Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[a smiling bearded cretin]

But what about the vastly more important, combat speed?

The question's been asked before, but none of us have answered it. Maybe we're happy with it as is? It's workable knowing your combat speed is relative to your opponent.

The better your observer skill, the more likely/often you can see the stages of an attack: distant => starting => incoming => IMMINENT (I believe there may have been a fifth, but am probably imagining it) and with even more observer, along with stage of the attack (only weapon) can be the attack type heavy/precise/(standard will have no indicator)/wild/quick. Each stage is one or more instances/ticks "," away from being delivered. If you see an opponent's distant attack and advance by one instant "," it advances to starting, etc. And the number of instances can be modified by the attack type as quick means fewer instances while heavy/precise would add more to the attack/recovery.

Also the quality of your attack menu will change by instant as well. What was once green the next instant/tick, becomes white, yellow or red.

[bahihs]

Very useful information smiling cretin! Thank you.

So it seems observer is actually very important now.

Based on what you said, each "state" is only one tick long regardless of anything? e.g Suppose my opponent is in recovery, if I wait 1 tick he will no longer be in recovery?

Also we do know that attack accuracy is affected by the state of the opponent, but how is it affected and to what degree? For example, if my opponents attack is "distant" and I check the attack menu and see that an attack to the head is "difficult, not very square" how will this change if I wait a tick, and my opponents attack is at "starting"? Or if I wait until my opponent is in recovery?

Another big problem I seem to have is actually landing a hit, regardless of the difficulty of the strike. From what I can see several things can happen in order for a hit to be ineffective (assume all of these are at "easy, square hit"):

1. "You miss!" <-- happens when your fighter/weapon skill is too low (?)
2. "[enemy] jumps away!" <-- Opponent dodged
3. "You attack is [adjective describing skill disparity(?)] parried by [weapon]" <-- Attack is parried, the adjective possibly showing the weapon skill of the opponent OR the disparity in your rolls.
4. "Your attack is blocked by [shield]" <-- Attack is blocked
5. "Your attack bounces off" <-- Happens with Bronze Colosetes
6. "Your attack is deflected by [armor]" <-- Combination of enemy armor user skill, and material of armor relative to your weapon

What is actually going on behind the scenes?

Also has anyone done any testing with jumping/attacking while hanging from something? Often times, you can dodge back then wait for the opponent to come to you (multi-attack is very nice for this). What if instead of wait you jump toward your opponent? Can you get struck while in the air?

Time to go test some of this!
 

[Dampe]

From the comments in the raws it looks like a speed of 1 tile per frame would be 8.100 but I've yet to actually get a creature capable of this speed, though I haven't tried non-humanoids, I have no idea what happens if you were to go faster but I assume your character implodes or turns into a wambler which then implodes or something like that.

That's basically what they thought about the sound barrier before Yeager proved them wrong.
(Except for the wambler part)

[a smiling bearded cretin]

Based on what you said, each "state" is only one tick long regardless of anything? e.g Suppose my opponent is in recovery, if I wait 1 tick he will no longer be in recovery?

Attack stages (distant => starting => incoming => IMMINENT) are one "," tick/instant long for dwarves (who are the slowest out of dwarf/human/elf). It may be possible for an extremely fast character versus a very slow one to see more instants per stage. Again, maybe my imagination, but I recall once seeing starting for two ticks. That needs to be confirmed.

Recovery is different than the attack stage. You can wait many ticks in recovery. Let's do my dwarf (observer at talented) versus NPC elf. So, say NPC elf attacks my dwarf with axe, a hack. No other info is given.

Spoiler (click to show/hide)

I choose to block. Elf performs the hack and the shot is blocked. This begins the elf's recovery cycle which I check on by looking at the attack menu (shift + a)

Spoiler (click to show/hide)

Then I follow this by moving combat forward by one "," instant before checking again, where balance is regained.

Spoiler (click to show/hide)

One "," instant later, he is talking.

Spoiler (click to show/hide)

one "," instant later, he still speaks.

Spoiler (click to show/hide)

One "," instant later and recovery is now over.

Spoiler (click to show/hide)

One "," instant later and the elf resumes attacking.

Spoiler (click to show/hide)

So there were 4 instances of being in recovery (5 if you count coming out of it). Axe is [ATTACK_PREPARE_AND_RECOVER:3:3]. Did the elf actually take a heavy/precise attack, but my dwarf's observer was too low to notice it? How much does agility modify the penalty that heavy/precise cost? From the perspective of utility, it doesn't matter if his recovery was more or less. With each instance (4), I could have selected to attack and check the next menu to see if there were are any green shot opportunities. Instance 1 where the elf was regaining balance (the result of shield block) would have been the best time to attack and because elf and dwarf are similar in size possibly to charge attack. If there were no good choices in this recovery cycle, I'd let the cycle repeat itself.

Unproven, but I always tend to assume that if there were penalties for being in recovery, they were worse at the beginning and less to toward the end, when the NPC is coming out of it.

Also we do know that attack accuracy is affected by the state of the opponent, but how is it affected and to what degree? For example, if my opponents attack is "distant" and I check the attack menu and see that an attack to the head is "difficult, not very square" how will this change if I wait a tick, and my opponents attack is at "starting"? Or if I wait until my opponent is in recovery?

No idea about the first question. If you choose to attack during the NPC's attack stage, you are countering. If you wait the one "," instant from "distant" to "starting" the attack to the head may become better or worse. You can keep moving forward one "," instant through each stage until IMMINENT where you should act (unless you opt to auto/passively-defend which will be random). Once the NPC's attack is performed he/she will be in recovery and you can advance combat forward one "," instant and check the attack menu each instant once again.

So it seems observer is actually very important now.

But here's the thing. Observer is very useful, but having more of it which means having more information, only means you have more to work with, not that you will be more successful. In the end, you could just have novice observer, always choose the most cautious fight option, {quick attack + dodge away) and it will serve you as well as the player with more choice who chooses to be more selective when attacking. I say that as someone who prioritizes observer.

[vjmdhzgr]

Very useful information smiling cretin! Thank you.

So it seems observer is actually very important now.

Based on what you said, each "state" is only one tick long regardless of anything? e.g Suppose my opponent is in recovery, if I wait 1 tick he will no longer be in recovery?

Also we do know that attack accuracy is affected by the state of the opponent, but how is it affected and to what degree? For example, if my opponents attack is "distant" and I check the attack menu and see that an attack to the head is "difficult, not very square" how will this change if I wait a tick, and my opponents attack is at "starting"? Or if I wait until my opponent is in recovery?

Another big problem I seem to have is actually landing a hit, regardless of the difficulty of the strike. From what I can see several things can happen in order for a hit to be ineffective (assume all of these are at "easy, square hit"):

1. "You miss!" <-- happens when your fighter/weapon skill is too low (?)
2. "[enemy] jumps away!" <-- Opponent dodged
3. "You attack is [adjective describing skill disparity(?)] parried by [weapon]" <-- Attack is parried, the adjective possibly showing the weapon skill of the opponent OR the disparity in your rolls.
4. "Your attack is blocked by [shield]" <-- Attack is blocked
5. "Your attack bounces off" <-- Happens with Bronze Colosetes
6. "Your attack is deflected by [armor]" <-- Combination of enemy armor user skill, and material of armor relative to your weapon

What is actually going on behind the scenes?

Also has anyone done any testing with jumping/attacking while hanging from something? Often times, you can dodge back then wait for the opponent to come to you (multi-attack is very nice for this). What if instead of wait you jump toward your opponent? Can you get struck while in the air?

Time to go test some of this!

Seems good, but very important correction. Armor user has no effect on the effectiveness of the armor, it only decreases the weight of it when worn. I've also heard that jumping can be pretty good as an attack, so that last idea could work.

[Max™]

You get the option to attack someone while in midair if you initiate a jump which could include an attack, I get it all the time when jumping rivers, and no, I don't need to stab that turtle, or even the alligator, thanks... but the carp and sponges should watch out.

[bahihs]

Very useful information smiling cretin! Thank you.

So it seems observer is actually very important now.

Based on what you said, each "state" is only one tick long regardless of anything? e.g Suppose my opponent is in recovery, if I wait 1 tick he will no longer be in recovery?

Also we do know that attack accuracy is affected by the state of the opponent, but how is it affected and to what degree? For example, if my opponents attack is "distant" and I check the attack menu and see that an attack to the head is "difficult, not very square" how will this change if I wait a tick, and my opponents attack is at "starting"? Or if I wait until my opponent is in recovery?

Another big problem I seem to have is actually landing a hit, regardless of the difficulty of the strike. From what I can see several things can happen in order for a hit to be ineffective (assume all of these are at "easy, square hit"):

1. "You miss!" <-- happens when your fighter/weapon skill is too low (?)
2. "[enemy] jumps away!" <-- Opponent dodged
3. "You attack is [adjective describing skill disparity(?)] parried by [weapon]" <-- Attack is parried, the adjective possibly showing the weapon skill of the opponent OR the disparity in your rolls.
4. "Your attack is blocked by [shield]" <-- Attack is blocked
5. "Your attack bounces off" <-- Happens with Bronze Colosetes
6. "Your attack is deflected by [armor]" <-- Combination of enemy armor user skill, and material of armor relative to your weapon

What is actually going on behind the scenes?

Also has anyone done any testing with jumping/attacking while hanging from something? Often times, you can dodge back then wait for the opponent to come to you (multi-attack is very nice for this). What if instead of wait you jump toward your opponent? Can you get struck while in the air?

Time to go test some of this!

Seems good, but very important correction. Armor user has no effect on the effectiveness of the armor, it only decreases the weight of it when worn. I've also heard that jumping can be pretty good as an attack, so that last idea could work.

Doubted this at first, but extensive testing with arena shows that you are absolutely correct, with some surprising results besides:

Spoiler (click to show/hide)

Effectiveness of Armor User on Combat Ability

Testing Conditions:

1. Two groups, 50 humans with average in all attributes (is there actually a way to change attributes?)
2. GM skill in swordfighter and GM skill in Armor User for group 1
3. GM skill in sword-fighter, but Competent Skill in Armor User for group 2
4. Both groups, fully armored (mail shirt, helm, greaves, gauntlets, boots) in iron, given Iron Long Swords

Results:

Performed two trials, both resulted in both groups hilariously falling unconscious from over-exertion. No casualties after a good 10(!) IRL minutes of fighting.
Also repeated the experiment a further two times, this time setting armor user skill to 0 for group 2, still the same results! The combat log shows that attacks are being deflected, even strikes on unconscious opponents.

Note that in both sets of trials, speed was not affected by armor (GM skill or no skill, speed was still 1.0 for both groups). I suspect this was because iron is relatively light. I address this in the second experiment (see below)

Experiment 2:

Testing Conditions:

1. Same as above except:
2. Used copper armor and arms instead of iron (to look at the effect of armor user on speed)
3. Used second set of trial conditions (GM Armor user vs. No armor user)

Results:

First off, this time armor user does have an effect on speed. GM armor users remain unhindered, while no skill armor users have a speed of 8.98. Not a significant deviation but nevertheless, as well shall see, a very important one.

This time, there were casualties: out of 50 GM armor users, 26 survive; out of 50 no skill armor users, 50 perish. The results are quite clear.

However, for the sake of completeness, I decided to do one more experiment. This time to determine the minimum level of armor user necessary to move unhindered in copper armor (the heaviest available armor) as well as to determine when the gap in armor user skill starts to become significant (which I hypothesized was directly related to speed penalties)

Experiment 3:

Testing Conditions:

1. I use one on one fights to establish some baselines, then 50 vs 50 to eliminate outliers
2. We use copper armor, since iron armor has no penalty even at 0 armor user skill; same setup as before, full set, copper long sword
3. Fights are initially done between 0 level armor users and variable level armor users, but then switched to variable vs. variable to determine if the necessary skill gap is consistent (i.e linear) at different levels

Results:

What I found was utterly bewildering:

1. It seems at 0 skill, speed is effectively random. Despite every person having the same amount and quality of armor, attribute and skill, speeds varied from 1.0 (totally unaffected by the armor weight) to .492 (significantly affected).

2. Things quickly stabilize as armor user is increased. However there still seems to be some level of randomness lingering around until about "proficient" level. At that point, out of 6 proficient armor users, all 6 had 1.0 speed (compare this to: 1 out of a group of 6 skilled armor users had a speed of .774). Before that point, however, there were units with less than 1.0 speed (at competent it seems to slow down).

Clearly more rigorous testing would need to be done to establish concrete percentages, but there is clearly some randomization at play.

3. Due to the above discoveries, trying to determine the "skill gap" necessary to be combat effective was abandoned. However I would stick with my original hypothesis given the above data since it seems speed is the only thing influencing the combat outcome.

4. After rechecking the units from experiment 2, I found that, indeed there were some units with 1.0 speed even with no skill in armor user. Nevertheless those results stand.

Conclusions:

1. Armor User has little contribution to overall combat effectiveness/movement speed for characters with at least average strength
2. However, in battles with opponents of at least equal weapon skill, arms and armor; speed becomes crucial and armor user may become significant as a result ("may", because the weight/material of the armor is far more important than the actual armor user skill)
3. There is some level of randomness to the armor-weight-speed penalty, but this disappears (more testing needed to see for sure) once the user approaches proficient skill.
 

[bahihs]

A quick experiment shows that long swords are superior to both 2H and short swords:

Spoiler (click to show/hide)

Sword size and combat effectiveness

Test conditions:

1. 30 v 30 v 30 average attribute humans with GM sword skills
2. Group 1 has short sword, 2 has long sword, 3 has 2H sword

Results
Only group two survives, they have 14 people left.

Conclusion:

Long swords are superior to short sword and 2H swords (the latter, at least with multigrasp)

[FallenAngel]

At request, reposting something from another thread...

So, in the raws, Bucklers have BLOCKCHANCE of 10, while Shields have BLOCKCHANCE of 20.
Therefore, a shield is approximately two bucklers, and two bucklers are approximately one shield.
So, in terms of full shields, my adventurer has ~21.5 shield's worth of protection.
Given that each shield's utility decreases by approximately the same factor as your first shield's (most practical way for the math to function - at or above two-thirds, the utility reaches and exceeds 100%). Assuming my adventurer has an approximate chance of blocking of 50% (due to being Legendary+Whatever, this is probably closer to 99.9999...%, but this makes the math easiest), the approximate overall utility of my 14 shields and 15 bucklers is very, very close to 1.
The utility exceeds 99.9% upon 10 shield's worth.
Therefore, the minimum number of shields to get near-complete protection is 10; near-complete, in this case, is defined as "99.9% PoB or higher".

EDIT: However, going by Platinum shields and bucklers, one buckler weighs approximately 81.5% that of a shield; therefore, while two bucklers are approximately equal to a shield in terms of raw blocking power, their weight is approximately 1.63 times that of a single shield. Bucklers are the lighter option, but only when one instrument of blocking is held - otherwise, shields are the most efficient, by far.

Well, with 14 shields and 15 bucklers, my speed (at extreme amounts of agility and strength), dropped down near 1/4 normal speed.
This would be a problem, but 0.40.01 had the movement/action split; there is no significant drop in attack speed no matter your movement speed. So, a cheetah attacks about as fast as a tortoise.
Therefore, it is perfectly safe to be completely encumbered; however, once battle is close (unless you're a shieldball), I advise waiting (using , or .).

I was told this was quite interesting.

[vjmdhzgr]

How many trials did you do for that sword based test, and did you make sure the fighters were placed evenly? Testing with groups of three has far more random aspects than testing with groups of two does so I'd recommend just going with several tests of short sword vs. long sword, short sword vs. two handed sword, and then long sword vs. two handed sword. In the respect of long swords though it's already well known that long swords are better than short swords, but what would be interesting is to test two handed swords. Because I've heard a lot of stories about how incredibly powerful they are, and aside from that I haven't heard much so some specific testing would be good. I kind of want to just do these tests myself, and think that I actually will after posting this. About that armor test though, the reason some people weren't slowed down by the copper armor is because for some reason in arena attributes are still randomized so some of the humans had higher strength than the others. Another thing I've realized is that it's a well known fact that bigger creatures get thicker clothing to the point where a demon's clothes are nearly impenetrable. However the clothes also get heavier, so I'm wondering if there's any significant difference between human armor, and dwarf, goblin, and elf armor.

[vjmdhzgr]

Short sword vs. two handed sword test 1:11 humans versus 11 humans with grand master fighter and swordsman and steel short or long swords. Two handed swords won with five survivors that didn't have any significant injuries.
Test 2: same set up two handed won with four survivors, two broken legs, and one cut off arm.
Test 3: same set up two handed won with seven survivors, one broken lower body, one broken lower and upper body, a broken leg, one severed hand and lower leg, one broken arm, and a severed arm.
Conclusion:steel two handed swords are a good bit better than steel short swords among masters with no armor involved.
Short sword vs. two handed sword, steel swords and armor, grand master fighter, swordsman, and armor user, 11 humans versus 11 humans, armor is solid armor everywhere and three mail shirts (highest armor possible).
Test 1:Concluded as a tie after 3 minutes all remaining combatants are passed out from exhaustion 6 survivors from team short sword, 7 survivors from team two-handed sword.
Test 2:Concluded after 2 minutes as all remaining combatants had passed out from exhaustion 5 survivors from team short sword, 4 survivors from team two-handed sword.
Conclusion:neither steel short swords or two-handed swords are good at getting through steel armor.
Short sword vs. two handed sword, copper swords, grand master fighter and swordsman 11 humans versus eleven humans.
Test 1:Team two-handed sword won with four survivors, two had red skulls, and the other two had red upper bodies.
Test 2:Team short sword won with eleven survivors, five relatively uninjured, one with a red skull and full red leg, one with red foot, one with red skull red arm and red leg, one with red arm, one missing both feet three red arm parts, and one red leg part.
Test 3:Team two-handed sword won with five survivors, I'll just simplify it and say all had red injuries, and one was basically dead.
Test 4:Team short sword won with seven survivors, two uninjured, three with one red injury, and two basically dead.
Conclusion:I don't want to do more testing right now, but it would seem that with copper swords short swords are actually better than two-handed swords. This is why you test with more than one material everybody.

[bahihs]

I applaud your efforts vjm but I must ask did you check to see if the 2h ers were all multi grasping? Would that even have an effect? Also I did check their str in the armor experiment and it was all average. If the attributes are changing they must be hidden. Finally could you test 2h vs long sword? I think that's the most important pair we need to check (its results could support or refute my earlier experiment)

Also thank you Angel for putting your shield info, much appreciated.

[Bloax]

I've also heard that jumping can be pretty good as an attack, so that last idea could work.

Jumping is rather hilarious at the moment, and actually one of the few ways you can take on a necromancer tower - albeit at the constant risk of instasplatting if any of the dozens of attacks you receive hit you.
It's very useful for one specific purpose though - getting your passed-out companions out of trouble, since they won't dodge your jump and are thus catapulted away in the direction you jumped in.

[Max™]

Waaaaaay out of trouble:



Bonus aquarescue technique:

[bahihs]

I've also heard that jumping can be pretty good as an attack, so that last idea could work.

Jumping is rather hilarious at the moment, and actually one of the few ways you can take on a necromancer tower - albeit at the constant risk of instasplatting if any of the dozens of attacks you receive hit you.
It's very useful for one specific purpose though - getting your passed-out companions out of trouble, since they won't dodge your jump and are thus catapulted away in the direction you jumped in.

Actually, having done some testing with a peasant character, jumping is absolutely suicidal unless your opponent is prone or knocked out. Seriously, you will die. Also it seems you can't actually attack while in the air (at least I've never had the option to).