is ok but 100 for a ground-based lancer is too low we can get away from that ratio. But because of the fact that VTOL lancer needs a rearming pad, I think it's ok that they are having a bigger base damage.

Bethrezen
Regular
Regular
Posts: 584
Joined: 25 Sep 2009, 02:05

Re: Help needed testing 3.2.x Campaign games!

Post by Bethrezen » 13 Jan 2018, 19:04

I think the reason why VTOL lancer having a bigger base damage than ground-based lancer is that they need to be rearmed at a rearming pad. A ground-based lancer is standing in front of his target and can shoot one salvo after the other without moving away. A VTOL lancer shoots his salvo and must be rearmed after that. That means that ground-based lancer can do more shoots in a defined time than VTOL lancer. And this disadvantage is compensated by a higher base damage.
Don’t get me wrong I don’t disagree with you on that point, however I think you are missing one very important detail.

Lets say for the sake of argument that v1.10 didn’t have the bug that means that damage upgrades are not applied even if that was the case lancer vtols in master would still be more powerful why ACCURACY and that is the key difference here lancer vtols on v1.10 are much less accurate than they are on master so even if lancer vtols had identical damage on both v1.10 and master the ones on v1.10 would still do less damage due to a higher miss rate, this is another unintended side affect of changes that have been made since the game went open source, so to deal with this fact you either need to fix the accuracy issue or you need to lower the damage to compensate for the higher hit rate, as it appears as though who ever who ever increased the accuracy of units failed to take in to account that you also need to reduce the damage of weapons proportionally in order to maintain proper balance. Now in the case of most weapons this isn't such a big deal as they are underpowered anyway but in the case of lancers and particularly vtol lancers this is a big problem that makes units way overpowered.

Therefore in order for the collective lancer vtols to not one shot units when all 4 of them focus fire on a single target then the damage needs reducing or there accuracy needs reducing if you go down the damage route then that means that the damage value of there lancer vtols has to be lower then ground based equivalents.

Unfortunate I can't see any other way to get around this other than buffing the armour of the players units at the start of beta which would have the same effect and would reduce the damage enough that when all 4 vtols focus fire a single target its isn't instantly destroyed the downside to this is that it requires more messing about than simply reducing the damage of the lancer vtols as that would require altering a whole bunch of other things to.

alfred007
Trained
Trained
Posts: 409
Joined: 31 Jul 2016, 06:25
Location: Stuttgart, Germany

Re: Help needed testing 3.2.x Campaign games!

Post by alfred007 » 13 Jan 2018, 21:17

I totally agree with you, that the base damage of lancers must be decreased. What I wanted to say is that even with decreased base damages the base damage of VTOL lancer should be higher than the base damage of ground-based lancer. Buffering the armour of the player's units would bring in more problems in my opinion because this improved armour would have an effect on all weapons of the collective. And at the moment the problem is just the way too overpowered VTOL lancer. So for a fast solution, I prefer decreasing the base damage of lancers. Doing a new balance for the whole game will be a huge project and will need a lot of time and work. So for the next release, we should have an acceptable result and for a really good result, we have to rebalance nearly all the weapons. Hironaru started such a project that is not active at the moment due to the fact that there are no testers. Maybe we should join that project after the next release because he has interesting ideas.

User avatar
Berserk Cyborg
Code contributor
Code contributor
Posts: 601
Joined: 26 Sep 2016, 19:56

Re: Help needed testing 3.2.x Campaign games!

Post by Berserk Cyborg » 13 Jan 2018, 21:53

I made VTOL lancer have 180 damage now in 354207ee89584d9320ea8db938aabf9047445db3. Now I want to see what one of those VTOL tank killers do (375 damage). Those are produced from factories so that is fair I think.

Finished playing through Beta 6. Can't say I saw anything weird in it and completed it in ~20 minutes. At least those Collective VTOLs hiding in the ground pits attack the player again for the first time ever should they survive a trip to the main base.

alfred007
Trained
Trained
Posts: 409
Joined: 31 Jul 2016, 06:25
Location: Stuttgart, Germany

Re: Help needed testing 3.2.x Campaign games!

Post by alfred007 » 14 Jan 2018, 19:47

I updated to master 97a251c and tested Gamma 04.
With the settings in the current master, it is nearly impossible to win. You get outnumbered before you can bring in your first reinforcements. As you can see in the picture below I have to fight 20+ enemies just with the units of my first transport. And because my LZ is compromised I can't get any reinforcements. It was a crushing defeat. And you can nothing do to prevent to get outnumbered because with insane difficulty the SE factory produces 10-15 units any 2:40 and my reinforcements needs 3:00 to arrive. Very simple math.
wz2100-20180114_152401-SUB_3_2.jpg
Then I changed the code in the js-file of Gamma 04 in that way that the SE factory starts production after I reach the rescueTrigger instead of starting production after reaching the phantomFacTrigger. The fast production of the NE and SW factories makes a rush nearly impossible and I had to fight my way south tile by tile. I needed nearly 20 minutes until I could rescue the alpha team. Maybe you can use a suicide squad for a rush but I didn't check that.
I think for insane difficulty it is no longer necessary to add any defences. But after the next release, we should check if it makes sense for normal and hard difficulty. Because when the factory production slows down with these difficulties it could maybe possible to rush again. But this can wait until the next release.

Surprisingly an old bug showed up again and I could build an HQ in this away-mission.
wz2100-20180114_170218-SUB_3_2.jpg
In the logs, we have still mapTile info and fpathRoute errors. In the attached file I included the logs from my first try with master setting, the logs from my second try with changed js-file, a saved game with the build HQ and the changed cam3-2.js file.
Attachments
gamma 4.zip
(128.15 KiB) Downloaded 11 times

User avatar
Berserk Cyborg
Code contributor
Code contributor
Posts: 601
Joined: 26 Sep 2016, 19:56

Re: Help needed testing 3.2.x Campaign games!

Post by Berserk Cyborg » 14 Jan 2018, 20:31

:shock: I finally found a way to stop you. Pushed 6e805bdce4dc0ae1c55ebdb0a6dd99c6ad68c3ab which should stop the command center/relay from being built offworld also which I must have broke in 47d6884c4a7cf18232526639d444f3e3a2ead14e.

alfred007
Trained
Trained
Posts: 409
Joined: 31 Jul 2016, 06:25
Location: Stuttgart, Germany

Re: Help needed testing 3.2.x Campaign games!

Post by alfred007 » 15 Jan 2018, 22:11

I updated to master 6e805bd and tested Gamma 05.

Works really good with just two small issues. The absorbed HQ still disappears. And when Nexus absorbed one of my sensor towers the sound message came several seconds after the tower gets absorbed. First, the tower gets absorbed, then the tower gets destroyed by my RR batteries and after a few more seconds Nexus told me that he absorbed a building. I had a close look at my mini-map when Nexus called something absorbed and only when he absorbed one of my units or RR batteries the sound came with the absorption. With my sensor towers several seconds too late.

I think 16-20 units for every reinforcement is a good choice. Less and it would become too easy, more and the scourge units would kill you immediately. Maybe 2 or 4 more could be possible but at the moment it is good as it is.

It looks to me as if the reinforcements came alternately from the two phantom factories. I did not pay attention the whole level, but the last 10 reinforcements came alternately. Am I right and if yes is that intended? I would prefer they come randomly, that would make this level less predictable.

I defined the assemblies for Gamma 06 and noticed that you forgot the southern cyborg factory. It's object 3 in the labels.json file, but she doesn't appear in the cam3-c.js file. I changed the cam3-c.js file by adding this factory and the assemblies for all factories. Attached are the log files of Gamma 05, the changed cam3-c.js file, the labels.json file for cam3-c, the changed cam3-2.js file ( I changed some comments, that was no longer up-to-date) and a saved game from the very beginning of Gamma 05, so that you can reproduce the reported sound error.
Attachments
gamma 5.zip
(82.66 KiB) Downloaded 18 times

User avatar
Berserk Cyborg
Code contributor
Code contributor
Posts: 601
Joined: 26 Sep 2016, 19:56

Re: Help needed testing 3.2.x Campaign games!

Post by Berserk Cyborg » 16 Jan 2018, 03:32

updated-Campaign.wz
(35.05 KiB) Downloaded 17 times
In summary:
1. Changes to Gamma 4 make the "return to LZ" message play and blip show up when you rescue Alpha team.
2. A couple of typos prevented Nexus from using sounds in eventObjectTransfer... which I misspelled.
3. Gamma 5 randomly picks a location for the phantom factory.
4. Gamma 6 I moved the limbo LZ area (where the units from Gamma 2 are placed) to attempt to prevent units getting stuck past the scroll limits. I also see that the original cleanup area missed a few structures for the southern base so I enlarged it. South cyborg factory produces stuff.
5. I made the reticule buttons update on a timer since there are too many edge cases that can cause problems (this change requires loading a new mission to notice it).

Only problem, aside from the current structure count being incorrect for players 1-10, is that the reticule buttons for manufacture and research no longer show when on Gamma 6. I think levelType is not being updated somewhere when the limbo mission is set to type expand.

alfred007
Trained
Trained
Posts: 409
Joined: 31 Jul 2016, 06:25
Location: Stuttgart, Germany

Re: Help needed testing 3.2.x Campaign games!

Post by alfred007 » 16 Jan 2018, 22:20

I replayed Gamma 04 and 05 and played a few minutes Gamma 06 to check the changes from your mod.
The reticule button for commander disappeared. And neither producing a new one nor recycling all commander and producing a new one helped.
wz2100-20180116_184534-SUB_3_2S.jpg
In Gamma 04 the "return to LZ" message and the blip shows up after rescuing Alpha team.
In Gamma 05 the location for the phantom factory is now picked randomly. But the error with the sound gets worse. After absorbing one of my units Nexus always say it twice. And after absorbing one of my buildings sometimes Nexus says it at the right moment, sometimes not, sometimes too soon. But now never too late. Weird. I also noticed that Nexus doesn't absorb one of my factories, what he always did in 3.1.5. Intention or coincidence?
In Gamma 06 the reticule buttons for manufacture and research no longer show as you noticed before, along with the now missing commander button.
The limbo units no longer get stuck. The produced VTOLs attacked the Gamma base in the west in opposite to 3.1.5 where they went north to attack my units there.

I think Gamma 04 is ok now, Gamma 05 needs some work with the sound and the disappearing HQ and Gamma 06 should show up all reticule buttons.

User avatar
Berserk Cyborg
Code contributor
Code contributor
Posts: 601
Joined: 26 Sep 2016, 19:56

Re: Help needed testing 3.2.x Campaign games!

Post by Berserk Cyborg » 16 Jan 2018, 22:49

alfred007 wrote:I replayed Gamma 04 and 05 and played a few minutes Gamma 06 to check the changes from your mod.
The reticule button for commander disappeared. And neither producing a new one nor recycling all commander and producing a new one helped.
I force the player to have a command relay now so it is the same as how manufacturing requires a factory or design requires an HQ.

Unfortunately, buildbot is broken again which means you can't test everything at the moment (unless you want to try compiling warzone). For Gamma 5 Nexus has another sound that comes from droid.cpp (which I have removed in my campaignJS branch) for unit absorbing. As for the sounds, some of them have a random chance of being played so that there is not sound spam going on constantly.

Nexus can absorb factories, but as you probably have noticed it does not focus solely on base objects now as seen after 8:33 in this video. I have to see how to reset the structure counts so that donateObject() works correctly. After that it is just getting the absorbed factories to produce something.

For Gamma 6 I introduced a function to retrieve the current mission type and removed the global levelType from the engine (changed mid-mission which levelType can not reflect). The reticule buttons should work on that mission again.

alfred007
Trained
Trained
Posts: 409
Joined: 31 Jul 2016, 06:25
Location: Stuttgart, Germany

Re: Help needed testing 3.2.x Campaign games!

Post by alfred007 » 16 Jan 2018, 23:46

Berserk Cyborg wrote:I force the player to have a command relay now so it is the same as how manufacturing requires a factory or design requires an HQ.
You could have said that to me before ;-)
Berserk Cyborg wrote:Unfortunately, buildbot is broken again which means you can't test everything at the moment (unless you want to try compiling warzone).
Me?? Compiling?? Great joke!!! :lol2:

What I forgot about Gamma 06 is that like in Gamma 04 the trucks of the rescued teams disappear if I already have 15 trucks. In 3.1.5 the truck in Gamma 04 became my 16th and in Gamma 06 they became number 17 and 18. And in Gamma 06 the factories and the research facility disappear if you have already 5 in your northern base. I remember in 3.1.5 you could only build 4 of each factory type and also only 4 research facilities from Gamma 01 until Gamma 05. Just from Gamma 06, when you got the 5th factory and research facility in gamma base you had again access to 5 base buildings. I think we should go back to this setting because without a factory and trucks the gamma base is unprotected against Nexus attacks and especially new players would be surprised and get unprepared to protect gamma base.

Bethrezen
Regular
Regular
Posts: 584
Joined: 25 Sep 2009, 02:05

Re: Help needed testing 3.2.x Campaign games!

Post by Bethrezen » 17 Jan 2018, 02:15

Ok so updated to the latest master
warzone2100-master-20180114-192518-6e805bd.exe (14-Jan-2018)

Started again at the alpha beta interlude and I have a few observations.

Beta 01

Ok so Beta 01 is proving to be a bit of a struggle at the moment due the research change also lowering the damage of your ground based lancers, so I'm wondering how best to handle this because I think with this change the player needs a little help, on insane.

With that in mind I'm wondering what armour upgrades to the collective get on Beta 01 ? because I'm of the opinion that there units are probably a little to tough so it might be an idea to delay giving them one of there armour upgrades till a bit later on to account for the lower damage that your units are inflicting

also how come the mini map is so small ?? as you can see in this screen shot and no i didn't zoom it out it started that small which it shouldn't do so i think the scaling is a little bit off

Image

also is there a way to stop the mini map becoming blurry when you increase its size?

one other thing I'm not sure if this is just my imagination or not but the collective units seem to be taking out my turrets quicker than before, so I'm not really sure what is going on there, because normally i would have time to build a second builder and send it to help the first keep the cannon turret that is to the left of your tank factory from being destroyed by there machine gunners but at the moment even when i use the reload bug to get my base back to 100% to save me a bit of time and effort getting stuff repaired they are still taking that out before i can build a second builder so how the hell is that happening ?? did there heavy machine guns get buffed ? or is that scanner somehow directing the mortar pits to fire on my turret which it shouldn't be able to do? because only commanders can command and control artillery turrets like that.

[edit]

Ok so there is definitely something odd going on there because I just looked at a slightly older version and those mortar pits to the north west of your base wouldn't normally target the cannon turret to the left of your tank factory, can't be sure but it looks like somehow that scanner is commanding those mortar pits to fire on my turret in the same way a commander can which definitely shouldn't be happening, well that explains why I can't keep that turret up even when the turret starts at 100%

there is definitely some sort of a behaviour change going on here.

I also notice a behaviour change from the tanks to the right of your base normally the ones that are accompanied by the mobile vtol radar when they take out the lancer turret normally they would back off but they don't seem to be doing that at the moment they just keep attacking stuff which wouldn't be a problem if not for the fact that I have nothing to send to deal with them because the units you get given are in no fit shape to fight and the units from my transport are busy else where.

alfred007
Trained
Trained
Posts: 409
Joined: 31 Jul 2016, 06:25
Location: Stuttgart, Germany

Re: Help needed testing 3.2.x Campaign games!

Post by alfred007 » 17 Jan 2018, 07:21

I tested Gamma 06 with the latest mod from Berserk Cyborg

Although the reticule buttons for manufacture and research don't show up you still can use your factories and research facilities. You just have to click on it. It's a bit uncomfortable but it works.
Gamma 06 looks good so far, the only thing we should possibly talk about is the fact that the produced VTOLs attack gamma base instead of flying north like they do in 3.1.5. And I adjusted the assembly for the southern cyborg factory.
I defined new assemblies for Gamma 07 and got a look at the code of Gamma 08. I noticed that in master the western factory produce hover tanks and cyborgs, but in 3.1.5 only hover tanks get produced in the west. So I deleted the cyborg templates for the western factory in the code for Gamma 08.

The attached file includes the logs from Gamma 06, adjusted labels for Gamma 06, new labels for Gamma 07, the cam3-ad1.js file with the assemblies for the factories and the cam3-ad2.js file with the changed templates for the western factory.
Attachments
gamma 6.zip
(11.94 KiB) Downloaded 14 times

Post Reply