Spell / Phantasm re-work

[SageAcrin]

You can max three stats by game end, and that still leaves enough points to buff up Con a bit for everyone's favorite Resist All talent.

In what way is it difficult for those three stats to be Dex, Mag and Cun?

Note that for the Shadowblade, Cunning also boosts Spellpower.

There's a lot of game before you are able to cap stats, you know.

Three high priority stats is actually pretty hard to juggle-and there's no real need to if you're not specifically trying to leverage Illuminate. You can easily just go for enough Magic for requirements when and as needed-not really a lot, in practice-until later in the game when you have more stat points...unless you're specifically abusing this.

That's a short term downside. Long term, yes, Illuminate's quite good. Still rather hard to get status to reliably land without a staff in hand, but...

True, but this does NOT break Stealth.

True. But it does mean you're at risk in hit-and-run tactics. It's not totally free.

We're only discussing how Shadowstrike works in conjunction with spells.

C'mon.

Next time, I'll know that it offends you when I admit my failings(Namely, that I know how Stealth mechanics work, but am in fact uncertain on an element of the topic of discussion), and not do so.

My apologies.

I don't know if this is reproduced on spells, however-it's a rather hard thing to code dive for. It fails at random and it fails just fine with everything else, however.

It actually has the same behavior for spells and mental attacks, UNLESS the mental attack goes through the melee code.

This is why my psychic Stealth guy uses Gesture of Pain, and why his main ranged attack power cannot crit.

Those are NOT broken in the same way as spells and other mental powers.

The code is spread out, sure, but it's not actually difficult to understand.

Didn't you say that Stealth never breaks? Or are you saying that mental attacks(that are not Gesture of Pain) also fail to break Stealth? Regardless, that doesn't really change my original point-that it sounds more like Stealth needs a chance to fail for skills that it does not fail for.

Also, yes, it's spread out... that's why I don't know off the top of my head. It isn't something I've grepped up, because it basically comes up for...this, and only this.

I really like ToME's code, but not enough to read Actor.lua for fun. (Or maybe it's in Combat?)

...Well, okay, not enough to remember all of the info from the times I've read Actor.lua for fun...

[Doctornull]

You can max three stats by game end, and that still leaves enough points to buff up Con a bit for everyone's favorite Resist All talent.

In what way is it difficult for those three stats to be Dex, Mag and Cun?

Note that for the Shadowblade, Cunning also boosts Spellpower.

There's a lot of game before you are able to cap stats, you know.

Three high priority stats is actually pretty hard to juggle-and there's no real need to if you're not specifically trying to leverage Illuminate. You can easily just go for enough Magic for requirements when and as needed-not really a lot, in practice-until later in the game when you have more stat points...unless you're specifically abusing this.

That's a short term downside. Long term, yes, Illuminate's quite good. Still rather hard to get status to reliably land without a staff in hand, but...

The big issue here is that Illuminate ideally does its thing from across the map. You probably have the spare turn to swap weapons back if you fail to vaporize a Rare or Boss (which you will), before jumping into melee to Flurry or whatnot. I usually had lots of time to spare in my tests, but yeah, I could have gotten Rushed or Blindsided.

Regarding stats -- it's not too hard to get 3 stats to decent levels, especially if you disregard Con for the short term, I think it's quite possible to have a Magic major Dex / Cun minor build, moving into a more balanced allocation over time. There are other classes which rely on 3 (or more) stats, so I don't think the stat spread is much of a limiting factor on the power of the tactic.

True, but this does NOT break Stealth.

True. But it does mean you're at risk in hit-and-run tactics. It's not totally free.

Right, but it's hit and run from across the map. I guess it's technically not free, but it's VERY cheap compared to any other options available, since your other options would have you engaging in melee with your target.

My apologies.

Accepted.

Please do trust me a bit, that I know what I'm talking about. It'll save us both a lot of time going forward

Didn't you say that Stealth never breaks?

Almost: I said that being briefly spotted by an enemy does not break Stealth.

Or are you saying that mental attacks(that are not Gesture of Pain) also fail to break Stealth?

They behave the same way as spells, which is to say they get a guaranteed mega-crit and then there's a check for Unseen Actions.

Regardless, that doesn't really change my original point-that it sounds more like Stealth needs a chance to fail for skills that it does not fail for.

That's possible, but since spells and mental powers work differently from melee attacks -- since there's one crit chance per spell-effect rather than per single-target strike -- it's not clear WHO would get a chance to see through your cunning camouflage.

If someone wants to figure that out, then that may be a viable long-term solution, but I'd like the game to be playable in the short term as well (ideally even playable in the PRESENT ), so that's what I'm looking for in this thread -- short-term solutions to fix this problematic talent interaction.

---

Right now the ideas are:

- Split up the Illuminate effect into "stealth-compatible" and "no-stealth", and allow lighting up a room without doing any damage if you're near an escort or paladin or whatever.

- Reduce the distance over which Illuminate can damage.

- Reduce the total damage Illuminate can do (which might make it less attractive to an Archmage, and for them it's already kind of niche, so that might need some testing).

Anything I'm forgetting?

[Crim, The Red Thunder]

If you take the last one, you might want to consider that's it's currently one of the only ways to reliably get the achievement 'ten at one blow'. Nerfing it would probably need to take that into consideration. (Other candidates are blood spray and blood boil, neither of which does the damage level of illuminate, or can be as strong, as quick. Fireflash takes too long to get running to one-hit everything at one blow reliably.)

[SageAcrin]

That's possible, but since spells and mental powers work differently from melee attacks -- since there's one crit chance per spell-effect rather than per single-target strike -- it's not clear WHO would get a chance to see through your cunning camouflage.

Honestly, the logical, easy solution is making a check for everyone hit, and if any of them fails, fail the Shadowstrike for the entire attack. (Or, since I'm not sure if it's sequential or simultanious-I think spells are sequential though-have it check for each in whatever sequence the blast radius is drawn in, and fail it individually for each. Either way.)

It's not terribly odd, either, just change the message to "(target) calls out a warning before the spell hits!", in the former case. It makes mass blasting through Stealth impractical...which is pretty logical, and seems like it fixes this problem.

Such a change is, honestly, necessary. Just nerfing Illuminate would be a stopgap, and quite a poor one; If a 180 spelldamage at cap spell is way too good under Stealth, then any attempts to make stealth based casters in the future would become extremely unbalanced creatures, requiring terrible mults and needing stealth to do any damage at all, or far too powerful in Stealth.

Considering that "stealth based caster" is one of the most popular addon ideas in general, and that it's not at all unusual for fan created content to get into the game, that's pretty major. May as well head off the real problem, no?

[Hachem_Muche]

A couple of of addition things to consider:

Currently. when a stealthed character casts an attack spell, the only way to break stealth is to fail an Unseen Actions (if applicable) check. The actual attack against the target does not trigger the normal chance for the target to see the attacker (which forces stealth to deactivate) like it does for a melee attack. This should probably be fixed, and I'm looking into what needs to be changed to do this.

Also, Unseen Actions works differently after the scaling changes. For 1.0.5, it takes distance, number of LOS opponents, and their stealth detection into account when making the check for failure. So the chance to stay stealthed when attacking a room full of enemies with AOE will increase with range and decrease with the number of targets. The stealth code for 1.0.5 is here: http://git.net-core.org/darkgod/t-engin ... tealth.lua

[Earwicker]

- Is there a scaling value which retains usefulness for an Archmage but disallows screen-wide vaporization by a Shadowblade? (I can't find one.)

Seems like my 2-lines post got lost into the flames? As I wrote earlier, you could just multiply damage by a factor that depends on the current stealth value. I suggested a simple (100 - current_stealth)%, but it could probably be improved upon.

At least it doesn't change anything to non-stealth classes that use Illuminate, or to the stealth mechanics. Plus, it's sort of thematic: the more you're stealthed, the less you inflict light damage.

[Hachem_Muche]

I think the bigger issue is what effect stealth bonuses have on non-melee attacks in general, including both spells and mind talents. There needs to be a fix that covers both shadow blades using channel staff and rogues sniping with mana-clash for example. Penalizing these attacks due to stealth feels like a kludge, that doesn't make any sense thematically.

I'm particularly concerned about stealthed random bosses/elites. Think about a level 95 stair guard in the High Peak with one stealth class and one ranged class, like ArchMage. It gets all of the stealth bonuses against the player with its ranged attacks and won't break stealth until it fails an Unseen Actions check (a very low chance in most cases).

[SageAcrin]

Currently. when a stealthed character casts an attack spell, the only way to break stealth is to fail an Unseen Actions (if applicable) check. The actual attack against the target does not trigger the normal chance for the target to see the attacker (which forces stealth to deactivate) like it does for a melee attack. This should probably be fixed, and I'm looking into what needs to be changed to do this.

Yeah, this was what I was referring to.

The fact that it's an issue on enemies is something I didn't think of, but it's a very good point as well.

[HousePet]

Perhaps Shadowstrike could have a reduced effect for spell and mind attacks.
Either no longer a guaranteed crit, just a chance boost, or reduced extra crit damage.

[wobbly]

Something I'd like to add to the discussion which is less of a balance issue & more of a "how fun is this class to play issue?" is:
Whenever I play a shadowblade most of the early game seems to be mass blasting the whole screen with illuminate, wait for illuminate to come off cool down, mass blast the whole screen again. Apologies if I have no good suggestions for fixing it at the moment, just thought I'd mention that how it stands now isn't particularly fun, balance questions aside.

[SageAcrin]

Personally, my most successful Shadowblade attempts largely ignored Illuminate and played them as a melee fighter(With Ambuscade and Essence of Speed backing things up, eventually).

It's basically a split build option.

They probably could use a L10 blasting spell tree to make a blasting Shadowblade less boring, though.

[Doctornull]

Also, Unseen Actions works differently after the scaling changes. For 1.0.5, it takes distance, number of LOS opponents, and their stealth detection into account when making the check for failure. So the chance to stay stealthed when attacking a room full of enemies with AOE will increase with range and decrease with the number of targets.

The increase with range means it may be BETTER for this tactic, though I guess that depends on how it scales vs. number of opponents.

- Is there a scaling value which retains usefulness for an Archmage but disallows screen-wide vaporization by a Shadowblade? (I can't find one.)

Seems like my 2-lines post got lost into the flames? As I wrote earlier, you could just multiply damage by a factor that depends on the current stealth value. I suggested a simple (100 - current_stealth)%, but it could probably be improved upon.

At least it doesn't change anything to non-stealth classes that use Illuminate, or to the stealth mechanics. Plus, it's sort of thematic: the more you're stealthed, the less you inflict light damage.

Hmm. At the extreme, that could make Stealth spells worse than regular spells if your Stealth score was very good. Your system would mean the strongest Stealth caster is the guy with a 1% Stealth chance... penalizing players of stealthy casters for being good at Stealth isn't very intuitive to me.

Perhaps Shadowstrike could have a reduced effect for spell and mind attacks.
Either no longer a guaranteed crit, just a chance boost, or reduced extra crit damage.

Even if it were just a guaranteed crit with no extra damage, that would be beefy and good and a bit less broken.

It's still potentially broken on a rare or boss, though in my experience NPCs are kind of dumb about Stealth.

Something I'd like to add to the discussion which is less of a balance issue & more of a "how fun is this class to play issue?" is:
Whenever I play a shadowblade most of the early game seems to be mass blasting the whole screen with illuminate, wait for illuminate to come off cool down, mass blast the whole screen again. Apologies if I have no good suggestions for fixing it at the moment, just thought I'd mention that how it stands now isn't particularly fun, balance questions aside.

I totally agree!

That's exactly the same experience I had, which is what caused me to start this thread in the first place.

The problem isn't just that the tactic is brokenly strong, because it won't solve the end game or anything, it's that the tactic is GOD DAMN BORING.

Finding effective talent uses is fun, and being successful in finding clever talent uses feels good, right up until you break the game and stop having fun. That's where Stealth casting is right now: broken to the point of being no fun.

[HousePet]

They probably could use a L10 blasting spell tree to make a blasting Shadowblade less boring, though.

I threw in Spells/Nightfall for some extra casting/debuff options and it works okay.

[Doctornull]

I threw in Spells/Nightfall for some extra casting/debuff options and it works okay.

This sounds like a good start.

IMHO the Stealth casters need talents to be designed with Stealth in mind, just like their melee trees are designed in a different way from non-Stealth melee trees.

[SageAcrin]

Spell/Nightfall 1.1 as a locked category would be a neat addition to Shadowblade. Certainly no danger; they have tons of high impact locked categories already.

Though it sounds like maybe fixing the lowered check amount (for spells, relative to physicals) would help, first.