Breakdown Silent Stalker for me?

[z7sakari]

I am unclear on how this talent works exactly.  It says "From Surprise" you reap all these excellent rewards, but are these rewards applied to:

1. Your first hit only?  
2. That first entire turn that you surprised them?  
3. Or do these rewards last through the entire battle assuming you had enough stealth to steak up on them?

Additionally, is there a way to see how much stealh is required to sneak up on monsters or do I just have to mess around with abilities and equipment until I consistantly score the surprise?

Thanks so much for your help.

[En1gma]

First turn, user only.

I typically put a few points into stealth, unless you are on higher difficulties you should be ok.. I'm on normal with 5 (9) stealth, I'm not wearing stealth armor, and I haven't lost a single roll leading up to and into the Vermin Lair. Not one roll.

[fallen]

z7sakari - it is on our list to try to give some indication of your ability to surprise monsters after you do or don't, so you have an idea of what you might need. Each monster type has a Spot+Stealth value (their ability to detect you / sneak up on you), and they are averaged together for the group and rolled against your leader's Stealth. So, as an example, spiders are high spot / stealth monsters. Armored Orcin, heavily armored soldiers - these are lower stealth monsters.

With my current Nightmare build, I had 12 Stealth through E1A2 and never failed to get the surprise round. I changed up my gear as I entered E1A2 and now have only 10 Stealth, and am getting some combats where I don't get the ambush.

My tactics with Silent Stalker include:
1) Build Tamilin to be successful. This means Stealth and Crit %. Use both Deadly Intuition and Black Night, and find a ring that is +Crit %.
2) If you get the Stealth round, do not move Tamilin. Send out Vraes to find the enemies. Then send Kjartan to Choking Ash them. Then move Tamilin 1 square so she can fire 3 times.
3) Do not use 3 AP bow for this, too few shots.
4) Try to get Tamilin +Move Point gear later in the game, this will help her get into position more successfully and still get 3 shots.
5) Attack with Silent Stalker, watch them fall.

The best combat I have seen in E1A2 so far is in the dungeons of Red Hill. With the right jump, Choking Ash, and 3 shots - Tamilin dropped 3 guards. Since I'm playing on Nightmare, I'm playing to survive, and so Vraes' damage is pretty low at this point. Combats can grind as only Fyona and Tami are really dropping bombs on people (Kjartan is full curse). So, to have a combat start with -3 Enemies from the first turn ... very happy. Sometimes, you'll get a complete failure and the passage will bend or something and Tamilin won't get a shot off, but sometimes its absolute bombs.

[z7sakari]

Awesome, thanks for all the info and advice.

Silent Stalker lasting for the surprise turn makes the most sense, but I wanted to make sure. I may put 1 point into it to see how it goes, but I see more lasting benefits coming from leveling Deadly Intuition. Your excellent tactic of scouting with Vraes, then firing off 3 volleys with Tamilin is definitely the best method I have found, but it is rare that enemies start clustered enough for me to score some good Ranged Onslaughts. Instead I typically need to herd them before laying down the AOE. I can see 1 point in Silent Stalker being worth it simply for the immediate -3 accuracy penalty, and certainly will not complain about the rest =D

[xdesperado]

z7sakari I cant speak on SS but as a general rule putting one point in a skill "to see how it is" tends to be disappointing and misleading.
Generally you won't see major impact/benefits of a given skill until you've made a bit of investment into it.
Also keeping them more focused will make the hero stronger than spreading your trains about all over with a point here and a point there.

[Cory Trese]

The build where i have the most success with Silent Stalker has 22 Stealth and 16% Critical -- I lead with Tamilin (buffed for Dodge, of course) into every battle. When I win on Stealth I advance Vraes and then usually can kill 2-3 mobs in the firs turn.

[tenbsmith]

One downside for Silent Stalker is that the most challenging fights are often scripted and thus you don't even have the chance to win a stealth roll.

[abysmal]

Something that I think would make stealth more viable is the ability to actually see the enemies I am ambushing. I mean, if I can get the first turn in an attack, my Selen has to wait 1-2 turns for the enemy, or waste SP on Sly Look which is pretty sketchy at best.

[tenbsmith]

My current build has Tamilin--level 8, stealth 4 (10)--lead and mostly win stealth rolls. Then send Vraes ahead to find enemies. This usually leads to 1-2 kills prior to the enemy having a turn. Another advantage is positioning heroes prior to enemy movement.

I think Tamilin's stealth enables this build to finish fights quicker while sustaining less damage. This means fewer potions used and less need to recharge at campsites/towns.

I use stealth but not Silent Stalker, preferring Deadly Intuition and a few points for Black night.

[Cory Trese]

Oct 11, 2014 10:30:10 GMT -5 abysmal said:

Something that I think would make stealth more viable is the ability to actually see the enemies I am ambushing. I mean, if I can get the first turn in an attack, my Selen has to wait 1-2 turns for the enemy, or waste SP on Sly Look which is pretty sketchy at best.

That's an interesting idea. Not sure how we'd implement it in a balanced way, but I will chat with fallen to see if their is anything fun we could add there.

[tenbsmith]

Cory Trese and fallen, it seems logical that upon winning stealth one would at least know the location of a one or more of the closest enemies, even when they are out of the range of vision or around a corner. In cases where enemies are out of vision range/around a corner, you might provide a generic icon in the location or approximate location of the nearest enemies.

I don't think something like that would change the balance of the game very much, but it would make stealth a bit more useful and, to me, seems realistic.

I prefer that to some sort of pointer, because the pointer wouldn't deal well with nearby enemies in more than one direction.

PS. Welcome back to the boards.

[Cory Trese]

Winning Stealth often does give the location of one or more enemies based on the LoS code -- however, Stealth isn't intended to be all about visual contact, it includes other senses as well.

Any change here definitely will have a balance impact -- range and AP costs were all calculated based on the rules as they stand.

We have to be very cautious to not nerf or OP builds as we make adjustments.

Stealth is more realistic as it is, especially given the very low AP cost of Sly Look, however we're going to talk about it and see if the stealth/exploration game element would be more fun for a wider majority of players with changes.

[tenbsmith]

Cory Trese, you guys understand the game best and I'll respect whichever way you go with this, I'm just having fun brainstorming.

For the heroes, I imagine noise being really important for winning stealth. Generally, when you hear something you get a sense of it's approximate location/direction. Noise might bounce and diffuse or misdirect in the caverns of Steel providing either an incorrect directional cue or none at all. Providing just direction with some probability of it being misdirection might be more work than it's worth.

Far Sight and Sly Look start at 3 AP and you have to spend 3 Talent Points to get them down to 1 AP. I don't recall builds posted here investing points in either of these Talents--maybe they do not get used very much.

[Cory Trese]

Our metrics show that a not insignificant number of people level up Far Sight, less use Sly Look.

One possible option is to reduce the AP cost of that Talent during Stealth actions.

[tenbsmith]

I like this idea. Decreasing the AP cost will make those those Talents more useful and attractive.

Since Tamilin is my primary damage dealer from surprise, I prefer to retain her AP for bow shots, but would change the algebra of my decision algorithm.