Why not remove 5 stars from scaling?

edgewriter

Just change the algorithm to ignore 5 star characters when deciding scaling for pve matches. Only a handful of players actually have 5 stars with more than two or three covers anyway. The vast majority of us maybe have one or two and that throws off the roster strength when determining difficulty.

I am reading about player selling their five stars so that they can actually be competitive. Isn't that the opposite of what the Devs want?

SnowcaTT

I've long advocated for scaling to just ignore your top "X" number of characters, where X = current number of released 5*'s. Seems like it would help everyone, from 1* rosters to 5* rosters.

But with their new test that obviously wouldn't go far enough - now all of your 3*'s are likely leveled well under the first node the first time you try it. Scaling should somehow take that into balance as well. Perhaps if PVE was more like DDQ - nodes that accepted 2*'s and lower only, nodes that accept 3* and lower only, and nodes that accept everything - that would let folks play all their characters again (if you could have those nodes not scaled crazy high).

Until rewards are removed from placement to progression, it will probably be impossible to fix scaling in PVE.

Unknown

SnowcaTT wrote:

Until rewards are removed from placement to progression, it will probably be impossible to fix scaling in PVE.

This.

But the problem with removing 5*'s from scaling is that people with 2* rosters or less could have characters 2-4x more powerful than their scaling. The top 10 will be nothing but the people who need those covers the least.

It's already happening in the new PvE. ->

Jwhitmire36 wrote:

Half of the top 10 in my bracket have 2* or less rosters, the person ranked 6th currently has over 22,000 pts with these top 6 characters, frustrating...

257 jean grey phoenix
78 cstorm
70 quake
70 thor goddess
68 b panther
55 beast

KGB

I've never understood why they are trying to scale to rosters instead of scaling each battle based on the 3 characters you use for the battle.

As someone who writes software for a living, I'd scale based on the characters you use for the battle rather than try and pre-scale the entire event based on rosters that can consist of 50+ characters of wildly different levels.

Here's how I'd do it:

Player selects 3 characters to use for a battle. The enemies are scaled as follows:
60% of highest character level + 20% of other 2 characters.

So if you use 3 characters of level 100, the enemies are scaled to .6*100 + .2*100 + .2*100 = level 100. If you try and use a level 270 + 2 level 1 characters the enemies are scaled to .6*270+.2*1+.2*1 = 164 (I picked 60% of highest level for this exact reason so players aren't using 1 270 + 2 L1 characters).

THEN the game can make node designations of trivial, easy, normal, hard and adjust scaling so that trivial starts at 60%, easy at 80%, normal at 100% and hard at 120%. The game can then increase enemy scales on subsequent clears by say 5% per clear.

Here's how it would work in the game.

You see a battle that's designated as 'easy'. You select 3 characters of L150 to use. The scaling of the enemies for this battle is then .6*150 + .2*150 + .2*150 = 150. Its an easy battle so they are reduced to 80% which is 120.

You win the battle and want to play again. But your team is damaged and you don't want to use health packs. So you look at your B team which is 3 characters of L100 and replay the battle. The scaling is now .6*100 + .2*100 + .2*100 = 100. It's an easy battle so it's 80% plus 5% more since you've beaten it once so the enemies are now 85% or L85. If you win this one and then switch back to the A team at L150, the enemies would be scaled back to 135 (90% of level 150 based on 80%+5%+5%).

What's not to like here? Players can use their entire rosters knowing each battle is scaled based on what they are using for this battle, what the designation is (trivial, easy, normal, hard) and how many times they previously cleared it. New players with weaker rosters can still play. Veteran players only advantage is deeper rosters with more characters (which seems a reasonable advantage). The big upside is you can once again use lower tier characters you haven't used in a long time and you aren't penalized for having some 5*'s on your roster than you are never using or a few high level 4*s that you reserve for PvP.

KGB

Killinstinct

@kgb: that is a bad idea because it completely removes the incentive to level up. And why would I use other characters when I can use my juggernaut and play vs level 40 enemies.

Unknown

KGB wrote:

I'd scale based on the characters you use for the battle

I think it's a great idea. Sure it will get abused and need tweaking, but I think it's a start in the right direction.

I really hate being at the mercy of boosted characters, I would find the game much more fun if I could just use the characters I enjoy playing rather than the most efficient based on *'s and boosts.

Killinstinct wrote:

@kgb: that is a bad idea because it completely removes the incentive to level up. And why would I use other characters when I can use my juggernaut and play vs level 40 enemies.

Level 40 Jugs + Level 40 fully covered Kamala...Of course it will get abused, but at least it will be strategic. I'd personally find it fun reading what crazy stuff people come up with.

PeterGibbons316

edgewriter wrote:

Why not remove 5 stars from scaling?

Because then you could roster a bunch of 5*s at 255 and get 1* or 2* scaling and dominate PvE

edgewriter

PeterGibbons316 wrote:

edgewriter wrote:

Why not remove 5 stars from scaling?

Because then you could roster a bunch of 5*s at 255 and get 1* or 2* scaling and dominate PvE

So you start an all new game with 1 and two stars? How do you get a "bunch of 5 stars" with that roster? Start out with a great roster you have built for a year and sell everything but the five stars. Then what? Boring. I doubt that would happen.

But alternatively, they could ignore five stars under a certain level cap. I have played with a level 270 1 star silver surfer and though match damage with his colors is nice, its not worth playing when you have fully covered three stars.

KGB

Mau-- wrote:

KGB wrote:

I'd scale based on the characters you use for the battle

I think it's a great idea. Sure it will get abused and need tweaking, but I think it's a start in the right direction.

I really hate being at the mercy of boosted characters, I would find the game much more fun if I could just use the characters I enjoy playing rather than the most efficient based on *'s and boosts.

Killinstinct wrote:

@kgb: that is a bad idea because it completely removes the incentive to level up. And why would I use other characters when I can use my juggernaut and play vs level 40 enemies.

Level 40 Jugs + Level 40 fully covered Kamala...Of course it will get abused, but at least it will be strategic. I'd personally find it fun reading what crazy stuff people come up with.

There is an incentive to level up for DDQ + PvP events. You just now have a reason to keep those 1* and 2* characters that mostly get abandoned once you reach the 4* stage of the game.

If using *too* low a level of characters is a problem the game could easily add a minimum level for enemies regardless of scaling. Say something like L30 for trivial, L40 for easy, L50 for normal and L60 for hard type thing.

KGB

P.S. I use maxed 1* Juggs/Hawkeye regularly in PVE against non board movers up to L120 or so. In the current PVE I used those 2 plus Ms Marvel to clear her node all the way to L130 before I wiped. 4 board shake colors does amazing things and Ms Marvel soaked up the occasional 1000-1500 pt hit I could not get to in time.

Druss

not sure "ignoring" works.

taking a very extreme example: an absolute noob going through the prologue with his Lv 10 IM35 compared with same scenario bet having a Lv 255 1 cover 5*.

Also, I have seen, very early on, mind you, in PvP the 2-3* transitioners using their 1/2 cover 5* - if they are using them then ignoring hardly seems fair as opposed to people in the same position who aren't/can't use them.

I know that PvP & Pve are different (well they are SUPPOSED to be) but I'm sure that the peeps with the 5* still use them, especially with this new format which forces you to only use your top 5/6 characters

IMO, what should happen is that the 5* should start a lot lower lv - say 140 ish & make levelling them for the first 100 levels or so very cheap (1 iso?)

Crowl

Druss wrote:

IMO, what should happen is that the 5* should start a lot lower lv - say 140 ish & make levelling them for the first 100 levels or so very cheap (1 iso?)

That definitely seems like a simple and workable solution, although even then they would need to advise players who don't frequent these forums of the issues they could encounter if they proceeded to level up their shiny new 5* as in most sensible games, if you get one of the best rewards you do not tend to have to jump through hoops to avoid it screwing up your game completely.

The real problem is that more accurate representations of the power of a character will probably add more overhead to the game than they would like as you'd have to consider star rating, level, number of covers, types of covers and that is before you get into other considerations like how good a character is if you really wanted to scale things fairly to everyone.

One thing I just don't get with their attempts at scaling is why their default position always seems to be away from the idea that a stronger roster should make things harder for you when logically you would expect that the challenge should be at worst kept at an even level for all with people being encouraged to keep expanding and leveling their roster with any iso and hp they can get their hands on.

madok

What if we were to combine KGB's idea along with getting rid of placement rewards by moving them into the progression rewards?

If that were to happen, I don't give a damn if you level your roster or not because I get to use my entire roster to have fun.

tizian2015

edgewriter wrote:

Just change the algorithm to ignore 5 star characters when deciding scaling for pve matches. Only a handful of players actually have 5 stars with more than two or three covers anyway. The vast majority of us maybe have one or two and that throws off the roster strength when determining difficulty.

I am reading about player selling their five stars so that they can actually be competitive. Isn't that the opposite of what the Devs want?

My suggestion would be, to get a trigger to lock out own chars for the event. So there are the 5*s ignored for scaling because they are not part of the event for the player.

The other way would be to start 5*s at lv 100 or so.

I´m hoarding my LT´s and cp´s because of the fear to get a 5*. What a game...

Eddiemon

KGB wrote:

As someone who writes software for a living, I'd scale based on the characters you use for the battle rather than try and pre-scale the entire event based on rosters that can consist of 50+ characters of wildly different levels.

How do you make money off your software? Do you give it away and hope to encourage people to buy features?

Your model encourages under levelling your roster, and small rosters incorporating just the best characters. (Unless there are still featured characters required for nodes). In many cases characters above 3 stars become simple liabilities who 'over level' your opponents and should never be used.

Which is all great for a free player but gives no reason to fund the game.

KGB

Eddiemon wrote:

KGB wrote:

As someone who writes software for a living, I'd scale based on the characters you use for the battle rather than try and pre-scale the entire event based on rosters that can consist of 50+ characters of wildly different levels.

How do you make money off your software? Do you give it away and hope to encourage people to buy features?

Your model encourages under levelling your roster, and small rosters incorporating just the best characters. (Unless there are still featured characters required for nodes). In many cases characters above 3 stars become simple liabilities who 'over level' your opponents and should never be used.

Which is all great for a free player but gives no reason to fund the game.

Actually using my model it makes no difference whether you under level or not since enemies are scaled to what you use. There is no incentive to under level at all unless you think have 13 covers on a L70 4* character is going to be that over powered in PvE. Of course you can do that now too as long as you keep your whole roster under leveled.

Anyway the current model promotes under leveling. There is no incentive to level up in PvE because the enemies constantly level up too. I have a roster full of characters I refuse to level up past L100 so I don't get scaled out in PvE because the rest of my roster is still in the 70-90 range.

If my model were adapted the designers could create PvE events (or even just a couple nodes per event) that had a high minimum enemy levels regardless of scaling.

KGB

Koffitok

KGB wrote:

There is no incentive to level up in PvE because the enemies constantly level up too.

According to this line in the notes:
"As your roster gets stronger, you'll see enemies that are relatively easier. (Their levels will still increase, but more slowly than yours do.)"

The incentive to grow your roster is that even though enemies are getting harder with you, the amount they increase gets less and less, so the power differential between your best team and enemy teams gets bigger with more levels. At least, that's what I got from this.

JVReal

Crowl wrote:

The real problem is that more accurate representations of the power of a character will probably add more overhead to the game than they would like as you'd have to consider star rating, level, number of covers, types of covers and that is before you get into other considerations like how good a character is if you really wanted to scale things fairly to everyone.

This is exactly what the test is doing.

"David wrote:

Moore"]Mission Difficulty
The first thing I want to talk about is how the difficulty of the missions are changing. Previously, the system presented you with opponents based on your characters’ levels. There was a couple of quirks with this system. In some circumstances, it could be disadvantageous to level characters, or to add a high-level character to your roster. Now, the game will take into account both your heroes’ power levels and heroes’ levels into account when establishing the difficulty of the mission.
...
* Mission Difficulty Change (commonly referred to as scaling)
o Mission difficulty

- We have changed the way the game determines which level enemies you should go against to be more accurate. Instead of basing difficulty solely on heroes’ levels, it takes your heroes’ power levels into account as well.

This is saying that previously they based scaling on your characters level... 5* with 1 cover was level 255... as damaging to your scaling as a 255 fully covered 4* in the cold calculating formula of the computer.

Now this test is saying if you have a 255 level character, how many covers does it have? (power levels). If you have a level 120 character, how many covers does that have?

I think the calculation is working out decently but I think it just needs to be scaled down to a lower starting point, but what it's actually looking at, seems to be correct. Looking at BOTH your characters level and number of covers (power levels) seems to be a decent way to assess it.

BearVenger

Why not remove 5-stars from PvE?

JVReal

BearVenger wrote:

Why not remove 5-stars from PvE?

because it's the only place I get to actually use them asides from the Big Enchilada.

I have a Phoenix with 1 cover, GG with 1 cover, OML with 2 covers, SS with 1 cover. With the prologue locking down, and the restricted DDQ roster... unless I'm playing seed teams in PVP, I have nowhere to try these guys out.

Without actual capped Trivial nodes as we had previously, I still don't have a place to test them out.

Crowl

Koffitok wrote:

According to this line in the notes:
"As your roster gets stronger, you'll see enemies that are relatively easier. (Their levels will still increase, but more slowly than yours do.)"

The incentive to grow your roster is that even though enemies are getting harder with you, the amount they increase gets less and less, so the power differential between your best team and enemy teams gets bigger with more levels. At least, that's what I got from this.

The problems is that there are clear limits to how much the middle of your roster can grow in levels, so once you have a few maxed 4*'s the scaling means the vast majority of your roster is rendered useless and you are forced to use just those 4*'s and eat through loads of health packs since even those characters rapidly notice getting 5k slashes every few moves.

BearVenger

JVReal wrote:

BearVenger wrote:

Why not remove 5-stars from PvE?

because it's the only place I get to actually use them asides from the Big Enchilada.

I have a Phoenix with 1 cover, GG with 1 cover, OML with 2 covers, SS with 1 cover. With the prologue locking down, and the restricted DDQ roster... unless I'm playing seed teams in PVP, I have nowhere to try these guys out.

Without actual capped Trivial nodes as we had previously, I still don't have a place to test them out.

Thanks for replying, JVReal, and being less of a tinykitty than I deserved. I, too, also have five 1-cover 5*s that I consider a currently useless 5,000-HP sink.

It would just seem like another "rich get richer" decision to not count 5*s in PvE scaling. They already can rule the roost in PvP; making a 360 OML/PHX/GG trio face the equivalent of 4* challenges in PvE makes it easier for them to speed-run a closing grind than a 4* team with similar opposition.