Scaling seems more broken than ever

simonsez

Malorick wrote:

It was a good fight with you

Yes, it was the toughest I've ever had. When I saw you pull away on the next to last night, even though I spent the entire super bowl grinding away, I knew you had me. Congrats!

simonsez

re: links from IceX, when he says "scaling" doesn't take into account levels, that's not to say the initial opposition levels aren't a function of your levels. We tend to use "scaling" a little too freely as a proxy for "difficulty", but specifically, "scaling" describes how the difficulty increases, and I never thought that was a function of level either. As phantron (I think) said, the scaling is most likely a fixed percentage for everyone, but when that rate is relatively low, those who started a low point will feel it much less.

Arondite

simonsez wrote:

re: links from IceX, when he says "scaling" doesn't take into account levels, that's not to say the initial opposition levels aren't a function of your levels. We tend to use "scaling" a little too freely as a proxy for "difficulty", but specifically, "scaling" describes how the difficulty increases, and I never thought that was a function of level either. As phantron (I think) said, the scaling is most likely a fixed percentage for everyone, but when that rate is relatively low, those who started a low point will feel it much less.

Odd then, how I can add 200 levels to my characters and my starting points remain the same. Hmm.

simonsez

Arondite wrote:

Odd then, how I can add 200 levels to my characters and my starting points remain the same. Hmm.

The same compared to what? Every event is different. Did you level your roster and then jump into your hot tub time machine and replay the event? Otherwise, I don't know what you're comparing.

Why not open another thread where everyone can post their roster screenshot, and a screenshot of their opposition levels? But please don't do that here.

Arondite

simonsez wrote:

Arondite wrote:

Odd then, how I can add 200 levels to my characters and my starting points remain the same. Hmm.

The same compared to what? Every event is different. Did you level your roster and then jump into your hot tub time machine and replay the event? Otherwise, I don't know what you're comparing.

Why not open another thread where everyone can post their roster screenshot, and a screenshot of their opposition levels? But please don't do that here.

Of course the events are different and use different characters, but these different characters seem to start universally at the same levels every event I've ever participated in. If there has been variation, it's been so slight as to go literally unnoticed when I'm making an effort to discern an effective difference.

dr tinykittylove

Arondite wrote:

simonsez wrote:

Arondite wrote:

Odd then, how I can add 200 levels to my characters and my starting points remain the same. Hmm.

The same compared to what? Every event is different. Did you level your roster and then jump into your hot tub time machine and replay the event? Otherwise, I don't know what you're comparing.

Why not open another thread where everyone can post their roster screenshot, and a screenshot of their opposition levels? But please don't do that here.

Of course the events are different and use different characters, but these different characters seem to start universally at the same levels every event I've ever participated in. If there has been variation, it's been so slight as to go literally unnoticed when I'm making an effort to discern an effective difference.

I've done the comparisons with my alliance mates at the start of events with no one having played any nodes yet. They get starting enemies at much higher levels than I do. Period.

It's improved somewhat in recent times - seems more set to an average of a number of your top characters than just the highest levels, so my flat levelled 150 roster can get similar enemies to someone with a couple of 166s and others all lower levels, but the 4* players still get at least 15-30 levels higher than I do TO START WITH.

Arondite

dr tinykittylove wrote:

Arondite wrote:

simonsez wrote:

Arondite wrote:

Odd then, how I can add 200 levels to my characters and my starting points remain the same. Hmm.

The same compared to what? Every event is different. Did you level your roster and then jump into your hot tub time machine and replay the event? Otherwise, I don't know what you're comparing.

Why not open another thread where everyone can post their roster screenshot, and a screenshot of their opposition levels? But please don't do that here.

Of course the events are different and use different characters, but these different characters seem to start universally at the same levels every event I've ever participated in. If there has been variation, it's been so slight as to go literally unnoticed when I'm making an effort to discern an effective difference.

I've done the comparisons with my alliance mates at the start of events with no one having played any nodes yet. They get starting enemies at much higher levels than I do. Period.

It's improved somewhat in recent times - seems more set to an average of a number of your top characters than just the highest levels, so my flat levelled 150 roster can get similar enemies to someone with a couple of 166s and others all lower levels, but the 4* players still get at least 15-30 levels higher than I do TO START WITH.

I haven't personally seen anything to confirm or support this, but let's operate under the assumption that this is 100% established fact.

If the guys are 15 levels or so higher and they use a lower level roster (like the people who are playing against the 15 level lower enemies) then they'll take less damage, resulting in a slower up-scale, meaning the enemies will look the same after a few clears.

Zen808

Simon & Arondite: I find it funny that you guys are at each others throats.

Simon is saying that scaling punishes guys for having good rosters.

Arondite is saying that you can manage scaling by not using your top characters.

You guys are basically saying the same thing.

fmftint

simonsez wrote:

re: links from IceX, when he says "scaling" doesn't take into account levels, that's not to say the initial opposition levels aren't a function of your levels. We tend to use "scaling" a little too freely as a proxy for "difficulty", but specifically, "scaling" describes how the difficulty increases, and I never thought that was a function of level either. As phantron (I think) said, the scaling is most likely a fixed percentage for everyone, but when that rate is relatively low, those who started a low point will feel it much less.

This is what confuses most players when scaling is discusses, there are 3 types of scaling:
1 Baseline - This is where you start and is determined by your roster when events begin
2 Personal - This is based on your performance
3 Community - This is based on everyone playing an event (unknown if isolated to your slice)

Arondite

808SpicyToro wrote:

Simon & Arondite: I find it funny that you guys are at each others throats.

Simon is saying that scaling punishes guys for having good rosters.

Arondite is saying that you can manage scaling by not using your top characters.

You guys are basically saying the same thing.

I wouldn't say that we are at each other's throats. I don't think either of us has been disrespectful to the other.

Unknown

There hasn't been a penalty for 'winning too easily' for a very long time. Part of the problem is that back when that existed it was trivial for the top rosters to dominate because anyone can turn an easy victory into a messy one. All you have to do is win like normal and then start doing dumb things when the enemy is down to one guy and you can instantly down him if needed. I remember when this worked I often have a commanding lead in PvE while still having lower absolute enemy levels, and I don't even have to fake those encounters because all you do is say if an encounter features level 200 guys, take 3 level non top tier 166s and you'll engineer your epic messy victory without even trying, because that's pretty much the only way you're going to win. If there's a node with say 3 level 150s then you just take 3 2*s instead, and so on. If you do this right you'll see enemy levels remaining static and sometimes even going down while you play, and it's far easier to engineer these kind of scenarios with a strong roster compared to a weak one. My best guess is that personal scaling is very low now because you can't even purposely try to drive up your scaling to 395 in most events if you tried and community scaling is the dominant form of scaling, and if community scaling is dominant it basically means nothing you do really matters because the community is what sets the difficulty. Even in high scaling events like Simulator Basic hard, it'd not be unusual to go through a full clear and see no noticeable changes in your enemy levels, and then come back in 8 hours and found everything went up by 50+ levels, which has to be community scaling at work.

Unknown

simonsez wrote:

re: links from IceX, when he says "scaling" doesn't take into account levels, that's not to say the initial opposition levels aren't a function of your levels. We tend to use "scaling" a little too freely as a proxy for "difficulty", but specifically, "scaling" describes how the difficulty increases, and I never thought that was a function of level either. As phantron (I think) said, the scaling is most likely a fixed percentage for everyone, but when that rate is relatively low, those who started a low point will feel it much less.

It was very exciting watching you duke it out with Malorick. Nice fight guys.

A few days ago, I sent a PM to IceIX asking him to clarify whether initial enemy levels is considered scaling or not. No reply yet. (Does he even reply to PMs?)

I don't have data on whether or not higher level rosters fight harder or longer fights. The only thing I know for sure is that HP grows linearly, so that for a level 100 roster, a level 200 fight should be on par with a level 200 roster fighting a level 400 fight. If skill and match damage grows proportional to HP, then this math should be correct.

Personally I don't think I'm having an easy time with a level 100 roster. Only the boosted characters and very few select team combinations can hope to bring down an enemy team that has a level 270 Ares. Basically the solution to most hard fights are 2*Magneto + 2*Storm + boosted attacker like LThor, or OBW + boosted attacker + another support or attacker, or 2*Magneto + Mystique + someone with a decent black ability. The breadth of the roster helps only in that you have a wider choice of boosted attackers to choose from to fit into one of these teams. If there are no better team compositions than these at the 3-star or 4-star tier, then you could be right that 2-stars have it easier.

Unknown

Davyx wrote:

Personally I don't think I'm having an easy time with a level 100 roster. Only the boosted characters and very few select team combinations can hope to bring down an enemy team that has a level 270 Ares. Basically the solution to most hard fights are 2*Magneto + 2*Storm + boosted attacker like LThor, or OBW + boosted attacker + another support or attacker, or 2*Magneto + Mystique + someone with a decent black ability. The breadth of the roster helps only in that you have a wider choice of boosted attackers to choose from to fit into one of these teams. If there are no better team compositions than these at the 3-star or 4-star tier, then you could be right that 2-stars have it easier.

A multiplier on levels preserves whatever the original balance is. It'd only be fair if the encounter started out fair for both. If it was unfair toward one person, then multiplying the levels by 2 is still just as unfair to the same person as before. Here's the problem. It's literally crazy talk to expect to look at a roster and say 'yep, level 172 sounds about right'. Someone has to be screwed in this arbitary process. Right now it's the max rosters getting screwed but it can easily be the other direction too, and if max rosters have an advantage here then multiplying everything by 2 is still in favor of them. People would normally expect the weaker players being screwed since that's usually what happens when you're weaker. And let's say for some reason the point is that people shouldn't have strong rosters for PvE. Even then, the process still feels arbitary. That is, you can see people with very strong rosters that still have significant difference in enemy levels. When you don't even know why the playing field is the way it is, that tends to make people angrier. If D3 just said 'if you got a guy at level 166 all enemies are always level 395', that'd obviously be pretty unfair but at least you wouldn't have to start guessing why, and oddly enough in this scenario it is consistent with the generally accepted notion of difficulty. That is, if you have just 1 level 166 everything is very hard for you, but the more guys you have at 166 or higher the easier everything becomes, so no one feels they're punished for having a strong roster (ignoring the very obvious choice of just leaving guys at 165 in this scenario).

Additionally, when you start increasing levels, certain characters ceases to be viable as options due to a lack of HPs versus match damage. The recent high level match damage nerf helped, but it still doesn't matter on goons. For example say you load up Magneto and Mystique and say my infinite combo can beat everything. That's great but a level 300 goon/villian combo can drop a 1K match 4 on you and Magneto only has 4000 HP. How sure are you about your infinite combo if Magneto takes a 1K match 4 as soon as the fight started? At this point the best outcome is likely you beat the fight while down one health pack because you probably end up accumulating about 2K worth of match damage on Magneto and you sure aren't starting the next fight with 2000 HP. You could simply panic and make the wrong move, or lose Magneto to an especially good cascade. This is because there isn't a 3* or 4* version of MN Mag to use instead to compensate, so when the enemies get higher he's just increasingly harder to use. This issue is mostly addressed now in villian only configuration, but villian/goon mix tends to be the hardest nodes where you do need your tricks.

Arondite

Phantron wrote:

There hasn't been a penalty for 'winning too easily' for a very long time. Part of the problem is that back when that existed it was trivial for the top rosters to dominate because anyone can turn an easy victory into a messy one. All you have to do is win like normal and then start doing dumb things when the enemy is down to one guy and you can instantly down him if needed. I remember when this worked I often have a commanding lead in PvE while still having lower absolute enemy levels, and I don't even have to fake those encounters because all you do is say if an encounter features level 200 guys, take 3 level non top tier 166s and you'll engineer your epic messy victory without even trying, because that's pretty much the only way you're going to win. If there's a node with say 3 level 150s then you just take 3 2*s instead, and so on. If you do this right you'll see enemy levels remaining static and sometimes even going down while you play, and it's far easier to engineer these kind of scenarios with a strong roster compared to a weak one. My best guess is that personal scaling is very low now because you can't even purposely try to drive up your scaling to 395 in most events if you tried and community scaling is the dominant form of scaling, and if community scaling is dominant it basically means nothing you do really matters because the community is what sets the difficulty. Even in high scaling events like Simulator Basic hard, it'd not be unusual to go through a full clear and see no noticeable changes in your enemy levels, and then come back in 8 hours and found everything went up by 50+ levels, which has to be community scaling at work.

See, you say that "that hasn't existed for a long time", yet ice referred someone to that post in the past week. Why would he bring up a post that was no longer relevant? His referencing said post is the only reason I discovered it - I certainly haven't even been playing that long.

Unknown

simonsez wrote:

Take a look at the rosters of the top scoring alliances in Thick as Thieves. If your shard is anything like mine, it's full of people closing in on 200k with 2* rosters. Looking at my bracket at the people fighting for top 5, two of them don't have a maxed 3, and another has just a handful, with no 4s. If scaling were balanced, you'd see a proper mix of rosters at the top end, but instead, it looks way more heavily-skewed towards weak rosters than before.

It used to be these weak rosters would eventually scale themselves out, but I don't see that happening here at all. If anything, I was the one who got scaled out. You can't do a proper end grind when you've got nodes that are a potential wipe, or take 15 minutes to clear with mystique/mag.

So is this the intended paradigm now? PvP is for the Xf/GT crowd, and PvE is for everyone else? Because based on the leaderboards I'm seeing, that's pretty much how this one is turning out.

This is how it's been since I've always been here (110 days). I read Polarity's guide Day 1 and kept my 3* at 130. Scaling was very easy for me. I have 5 3* & 1 4* fully covered but at 100-130 lvl. This seems to be the sweet spot at doing well in PVE and being able to compete in PVP. For what it's worth.

I LOVE that 2* dominate PVE. It's literally the only thing they can do in game. But as someone said, it doesn't help them. They have a 3*/4* with a few covers, can't get any other covers, barely have iso to lvl it, if they do lvl it it screws them on PVE......which is literally the only thing they can do in game.

Instead of softball Q&A videos, Devs should be addressing this.

Unknown

Arondite wrote:

See, you say that "that hasn't existed for a long time", yet ice referred someone to that post in the past week. Why would he bring up a post that was no longer relevant? His referencing said post is the only reason I discovered it - I certainly haven't even been playing that long.

Even with high scaling, most games with X Force do look like complete demolition, partly because if you didn't totally demolish a 300+ team it's likely the reverse happens as soon as they have enough to use any of their moves, but I can't even get my scaling to 395 even if I tried on most recent events not counting any nodes with built-in especially high scaling modifier (goons only or shorthanded). The 'winning too easily' is like back when Magneto can infinite combo on turn 2, you notice you beat a node and then it went up by 30 levels immediately. In fact, that was so obvious that you quickly learn to switch to a strategy like 'match only top row' to make sure you took enough damage to not trip it before you win the game. For the weaker rosters, it's a good thing they got rid of these things because the more powerful your roster is, the easier it is to fake any difficult battle you want.

Unknown

simonsez wrote:

re: links from IceX, when he says "scaling" doesn't take into account levels, that's not to say the initial opposition levels aren't a function of your levels. We tend to use "scaling" a little too freely as a proxy for "difficulty", but specifically, "scaling" describes how the difficulty increases, and I never thought that was a function of level either. As phantron (I think) said, the scaling is most likely a fixed percentage for everyone, but when that rate is relatively low, those who started a low point will feel it much less.

Why would he say this:
viewtopic.php?f=7&t=5451&p=95493
If your roster affected "initial" pve levels? That makes no sense and you're desperately grasping for a nonexistent semantic difference. We have tons of evidence (people with the same rosters seeing very different initial levels in the same event, statements from devs) that your roster levels do not directly affect enemy levels or mmr, and zero evidence that they do

Unknown

Malorick wrote:

I notice that you have delay some of your refresh and this is really what make the difference, this is why we were switching place all the time but I was one step ahead nothing to do with my "poor" roster.

Malorick wrote:

it was not easy for me to face 395 lvl without any 4* like you, I had to play smart and consistent

Hope that doesn't get lost in the shuffle. This was about someone being more efficient, not having it easier. It was actually the other way around, more difficult for him when the nodes maxed out.

simonsez wrote:

And, when you have a 2* roster, you can probably grind those nodes 5 or 6 times each.

Judging from what I saw, all of the strong rosters made up huge ground when it came to the grinding at the end of subs. I forget where I saw it but someone just mentioned being able to grind all 6 stacks in 2 hours. Show me a 2* player that can do that against level 260+ nodes without a massive HP/ISO spend.

I find it hard to believe 270 X-Force/4thor against level 375's is less efficient than trying to Storm/Neto a node 3x their level.

simonsez

Mau-- wrote:

I find it hard to believe 270 X-Force/4thor against level 375's is less efficient

Might've been, but she wasn't an option in the essentials. Buffed LT did nice damage, but he's slow as hell. And how many CotS does it take to kill a 32k health Juggs? And how many headbutts will he get off while you're collecting 70 green AP?

Unknown

TaT's final sub was likely one of the few in favor of strong roster since Magneto/The Hood starts with the shorthanded extra scaling. That is, if they were 3 guys they'd probably still be level 300 for a max roster so fighting 2 guys at 395 is arguably not even harder than 3 guys at 300, while a weaker roster would definitely feel the full weight of the extra 50% levels tacked onto Magneto and The Hood for being shorthanded. Two of the essentials are goons only, which not only has the extra scaling factor for being all goons but also a strong roster is likely to have far more options to deal with CDs than a weak one which renders the enemy levels irrelevent.