Sil 1.0.2

[half]

It's mentioned a lot that parry would be broken if it doubled the evasion bonus of your offhand weapon as well... but would it actually be? Keep in mind that you're giving up a shield/two handed weapon, spending exp on parry which will thereby making future evasion skills more expensive to get (and evasion has soooo many good skills - sprinting, flanking, controlled retreat, etc etc)...

The problem we see is that this only works in conjunction with Two Weapon Fighting, and we see that ability as already being a bit too strong, and leaning a bit too much towards choosing weapons for their defensive bonuses or abilities instead of their offence. I agree that Parry is on the weak side, typically offering only a single skill point (which would be good if you had very few evasion abilities, but is often a poor deal). Its main benefit is leading to Riposte. If there were a nice clean way to make Parry more interesting we'd certainly consider it.

[Patashu]

+N evasion only when doing a normal melee attack? Like an anti-dodging

[debo]

The problem we see is that this only works in conjunction with Two Weapon Fighting, and we see that ability as already being a bit too strong, and leaning a bit too much towards choosing weapons for their defensive bonuses or abilities instead of their offence. I agree that Parry is on the weak side, typically offering only a single skill point (which would be good if you had very few evasion abilities, but is often a poor deal). Its main benefit is leading to Riposte. If there were a nice clean way to make Parry more interesting we'd certainly consider it.

Stupid question, but what about giving a small % chance that an attack that has already been determined to do damage will whiff entirely?

I personally think parry is just fine the way it is now... it's situationally great when you find an early Defender weapon. It's really only helpful in the early game, but it leads to Riposte which is awesome in the lategame.

[jujuben]

Here's a thought to keep parry relevant longer without using the second weapon or making it too powerful too early. Either in place of the current version, or in addition to it, give a defense bonus equal to some fraction of melee score, adjusted by weapon weight. This works for me thematically, as parrying is basically hitting a weapon, rather than a monster. Something like [defense bonus] = floor([melee]/[n + weapon weight]) where n is whatever it needs to balance the skill's power.

[Scatha]

Thanks all for the suggestions. Some comments:

+N evasion only when doing a normal melee attack? Like an anti-dodging

Something like this could work (and it's certainly elegant), but is it interesting? Dodging rewards you for doing something slightly unusual (moving in combat), but attacking is something so many characters do when they'll be attacked back that it seems a bit too much like a static bonus. To be fair this is also a problem with the current ability, which just slightly rewards certain weapons!

There's also an issue of rewarding players for behaviour that's often good anyway. We wondered about Parry giving a bonus when you were fighting just one opponent, but since it's already really good to only fight one-on-one thought that could be a bit silly. Looking at it now, I wonder: perhaps it's okay to make these fights easier, as it's very hard to keep to only this type.

Stupid question, but what about giving a small % chance that an attack that has already been determined to do damage will whiff entirely?

This doesn't fit elegantly into the way we do abilities in the game, so we'd rather do it like this unless it really gave a cool effect we couldn't get otherwise. I also don't think it would play that interestingly -- it's a bit like another point of Con that only works against melee attacks.

Here's a thought to keep parry relevant longer without using the second weapon or making it too powerful too early. Either in place of the current version, or in addition to it, give a defense bonus equal to some fraction of melee score, adjusted by weapon weight. This works for me thematically, as parrying is basically hitting a weapon, rather than a monster. Something like [defense bonus] = floor([melee]/[n + weapon weight]) where n is whatever it needs to balance the skill's power.

Hmm. We definitely thought about some tie to melee (though I don't think to weapon weight -- I like it), and I can't remember exactly what decided us against. I do have a couple of concerns:

Since Parry is available early in the tree, it should do something nontrivial when you get it, which might be hard to balance with this version while still being fair later. It could potentially be moved later in the tree, but there is something very natural about Parry --> Riposte.

It might also suffer from the being too "samey" across different characters: more or less everyone puts points in melee. The spread on weapon weight does a good job of mitigating this, though.

One could also go for something like Versatility, letting melee partially substitute for Evasion. I think this would have interesting gameplay, but also change the flavour as it would not be for Evasion specialists, and perhaps shouldn't be in the Evasion tree.

[jujuben]

Since Parry is available early in the tree, it should do something nontrivial when you get it, which might be hard to balance with this version while still being fair later. It could potentially be moved later in the tree, but there is something very natural about Parry --> Riposte.

It might also suffer from the being too "samey" across different characters: more or less everyone puts points in melee. The spread on weapon weight does a good job of mitigating this, though.

One could also go for something like Versatility, letting melee partially substitute for Evasion. I think this would have interesting gameplay, but also change the flavour as it would not be for Evasion specialists, and perhaps shouldn't be in the Evasion tree.

On the first point, that's why I was thinking this might be in addition to, rather than in place of the current version of parry. Obviously, n should be fairly high if you go that route.

The weight spread was an attempt to keep things flavored - I was imagining swashbuckling/Errol Flynn/Zorro type rapier/smallsword fighting. This version doesn't really accurately describe the way a skilled staff or polearm fighter can defend, but I can't think of a version encompassing both that doesn't fail either the complexity test or the sameness test.

I like the third idea quite a bit, especially as a midgame option for protection specialists. Something to keep the cats and vamps off if you don't feel like taking will/crit resistance.

One other option I thought of: keep the current value of parry, and in addition, double the protection value of the hand slot.

[Scatha]

Yes, the flavour of using the weight is very clear, which is nice.

I appreciate the thought of adding it to the current bonus; the issue is that adding on one of these extra bonuses to the current rule just makes the whole thing a little bit too complicated, I think. We try quite hard to have clean rules for all of the abilities; sometimes it's okay to spend extra complexity when there's particularly clear flavour or it just makes the gameplay work well, but I don't think either of those would apply here.

It might work to take the ceiling instead of the floor, but almost everywhere in the game uses floors (Song of Slaying is the only exception I can think of, and that's taking the ceiling of a hidden variable).

[Psi]

Which brings me to the new plans. We are planning to extend this kind of feature for forges (and this is basically the last thing to be added to the current development version before it gets playtested and released). I haven't started on that yet, but the idea is to smooth out the number of forges per game. Again, I don't like the dependence on what has gone before, but I think that it can help avoid certain un-fun situations (for example, I really wanted to play a smith and made a smith character, but didn't find a second forge until 950 ft).

Quick question on this... so what happens if you dive quickly through the first 750' where there happened to be a plethora of forges which you missed. Do you suddenly find it almost impossible to find a forge?

[fizzix]

Quick question on this... so what happens if you dive quickly through the first 750' where there happened to be a plethora of forges which you missed. Do you suddenly find it almost impossible to find a forge?

There is a pretty hackish solution that might work for good gameplay. Let the probability of a forge appearing be a function of level and number of forge uses that have so far been used. The function should have thresholds so that you can't exhaust all your forge uses for the game at 500' or so. In order not to made diving the default best option, you can let an unused forge be equal count a bit to the used forge values. I can't offer a good function, but something like this should ensure that every game has some roughly even number of forge uses and penalizes (slightly) players that skip forges.

[Scatha]

Quick question on this... so what happens if you dive quickly through the first 750' where there happened to be a plethora of forges which you missed. Do you suddenly find it almost impossible to find a forge?

We haven't finalised a mechanic, but that sounds problematic and shouldn't happen. It is quite possible that we will use something with a short memory, so (for instance) it only remembers when you last generated a level with a forge.

I should warn you, however, that we think you find slightly too many forges at the very bottom at the moment, and this will probably be reduced as part of the smoothing.

There is a pretty hackish solution that might work for good gameplay. Let the probability of a forge appearing be a function of level and number of forge uses that have so far been used. The function should have thresholds so that you can't exhaust all your forge uses for the game at 500' or so. In order not to made diving the default best option, you can let an unused forge be equal count a bit to the used forge values. I can't offer a good function, but something like this should ensure that every game has some roughly even number of forge uses and penalizes (slightly) players that skip forges.

I think that you're right that this could lead to interesting gameplay! However, it does so by giving the player more interesting decisions, and because there isn't really any flavour justification for why forges should appear like this, we'd really prefer the forge mechanic to be one that fades into the background and people don't need to know about (and hopefully don't even notice during play unless they keep stats and notice it's smoother than random!).

[debo]

I should warn you, however, that we think you find slightly too many forges at the very bottom at the moment, and this will probably be reduced as part of the smoothing.

I have a feeling that the "you" in this sentence was directly targeted at Psi

[Psi]

I should warn you, however, that we think you find slightly too many forges at the very bottom at the moment, and this will probably be reduced as part of the smoothing.

At this point though I am more or less forge scumming with a _Revelations. Forges are not that common, but play enough levels and you will find them.

[half]

At this point though I am more or less forge scumming with a _Revelations. Forges are not that common, but play enough levels and you will find them.

Yes, we will think about this. We've tried hard to have a balanced smithing system (which is often broken in roguelikes with it) and I think we've done well, though it can be very much broken by a couple of things at the moment:

1) Staves of Revelations
2) Staying in Angband after getting a Silmaril

Number (2) is because you stop being forced down. It is a bit hard to know what to say about this. Perhaps we should stop it somehow, but it is not clear that it is the forging system's fault.

Re (1), I really like the clean mechanic of Staves of Revelations, and wouldn't want to simply remove them. Using them to find forges is also quite fun and natural, but it can lead to silly amounts of forges and it is hard to know what to do about that.

Both (1) and (2) allow much more of forging smithing gear and climbing up in terms of power. I like the balanced nature of this for people who find a normal amount of forges: there is an interesting delayed gratification aspect which keeps it interesting, but if the number of forges goes up a lot, it gets too strong.

[Derakon]

Yes, we will think about this. We've tried hard to have a balanced smithing system (which is often broken in roguelikes with it) and I think we've done well, though it can be very much broken by a couple of things at the moment:

1) Staves of Revelations
2) Staying in Angband after getting a Silmaril

Number (2) is because you stop being forced down. It is a bit hard to know what to say about this. Perhaps we should stop it somehow, but it is not clear that it is the forging system's fault.

It would make sense, to me, for monster spawning to get more and more aggressive the more time has passed since you got the Silmaril. Morgoth should be sending everything he has after you. Thus, sticking around to scum for stuff wouldn't be a great idea...unless you're powerful enough to survive even among hordes of monsters.

[Patashu]

I'm not sure 2) is necessarily a bad thing. You already 'beat the game' if you get a silmaril and can scum the dungeon, why not let the player have some fun?