[1.6.0] The game feels entirely overtuned/general feedback

[richardhawk]

Right now because of the disruption shield change in 1.6.1 you can get hit for 8,4 K in nightmare in the final fight so playing on insane isn't even worth it before 1.6.2 comes out if you are planning on trying to win.

That's rough.

*quietly shelves Insane Demonologist*

[GlassGo]

So.
Was playing with this character https://te4.org/characters/35987/tome/3 ... ccf7bd6d40 throught EoR.
Everythign was fine - nobody could even scratch me in lategame, like in mid-game I had just a couple dangerous situations with cultist-based mobs and that's all. My char wasn't bad at all - 100% crit shrug off, 70% res fire, 45% res all, 160 flat damage resistance, and on second time I had 60% fire affinity - which doesn't matter because for it to be useful you need to survive attack.

And suddenly... you already guessed right?



This thing not only oneshotted my char through all defences and resists (1600 and 1800 fire damage).
That Furnace Vent that have 21 raw skill level and 27 effective not only have high damage - you can't even guess not the range of it nor the cd it has.
15! I learned it when it fried my char second time when I could only see it on map via telepathy presumable far, far away from my character.

Also this



and this (and the rest you can figure out for yourself)



Tell me I'm supernoob and could somehow deal with it - I'll gladly hear you out.

For now it all looks like bullshit.

[Snarvid]

Well, that's simple. All you do is...

I mean, you just...

Holy hell. (I'm a little sad you didn't post the screengrab of what a 53.3 level Kinetic Aura does, or the DR from 41.6 levels of Molten Metal, because damn.)

Can we just get some simple and clearly listed talent hardcaps per difficulty level? I mean, something like Normal is 6.5, same as the character, and Nightmare is... I dunno, 8, Insane is 9 or 10, and Madness is 12? That won't help with the early game screwage from having a boss tier level 20 frantic summoner show up when you're 12th level and drown you in War Hounds and Spiders pinning you down, or when the cross-class talent synergies that create accidental Adventurer kill-combos, or whatever is happening with crits, but it would put an end to this sort of monstrosity.

[starsapphire]

Well, that's simple. All you do is...

I mean, you just...

Holy hell. (I'm a little sad you didn't post the screengrab of what a 53.3 level Kinetic Aura does, or the DR from 41.6 levels of Molten Metal, because damn.)

Can we just get some simple and clearly listed talent hardcaps per difficulty level? I mean, something like Normal is 6.5, same as the character, and Nightmare is... I dunno, 8, Insane is 9 or 10, and Madness is 12? That won't help with the early game screwage from having a boss tier level 20 frantic summoner show up when you're 12th level and drown you in War Hounds and Spiders pinning you down, or when the cross-class talent synergies that create accidental Adventurer kill-combos, or whatever is happening with crits, but it would put an end to this sort of monstrosity.

There is a balance problem here.
As game process on, the need for 'stronger enemy' on is always here, and it's a hard demand.
Talent level is the most simple way to achieve it, yes, just raise the number, as those numbers on player raises.
It's not a very good answer, but the logic is very consistent. The player is getting stronger, so enemy should get stronger.
TL 20 is not an illogical answer, considering it's end game, the enemy should always be stronger than previous region, and each region's enemy should at least be able to be theoretically 'threatening' to you, not completely meat on cutting board. If TL 10 in middle game is completely okay, TL 20 in end game is nothing strange. They 'should' be that high.

If you want to hard cap the talent level, but still need the enemy to be 'stronger'. Considering they don't have something like uber talent, you don't want them to be equipped players' deadly combo or achieve players' multi-layer defense options, then the end result will be 'learn more talents' or something like this, however, this strategy would not scale well linearly.
That is, when you enter endgame, the enemy talent list will be likely to be more 'fullfilled' than before. Which means:

• Largely identical skill set.
• Since AI can't deal with too much talent very well, they maybe even weaker with their 'inferior' version with fewer talents at times, but perform strange deadly combo in rare cases.
• Undertested power creep by many cross-class multiple sustain/passives.
• Increased the time to check enemy talent list.
• Some problematic talent, that had been rare before because they are T4 (e.g. Chromatic Fury, Fearscape, high resist penetration elemental mastery), will appear on much more, if not nearly all enemies, and we may need to tweak all of them.

For me, it's mainly because of the first reason: We want boss to perform well their specific area, or fighting with all of those enemies with similar 'jack-of-all-stats' talent table will be repetitive and tiring. Recent changes in the game is encouraging more varied enemies, with their own weakness and strong point, not the 'all [class xxx] enemies are all the same' type.

But in fact, it's a problem that I found hard to answer: What would be the enemy I hope to see in the game actually is? With all those fresh 1.6 changes, AI changes, talent level changes. When we are talking about 'game balance' or something like this, what enemy do I or the community really want? Clearly, nobody want the enemy to be 'completely fully predicatble', that the process of game it is just follow the manual and see the outcome that you have already known. Currently, we are more like in that 'nerf that awful skill because it kills me for no reason' part of it, but there are still things that areimportant and haven't achieve any community consensus, and they would really need a strong-willed developer to think about and take long term plans about.

[GlassGo]

Well, that's simple. All you do is...

I mean, you just...

Holy hell. (I'm a little sad you didn't post the screengrab of what a 53.3 level Kinetic Aura does, or the DR from 41.6 levels of Molten Metal, because damn.)

It added 160 damage on hit, or 1200 on deactivation lol.
Also, I did see this monster crit for 4 K my toxic canister - wich had I belive 80% resist all, not sure about armor.
Molten Metal reduced damage by 300.
This thing had 300+ steampower from Steam Power. And 600+ steam.

In retrospective - I think Improved Geshtalt and Instant Channeling could give me needed damage buffer. Turned out it is important and useful skill.
Or maybe not - I still have no idea how Furnace Vent is working and what it do - nothing in the description tell it has big radius and that mob can trigger it at will. And does it have cooldown or not. This is the problem too in this particular case at least.

It's not a very good answer, but the logic is very consistent. The player is getting stronger, so enemy should get stronger.
TL 20 is not an illogical answer, considering it's end game, the enemy should always be stronger than previous region, and each region's enemy should at least be able to be theoretically 'threatening' to you, not completely meat on cutting board. If TL 10 in middle game is completely okay, TL 20 in end game is nothing strange. They 'should' be that high.

Agree, but here is the problem of that particular case - in lategame nobody could do to me anything.
Literally, those bosses made from giants with whom you fight before entering Slumbering Caves didn't manage to deal to me even 50 hp.
This thing didn't one-shotted me suddenly only because I decided to test various shells and like, checked every moB and thought which shell is better here.
There was no danger at all in lategame - and then this "developer's ambush".
Danger curve shoudl be little more consistent I think.

[Snarvid]

I mean, if you want the enemy to be more powerful on higher difficulties, there are four stats that literally have the word “power” in them (spellpower, mindpower, etc) that the game could buff. Power stats scale consistently, and would not immediately obviate player choices in the way that e.g. uncapped (or even your suggested TL 20) respen talents do, nor open up the same broken inter-talent synergies that very high levels of talents from multiple classes do, and it might filter out a bunch of the unkillable enemies from stacked uncapped defensive skills at the same time. This solution would also avoid the “players never get to land any statuses” effect that relying solely on raw stat boosts (and thus save boosts) would cause, which would destroy a significant number of classes ability to interact with enemies meaningfully. If this were still insufficient you could add a power multiplier per difficulty level and/or per enemy tier (norm/elite/rare/unique/boss).

The player is capped at talent level 6.5 outside of rare circumstances. I have no expectation that anyone actually signed off on GlassGo’s example’s Molten Metal talent and said, “yeah, that’s what it should do at level 41.6, seems fair and balanced,” and I would imagine with would be easier to tinker with and tune power across the board than to make this kind of adjustment on a talent by talent basis. So: boost power, hardcap talents (it doesn’t have to be 6.5, but not far north of it).

Things that tend to scale frequently with talent level, sometimes with stats, and very rarely with power: resistances, global speed, respen, duration, cooldown. These can create much larger problems than would damage or shield numbers (or the existing significant boosts they get to hp) - when uncapped talent levels give a monster e.g. a significant resist all talent whose CD is now lower than its duration, you have a balance problem in a monster that has never been tested in a player.

E: thanks for the numbers, GlassGo! And tried to remove repetitive elements from core power buff suggestion

[whitelion]

Even if you can possibly get around this stuff, should the higher difficulties essentially require you to break the game, or at least play the game in a totally different way, to win? That was the case with 1.5 madness, and I thought the idea of 1.6 was to move away from that model. But it does seem like the current insane meta is that you need to build around being able to survive a small number of cases of massive damage spikes of several thousands of damage, while you can pretty much laugh at the enemies the rest of the time for being useless at killing you (although some may also be unkillable themselves.)

Since you battle your way through a huge number of monsters in a game of ToME, crazy outlier cases are actually fairly likely to happen during a playthrough, and I think that the design needs to take that into account somehow.

[Peep]

Unfortunately, Talent Levels simply do not work unless the spells monsters are allowed to get are greatly restricted.

Ultimately, there is no defense against any sort of resistance penetration or overwhelming damage. I've been hit with Wave Of Powers that strike twice for 4k through 70% Res just by sheer power alone. So, too, have I died to an Invoke Darkness with over 100% Darkness Pen.

Not all talents are made equal, unfortunately, and jacking up the levels is far more disastrous on some then others.

[Snarvid]

Even if you can possibly get around this stuff, should the higher difficulties essentially require you to break the game, or at least play the game in a totally different way, to win? That was the case with 1.5 madness, and I thought the idea of 1.6 was to move away from that model. But it does seem like the current insane meta is that you need to build around being able to survive a small number of cases of massive damage spikes of several thousands of damage, while you can pretty much laugh at the enemies the rest of the time for being useless at killing you (although some may also be unkillable themselves.)

Since you battle your way through a huge number of monsters in a game of ToME, crazy outlier cases are actually fairly likely to happen during a playthrough, and I think that the design needs to take that into account somehow.

I hope it's the idea to move away from it for 1.6, since they have removed a bunch of the interactions that allowed Madness to be winnable.

I posted a poll about boosting power and capping talent levels here. I don't think it's a particularly elegant solution, but if implemented well it would (in conjunction with necessary bugfixes) hopefully patch a preexisting problem that 1.6 significantly exacerbated.

[Drazhya]

I would prefer diminishing returns on enemy talent level gains, rather than a hard cap - and reverting to 1.0 mastery for enemies across the board. So the current 6.5 would be 5.0, the current 13.0 might be 8.0 and so on. Then slap a blanket -70% res pen on enemies across the board so they stop breaking itemization, and I think that more-or-less solves that. Higher-level enemies would continue to get higher-level talents, and extremely high level enemies can sit in out-of-the-way areas as optional challenges.

Things were much better with 1.0 as the world's baseline mastery and 1.3 as the world-saving hero's baseline. Maybe have 1.3 for baseline talents, such as a snake's bite, or maybe only for bosses, but everything else, including randomly-rolled boss classes would be 1.0.

[Snarvid]

I would not be behind a blanket 70% respen for all enemies. At that point, just throw out damage resist as a concept.

Most talents do get diminishing returns at higher levels already. You suggest some sample scale downs, but it really matters how much they get scaled back if we’re going to keep level 50+ talents in the game.

[St_ranger_er]

I would not be behind a blanket 70% respen for all enemies. At that point, just throw out damage resist as a concept.

It's minus 70%, not plus. This way, to make any use of res pen, monsters should have at least 71%+ of penetration.

[visage]

I would not be behind a blanket 70% respen for all enemies. At that point, just throw out damage resist as a concept.

To be honest, I've yet to see a game with the concept of "resistance penetration" or "resistance reduction" where it did not turn into the primary offensive stat, and in my mind that means that the game's entire concept of $DAMAGE_TYPE resistance is broken.

It's particularly broken in ToME where you can run into enemies with 100%+ Resist All, and where you can accumulate enough res-pen that the defender's resistances are nearly or completely negated. Once enemies have very high Resist All, the individual resistances that should be making their defenses distinctive are irrelevant; similarly, once you're running around with 70+% resistance penetration in your chosen damage type you no longer care about what the enemy resistances are. The whole system exists to make itself irrelevant outside of focusing a character on a single damage type. ...and that's actually kind of boring.

Given that enemies can have very high resistance penetration/reduction, any player build focused on building up resistances is playing towards optimizing the average case... while leaving you totally vulnerable to dangerous damage spikes.

I'm quite fond of ToME overall, but this is one of the places where I think the design is poor.

[Snarvid]

Got it on the negative 70%. My bad - doing most of this on my phone and scanned it as a tilde.

I think the only game I’ve seen statuses handled in a way I really liked was Divinity Original Sin 2. It took the notion of “save or die” out of the equation by making resistance to statuses a product of your defenses/damage shields, which every player had - while you had physical or magical shielding up you were effectively immune to (almost) all statuses, and once they were down you were vulnerable to all, which made when and how you stripped defenses predictable and something you could plan for in an individual combat rather that something you build your character for followed by a die roll where you hoped to get lucky.

That’s a whole different system that would not map well onto TOME without making it a whole new game, and I’m hoping for the time being to just get it back to 1.5 level playability.

[Drazhya]

I suppose my problem with capped talent levels is that the game is entirely built around things scaling up with levels, so responding to things scaling too high with 'kill the scaling after a set point' seems out of place.

...Although since most things scale with power anyway, just increasing that that might be fine. Still doesn't really sit right with me, but I'm not overly concerned. I'd rather pull down talent scaling than give it a cut-off point... although talents that negate gameplay mechanics are a problem there. The Unstoppable talent and resistance penetration come to mind. But there I wouldn't say 'no talent levels higher than X', I would just make those talents level off, so for example Unstoppable would stop gaining duration at some point and only add healing.