[LY] Balance adjustments (live testing)

This section is for the organised reporting of bugs or any sort of issues found in the test server.


User avatar
jesusalva
Administrator
Administrator
Posts: 1150
Joined: 14 Nov 2016, 22:20
Location: Brazil
Contact:

[LY] Balance adjustments (live testing)

Post by jesusalva »

Authoritative; yet somehow experimental.

Green line is the old formula, red line is the new one.
Purple line is for comparing with Moubootaur Legends.

Y axis is the real attack speed which will be applied, while X axis is the theoretical one.

Please be my guinea pigs and report how does the new Attack Speed rules fare out in production. While the formula I applied is much superior to the one I used in Moubootaur Legends (see purple line), I, you, everyone else is invited to write down their feedback regarding these changes (specially mages). If changes are deemed necessary, they'll be applied before Christmas event.

ASPD_Formula_Cp.png
ASPD_Formula_Cp.png (29.53 KiB) Viewed 358 times

Who is NOT affected?
Players which attack speed is less than 300ms, such as newbs.
Players who have a full dedicated attack speed build, speed focus maxed out, and are using the best pots to top that.

Who is affected?
All mages are affected, and I really wanna feedback on this, as this is the most drastic change.
Players of any class whose attack speed is currently less than 300ms.

What am I really doing?
This reduces the flexibility on builds at maximum attack speed, meaning that if you are a specialist in attacking blazing fast, you cannot specialize in anything else. Currently, you have enough flexibility to attack blazing fast AND powerfully. This essentially forces you to calculate your trade-off carefully, and due to the flattening effects, reduces the disparity between high level players.
This has a tendency to lower the DPS for most high level players (even for critical-based builds), you may want to try some different builds on test server and redistribute your status points in a more effective manner.

However, adding ASPD calculations to magic is just a fun experiment which I was advised to do. This is what worries me the most: It received zero testing and will either make mages utterly OP machine guns which demolish literally anything, or be actually reasonable idea with just minor tweaks.

In any case, PvP is now stupidly broken (it was already broken, but right now a mage can use a dark concentration potion and become a machine gun), so I'll temporarily boost asorm. It's more powerful and lasts longer but I still do not advise PvP until it is balanced properly. Feedback will determine the next course of action regarding this.

PLEASE GIVE ME FEEDBACK.
Changes will be applied today at 04:00 UTC on live and test server because if I don't apply on both at same time, there's a risk of Segmentation Faults and data loss. Sorry!

Spoiler:

Code: Select all

import matplotlib.pyplot as plt
import matplotlib as mpl

i=0; x=[]; a1=[]; a2=[]; a3=[]
while i < 500:
    x.append(i)
    a1.append(max(i, 190))
    if i < 300:
            a2.append(max(min(300+((i-300)/2), 300), 190))
            a3.append(max(min(i*1.5, 300), 190))
    else:
            a2.append(max(i, 190))
            a3.append(max(i, 190))
    i+=1

mpl.style.use('seaborn-v0_8')
fig, ax = plt.subplots(figsize=(3, 3))
ax.set_title("ASPD Formula", color='C0')
ax.plot(x, a1, "C1", label='Legacy Formula')
ax.plot(x, a2, "C2", label='New Formula')
ax.plot(x, a3, "C3", label='Moubootaur Legends')
ax.legend()
plt.show()
Jesusalva (aka. Jesusaves)
Donate to the project! ─ (Note: If you want to support me instead, Buy me a coffee!)
User avatar
Hello=)
TMW Game Master
TMW Game Master
Posts: 605
Joined: 11 Jun 2009, 12:46

Re: [DIS] Phylactery

Post by Hello=) »

I'm a bit late with feedback but "usual" mages weren't really dependent on atk spd as such, and that's why I wasn't super-interested in this.

  • Things like ingrav install custom-computed attacks by scripts and these have their own idea on atk delay and damage.
  • confringo, etc only used as fallback or default noob spell as it drains mana very fast and once mana drained mage turns into useless dummy for a while.

But according to jes now it sould also apply changes by pots, etc to magic. If its true, it could change hell a lot of things in balancing I guess and I can even imagine at some point some mages could even suddenly dominate battlefield. Quite a subject for testing and measurements to me.

New formula as such looks quite good to me though. However how it now interacts with magic... remains to be seen...

kereszte
Peon
Peon
Posts: 8
Joined: 09 Nov 2022, 12:10

Re: [DIS] Phylactery

Post by kereszte »

jesusalva wrote: 09 Nov 2022, 02:53

Authoritative; yet somehow experimental.

Green line is the old formula, red line is the new one.
Purple line is for comparing with Moubootaur Legends.

Y axis is the real attack speed which will be applied, while X axis is the theoretical one.

Please be my guinea pigs and report how does the new Attack Speed rules fare out in production. While the formula I applied is much superior to the one I used in Moubootaur Legends (see purple line), I, you, everyone else is invited to write down their feedback regarding these changes (specially mages). If changes are deemed necessary, they'll be applied before Christmas event.

ASPD_Formula_Cp.png

Who is NOT affected?
Players which attack speed is less than 300ms, such as newbs.
Players who have a full dedicated attack speed build, speed focus maxed out, and are using the best pots to top that.

Who is affected?
All mages are affected, and I really wanna feedback on this, as this is the most drastic change.
Players of any class whose attack speed is currently less than 300ms.

What am I really doing?
This reduces the flexibility on builds at maximum attack speed, meaning that if you are a specialist in attacking blazing fast, you cannot specialize in anything else. Currently, you have enough flexibility to attack blazing fast AND powerfully. This essentially forces you to calculate your trade-off carefully, and due to the flattening effects, reduces the disparity between high level players.
This has a tendency to lower the DPS for most high level players (even for critical-based builds), you may want to try some different builds on test server and redistribute your status points in a more effective manner.

However, adding ASPD calculations to magic is just a fun experiment which I was advised to do. This is what worries me the most: It received zero testing and will either make mages utterly OP machine guns which demolish literally anything, or be actually reasonable idea with just minor tweaks.

In any case, PvP is now stupidly broken (it was already broken, but right now a mage can use a dark concentration potion and become a machine gun), so I'll temporarily boost asorm. It's more powerful and lasts longer but I still do not advise PvP until it is balanced properly. Feedback will determine the next course of action regarding this.

PLEASE GIVE ME FEEDBACK.
Changes will be applied today at 04:00 UTC on live and test server because if I don't apply on both at same time, there's a risk of Segmentation Faults and data loss. Sorry!

Spoiler:

Code: Select all

import matplotlib.pyplot as plt
import matplotlib as mpl

i=0; x=[]; a1=[]; a2=[]; a3=[]
while i < 500:
    x.append(i)
    a1.append(max(i, 190))
    if i < 300:
            a2.append(max(min(300+((i-300)/2), 300), 190))
            a3.append(max(min(i*1.5, 300), 190))
    else:
            a2.append(max(i, 190))
            a3.append(max(i, 190))
    i+=1

mpl.style.use('seaborn-v0_8')
fig, ax = plt.subplots(figsize=(3, 3))
ax.set_title("ASPD Formula", color='C0')
ax.plot(x, a1, "C1", label='Legacy Formula')
ax.plot(x, a2, "C2", label='New Formula')
ax.plot(x, a3, "C3", label='Moubootaur Legends')
ax.legend()
plt.show()

you just totally destroyed assassin upmarmu build... was this intended? I was playing for an hour earlier, then you performed a server restart and now it's totally useless.

User avatar
jesusalva
Administrator
Administrator
Posts: 1150
Joined: 14 Nov 2016, 22:20
Location: Brazil
Contact:

Re: [DIS] Phylactery

Post by jesusalva »

The upmarmu bug has been "fixed". Well, to be honest, to actually fix it I need to make it honor global limits (pc.cpp:795)

Code: Select all

sd->aspd = std::max(speed, battle_config.max_aspd);

But that would bring back the "bug" so for now I just reverted this part and will leave it as is.

Also, after some live testing, we tweaked the formula, so the nerf is now 10% weaker than earlier today. We keep monitoring and hearing feedback, feel free to poke me on Hurnscald Square anytime. This is a community game, of course your voice will be heard.

Once we're confident enough with the new formula, we'll add new weapons, so keep your eyes open :3

Jesusalva (aka. Jesusaves)
Donate to the project! ─ (Note: If you want to support me instead, Buy me a coffee!)
kereszte
Peon
Peon
Posts: 8
Joined: 09 Nov 2022, 12:10

Re: [DIS] Phylactery

Post by kereszte »

I'm reading that this is supposed to be a 'minor' nerf.

Is 33% small to you? that's huge... If games like WOW did this then it would no longer exist.

Also, why nerf chest during 2x Xp? That was one of the most exciting things about this game. Get chest -> Illia

Now you need to create content to fill in this gap.

Last edited by kereszte on 09 Nov 2022, 17:12, edited 1 time in total.
User avatar
jesusalva
Administrator
Administrator
Posts: 1150
Joined: 14 Nov 2016, 22:20
Location: Brazil
Contact:

Re: [DIS] Phylactery

Post by jesusalva »

kereszte wrote: 09 Nov 2022, 17:09

I'm reading that this is supposed to be a 'minor' nerf.

Is 33% small to you? that's huge... If games like WOW did this then it would no longer exist.

It is "from 0% (min) to 33% (max)" exp/minute nerf. Ginosuper is now generating 100k exp/min less, and that is equivalent to 33% for him. But some players don't even generate that much experience in first place, and the nerf for these builds might be simply non-existing.

Nonetheless, I never said it was a "minor" nerf:

jesusalva wrote: 09 Nov 2022, 02:53

However, adding ASPD calculations to magic is just a fun experiment which I was advised to do. This is what worries me the most: It received zero testing and will either make mages utterly OP machine guns which demolish literally anything, or be actually reasonable idea with just minor tweaks.

In other words, applying ASPD to magic would either be broken or work with minor tweaks. That holds true (it is broken as I anticipated).

kereszte wrote: 09 Nov 2022, 17:09

Also, why nerf chest during 2x Xp? That was one of the most exciting things about this game. Get chest -> Illia

Now you need to create content to fill in this gap.

For two reasons. First is the frequency on which 2x exp events happen. Second is the amount of experience the chest gives. (This is off-topic btw)

If the chest doesn't give enough EXP, the real solution is to increase how much experience it is giving out, not to say you must login at a time that Hello=) is feeling good and decides to make a 2× experience event. After all, experience events can only be noticed by experimentation or by broadcasts ─ so if you only play when GMs are sleeping, you're pretty much excluded from this, as even if the exp bonus is left during the night, you'll not be aware of it.

Making quests rewards invariable is more democratic. If they're insufficient, that's an entirely different matter.
Do you think the swamp chest should give 40 million exp points? If yes, just say it already, we don't bite :alt-9:

And obviously I have content to fill in the gap :alt-5: who do you think I am? :alt-0:
A couple weapons and apparently I'll get add the Witch JackO too :alt-7:
I am hoping to be done with this by Friday but it still needs more testing.

By the way, why no one noticed I increased Phylactery ASPD bonus from 25% to 30%? I wish people would notice buffs more too :alt-4: Even the 10ms on Setzer did not escape the players hawk eyes :lol:

Jesusalva (aka. Jesusaves)
Donate to the project! ─ (Note: If you want to support me instead, Buy me a coffee!)
kereszte
Peon
Peon
Posts: 8
Joined: 09 Nov 2022, 12:10

Re: [LY] Attack speed adjustments (live testing)

Post by kereszte »

I was just reading from here:
https://github.com/themanaworld/tmwa-se ... -10-02.txt
Perhaps this is the incorrect place? It was updated 14 hours ago.

It was Exciting to have to run to the chest at double xp time. More than just going to a chest at any time. Then make the chest more xp. That should be quite simple in the script.

User avatar
Ledmitz
Developer
Developer
Posts: 460
Joined: 17 Aug 2011, 22:40
Contact:

Re: [LY] Attack speed adjustments (live testing)

Post by Ledmitz »

I haven't tested anything live, but for years I knew the setzer made no sense. It required more materials, was made smaller and lighter than it should have been and only alchemists were involved in its creation, so calling it a magical item also made no sense. There were no pluses to attack speed to indicate its speed either... it was just made faster behind the scenes. It also makes sense that casting spells should be affected by all weapon speeds, but I would give exception to magical weapons like the wand and staff. On the other hand, a 33% change would be drastic, but back to the other hand 300k a min is a bit much too. Many of the things players got used to, were never intended to be. upmarmu, even after the initial nerf was still far too powerful. The purpose of an RPG is not to get to a max level, the purpose is to gain power over long periods of time and yet players would still reach max level in time. We used to have a level 99 cap, but it was increased to have players not need to retire characters, but then super high exp stuff was added after that. Also remember that the change affects everyone at the same time, so no one is actually at a disadvantage. They would just take longer to gain exp. I am ok with other changes being made to the setzer to explain its speed better, etc, but let's face it. We are all so busy. Nerfing is the only reasonable option. Tweaking it over time requires testing, feedback and fine tuning. Its a constantly evolving game. We can't be "Evolved" if we don't evolve. I'd have to test it out now to know for certain how drastic it is though.

Ledmitz = Mystic = Mystical_Servant = Ardits = BoomBoom = KillerBee
kereszte
Peon
Peon
Posts: 8
Joined: 09 Nov 2022, 12:10

Re: [LY] Attack speed adjustments (live testing)

Post by kereszte »

Ledmitz wrote: 09 Nov 2022, 19:31

I haven't tested anything live, but for years I knew the setzer made no sense. It required more materials, was made smaller and lighter than it should have been and only alchemists were involved in its creation, so calling it a magical item also made no sense. There were no pluses to attack speed to indicate its speed either... it was just made faster behind the scenes. It also makes sense that casting spells should be affected by all weapon speeds, but I would give exception to magical weapons like the wand and staff. On the other hand, a 33% change would be drastic, but back to the other hand 300k a min is a bit much too. Many of the things players got used to, were never intended to be. upmarmu, even after the initial nerf was still far too powerful. The purpose of an RPG is not to get to a max level, the purpose is to gain power over long periods of time and yet players would still reach max level in time. We used to have a level 99 cap, but it was increased to have players not need to retire characters, but then super high exp stuff was added after that. Also remember that the change affects everyone at the same time, so no one is actually at a disadvantage. They would just take longer to gain exp. I am ok with other changes being made to the setzer to explain its speed better, etc, but let's face it. We are all so busy. Nerfing is the only reasonable option. Tweaking it over time requires testing, feedback and fine tuning. Its a constantly evolving game. We can't be "Evolved" if we don't evolve. I'd have to test it out now to know for certain how drastic it is though.

At lvl130, how long do you think it takes to gain one level (at 350k/min)? What else is there at that level besides just farming in swamp? Even if new stuff gets added, it won't target this player...

Edit(again): Oh yea, and the news states that it will only be 'slightly affected. :lol:

Last edited by kereszte on 09 Nov 2022, 19:54, edited 1 time in total.
User avatar
WildX
Forum Admin
Forum Admin
Posts: 1930
Joined: 07 Aug 2010, 14:13
Location: United Kingdom
Contact:

[DIS][LY] Balance adjustments

Post by WildX »

This is the discussion for changes related to general balance tweaks and tests.
Live test announcement: viewtopic.php?t=21497

jesusalva wrote: 09 Nov 2022, 03:30

Miscellaneous Changelog:

  • Setzer is now 10ms slower

  • Phylactery now gives 5% more ASPD

  • Attack speed bonuses are weaker when you're faster than 300ms

PLEASE NOTE: This type of discussion is intended to let people express their constructive feedback and communicate productively with developers. Complaints or any alternative suggestions that are not deemed on-topic will be moved to the feedback forum. This is a development forum.

User avatar
jesusalva
Administrator
Administrator
Posts: 1150
Joined: 14 Nov 2016, 22:20
Location: Brazil
Contact:

Re: [LY] Attack speed adjustments (live testing)

Post by jesusalva »

This is the new formula for v2022.11.10

ASPD_Formula_2.png
ASPD_Formula_2.png (25.86 KiB) Viewed 215 times

I don't think more buffs to the general formula will be needed.
Nonetheless, help testing. I also modified upmarmu and fixed pots on magic.
So tell me how fast an ingrav with a concentration potion is :alt-7:

TL;DR Unless upmarmu or potions-on-magic needs further nerfs, I am done nerfing and plan in adding new content. Unless people want me to spend more time revising nerfs instead of adding stuff.

Jesusalva (aka. Jesusaves)
Donate to the project! ─ (Note: If you want to support me instead, Buy me a coffee!)
kereszte
Peon
Peon
Posts: 8
Joined: 09 Nov 2022, 12:10

Re: [LY] Attack speed adjustments (live testing)

Post by kereszte »

Why are they nerfing at all? Can the server not handle the load???

I mean 0-day testing is great and all.... but you know

Edit: gotta add this:

"I don't always test my code. But when I do, it's in production."

please leave this post... It will last a looooong time.

Last edited by kereszte on 10 Nov 2022, 02:32, edited 1 time in total.
kereszte
Peon
Peon
Posts: 8
Joined: 09 Nov 2022, 12:10

Re: [DIS][LY] Balance adjustments

Post by kereszte »

In general, people are unhappy about the setzer nerf. And of course the aspd nerf.

Since this is referencing another post, I need to add here:

How often do you think large companies 'live' test? To say it's ok because it's 'a thing' is nonsensical. It goes along with the lack of transparency that you'd expect from an open source game. And if you can't put on a test server because you're afraid of a seg fault - then you obviously don't fully understand what you're doing. Do you need help?

Don't get me wrong, I love this game (and others do too). That's why people are upset. Right now, people don't want to do anything until the new material comes out.

User avatar
WildX
Forum Admin
Forum Admin
Posts: 1930
Joined: 07 Aug 2010, 14:13
Location: United Kingdom
Contact:

Re: [DIS][LY] Balance adjustments

Post by WildX »

"People are unhappy" - of course they are they had access to a completely unbalanced mechanic that gave them an advantage. Players are never going to be "happy" when these things are changed.

The test server issue is for Jesusalva. I do agree a test environment would be a lot better especially because everyone can test on a max level character and spawn the equipment they need, reset their stats and do a lot of other useful things that make testing much easier.

kereszte
Peon
Peon
Posts: 8
Joined: 09 Nov 2022, 12:10

Re: [DIS][LY] Balance adjustments

Post by kereszte »

WildX wrote: 10 Nov 2022, 17:51

"People are unhappy" - of course they are they had access to a completely unbalanced mechanic that gave them an advantage. Players are never going to be "happy" when these things are changed.

The test server issue is for Jesusalva. I do agree a test environment would be a lot better especially because everyone can test on a max level character and spawn the equipment they need, reset their stats and do a lot of other useful things that make testing much easier.

I'm intrigued... How was this completely unbalanced? It was just normal... It's not like warriors make the most xp/min in the game (that's assassins). If they had a distinct advantage over other classes, I'd agree, but they don't (didn't). The only thing I can think of is survivability in the swamp (which has nothing to do with xp/min and didn't change).

So, please inform me how it was completely unbalanced.

Last edited by kereszte on 10 Nov 2022, 18:07, edited 1 time in total.
Post Reply