Store Page
Symphony of War: The Nephilim Saga
All Discussions Screenshots Artwork Broadcasts Videos News Guides Reviews

Symphony of War: The Nephilim Saga

Store Page
View Stats:
Global Achievements
purplecharmanderz 22 Jan @ 4:55pm
Lithe Assault: A prelude to accuracy documentation
Writing this as simultaneously a little prelude to a larger guide I'm working on regarding documentation of the accuracy system, and as a more quickly googleable way to find an answer for an otherwise somewhat vague trait.

Lithe assault is a trait that as far as ingame documentation says a portion of bonuses to accuracy also boost your crit rate, up to 15%. The popular interpretation for this is that any accuracy above 100% is converted to extra crit chance, up to 15%.

This however is not how it works. Lithe assault instead just adds a flat 15% chance to crit, if an attack is determined to be a "strong hit".

Hits in SoW are determined based on the results of 3 seperate "attempts" to hit based on the accuracy value.

- If 0 of the attempts succeed, it is a miss. (Dragons upgrade this to a glancing blow)
- If 1 of the attempts succeed, it is a glancing blow.
- If 2 of the attempts succeed, it is a normal hit.
- If all 3 attempts succeed, it is a strong hit.

Critical hits then are a seperate roll, which can only occur if an attack is a normal hit or a strong hit.

One important thing to note is cannons ignore the hit attempts, and are always normal hits.

The whole accuracy system has some more intricacies in terms of how accuracy is actually calculated - and will be explored further when i finally finish that guide.
< >
Showing 1-7 of 7 comments
Toriaz 23 Jan @ 6:18pm 
I have read this information before somewhere, but I had not been able to find it again, so thank you for posting it here.
#1
Jab 24 Jan @ 9:53am 
Great writeup, excited to see more.
To add to this: As far as I remember the "lucky" trait gives each of those chances 2 "dice rolls" to be attempted instead of just one.
#2
purplecharmanderz 24 Jan @ 10:16am 
Originally posted by Jab:
Great writeup, excited to see more.
To add to this: As far as I remember the "lucky" trait gives each of those chances 2 "dice rolls" to be attempted instead of just one.
Its technically a bit more complicated than that - but is the right idea.

Lucky causes 2 sets of "attack rolls", taking the best total successes if the attacker has it, and the worst if ***ONLY*** the defender has it. This only impacts the normal hit rolls however, and no impact on another "roll" also involved in this equation. (And is similarly ignored when cannons just get set to normal hits)

Lucky then also causes 2 crit rolls if the attacker has it, using the best result. (No benefit if the defender has it)
Last edited by purplecharmanderz; 24 Jan @ 10:20am
#3
Jab 25 Jan @ 10:59pm 
Originally posted by purplecharmanderz:
Its technically a bit more complicated than that - but is the right idea.

Lucky causes 2 sets of "attack rolls", taking the best total successes if the attacker has it, and the worst if ***ONLY*** the defender has it. This only impacts the normal hit rolls however, and no impact on another "roll" also involved in this equation. (And is similarly ignored when cannons just get set to normal hits)

Lucky then also causes 2 crit rolls if the attacker has it, using the best result. (No benefit if the defender has it)
Interesting, so it will increase chance to crit but only if the attacker is able to NATURALLY roll a strong hit with high skill?

Edit: Oh never mind just saw crits roll for normal hits as well.

So basically lucky gives two rolls to:
Chance to not miss
Chance to crit
Chance to dodge

And nothing else?
Last edited by Jab; 25 Jan @ 11:01pm
#4
purplecharmanderz 26 Jan @ 8:34am 
Originally posted by Jab:
Originally posted by purplecharmanderz:
Its technically a bit more complicated than that - but is the right idea.

Lucky causes 2 sets of "attack rolls", taking the best total successes if the attacker has it, and the worst if ***ONLY*** the defender has it. This only impacts the normal hit rolls however, and no impact on another "roll" also involved in this equation. (And is similarly ignored when cannons just get set to normal hits)

Lucky then also causes 2 crit rolls if the attacker has it, using the best result. (No benefit if the defender has it)
Interesting, so it will increase chance to crit but only if the attacker is able to NATURALLY roll a strong hit with high skill?

Edit: Oh never mind just saw crits roll for normal hits as well.

So basically lucky gives two rolls to:
Chance to not miss
Chance to crit
Chance to dodge

And nothing else?
There's about 3 other portions of rng that even have the scope to potentially involve lucky - and those are as follows:
- status application, which I'm not currently familiar enough to confirm nor deny while garunteeing accuracy.
- stat growth, which i can garuntee is not impacted in the slightest.
- "True Dodge", a portion of accuracy calculations i didn't mention here due to that alone needing its own write up just to distinguish it from evasion boosts. It is unaffected by Lucky directly.

Edit: had a chance to look into status more - Lucky ***does*** give a second roll for that if its applicable (attacker only)

Now what impact this actually has depends on a few things:
- attack in question
- tech you have unlocked
- target
- whether or not its a glancing blow

Ice mages for a brief example have a chance to chill
with hex mastery, its a 100% chance, without it its a 50% chance
Some targets have resistance to some status (although chill is not amongst those that i can find)
and if its a glancing blow - we have 1/4th the normal chance to inflict.

Leads to a case with Ice mages specifically as an example - where lucky only really helps increase our odds of successful application if:
- we don't have hex mastery
- its a glancing blow (which lucky reduces the odds of)
Last edited by purplecharmanderz; 26 Jan @ 9:51am
#5
Jab 27 Jan @ 10:13am 
Originally posted by purplecharmanderz:
There's about 3 other portions of rng that even have the scope to potentially involve lucky - and those are as follows:
Not to haggle you too much but since you've got access to the code i'm interested to hear your thoughts:

I've always thought of gunners' main advantage of being that their damage is not bad against anything, sort of a jack of all trades.
Against heavy inf they're obviously great due to armor pen, against dragons they're good due to ignoring strength and magic defense, against supports they ignore magic and usually one shot etc.

Against light infantry I THINK they have a slight advanatage over even other light infantry due to not being dependant on crits whatsoever and this is where my question comes in.

From my observations a defender having very high skill makes them likely to dodge, but almost certain to avoid being crit, even by other high skill units. I very rarely see my swordmasters crit light infantry units. My gunners will occasionally miss them, but that rate of missing seems to be less common than how often swordmaster fail to crit.

Since swordmasters damage (when compared to gunners) comes almost entirely from crits, i wonder if gunners are a more consistent form of damage output even against enemy light infantry than swordmasters are?

Is this all due to the fact that while swordmasters have a higher chance per roll, gunners need to only succeed one roll to reach maximum damage while light infantry need to complete 3 rolls to hit their own?
Last edited by Jab; 27 Jan @ 10:14am
#6
purplecharmanderz 27 Jan @ 10:50am 
Originally posted by Jab:
Originally posted by purplecharmanderz:
There's about 3 other portions of rng that even have the scope to potentially involve lucky - and those are as follows:
Not to haggle you too much but since you've got access to the code i'm interested to hear your thoughts:

I've always thought of gunners' main advantage of being that their damage is not bad against anything, sort of a jack of all trades.
Against heavy inf they're obviously great due to armor pen, against dragons they're good due to ignoring strength and magic defense, against supports they ignore magic and usually one shot etc.

Against light infantry I THINK they have a slight advanatage over even other light infantry due to not being dependant on crits whatsoever and this is where my question comes in.

From my observations a defender having very high skill makes them likely to dodge, but almost certain to avoid being crit, even by other high skill units. I very rarely see my swordmasters crit light infantry units. My gunners will occasionally miss them, but that rate of missing seems to be less common than how often swordmaster fail to crit.

Since swordmasters damage (when compared to gunners) comes almost entirely from crits, i wonder if gunners are a more consistent form of damage output even against enemy light infantry than swordmasters are?

Is this all due to the fact that while swordmasters have a higher chance per roll, gunners need to only succeed one roll to reach maximum damage while light infantry need to complete 3 rolls to hit their own?
So there's a few moving parts here.

1) accuracy and crits are both dependent on the difference in skill, however importantly - accuracy has a high base from the attacks in question, while crit has a much smaller base. And crit has ***Double*** the penalty for skill being lower than your targets however.

2) glancing blows have a significant decrease in damage (bout 60% lost) compared to normal hits.

3) strong hits on their own only have a 10% damage boost over a normal hit. (This stacks multiplicatively with the crit multiplier if you land both)

4) gunners don't just ignore armour, but they also ignore a good sized portion of strength as well when compared to other infantry units.


This does lead to gunners having high consistent damage in this comparison - and just as a comparison for how consistent this can land - rough numbers assuming 85% accuracy, and 10% crit rate:

Normal hit rate, 32.5% - 10% of this is also a crit, or 3.2%.
Strong hit rate, 61.4% - 10% of which is a crit (ignoring lithe assault) for an extra 6.1% crit rate

Leads to a crit only 9.3% of the time, vs gunners hitting their higher average 61.4% of the time in comparison.

Realistically from what was described we're also dealing with smaller numbers than this example (since miss rate at this example is 0.3%)
#7
< >
Showing 1-7 of 7 comments
Per page: 1530 50

Symphony of War: The Nephilim Saga > General Discussions > Topic Details
Date Posted: 22 Jan @ 4:55pm
Posts: 7

Discussions Rules and Guidelines