ToME: the Tales of Maj'Eyal

DISCLAIMER: I'll be speaking of balance concerns, and some things I consider winning requirements. These apply primarily to Insane difficulty, but to a lesser extent to others. Keep in mind that I only play insane, so my opinions on the relative quality/importance of things is from that perspective.

Specifically, these two changes: "Orc babies do not drop loot", and "Ruined Dungeon orb summons do not drop items"

I'm going under the assumption that DG dislikes farming as a design philosophy, which in general I'm all for. However, the nature of the game currently is that if all of the potential for farming is removed... the net result is just a wild increase in randomness of your char's ability to win. Specifically, depending on your class/race, there are certain items that you need in order to progress past certain points on Insane. Currently, the only way to get them is to farm up a storm (a couple other places possible, those two being the most notable though).

Some examples of critical items follow. Mind you, all of the following examples aren't required to win, but a large subset depending on your race/class are, and I cannot conceive of a character without any of them having a chance:

Eden's Guile
Eggsack
Shield of Earth (talent level 4.0+)
Spellhunt Remnants
Wanderer's Rest

There are quite a few others, but will forgoe listing them all for brevity. The problem being that there is no other method to acquire these items (aside from vault transfer, which actually just makes the farming worse as you'd have to just keep starting characters till one got them). Other roguelikes tend to have farming spots at various places to solve this problem, or some non-random means of generating specific things (Wish, generally). Also note that several of the critical items are low-tier artifacts, so without ruined dungeon there's only a narrow window for most of them.

So, bottom line: removing the ability to farm doesn't help your average player in any way (who won't ever do so anyway), all it accomplishes is wildly increasing the randomness (and/or tedium) of those who play on higher difficulties. I'm mostly curious on the rationale behind changes such as this, and whether or not any means of addressing the issue of wild randomness will be addressed at some point. Because, while there ARE definitely better solutions than allowing farming, they're all way more of a pain to implement. Some possibilities follow.

Potential solutions:

1) Make everything so balanced that certain things wouldn't be required. Note: Almost impossible, certainly in the short term. First, wild class imbalance would have to be addressed, coupled with all the really good artifacts either being nerfed or all the really crappy ones being buffed to par.

2) Some non-random way of generating critical things (Equivalent of merchant that allows for super expensive purchase of the regular arts, or some sort of Crafting/Wish system)

3) Probably the most viable, keep some method of farming in for something that explicitly solves this. Ideally, since it's already there, Farportal. Currently it doesn't serve that purpose because several of the critical things are tier 1-3 artefacts, and even if they can drop in there (not sure), it's insanely improbable. So, farming, but manly, dangerous, roguelike farming!

grmblfzzz wrote: 1) Make everything so balanced that certain things wouldn't be required. Note: Almost impossible, certainly in the short term. First, wild class imbalance would have to be addressed, coupled with all the really good artifacts either being nerfed or all the really crappy ones being buffed to par.

This has always been the approach in ToME. It's the hard route, but it's the best route. The problem is players don't report them enough, and will happily spend ages farming for items instead of suggesting an improvement to the system. Say what's overpowered and underpowered and they can be fixed.

Are you serious, Grey? Because I could easily make an exhaustive list of the artefacts. In general, they fall into three catagories:

1) Overpowered/potentially necessary (ten items or so, probably)

2) Situationally handy (as in useful for certain classes, at certain points in the game) (maybe a third of the rest of the arts)

3) Garbage. Worse than any non-horrible non-artefact of the slot, doesn't fill any unique role. (all of the rest of the arts)

Could also do the same with classes/races/prodigies, for that matter. I've always assumed that tweaking items/classes was on a fairly low priority, excepting the flagrant extremes of class balance (way too good or unusable). Because there are some clear but significant issue's relating to the fundamental quality of items/classes/prodigies that seem to only change glacially if at all.

I support this. The item tier change is particularly important. Think of it this way--these items are all too ridiculous for their tiers anyway. Spellhunt being tier 1 is particularly bad.

Farportals seem to be the best mechanic for 'grinding', and work pretty well as is. The bosses are always doable when prepared despite the randomness, but the potential for a surround on entry makes the risk/reward bad for some higher difficulty builds. 2-3 second invuln or shield buff on entry would be a simple improvement to address that.

If DG is receptive to it I'm sure we could come up with some other simple proposals to change or add (maybe egos?) items to alleviate the required drop problem in other ways, too.

A class balance revamp to make different classes at least somewhat more equal is a bit ambitious, but item changes are easy. I'm sure some of us would love to give more detailed balance feedback given the opportunity though.

Even if these changes go through, there's rare bandit lords and elder vampires. I'm just pretty confused since only crazy people farm, and the ruined dungeon is dangerous at any level (void horrors have killed more than a couple of my characters).
I haven't ventured into farportals since I lost a character there (used to enter them searching for blank voratun amulets).

Eden's Guile is 150+% global speed for 9 turns than can be used multiple times. Nerfing that would change some, but like grmbl says, there's a ton of essential items.

There needs to be another method of acquiring them if farming is frowned upon.

grmblfzzz wrote:Are you serious, Grey? Because I could easily make an exhaustive list of the artefacts.

Then please do! Honestly, these things won't improve unless people who play the game most speak up about them.

Class balance is less relevant though. As long as each class is playable, interesting and fun there's no problem. The serious issue is when pre-designed items are deemed "necessary" in a random item drop roguelike. There's no way we want a Nethack-style "ascension kit" in the game.

Grey wrote:

grmblfzzz wrote:Are you serious, Grey? Because I could easily make an exhaustive list of the artefacts.

Then please do! Honestly, these things won't improve unless people who play the game most speak up about them.

Class balance is less relevant though. As long as each class is playable, interesting and fun there's no problem. The serious issue is when pre-designed items are deemed "necessary" in a random item drop roguelike. There's no way we want a Nethack-style "ascension kit" in the game.

The items you normally need to win nightmare (and I assume insane) are exactly what are listed above: wanderer's rest, etc. I think it is unfortunate that such a great game relies heavily on a few critically important "random" artifacts - the randomness being purely if they drop or not.

What is it about wanderer's rest that makes it so powerful? It is the use in vaults for one, where so many other movement methods fizzle (and waste a turn, a feature change that has been requested many times but not implemented afaik), the accuracy of the movement, and quick cd. It also shows that tome is designed in such a way that being surrounded is the single most dangerous situation, alongside certain boss combos. Teleport and phase door sometimes don't solve this problem, so wanderer's rest is vital.

The other class of "essential" items, particularly if you study stiition's approach, is anything that summons: spiders, vampires, rats, tentacles, etc. I too find these immensely helpful because they too reduce chance of being surrounded and draw enemy fire. And talents like corona, mindstorm, etc that radiate out damage towards enemies are also very important. And likewise, enemies that summon tend to cause the biggest problem.

I'm not sure what this means for the future development of tome.

Anyway, these are some of my thoughts on this matter.

grmblfzzz wrote:Are you serious, Grey? Because I could easily make an exhaustive list of the artefacts. In general, they fall into three catagories:

1) Overpowered/potentially necessary (ten items or so, probably)

2) Situationally handy (as in useful for certain classes, at certain points in the game) (maybe a third of the rest of the arts)

3) Garbage. Worse than any non-horrible non-artefact of the slot, doesn't fill any unique role. (all of the rest of the arts)

Could also do the same with classes/races/prodigies, for that matter. I've always assumed that tweaking items/classes was on a fairly low priority, excepting the flagrant extremes of class balance (way too good or unusable). Because there are some clear but significant issue's relating to the fundamental quality of items/classes/prodigies that seem to only change glacially if at all.

Speaking as the guy who makes a lot of artifacts, I'd like to ideally get rid of the third set entirely. Any particularly bad stand outs, in your opinion? Or indeed, the offered exhaustive list.

Infinite talent scaling will be introduced next version(multiple patches have already gone through for this).

Infinite talent scaling produces consistently diminishing returns past normally-PC-obtainable levels for all the talents in the game.

Please wait until the results on Nightmare/Insane-difficulties that have huge amounts of 10+ talent level enemies floating around-come through before making overcompensation for the grinding nerfs; The net effect may be an actually easier Nightmare/Insane after these changes, despite the grinding alterations, not a harder one.

Eden's Guile is 150+% global speed for 9 turns than can be used multiple times. Nerfing that would change some, but like grmbl says, there's a ton of essential items.

Doesn't it only hit that around 150 Cunning?

1) Overpowered/potentially necessary (ten items or so, probably)

2) Situationally handy (as in useful for certain classes, at certain points in the game) (maybe a third of the rest of the arts)

3) Garbage. Worse than any non-horrible non-artefact of the slot, doesn't fill any unique role. (all of the rest of the arts)

I'm working on 3, gradually, as well. I can list some garbage artifacts but I'm currently stuck on how to fix them. (Earlygame artiwhips are a problem. An issue of the system is that they basically mean -30 physical power to good players, and it's very hard to figure out where to set their power for this. Just off the top of my head.) If you can list some more, that'd be nice.

I think that having some occasional standouts(As in 1) really makes item generation more interesting to people, and I strongly object to having every item raised to that level, as well as to every item in that tier being nerfed to average.

However, in a more general sense... we are talking the highest difficulty possible here. Should it be consistently beatable? If so, why do you feel that way? Most Roguelikes don't feel that way about their highest challenges.

It's unbelievably hard-possibly impossible-to make a ridiculously high challenge that can always be cleared. Maybe it should only be possible to clear Insane with the best classes or with the best gear-is there something wrong with that idea, as a challenge level? And even then, there is the item vault as a secondary way around this.

As a final note; I cleared Nightmare with a Ghoul Archmage-not one of the best combinations ever, but pretty good on class.

I did not get a single item of these so called "necessary" items, that anyone has listed. Not. One.

Think carefully about this, and remember Insane's description; That it is literally for the mentally ill. It is not meant to be fair. At all. It says so. Think carefully here, when talking about nerfing it downwards.

Edit:

Could also do the same with classes/races/prodigies, for that matter. I've always assumed that tweaking items/classes was on a fairly low priority, excepting the flagrant extremes of class balance (way too good or unusable). Because there are some clear but significant issue's relating to the fundamental quality of items/classes/prodigies that seem to only change glacially if at all.

Rebalancing classes and races to perfect balance isn't really desireable, necessarily; It allows for a sliding level of difficulty.

Prodigies shouldn't be all equally good for all classes(mostly because that's boring), but they should all be good for someone. I've been trying to find reasonable ways to make some of the losers better, but it's not easy.

I hope insane doesn't get easier. I'm only confused about the farming nerfs since they only effect a select group (who orc baby farms?), and I can already think of several similar ways of farming. It'll be an endless chase of trying to stop my farming, but I tell you it won't work!

There's only 2 artifacts I consider absolutely necessary to beat the game: Eden's Guile and Burning Star. I'm helping make a list of the really powerful and really bad ones with some others.

Since the item vault, the only character on insane that I've lost is an alchemist to tannen, an event that I avoid normally but got careless. So probably 1 out of 5 characters died after daikara levels.
This is why I think it needs to be made more difficult, I'm starting to get sloppy in my play and am not getting punished (my cursed lost his blood of life to subject Z before dreadfell, something I would normally do after returning from the east).
Maybe this is a product of learning how to play the difficult zones (tunneling/rushing up high peak), but I think we'll complete Darkgod's challenge in the next month or so.

Hmm, some excellent replies. I'll take them in about the order they were submitted:

@Grey: After getting your response/talking about this with DG a little last night, I made a list of all the artifacts, and am (In conjunction with Stition and Shibari), doing a rough categorization and we'll be adding some comments/suggestions to the ones that stand out as needing some love. Just so we don't throw a huge list of stuff at you while saying "garbage, fix!", but can offer some more constructive possibilities to potentially lower the workload.

I'll be making a new thread with that later today. Warning: It'll be something of a beast.

@Jenx: Yeah, what you said about sums up the situation. Certain items are key, often not in that they're imbalanced per se, but in that they offer some critical utilities to classes that lack them. What it boils down to is that on harder difficulties, there's a few key things that the stronger classes have but the weaker ones need items for.

In no particular order, those are: Meatshields (pets of any sort, to reduce the constant damage/agro on you), Escape mechanisms (Ideally controlled, and not subject to crazy restrictions, as you pointed out with wanderers rest), General Utility (Digging/Earthrunes/projections/track/arcane eye/earths eye/etc) which some classes have in spades, and some not at all, and probably most importantly 1-shot protection (boneshield/unstoppable/damage smearing/Second life that's not all broken/etc). If your class is missing some of these, need them through items. Ideally, make each class have some rough equivalent for most categories.

@PureQuestion: As I said to grey, going to post a list of all the arts that were on the wiki, and if I spot any that weren't I'll add those as well. I could tell you right now that there are a few categories and types that stand out. Lots of the weapons, especially low level/legacy ones, and the heavy body armors. Would be glad to have any contributions once I post the list, will try to get it into a format where people can contribute easily.

@SageAcrin: Lots of stuff here. First off, would like to clarify a certain point you made.

...The net effect may be an actually easier Nightmare/Insane after these changes, despite the grinding alterations, not a harder one.

I know... which makes me sad. As you may know, a few of the Insane players have been lobying DG for a harder difficulty than insane! We don't want things to be easier, and power creep will just inevitably cause the game to become so as time goes on. The thrust of this discussion is that I'm opposed to increased randomness in the game!

But yes, there are definitely better methods than farming, and since the most desirable one relates to balance, I'll be doing what I can to offer my perspective on the various aspects.

Onward: Eden's guile, yes, 150% is kind of crazy, and would require some massive cunning. However, if you'll note that all of my and Stitions winners have cunning as our secondary stat, so that it's naturally maxed out... it's only a few items that make it go to like 120%, my oozemancer hit that only swapping jewelry. Cunning, by the way, is so necessary because of another balance issue that is rather subtle. Some form of track is 100% required on Insane. You simply cannot, I mean utterly cannot, just auto-explore on Insane. You need to be able to see/plan encounters, which requires track or some Arcane eye. Most classes naturally lack it, so that means through items. Either way, it means stacking cunning.

About general design goals/philosophy for Insane/any highest difficulty. You ask, should it be consistently beatable? Yes, with perfect play! I don't mean pretty good play, I mean perfect play. If you can't win while playing perfectly, there's just too much randomness going on. It shouldn't be a crap-shoot. As I've said, I'm not opposed to difficulty in any form, just randomness. Also, never played on nightmare, but I hear it's a fairly big jump to Insane. You should give it a try, I recommend it!

Which brings us back to general balancing. Since I've already gone all crazy and am working through the items, over the next couple days I figure I might as well go all the way and make seperate posts about prodigies/infusions/races/classes, and point out what I think the biggest offenders are, and get some discussion going relating to these issues.

In no particular order, those are: Meatshields (pets of any sort, to reduce the constant damage/agro on you), Escape mechanisms (Ideally controlled, and not subject to crazy restrictions, as you pointed out with wanderers rest), General Utility (Digging/Earthrunes/projections/track/arcane eye/earths eye/etc) which some classes have in spades, and some not at all, and probably most importantly 1-shot protection (boneshield/unstoppable/damage smearing/Second life that's not all broken/etc). If your class is missing some of these, need them through items. Ideally, make each class have some rough equivalent for most categories.

If pets are required, then next version is going to get very interesting, due to pets redirecting an enemy's targetting to the player after death. On Nightmare, all the item based pets explode violently to a hit or two. I can't imagine Insane's any different.

Regardless, this strikes me as a reason to remove pet based items(currently about three items) and instead retool the game to no longer require them, rather than a reason to spam pets constantly. Pets are both not a common advantage, and cannot be made one thematically, not with any sanity involved in the matter, and many players dislike pet oriented play; If it's being required for the higher difficulties, that seems like a problem.

Escape mechanisms are available in common, infusion/rune slot, forms. General utility has a pretty solid list, I think, of options, including some quite common items like detection wands. Oneshot protection has inscription forms in Heroism and Shielding, though I suppose a Skeleton might have some problems if they're not a specific heavy shields type. Regardless...

About general design goals/philosophy for Insane/any highest difficulty. You ask, should it be consistently beatable? Yes, with perfect play! I don't mean pretty good play, I mean perfect play. If you can't win while playing perfectly, there's just too much randomness going on.

"too much randomness", eh? "perfect play"?

Define those for me and I'll define the problem here.

Or, perhaps by asking for the definition, I've already defined it.

I've heard the trite "perfect play" statement enough to be made nauseous. Perfect play often involves playing the odds, learning to guess when things will happen and when they won't, and sometimes you just can have a run of bad luck. This is true of any game with any shred of randomization.

(Check out things like speedruns, and look at the difference between a fairly optimized human's play and a tool-assist that rigs the odds. The former might be "perfect play", but they cannot consistently account for the odds every single time, and you can't have the odds always go right, no matter how much you've broke the odds calculations down, even to the metal of knowing exactly every calculation's odds. Unless the game has a screwed up RNG, anyways.)

And ToME becoming totally non-randomized would ruin any replay value and any interest it has for me. If I want that, there's literally hundreds of other RPGs I can, and have, played. I didn't play them to completion a few dozen times. I can't imagine I'm the only one that feels this way.

It's a nice wish, and a good thing to try to strive for... But it's not an achievable, real design for any game that randomizes at all. Every Roguelike I've ever played has fallen far from this mark, no matter how hard they try; You just can get bad luck compounded by bad guesses, even though the guess was well constructed and took into account all the possibilities.

There's no way to construct a randomized system so that it's always winnable, and have it have any challenge at all. Let alone extremely high challenge.

ToME compounds this by giving enemies access to player skillsets at random; Any five enemies can be rares at a time you need them not to be, and careful scouting won't save you have to teleport from a Pride entrance into said pack of five, not on higher difficulties. This is an element of ToME I enjoy-the theoretically even playing field, mostly held back by enemies not being as smart as you. Removing it would also not be worth it to me.

I know... which makes me sad. As you may know, a few of the Insane players have been lobying DG for a harder difficulty than insane! We don't want things to be easier, and power creep will just inevitably cause the game to become so as time goes on. The thrust of this discussion is that I'm opposed to increased randomness in the game!

As randomization decreases, formulaic play becomes optimal-knowing exactly what you want to get for items, exactly when you do dungeons, exactly how to handle everything.

I don't agree with that being a good thing. I think the way most Roguelikes handle it(incredibly randomized skillsets/items, or worse, incredibly random items with a few fixed good items that are obvious to build around) is terrible, too, mind you; RNG should be minimized, and flexibility should be encouraged, not luck.

But turning a Roguelike into a series of "Do X to win game" is terrible.

Also, never played on nightmare, but I hear it's a fairly big jump to Insane. You should give it a try, I recommend it!

If it encourages this kind of play mentality? Expecting to win every time if you've got the math down? Grinding for hours on end being better than not having the grinding option? Expecting lower difficulties to get skewed and heavily rebalanced for me to have a better shot?

No thanks. I'd rather see it removed from the game than play it, if that's so.

Edit: Looking back at it, this is a bit harsh and bitter, my apologies.

"perfect play" is a button for me; I think it's a poor argument, because it assumes someone that you are designing for someone that doesn't exist, and requires a designer that is this theoretical perfect player to create(as, how do you know the perfect options without being the perfect player?). It never ceases to annoy me, as such.

SageAcrin! I suspect this is one of those cases of mostly fundamental agreement, but getting tied up in semantics and minor misunderstandings. I'll try to go into depth on the seeming sticking points, here.

1) My quick and dirty list of requirements. Once again, the term requirements is kind of loose. These are general tools and utilities that open up strategies, they aren't necessarily all required, but some classes have almost all of them, some have none. This is really a overall class balance issue, as some of the key weaknesses aren't really exposed until they're pressed to the breaking point on Insane.

For pets on insane, it's certainly never a matter of your pet killing stuff, and they explode in a hit or two as you say. It's a method of mitigating your damage for a turn or two, so you don't eat all of those cryomancer freezes or skele master archer shots. Also not a hard requirement, as classes can make do with EITHER good 1-shot protection, or something along these lines. Can't just run up and tank things with your face, is the gyst of it.

Escape mechanisms: Those available through runes and infusions are critical, but definitely not enough. Look at the infusions/runes of all the insane winners, you'll note 2 movement and a heroism on almost all of them. Once again, it's a matter of some classes having up to 3 or 4 of those built in (in addition to runes/infusions), while others have none. Those others are somewhat item dependant.

Oneshot protection: Yes, shielding/heroism are good and handy and should be on everyone. Once again, simply not enough. The good classes have a means of not being killed instantly (from unstoppable/boneshield/smearing/Aegis tree/etc), the others have to make do with items. Note, no amount of just heroism and shielding runes (without aegis tree) will save you from 5k freeze crits. Thus item dependance for classes that don't have de-aggro/good 1-shot protection.

2) Perfect play. Hmm, seem to have struck a nerve/been taken as being somewhat more extreme than I intended here. First off, of course some amount of randomness makes things interesting. In fact, I play a lot of poker and thoroughly embrace it into my life. Things like comparing defense to accuracy, or in a controlled teleport counting the grids that would screw you vs not and calculating the odds, are perfectly sweet things to have.

Good randomness has the ability to calculate some sort of risk/reward and make informed/optimized/intelligent decisions based on it. Bad randomness is completely uncontrollable/incalculable/unavoidable potential for getting screwed. Once again, this can be addressed by general balance, but there's quite a ways to go (which, I should point out, is perfectly understandable as ToME is so complex).

So, by perfect play in this context, I mean being able to make somewhat optimal decisions every turn (roguelikes being turn-based are more or less designed for this). Should the game be beatable, some large percentage of the time, if someone tries to optimize every single move and maximizes their equipment/skills/etc? I don't think any reasonable argument for a "No" here would be possible. Sure, sometimes it wouldn't work, but it should never be to the point of there being no chance of winning even with perfect play (NOT counting situations where you've already screwed yourself with clearly un-perfect play). That, once again, is a balance issue. Basically, some classes are so far behind the good ones that they are almost totally item dependent.

And finally, yes, I still encourage people to give Insane a try if they havn't, particularly if they're already comfortable with nightmare. I, and some significant amount of other players, simply enjoy trying to be optimal and maximizing things, trying to think of clever stratagems to get out of or past situations where just mashing into people won't work.

Expecting to win if you're super comfortable with all of the game/enemies/abilities/classes? Sure. We do have a remarkable success rate, which I once again point out I would rather NOT have as it's a little too easy.

Grinding for hours being better than not having a grinding option? First, options are almost always good. Second, Of course I would prefer some option that didn't require grinding (I assure you I hate grinding more than almost 100% of people). Thus, my attempts to help address various balance concerns (which is the entire point of this).

Expecting lower difficulties to be skewed and rebalanced to have a better shot? Once again, no, I never said anything like this and it is against all that I propose. Would I like more balanced game in general? Yes. The clear dichotomy of balance on Insane definitely carries over to lower difficulties, it's just masked. There would be tons of ways to fix it without skewing anything at all. Further, I have no problem with widely varied class difficulty in general. Difference is good. However, I suspect that DG and other devs don't want it quite as varied as it currently is!

Can't just run up and tank things with your face, is the gyst of it.

I'd like to point out that this is probably a stronger statement on what you're seeing for class imbalance than anything else, as likely as not.

I've definitely seen higher difficulty players trash the resistance/defense/armor tanks. This is because accuracy and damage keep going up, but your own defense and armor can't really be boosted easily to compensate. Resistances are totally torpedoed by L8(11 on Insane) Skirmisher being everywhere.

These classes work find on Normal, though. It can be read as them "not surviving stress testing", but the fact is, any subtractive defense system will fall apart against values that are far higher than they were tuned to. Skirmisher probably could use a 25% cap, but that does notably make the game easier for everyone, which you don't want.

Oneshot protection: Yes, shielding/heroism are good and handy and should be on everyone. Once again, simply not enough. The good classes have a means of not being killed instantly (from unstoppable/boneshield/smearing/Aegis tree/etc), the others have to make do with items. Note, no amount of just heroism and shielding runes (without aegis tree) will save you from 5k freeze crits. Thus item dependance for classes that don't have de-aggro/good 1-shot protection.

It's interesting that you highlight freeze crits and yet are saddened by the potential nerfing of Insane, as this problem will basically be fixed next version, by eventually capping out Freeze's potential damage. (Not at a low value, either! It's around L20. You certainly can still eat massive Freeze crits with that value...just not 5000.)

The rebalances so far were made by Hachem-if you've looked at Infinite 500, you may have seen his work. I don't think he's interested in massive nerfs at all, just from keeping certain skills from going out of control at higher levels-like Freeze.

If these are in fact creating the required equipment...then Hachem's rebalances are in fact fixing the lack of grinding problem you describe. This is actually kinda an interesting thought, and certainly means that this might be better explored when 1.05 comes out.

So, by perfect play in this context, I mean being able to make somewhat optimal decisions every turn (roguelikes being turn-based are more or less designed for this). Should the game be beatable, some large percentage of the time, if someone tries to optimize every single move and maximizes their equipment/skills/etc? I don't think any reasonable argument for a "No" here would be possible.

This is still assuming that someone can make perfect value judgements. I'm still very questioning of this concept, and that anyone's unbiased enough to have a good idea of what these plays are. I don't think I am.

It's sorta hard to design for an optimal play if deciding what one is can only be done for certain by hundreds of sample tests, for what option works the best in a given situation out of the ones you're going to have...tests that no one's going to make.

That, once again, is a balance issue. Basically, some classes are so far behind the good ones that they are almost totally item dependent.

Again, see above about higher difficulties minimizing some classes advantages.

Another example of this is Rogue; On Normal, enemy stats are not good enough to constantly pierce Stealth. Hiking Stealth-the focus of the class-enough to make it reliable on Insane would make it ridiculously powerful on Normal.

And finally, yes, I still encourage people to give Insane a try if they havn't, particularly if they're already comfortable with nightmare. I, and some significant amount of other players, simply enjoy trying to be optimal and maximizing things, trying to think of clever stratagems to get out of or past situations where just mashing into people won't work.

As a note; You and Stition aren't a significant amount of players.

There are exactly two other players who have gotten one Insane clear who even could have been legitimate, in the vaults. (One of those went on to play a 25 deaths Exploration run on Normal shortly after their Insane run, so I kinda doubt that one. I think the other one's a Mindslayer from when Conduit could nail enemies from outside of Mindlash range...)

Nightmare's enough for me, thanks.

The clear dichotomy of balance on Insane definitely carries over to lower difficulties, it's just masked.

You've never actually cleared Normal, correct?

The gap on Nightmare is spectacularly larger than the gap on Normal. It's not masked; The game is actually horrifyingly imbalanced on enemy design, strongly encouraging ways to prevent death to single huge damage spikes that enemies can only periodically make, removing buffs that were designed to not be required to be removed, dealing with 20-30 turn status that were never designed to last that long, etc.

While skills that normally function strongly throughout the game are terribly minimized, due to a focus on subtraction defenses, resistances that barely or often completely fail to function, enemy statistics punching straight through them, etc.

There literally are problems that never exist on Normal. And that's in Nightmare. In Insane, it can only be magnified. It's probably hilarious for an Insane player to hear that a Bulwark is one of the better classes on Normal, but it's true. On Nightmare, a Bulwark is a roadbump, and redesigning that to not be true-without making them incredibly overpowered on Normal-is very nontrivial.

Difference is good. However, I suspect that DG and other devs don't want it quite as varied as it currently is!

Considering that Edge likes Temporal Wardens/Paradox Mages and Tiger_Eye likes Rogues(and doesn't abuse Gravitic Trap going through walls, which I believe was Stition's strategy)? I think you'll be quite surprised at the range of characters the devs are happy with.

I personally feel the balance is perfectly fine on Normal, and if it shatters like an eggshell on higher difficulties...well, it's not the first game to have that happen.

Don't get me wrong, I'm not discouraging ways to make the lower end characters specifically better on higher difficulties, but not at the expense of them gaining nonsensical and athematic abilities, or the game turning into a full set of Archmage-power-level classes.

As a final note; To me, it sounds like Insane mostly just kills those that diverge from extremely binary playstyles.

Not using pets or something similar to absorb blows? Trying to judge how much durability you have instead? Too bad, won't work, die-HP values, normal durability values like armor, defense, armor hardiness, resistances, all of it useless.

Trying to use various abilities that work on Normal? They don't work here, except for those that do-hope you know the mechanics well enough to tell which is which, much of which is not immediately apparent without code diving. If you don't do that? Die.

Change things so that you can't get these options? Well, everyone dies.

This is sounding more and more like a flawed and unworkable difficulty and that its concept needs to be overhauled, rather than merely nerfed or buffed. Adding damage and durability and bigger numbers to Nightmare seems to just enhance its flaws, rather than produce anything notable, and it sounds like it's not "a big jump" for strategy, but instead do the things that work, even moreso, with any divergence from that optimal path being death.

I honestly doubt Insane was ever intended to be focused on and seriously played. And I am starting to doubt that it can be balanced at all without changing the fundamental concept of "more levels/talents".

Hmm, this has turned into something of an Insane Balance discussion, which wasn't really intended, but I'm always game for it!

First, a general statement/caveat that sums up the main thread of this discussion: No, I don't think the current situation is ideal with item dependence. A lot of this can be fixed with some fairly minor changes to artifacts, which I'm currently collaborating with Stition/Waladi/Shibari to look at (if anyone else wants to poke in before we finish and make a post, I can give you access). This post was primarily to point out the item-dependant nature and what these farming changes would mean, and just to point out that it's a fix that while understandable (no one likes farming), raises some other issues.

Some of it is more basic class/mechanic issues which are harder to deal with. You're right in that there would be no easy, sweeping changes that would fix it, nor is it necessarily good or desirable to do so due to the critical nature of variety. I have a lot of ideas on this topic, and I think there are a bunch of ways that you could do most of it without radically effecting the balance on normal difficulty, but I'll go into them in another post.

Overall, I'd like to point out that the balance is currently better even on insane than I might have seemed to be indicating. Almost half the classes now have clears, and they still play in a wide variety of ways, with quite a bit of interesting variations. It's certainly not a binary play-style, you really have to maximize the advantages inherent to each class. Things like rogues losing focus on melee/sneaking and tending to trap are fine in my opinion (note, it was mostly traps in general, gravity trap is too scary to use on things that can insta-gib you!). It's just that the basic lack of mechanisms on the lines I was discussion forms a clearer dichotomy of power than on normal/nightmare.

Havn't looked at infinite 500, will do so at some point, also don't keep up 100% with all the SVN changes. I'll definitely look into it. But things like the freeze change are probably for the good, although I'll note that there was already an (item based!) way to deal with it on like every class, that being cold wards Oh, and only the last boss can crit for ~5k, regular crymancers are in the much more reasonable region of 2k. I think things like finding/carrying around warding is perfectly viable and fun, but it goes to the whole item dependency issue (also the AM vs normal issue, which is also not the focus here).

Lastly, as to the whole perfect play issue, I think we're getting caught up in semantics again. Bottom line: Playing very well/strategically/intelligently throughout the game should net you some good win percent. It should require exponentially better play to require the same win percent as you go up difficulties, but in general the amount of randomness should be poker-style instead of getting-hit-by-lightning style