The Declaration of RIPLIB

[claus]

You should add Pikeman->Halbardier (10-3 pierce to 11-4 blade and 9-4 pierce).

[Elvish_Pillager]

You should add Pikeman->Halbardier (10-3 pierce to 11-4 blade and 9-4 pierce).

Oh yeah, that ridiculous thing. That was changed a while ago and no one paid any attention to the complainers. I keep forgetting that change.

[Tippsey]

Meh about this whole conversation I think everyone is missing the points of the darn examples yes everyone knows 4-3 is low. BUT FOR GOD SAKES it's an example not a somthing you shall see in the game therefore why not the examples agisnt riblip as an example be 6-3 or higher and the side for riblip LOOK PAST THE NUMBERS USED FOR THE BASE EXAMPLE! Jesus this entire thread has been Pillager whining about the 4-3 nothing is going to be proven either way if all that can be done is fighting over how low the numbers are in the base.

[Elvish_Pillager]

What does God have to do with this? It's a bit late for such an extreme response. I said it was weak, but that was only a side note to my main point, to explain why I wasn't using the original values.

[Tippsey]

Great now your hung up over god...... bloody hell

[Chris Byler]

Long thread... however, it has made me understand more fully the RIPLIB principle particularly as it applies to forked advancement trees.

A few things to add to your list of things that are considered to not violate RIPLIB even though they clearly do (along with upkeep and exp-given): first, leadability. Leveling a unit sometimes causes it to lose some or all of the leadership bonus it is receiving, which can make its attacks weaker per-hit or even overall, but this is not considered a violation of RIPLIB even though it *can* lead to "I would be better off - at least locally and in the short term - if that unit of mine hadn't leveled right then" which is what strict RIPLIB is supposed to avoid.

The other obvious violates-RIPLIB-but-doesn't properties are loss of advancement flexibility (sometimes it may be better to choose a path later than to choose it now because you will have more information later e.g. which of your other units survived), and loss of nearness-to-advance-heal (you can use about-to-level units in ways that you can't necessarily even use a similar stronger unit). But those are properties of the leveling process itself and not the stats of any particular unit.

[Jetrel]

My take on this:

RIBLIB is true as a general principle, but should only be followed as such. Certain slight exceptions, like the shyde, which trades a -10 addition to impact resistance for a +10 to grassland defense, are fine, and are even desirable.

It's a delicate thing though - if such an exception goes beyond the "subtlety" point, it's bad. Things like the shyde are ok because the lowering of power is dubious - the unit might take 10% more damage from a damage type that is not the most common on the battlefield, but also has at least 10% more hitpoints (actually something like +22%, in her case).

This statement of mine is not subject to discussion, though - that's what I think, and after studying this subject for a year, when it first came up with the paladin, I'm not going to change my judgement.

[Elvish_Pillager]

The first post in this thread has been updated to reflect the changes to the General (WOOHOO!) and the Elvish Rider (SHEESH!)

[Soliton]

1b) It does not generally make sense to say that a unit violates RIPLIB based on anything that applies to all units, for instance upkeep and experience-to-the-enemy. However, if the upgrade in a unit's attack is so slight that it is often better to do less damage and give less experience, then it DOES violate RIPLIB.

---snip

*Elvish Scout (Elvish Rider-Once again too weak to compensate increased XP to enemies)

Could you elaborate on that?
If you make such vague statements it would help if you at least state what kind of stats you would like to see instead.

Also it looks to me as if you base your post on trunk and not 0.9.2 or something. (which the Demilich statement doesn't reflect though)

On a related note:
* synchronized the Ghost's and it's advancements physical resistances to 50%
* increased Master Bowman melee damage from 7-3 to 8-3

[Sly]

*Knight (Paladin-2 damage decrease in twilight)

In terms of figures, it does not : you're taking in the usual 20% Holy resitance of non-undead units.

Mage(Red Mage-less damage per ranged strike during Day, White Mage-less damage per ranged strike)

8-4 is more than slightly better than 7-3 +25% in my opinion.

White Mage (Mage of Light-Illuminates)
*Mermaid Priestess (Mermaid Diviner-Illuminates)

I don't agree with youre idea tha Illuminates is a reduction in power.

*Mermaid Enchantress (Mermaid Siren-gains magical)

How come it's a reduction in power.

Ghost (Wraith-loses resistances, Shadow-loses ranged attack)

All right, you can keep both the resistance and the ranged attack, but I like this choice principle.

[Galsiah]

*Knight (Paladin-2 damage decrease in twilight)

In terms of figures, it does not : you're taking in the usual 20% Holy resitance of non-undead units.

That's because RIPLIB means that there needs to be an advancement which is at least as good in all situations. Paladin does less damage per strike than a knight in situations where the 20% resistance exists.

8-4 is more than slightly better than 7-3 +25% in my opinion.

But not in all situations: 7-3 + 25% = 9-3. If attacking a powerful ranged unit with 9hp left when you've got few hit points, the standard mage would be preferable to the Red Mage. Uncommon, yes - but still against RIPLIB.

I don't agree with youre idea tha Illuminates is a reduction in power.

It is a reduction in power in some situations. RIPLIB states that power should never be reduced (for at least one possible advancement).

How come [magical is] a reduction in power.

It is a reduction when attacking a unit which would otherwise have only 20% defence (and now has 30%).
Personally I view this as a game bug rather than a unit problem. I think magical should guarantee at least 70%, not exactly 70%. I'm not sure what the general view on this is.

All right, you can keep both the resistance and the ranged attack, but I like this choice principle.

I think that the idea is that if a player chooses unit X to perform function Y, then if X levels, there should always be an option which also performs function Y. Levelling should not negate the reason the unit was chosen in the first place.

Choice is good IMO, but that should include the choice to stick to the decision the player made in choosing his units. Generally there is no reason to reduce stats when a unit levels. The only real difficulty I can see satisfying RIPLIB is with "advantages" like illuminates. In that case you just need an alternative advancement path.

[Sly]

Allright on most points, even if I find them nonexistent problems (all right the RIPLB strict definition is matched but the effect is totaly neglectable).

But I don't understant your answer on this :

All right, you can keep both the resistance and the ranged attack, but I like this choice principle.

I think that the idea is that if a player chooses unit X to perform function Y, then if X levels, there should always be an option which also performs function Y. Levelling should not negate the reason the unit was chosen in the first place.

The ghost have both very good res and a ranged attack (X perfoms Y and Z). When levelling I can only keep one of the 2 (if I used X for Y I can still use it, the same for Z)

[turin]

BUT if you used X for both Y and Z, X LOSES its usefulness when it advances.

Unfortunately, magical does result in a reduction of power. It is a stupid, easily removed reduction, but for some reason it remains. The supposed rationale for magical not aways being better is that it needs to be separated more from marksman, and that since mostly mages have magical, and most mages have fire, and the main place units have 20% or 10% defence is in snow, swamp or water, that simulates fire being worse against units in water rather nicely.

IMHO, that's BS.

[irrevenant]

BUT if you used X for both Y and Z, X LOSES its usefulness when it advances.

Unfortunately, magical does result in a reduction of power. It is a stupid, easily removed reduction, but for some reason it remains. The supposed rationale for magical not aways being better is that it needs to be separated more from marksman, and that since mostly mages have magical, and most mages have fire, and the main place units have 20% or 10% defence is in snow, swamp or water, that simulates fire being worse against units in water rather nicely.

IMHO, that's BS.

I agree that rationale is BS.

OTOH, I think it makes sense for magic to be a fixed 70%, rather than a minimum 70%. The attacks themselves may be 100% accurate, but that 30% represents that the act of invoking magic itself is an ultimately fickle business...

[Elvish_Pillager]

Could you elaborate on that?
If you make such vague statements it would help if you at least state what kind of stats you would like to see instead.

The current stats are approximately what that stats were a long time ago when there were a ton of complaints about how weak the Elvish Riders were. In particular, I found that the Siege of Elensefar was easier to win with Elvish Scouts than with a mix of Scouts and Riders.

After the changes that FINALLY came about, I tried HttT scouts-only and found Riders to be decently powerful.

If your concern is that they'd be too powerful in multiplayer with the older stats, well, why NOT let a player get a powerful unit if they manage to level up an Elvish Scout?

Also it looks to me as if you base your post on trunk and not 0.9.2 or something. (which the Demilich statement doesn't reflect though)

Ah, the Demilich. Thanks for pointing out my two errors. (post corrected)

On a related note:
* synchronized the Ghost's and it's advancements physical resistances to 50%
* increased Master Bowman melee damage from 7-3 to 8-3

Three cheers for Soliton!