Update #39: Non-Core Classes, Cooldowns, Attack Resolution, Damage vs. Armor and a Tileset!

[J.E. Sawyer]

50% seems like a really harsh penalty. I would assume crushing weapons aren't getting the -50% applied against light/medium armor, which would mean carrying a slashing weapon would be a huge liability. If one third of the time it does moderately higher damage than piercing/crushing, but two thirds of the time does 50% less damage + potentially misses DT altogether, you're going to end up with a weapon that is basically useless. I can't imagine any reason to use anything but clubs and mauls in this scenario.

 

On the other hand, if crushing and piercing weapons ARE penalized against lighter armor, you have basically a game of rocks-paper-scissors.

 

I don't see what the problem was with the original system. Maybe you wouldn't immediately know what the optimal weapon would be in any given situation, but you could make a pretty good guess, and observing the combat log would basically seal it. I don't think it's intuitive at all for some weapons to just not work against some armor. In the old system, it sounded like you would be able to get by with a suboptimal weapon. In this system you HAVE to match up the damage type and armor type.

 

Josh and Tim obviously know better than me, since they've been playing with the system; this is just how it appears to me.

 

Well, we're going to try this system out, but I do admit that I'm not 100% sold on it.  I was really having a lot of trouble working out the wide variation in values in the original system (I made a post with the spreadsheet on the previous page) after adjusting formulae and values for weeks.  If we do "go back" to the original system, it will need to include some remedies for communicating relative damage in the interface.  At the high end of the DT spectrum, the decrease in damage is 90% over initial values for fast Slash/Pierce weapons, the worst of those (the fast Pierce, e.g. Stiletto) being less than 1/6th the value of the ideal weapon: the two-handed Crush (e.g. Maul).

[Gyges]

Looks gewd. Whatever you figure out is best i hope it opens up for a lot of theorycrafting for those who are willing to put in the brainpower to create real beasts. Nothing like carefully sharpening your skills, stats and gear into a cutting edge.

 

Also hoping for a preview of combat animations in the near future, a topic that needs more attention and love. It will determine whether combat will be trite intermission or a big reason we replay the game for the 10th time.

Edited January 24, 2013 by Gyges

[rjshae]

Hi Josh,

 

I have a quick question: regarding the 50% damage penalty for certain weapon/armor combos, might it not make more sense to double the DT instead? That way critical hits can still remain relatively effective.

[SunBroSolaire]

 

50% seems like a really harsh penalty. I would assume crushing weapons aren't getting the -50% applied against light/medium armor, which would mean carrying a slashing weapon would be a huge liability. If one third of the time it does moderately higher damage than piercing/crushing, but two thirds of the time does 50% less damage + potentially misses DT altogether, you're going to end up with a weapon that is basically useless. I can't imagine any reason to use anything but clubs and mauls in this scenario.

 

On the other hand, if crushing and piercing weapons ARE penalized against lighter armor, you have basically a game of rocks-paper-scissors.

 

I don't see what the problem was with the original system. Maybe you wouldn't immediately know what the optimal weapon would be in any given situation, but you could make a pretty good guess, and observing the combat log would basically seal it. I don't think it's intuitive at all for some weapons to just not work against some armor. In the old system, it sounded like you would be able to get by with a suboptimal weapon. In this system you HAVE to match up the damage type and armor type.

 

Josh and Tim obviously know better than me, since they've been playing with the system; this is just how it appears to me.

 

Well, we're going to try this system out, but I do admit that I'm not 100% sold on it.  I was really having a lot of trouble working out the wide variation in values in the original system (I made a post with the spreadsheet on the previous page) after adjusting formulae and values for weeks.  If we do "go back" to the original system, it will need to include some remedies for communicating relative damage in the interface.  At the high end of the DT spectrum, the decrease in damage is 90% over initial values for fast Slash/Pierce weapons, the worst of those (the fast Pierce, e.g. Stiletto) being less than 1/6th the value of the ideal weapon: the two-handed Crush (e.g. Maul).

Thanks for the response. One more question I have is whether you plan on displaying DT in the combat log? It seems like that would help make it clear to players why their stiletto isn't getting through the 35DT mail + Ring of Damage Threshold. Just something like "player hit goblin for 5 points of damage, DT negated 40 damage" would make it clear that the stiletto is a bad choice.

Edited January 24, 2013 by SunBroSolaire

[Gumbercules]

Thinking about this some more, here's how you can divide up the information you give the player (with credit to Kaz and SunBroSolaire for some of the ideas):

 

• A detailed weapons screen that gives you all the information in the spreadsheet from the previous page plus anything else that could possibly be needed. Allow the player to compare the selected weapon to any other weapon in the inventory or to weapons encountered in shops or dungeons. Maybe make it so that you can even compare weapons that were already encountered but that the player currently doesn't have access to. On this screen, everything is shown in real numbers rather than abstractions like Good or Poor, but you can use color-coding or bolding to emphasize better values.
• The main UI should probably have a way of indicating what weapon each party member has equipped. Mousing over this icon in combat displays the Good, Average, Poor comparisons I talked about earlier. Opening the inventory in combat and mousing over the other available weapons displays the same information for the other weapons.
• Selecting a party member in combat and pressing a hotkey displays a color-coded circle around each enemy corresponding to the same Good, Average, Poor comparisons mentioned before.
• The combat log displays DT information.

I think the key is a mix of detailed, accurate, absolute information for strategic decisions (purchases, crafting, weapon loadout) and quick, approximate, relative information for tactical decisions (which enemy do I fight with which character with which weapon?).

[IndiraLightfoot]

We're still playing with the relative size of characters on screen.  Adam and I both agree that they are too small at the current scale, so we will be trying out a closer view soon.

 

Great! 

Hopefully, you'll soon find a perfect balance between character size and reasonably wide space for combat and stealth tactics.

[Doppelschwert]

Hi Josh,

 

I have a quick question: regarding the 50% damage penalty for certain weapon/armor combos, might it not make more sense to double the DT instead? That way critical hits can still remain relatively effective.

Not only critical hits.

50% Damage Penalty means:

0.5*Attack-DT=Damage

whereas 100% additional DT means

Attack-2*DT=2*(0.5*Attack-DT)=2*Damage.

 

I agree that this would be better if we left the old system. However, I think It might be better to use the old one with labelled indicators how good weapons are, as others are suggesting.

[Shaewaros]

Graze system seems really exciting. Overall The AC system seems to work just like the way I want it to work. If cooldowns end up being rest based I will be even happier.

Edited January 24, 2013 by Shaewaros

[Archmage Silver]

Great update! I like the idea of damage type vs. armor type, it's definitely a lot more intuitive than the previous version of the system. The "cooldowns" are shaping out nicely too, I'm perfectly fine with this plan (I was opposed to the individual ability CDs per use).

[SunBroSolaire]

But your attack will be decided if it hits or not:

 

1.8 seconds - Your Attack

- Dice rolls (Turn)

2.4 seconds - Enemy Attack

- Dice rolls (Turn)

3.6 seconds - Your Attack

- Dice rolls (Turn)

 

Also, I did a little touch up of what I see in that mysterious fellow Aumaua or Orlan. I think it looks like a monkey (I'm not a great artist, just wanted to convey the concept of what I see)

 

versus

Nice work! I think you're right that this could be the Orlan. We know they're suppose to have big ears and be furry. The Aumua didn't have ears or side burns like that. Gotta say, I thoroughly support the way both these races look like Primates. From the desciption of Orlans I was afraid of something like the Elin from Tera I've loved all the art concepts so far. The screenshot looks excellent, too.

[Valorian]

And for aumaua example. If for example two-handed war-mace does 10-20 damage against medium and heavy armor, but is bad against light armors and there fore does 5-10 and dagger does 1-4 damage against light and medium armors and 0-2 against heavy armors, then there is no reason to switch mace to dagger.

Making AI react intelligently to this system is not more complex than make AI uses their boost/defense/heal/ abilities or spell in intelligent manner.

 

You're forgetting to factor DT into the equation.

 

Making a challenging AI for such a system is a very complex task. Because it involves calculating dynamic positioning relative to the risk of attacks of opportunity and damage-type effectiveness.

Well, we're going to try this system out, but I do admit that I'm not 100% sold on it.  I was really having a lot of trouble working out the wide variation in values in the original system

Have you considered designing DT and damage to not have such a wide range, from when you start the game to end levels? Instead of 0 - 50, to have a range of 0 - 25, for example? I'm not sure how damage/DT scales as you level up and find better loot, but having huge leaps is not good for balance.

[rjshae]

 

Hi Josh,

 

I have a quick question: regarding the 50% damage penalty for certain weapon/armor combos, might it not make more sense to double the DT instead? That way critical hits can still remain relatively effective.

Not only critical hits.

50% Damage Penalty means:

0.5*Attack-DT=Damage

whereas 100% additional DT means

Attack-2*DT=2*(0.5*Attack-DT)=2*Damage.

 

I agree that this would be better if we left the old system. However, I think It might be better to use the old one with labelled indicators how good weapons are, as others are suggesting.

 

The other issue I have with this 50% damage penalty is that it might not scale realistically in cases where there is a lot of power  behind the attack. Once your weapon has broken through the armor, there isn't much left to stop it. So why does it still take a 50% penalty? It may be more realistic to scale the armor DT based on the weapon type.

 

But perhaps I'm misunderstanding the physics of the impact?

Edited January 24, 2013 by rjshae

[Ieo]

I really enjoy combat systems with deep mechanics, so I have had a knee jerk reaction to this. Little disappointed. Please don't feel you have to simplify combat because all players may not want to go to the trouble of looking at numbers when selecting weapons and armour. Some of us like that sort of thing! It may also add depth to combat, as there are more variables to consider.

 

Besides which, won't casual players just use a simpler table with the numbers removed and qualitative ranks put in? They can make a decision on what weapons and armour to use with ranks such as "good", "poor", "excellent". n.n

You have to realize that there were never granular weapon/DT/armor calculations in the old IE games, so the vast majority of players (yes, I dare say it) simply wouldn't expect to have to do that extra work on their own in an IE-inspired CRPG. Thing is, there are many aspects of a tabletop experience, including looking up all those tables and whatnot, that simply don't translate to a fluid CRPG experience.

 

And yes, it's good to keep the difficulty levels in mind, because breaking out multiple types of weapon/DT/armor tables for different difficulty levels would be more work (for Obsidian) too. I suspect it'd be easier to adjust the damage values up and and down for easy mode rather than make a "simpler table."

 

 

@Josh, I'd like to reiterate from earlier in the thread that keeping the original BG-style selection circles would be my preference as well (green for party, red enemy, blue NPC, yellow for some state changes) with potential upgrades. Maybe selection circle thickness could represent relative armor level (I have no idea if that's visually doable). Just please don't make each party member's circle a different color or worse, as in BG:EE, some the same color as others; or moddable would be nice!

 

 

Edit to add: 

 

You have to take into consideration that BG, EE or not, is still an old game. BG:EE is just a modernized mechanical update. It's tons of new stuff too and it's great imo.

 

I think you should compare (by graphical standard today) DA:O's zoom. Where it looks good zoomed out, and it looks good zoomed in.

 

Or do I not understand what zoom means?

Zoom in on this

 

EDIT: Tiles! Props    didn't see that before.

You don't understand. DA:O is a completely different environment, fully 3D rendered. The IE games including BG:EE have 2D hand-painted final art assets, which means rasterized, which also means that zoom capabilities will always be crippled in terms of usefulness. The only purpose for zoom in an IE-type game with hand-painted backgrounds is for future-proofing when screen sizes and resolutions are so big that characters are .5 cm tall. If you're going to compare mechanical game visuals, do it apples to apples.

[Osvir]

But but... zoom in on the link, and how much do you need to zoom in until you can't distinguish what is what?

In BG:EE, zooming in is ugly. Why? Because the rendering or whatever doesn't adjust, so you get pixely stuff.

DA:O had an awesome zoom in and zoom out feature, DA2 had a terrible zoom and they could've done without it. I am not talking about the zoom in going "3rd Person" on you just because you zoom in.

ZOOM IN = Zooming in, take any picture and use the "Zoom in" tool and you'll get what I am going for. Not "Hey let's change the perspective as well" like DA:O.

Zooming in on the picture (that I linked), it doesn't get pixely. It doesn't turn "ugly" (try it, the link is in your post), in DA:O, zooming in, it doesn't turn "ugly".

Apples with apples right?

Edited January 24, 2013 by Osvir

[Hormalakh]

So Josh, i've been playing with your excel sheet. Which of these are dependent and independent variables? Which one of these deals directly with the general type (piercing/slashing/crushing) of weapon you choose and which of these deals with the specific weapon you use?

 

I realize min-max will be different for specific weapons (independent variable).

 

Everything else is a dependent variable, right?

 

-(Rate is dependent on whether it's Fast, one-handed, two-handed ONLY)

-MDTDT is dependent on whether it's F/1H/2H AND whether crushing/non-crushing

-DWF is dependent of whether it's 1H/2H.

 

The reason I ask this, is will this change significantly between weapons of the same type/handling? Will there be Fast Slashing weapons with rates that aren't 0.9, DWF that isn't 1/2, and MDTDT that isn't 0.1?

 

Are you changing weapons only based on DMG*?

 

 

One issue I recognize immediately, is that the equations you're using for your crushing/slashing vs. piercing is different. I would try to find something that has the same equation all around, but where the dependent variables give you the nice gradients you're looking for. The first that came to my minds was Michaelis-Menton equations for me (but that's because I work with these everyday)

 

Finally, it's really non-intuitive how your piercing weapons do the least damage and yet you have some sort of middle armor tier that it works out for. It would make more sense to have your crushing weapons do the least damage, with the highest MDTDT (it's weird that it goes all the way up to 0.6 - I tried 0.45-0.5 and it worked better.

 

Sorry if I'm overstepping here All the best.

Edited January 24, 2013 by Hormalakh

[Somna]

 

[...]

Finally, it's really non-intuitive how your piercing weapons do the least damage and yet you have some sort of middle armor tier that it works out for. It would make more sense to have your crushing weapons do the least damage, with the highest MDTDT (it's weird that it goes all the way up to 0.6 - I tried 0.45-0.5 and it worked better.

 

[...]

I thought Piercing weapons were supposed to have the capability to ignore some DT. 

[J.E. Sawyer]

 

 

 

Have you considered designing DT and damage to not have such a wide range, from when you start the game to end levels? Instead of 0 - 50, to have a range of 0 - 25, for example? I'm not sure how damage/DT scales as you level up and find better loot, but having huge leaps is not good for balance.

 

 

I've tried the system with many different ranges.  If the range is the problem (it doesn't seem to be), the larger problem is scaling.  I really don't like systems that terminate their scaling, which is why I'm avoiding DT as a percentile reduction.

 

I've tried lots of values for weapons, for armor, for DT bypass, for MDTDT, linked or unlinked to damage types/handedness/speed in different ways.  I think the model itself has too many input variables/factors to be elegant and clear.

 

I had another idea last night that I discussed with Tim this morning that I'll be playing around with.  I think it will be more intuitive than either system I've talked about, but I want to play around with it for a while.

[Hormalakh]

 

 

[...]

Finally, it's really non-intuitive how your piercing weapons do the least damage and yet you have some sort of middle armor tier that it works out for. It would make more sense to have your crushing weapons do the least damage, with the highest MDTDT (it's weird that it goes all the way up to 0.6 - I tried 0.45-0.5 and it worked better.

 

[...]

I thought Piercing weapons were supposed to have the capability to ignore some DT.

 

 

And crushing ignores more of that DT: hence why MDTDT is higher for crushing than for piercing. I could be seeing this whole thing wrong though...

[Hormalakh]

I've tried the system with many different ranges.  If the range is the problem (it doesn't seem to be), the larger problem is scaling.  I really don't like systems that terminate their scaling, which is why I'm avoiding DT as a percentile reduction.

 

I've tried lots of values for weapons, for armor, for DT bypass, for MDTDT, linked or unlinked to damage types/handedness/speed in different ways.  I think the model itself has too many input variables/factors to be elegant and clear.

 

I had another idea last night that I discussed with Tim this morning that I'll be playing around with.  I think it will be more intuitive than either system I've talked about, but I want to play around with it for a while.

I sort of agree. It's not that there are too many inputs, it's that the inputs are unintuitive in how they change the system. I quickly knew what to do to get the right equation for your slashing and crushing weapons. But when it came to piercing, I got stuck. It wasn't intuitive.

[J.E. Sawyer]

 

@Josh, I'd like to reiterate from earlier in the thread that keeping the original BG-style selection circles would be my preference as well (green for party, red enemy, blue NPC, yellow for some state changes) with potential upgrades. Maybe selection circle thickness could represent relative armor level (I have no idea if that's visually doable). Just please don't make each party member's circle a different color or worse, as in BG:EE, some the same color as others; or moddable would be nice!

 

Tim and I, who are both color blind, would prefer to not use green/red for allied/enemy colors.

[Kaz]

Hey Josh, I was curious how large of a role damage type will play among the many tactical decisions the player will face in combat. Is the intention to make it a dominating factor, or something that is sprinkled in on occasion for flavor?

[IndiraLightfoot]

Josh: As for marking enemies and allies. Instead of those red and green circles or what not, is it not possible to use some highlight-function similar to the Z-key in NWN2, but obviously just highlighting your party or the enemy group? As long as it doesn't become anything like that weird glow you saw on enemies in Diablo 3, it was pretty weird and glaring. Just a highlight, that's all.

Edited January 24, 2013 by IndiraLightfoot

[Hormalakh]

Along with color, use a different "shape" or pattern. Unlikely as it may be, players who are completely color-blind can still be albe to tell.

Edited January 24, 2013 by Hormalakh

[Valorian]

 

Have you considered designing DT and damage to not have such a wide range, from when you start the game to end levels? Instead of 0 - 50, to have a range of 0 - 25, for example? I'm not sure how damage/DT scales as you level up and find better loot, but having huge leaps is not good for balance.

 

 

I've tried the system with many different ranges.  If the range is the problem (it doesn't seem to be), the larger problem is scaling.  I really don't like systems that terminate their scaling, which is why I'm avoiding DT as a percentile reduction.

 

I've tried lots of values for weapons, for armor, for DT bypass, for MDTDT, linked or unlinked to damage types/handedness/speed in different ways.  I think the model itself has too many input variables/factors to be elegant and clear.

 

I had another idea last night that I discussed with Tim this morning that I'll be playing around with.  I think it will be more intuitive than either system I've talked about, but I want to play around with it for a while.

 

I don't like percentile reductions (as a base armor system) either.

 

Anyhow. I'll offer you my armor mechanics system, because... I'd really like your project to succeed so if I can help in any way, I'd be glad. I'm a little bit jealous of it and I'd love to use it some day in an imaginary rpg that I'd be working on, but since that's most likely not going to happen, here it is:

 

 

Armor has two stats:

 

DT (or DR) and AQ (armor quality).

 

DT would be a flat number and I'd have it work with smaller numbers (and differences) than your example;

For example. Tunic: 3 DT Leather: 6 DT Plate: 9 DT (It would of course vary depending on material type, magical properties etc.)

 

AQ is also an extremely important stat. It would be expressed in a range, a roll, (e.g. 2-4) and would represent the quality of armor and the fact that different parts of armor have a different chance to be penetrated.

 

For example. Average leather armor could have an AQ value of 2-5. Plate 3-8. Etc.

 

 

All weapons would have an armor penetration stat, in addition to other stats, also expressed in a range:

 

For example, a dagger could have an armor penetration value of 1d5 (1-5), a maul 2d6 (2-12) etc.

With your str modifier which should be something reasonable and not too hight (and dex modifier in the case of finesse weapons, but I'll leave that for later).

 

 

How would it work:

 

Let's say you have an enemy with leather armor with the following stats:

DT: 6 AQ: 3-8

 

And an attacker with a mace:

Damage: 4-8  AP: 3-5 (+2 str bonus)  -for finesse weapons the bonus would be drawn from dex

 

If the AP roll is greater than the AQ roll, DT is halved for this particular hit. For finesse weapons (like daggers and stilettos etc.) the DT would be reduced to 1/3.

 

In practice this would look like this.

 

Attacker with the mace rolls 7 damage. Rolls an AP value of 6.

Defender in leather has base 6 DT. Rolls AQ value of 5.

AP > AQ and therefore DT is halved from 6 to 3.

7 - 3 = 4 damage is dealt.

If the AP roll wasn't higher than the AQ roll. Only 1 damage would be dealt. 7 - 6 = 1.

 

These are just example numbers.

 

It's not rock-paper-scissors like your armor system (especially the new one with the 50% damage penalty) because every weapon has a chance of penetrating, but the chance varies depending on the type of weapons, your str or dex score and other factors such as magical properties.

 

A sword could have a lower AP of let's say 1-4 (+str bonus), but higher damage than a mace.

 

Minimum of 1 damage dealt, regardless of armor, would be good.

 

PS. Although, I do think that having DT be a range would be preferable, perhaps. Instead of 6 DT, it could be 4-7 DT etc. And I know it seems complicated at first.

[Somna]

 

 

 

[...]

Finally, it's really non-intuitive how your piercing weapons do the least damage and yet you have some sort of middle armor tier that it works out for. It would make more sense to have your crushing weapons do the least damage, with the highest MDTDT (it's weird that it goes all the way up to 0.6 - I tried 0.45-0.5 and it worked better.

 

[...]

I thought Piercing weapons were supposed to have the capability to ignore some DT.

 

And crushing ignores more of that DT: hence why MDTDT is higher for crushing than for piercing. I could be seeing this whole thing wrong though...

Not in the same way.

 

Piercing flat out ignores DT up to a certain point. On the other hand, there's a cap to how much Crushing damage gets reduced by DT.

 

Using the chart's examples, Crushing's damage getting reduced by DT caps out at 15 (20 for the "Fast" category), while Slashing just keeps going downwards more and more. Piercing, on the other hand, doesn't start dropping until 25 DT, but when it does, it drops like a rock.