27 replies to this topic

[qeurul]

Hello,

I found apparent discrepancy concerning Targeting Computers bonus to critical percentage.

Lets lay some ground rules first. In normal unmodified conditions the crit chance is static 42% on any given weapon.

This great illustration is borrowed from this crit-related topic:
http://mwomercs.com/...-a-brief-guide/


Now to the real question:
What is the total critical chance with the Targeting Computer?

Why I ask about the total critical chance, and not only the additional chance? The following apparent discrepancy got me thinking if I had understood the whole crit groundwork wrong all together, and so these differentiating percentages could somehow still mean the same.

In this case I use the Targeting Computer VI as an example. The MWO in-game mechlab tooltip shows as follows (29.9.2016):


This tooltip gives us three critical chances: 7,5%, 4,25% and 0,925%. Which makes up to total of 12,675% additional crit chance.

Combine 12,675% with static 42% and we get grand total of 54,675% of critical chance with TC VI.

On the other hand the MWO wiki http://mwo.gamepedia...geting_Computer tells us the following about TC VI:
Increased crit chance for projectile and beam weapons: [+23.5]%

So if we use this value instead, we would get grand total of 65,5% of critical hit chance with TC VI.


So which one is it? There is apparent 10,825% difference between the crit chance of these two sources.

Over ten percent is huge amount when weighing the pros and cons of using six tons and six critical slots for TC. And if you are wondering why TC VI, it's for Night Gyr with gauss and two erPPC's Critical ripping FTW!

[Tristan Winter]

Interesting. But more importantly, instead of looking at TT, look at the percentage of Clan mechs (each weight class separately) that are sacrificing weapon tonnage to carry big TCs. How many Adders carrying two ton TCs, how many Kodiaks carrying 5+ ton TCs.

[Lostdragon]

It is weird how TC stats are listed because whether you get 1x, 2x, or 3x is what it implies is impacted. The default % on those should add up to 100% (they do with rounding) because there are two seperate rolls, one to see if you crit then on a successful crit another to see how many slots you crit. The second roll would seem to be what the stats say is impacted by the TC but for that to be useful it needs to lower the chance of a 1x roll and increase the others while maintaining a total for all three of 100%.

If you just add the TC values to the listed chances of the second roll you wind up with values over 100, so that doesn't make sense and increasing the base 1x chance would actually be a nerf, not a buff to crits. The TC must be impacting the first roll in some way to make the 42% crit chance higher and then there may be something going on to impact the second roll but the tooltip info is as clear as mud.

Maybe McGral can share some insight on how the code is setup on the backend for these?

Edited by Lostdragon, 29 September 2016 - 06:39 AM.

[Monkey Lover]

When it does a crit roll does it do it 1 per-alpha or 1 per weapon?

My view it does it per weapon so really the extra few percent really didn't matter. Fire 8 clan lasers at something it's going to crit. I normally take the 1 ton for the beam range anyway.

Edited by Monkey Lover, 29 September 2016 - 07:09 AM.

[Mawai]

Hi,

In MWO, most effects are multiplicative and not additive.

For example, if something says that the TC provides a 23.5% boost to crits then it probably means:

Crit chance = 0.42 + 0.42*0.235 = 0.5187

This method prevents multiple effects that boost the same stat from getting out of hand whereas if you simply add them up it can get unbalanced.

Ok .. so the TC pop up lists 7.5, 4.25 and 0.925%

The base crit probabilities are

Total crit chance = 0.42

0.42*0.595 = 0.2499 *1.075 = 0.2686
0.42*0.333 = 0.1399 *1.0425 = 0.1458
0.42*0.0714 = 0.0299 *1.00925 = 0.0302

Total crit chance after TC if multiplicative = 0.4446 = 5.85%

0.42*(0.595 +0.075) = 0.2814
0.42*(0.333 + 0.0425) = 0.15771
0.42*(0.0714 + 0.00925)=0.0387

Total crit chance with TC if additive: 0.4778 = 13.7% increase

So whether applied additively or multiplicatively, the numbers listed on the in game pop up are significantly less than were originally advertised for TC ... which probably means that TC crit chances were nerfed at some point (though all of this is a guess as to how it is actually implemented in the game ... I'd also add that PGI has demonstrated in the past that math is not their strong suit).

[Tatula]

This bring out the silliness of TT Battletech. What computer weighs 7 tons?!?!? Are they using vacuum tubes and hydro cooling? I guess Lost Tech.

[Dread Render]

to me the bottom line is TCs should be worth their weight, in weapons, armor, heat sinks, whatever.
I'm not convinced they are.
I only use them when i have a slot left that i cant put anything else in.
...If they get a boost, I'll think again,

[Requiemking]

The interesting thing about TCs is that, while they boost the crit chances of most projectile weapons and all energy weapons, the dedicated crit-seeking weapons(LBX AC and MG) get nothing from them. I can understand with MGs, they weigh next to nothing and are terrifying once your armor is gone, but LBX need a buff anyways. Why not let them take advantage of the Crit boost? The boost isn't even noticeable unless you are using a TC Mark IV and up, and Clan LBX are notorious for being bad to begin with. I understand that this might make LBX OP, but really, when the crit boost doesn't really mean much unless you spend at least four tons and four crits on a TC? I think it could be the buff LBX ACs need to make them worth-while.

Edited by Requiemking, 29 September 2016 - 07:52 AM.

[Lostdragon]

Mawai, on 29 September 2016 - 07:15 AM, said:

Hi,

In MWO, most effects are multiplicative and not additive.

For example, if something says that the TC provides a 23.5% boost to crits then it probably means:

Crit chance = 0.42 + 0.42*0.235 = 0.5187

This method prevents multiple effects that boost the same stat from getting out of hand whereas if you simply add them up it can get unbalanced.

Ok .. so the TC pop up lists 7.5, 4.25 and 0.925%

The base crit probabilities are

Total crit chance = 0.42

0.42*0.595 = 0.2499 *1.075 = 0.2686
0.42*0.333 = 0.1399 *1.0425 = 0.1458
0.42*0.0714 = 0.0299 *1.00925 = 0.0302

Total crit chance after TC if multiplicative = 0.4446 = 5.85%

0.42*(0.595 +0.075) = 0.2814
0.42*(0.333 + 0.0425) = 0.15771
0.42*(0.0714 + 0.00925)=0.0387

Total crit chance with TC if additive: 0.4778 = 13.7% increase

So whether applied additively or multiplicatively, the numbers listed on the in game pop up are significantly less than were originally advertised for TC ... which probably means that TC crit chances were nerfed at some point (though all of this is a guess as to how it is actually implemented in the game ... I'd also add that PGI has demonstrated in the past that math is not their strong suit).

After doing some more research I think I was wrong the way it actually works is there is one roll, not two. Baseline is 25% 1 crit, 14% 2 crit, 3% 3 crit.

Multiplicative it would be:

0.25*1.075=0.26875
0.14*1.0425=0.14595
0.03*1.0925=0.032775

Which would be 44.75% crit chance or just 2.75% increase.

But I think it is more likely to actually be additive and yield:

0.25+0.075=0.325
0.14+0.0425=0.1825
0.03+0.00925=0.03925

Or total crit chance of 54.7 for a 12.7% increase.

Edited by Lostdragon, 29 September 2016 - 08:04 AM.

[Lostdragon]

Also, @ OP, info on MWO gameopedia seems to be based on outdated original values for TCs.

http://mwomercs.com/...ommand-console/

[Ghogiel]

The MWO wiki is likely out dated. The tooltips are likely updated, or hopefully pulled from game files.

[qeurul]

Thank you really much for you input!

Now if we just could get some blue input on this.

As I was afraid, as of now we have 1-7 tonnes of potential crap, because nobody actually knows what they deliver. Even in TC1 it would be nice to know the actual crit% bonus it provides.

[Navid A1]

I believe the Crit chance system works like this:

1x,2x,3x chances after a crit is called are not modified. They all remain the same as base values adding up to 100%.
The numbers in TC stats popup for each crit type is the total increase of each crit type after being multiplied by the total crit probability

• First, 1x Crits:

1x Crit chance with TCIV = ( 0.42 * 0.595 ) + 0.075 = 0.3249

Since 1x crit always has 59.5% crit chance after a crit is called, the original crit chance would have been as following:
0.3249 / 0.595 = 0.546 (54.6%)

• Now lets check the 2x crit type:

2x Crit chance with TCIV = ( 0.42 * 0.3333 ) + 0.0425 = 0.1825

Since 2x crit always has 33.33% crit chance after a crit is called, the original crit chance would have been as following:
0.1825 / 0.3333 = 0.547 (54.7%)

• And finaly, 3x crits:

3x Crit chance with TCIV = ( 0.42 * 0.0714 ) + 0.00925 = 0.03924

Since 3x crit always has 7.14% crit chance after a crit is called, the original crit chance would have been as following:
0.03924 / 0.0714 = 0.549 (54.9%)

As you can see, the original crit chance leads to the same number which is the actual increase in crit chance. crit type chance is then calculated using the base 59.5%, 33.33% and 7.14% after a crit is called)

Rounding errors aside, the TC mk VI increases the crit chance from 42% to a total 54.7% (which is about 13% additive increase, or 30% multiplicative increase)

Edited by Navid A1, 29 September 2016 - 09:06 AM.

[Lostdragon]

Navid A1, on 29 September 2016 - 09:00 AM, said:

I believe the Crit chance system works like this:

1x,2x,3x chances after a crit is called are not modified. They all remain the same as base values adding up to 100%.
The numbers in TC stats popup for each crit type is the total increase of each crit type after being multiplied by the total crit probability

• First, 1x Crits:

1x Crit chance with TCIV = ( 0.42 * 0.595 ) + 0.075 = 0.3249

Since 1x crit always has 59.5% crit chance after a crit is called, the original crit chance would have been as following:
0.3249 / 0.595 = 0.546 (54.6%)

• Now lets check the 2x crit type:

2x Crit chance with TCIV = ( 0.42 * 0.3333 ) + 0.0425 = 0.1825

Since 2x crit always has 33.33% crit chance after a crit is called, the original crit chance would have been as following:
0.1825 / 0.3333 = 0.547 (54.7%)

• And finaly, 3x crits:

3x Crit chance with TCIV = ( 0.42 * 0.0714 ) + 0.00925 = 0.03924

Since 3x crit always has 7.14% crit chance after a crit is called, the original crit chance would have been as following:
0.03924 / 0.0714 = 0.549 (54.9%)

As you can see, the original crit chance leads to the same number which is the actual increase in crit chance. crit type chance is then calculated using the base 59.5%, 33.33% and 7.14% after a crit is called)

Rounding errors aside, the TC mk VI increases the crit chance from 42% to a total 54.7% (which is about 13% additive increase, or 30% multiplicative increase)

You are assuming the same thing I did, that you roll to see if there was a crit then you roll again on a successful crit to see how many crits there are. I believe that is incorrect based on how PGI has said crits work, I think it is just one roll with a total crit chance of 42%, so if you roll >58 and <=83 you get one crit, >83 and <=97 is 2x, >97 is 3x.

Quote

Each time the internal structure of a Mech takes damage, there is a chance that the hit will cause at least 1 critical hit. There is a 25% chance of causing 1 critical hit, a 14% chance of causing 2 critical hits, and a 3% chance of causing 3 critical hits (for a total of a 42% chance of any sort of critical hit). Each critical hit will randomly hit a weapon or piece of equipment in that location; the chance of a particular piece of equipment being hit is proportional to how many critical slots it occupies. Each critical hit damages the weapon/equipment an amount equal to the damage that caused the critical hit.

Source: http://mwomercs.com/...14#entry1474814

Edited by Lostdragon, 29 September 2016 - 09:23 AM.

[Navid A1]

Lostdragon, on 29 September 2016 - 09:21 AM, said:

You are assuming the same thing I did, that you roll to see if there was a crit then you roll again on a successful crit to see how many crits there are. I believe that is incorrect based on how PGI has said crits work, I think it is just one roll with a total crit chance of 42%, so if you roll >58 and <=83 you get one crit, >83 and <=97 is 2x, >97 is 3x.



Source: http://mwomercs.com/...14#entry1474814

Thanks for the source regarding the crit calls
However, the calculations are still valid

[Mcgral18]

qeurul, on 29 September 2016 - 08:57 AM, said:

Thank you really much for you input!

Now if we just could get some blue input on this.

As I was afraid, as of now we have 1-7 tonnes of potential crap, because nobody actually knows what they deliver. Even in TC1 it would be nice to know the actual crit% bonus it provides.

Check the game files themselves

-<Module faction="Clan" CType="CTargetingComputerStats" name="TargetingComputerMkI" id="9013">
<Loc iconTag="StoreIcons\TargetingComputerMkI.png" descTag="@TargetingComputerMkI_desc" nameTag="@TargetingComputerMkI" shortNameTag="@ui_TargetingComputerMkI"/>
<ModuleStats health="10" tons="1.0" slots="1" amountAllowed="1"/>

-<TargetingComputerStats moduleDestroyedHudWarning="eHUD_TARGETING_COMPUTER_DESTROYED">

-<WeaponStatsFilter compatibleWeapons="SmallLaser,MediumLaser,LargeLaser,ClanERSmallLaser,ClanERMediumLaser,ERLargeLaser,ClanERLargeLaser,SmallPulseLaser,ClanSmallPulseLaser,MediumPulseLaser,ClanMediumPulseLaser,LargePulseLaser,ClanLargePulseLaser" tag="BeamWeapons">
<WeaponStats critChanceIncrease="0.05,0.0275,0.0055" operation="+"/>
<WeaponStats operation="*" maxRange="1.04" longRange="1.04"/>
</WeaponStatsFilter>

-<WeaponStatsFilter compatibleWeapons="AutoCannon2,ClanAutoCannon2,AutoCannon5,ClanAutoCannon5,AutoCannon10,ClanAutoCannon10,AutoCannon20,ClanAutoCannon20,GaussRifle,ClanGaussRifle,ClanUltraAutoCannon2,UltraAutoCannon5,ClanUltraAutoCannon5,ClanUltraAutoCannon10,ClanUltraAutoCannon20,PPC,ERPPC,ClanERPPC" tag="ProjectileWeapons">
<WeaponStats critChanceIncrease="0.05,0.0275,0.0055" operation="+"/>
<WeaponStats operation="*" speed="1.05"/>
</WeaponStatsFilter>
</TargetingComputerStats>
<Audio OnDestroyedDialogue="BB_Targeting_Computer_Destroyed"/>
</Module>

Note the additive Crit multiplier, and Multiplicative Range
You have your answer

[Lostdragon]

Navid A1, on 29 September 2016 - 09:39 AM, said:

Thanks for the source regarding the crit calls
However, the calculations are still valid

Yeah, your math is good but I think the way the game actually handles it is simpler than the method you used.

[Widowmaker1981]

To be honest, crits are a secondary enough feature of gameplay that i don't think large TCs are worth the tonnage when only considering that. However, the linear increase in projectile speed is what i buy them for.. i find PPCs and UAC10s especially benefit from large TCs.

If im using primarily PPCs or UAC5/10/20 i fit a large TC if i have the tonnage. If im using lasers / gauss i fit a TC1 for the 3% range/small crit boost. I do not use a TC for (S)SRM builds, and i do not use LRM, LBX or MG builds.

[Felicitatem Parco]

I don't give a rip about Criticals.

I use TCs as a way to buy Projectile Velocity and Sensor quirks with tons and slots.

[Mcgral18]

Widowmaker1981, on 29 September 2016 - 10:18 AM, said:

To be honest, crits are a secondary enough feature of gameplay that i don't think large TCs are worth the tonnage when only considering that. However, the linear increase in projectile speed is what i buy them for.. i find PPCs and UAC10s especially benefit from large TCs.

If im using primarily PPCs or UAC5/10/20 i fit a large TC if i have the tonnage. If im using lasers / gauss i fit a TC1 for the 3% range/small crit boost. I do not use a TC for (S)SRM builds, and i do not use LRM, LBX or MG builds.

Prosperity Park, on 29 September 2016 - 10:21 AM, said:

I don't give a rip about Criticals.

I use TCs as a way to buy Projectile Velocity and Sensor quirks with tons and slots.

I use them for 2600M/s Goose Waffles, and the 20% chance to blow faces out.