Ships & Modules

So a few days ago I noticed that EFT seemed to be showing wrong numbers for applied DPS. At least according to the EVEuni wiki info, which CCP has quoted as correct recently. So I made a quick spreadsheet to test damage application, and EFT was overestimating by 15%. I thought I should check PYFA as well, and it was showing the same thing.

So, the question then is, am I wrong, or are both EFT and PYFA wrong?

(Large Images)
PFYA Graph
EFT Graph

Excel Sheet - Added Pyfa Implementation for reference, updated for error fix.

Ok, so a quick look on the pyfa code shows:

multiplier = (chanceToHit ** 2 + chanceToHit + 0.0499) / 2

My implementation is:

multiplier = (chanceToHit - 0.01) * [ (0.50 + chanceToHit + 0.49) / 2 ]

So at 100% hit, pyfa shows 102.495% while I (and EvEuni) show 101.505%.

This should be 1% * 300% plus 99% * AverageHitDamage%
Where AverageHitDamage% is (LowHit% + HighHit%)/2
LowHit is 50% always
HighHit is 50% to 149% depending on accuracy
HighHit thus = 0...0.99 + 0.5 or chanceToHit + 0.5 (it is not 0-1 since that is a 'crit')

This makes:

AverageHitDamage% = (0.50 + chanceToHit + 0.49) / 2

The 0.49 is a 49% damage bonus for maximum hit quality, the 0.50 is the 50% minimum damage%.
And if we go back to the 100% hit% from above, 99% is the chance to hit that remains from non-crit hits. Or rather:

Average Crit = chanceToCrit * CritDamage%
Average Crit = 1% * 300%
Average Hit = (chanceToHit - chanceToCrit) * (AverageHitDamage%)
Average Hit = (chanceToHit - 0.01) * ( (0.50 + chanceToHit + 0.49) / 2 )
Average Damage = Avg.Crit + Avg.Hit
Avg,Dmg = 3% + (chanceToHit - 1%) * (AverageHit)

Average Damage is a modifier, so you can apply it where you want.

AverageDPS = ShipDPS * ( 0.03 + ( ( chanceToHit - 0.01 ) * ( ( chanceToHit + 0.50 + 0.49 ) / 2 ) )

Well the story deepens, as I am still getting a discrepancy...
So I get the same values for Hit%, Angular Velocity, etc. I suspect PYFA is doing something with the SigRadius.
In addition to all the above (PYFA damage% error), there is an error with skills affecting things they should not. (Either TargetSig or the 40,000 base W.A.S target size) See picture for Lv0 skills test showing PFYA under-estimating applied dps EFT has the same issue as well. (EFT numbers are slightly different because it enforces min-skill levels)

There was an error in another part of my sheet, but the above issue is still present in PYFA and EFT, the difference is only about 1% though. Interesting that it seems both EFT and PYFA have the same error. The red/blue bars in the sheet show the relative difference.

Is there an ETA for the June release?

Splendid.

I wasn't using any specific fit, just grabbing any of the FAX's for example and sticking a T2 triage on their, now without any links or anything Pyfa says default lock range on the Lif with Triage is 125km, which is accurate.

If I apply 1 T2 remote sensor damp range scri[ted Pyfa says I should get 86km lock range, and 63 with 2 on it. If I apply this in game I get 113 and 104km lock range respectively.

I have the same issue with remote sensor boosters as well.

If you would prefer I'll post this issue on Github rather than here.

lock range of carriers:
as soon as i open a carrier fit is the right one. when i click any module/change fit, it goes back to 300 km.

Feature Request: Outgoing Damage Profile
Sometimes it's not clear-cut (without doing the math) what your damage output profile is. For example, a rattlesnake. What percentage of your damage goes into each type?

I envision it visually exactly as the incoming damage profile, 4 percentage bars in the firepower panel.

Bug Report: Reactive Armor Hardener -- Resistance Phasing skill
When I have my own character selected (who does not have the skill injected) and right-click on the module to go into the module skills, there is only one skill listed: Thermodynamics. When I select the All 5 character and right-click on the menu into the module skills, Resistance Phasing shows up.

Is there any plans to add citadels? Since they fit like ships I would think that they, and the future structures would be something that everyone would be interested in.

I'm liking some of the recent changes...compare thing built into the module stats, drag & drop replace modules...

=/
=\

Good work.

3 standing requests still:

> All Level 3 character, All Level 4 character (it's a huge PIA to make the plan in EVEMon. I started twice, aborted twice...massive PIA)
> Capacitor values (+values) next to the module, in the same way that there are minus values for items like armor reppers, hybrid guns, hardeners, sensor boosters, etc. Some + value from a cap battery or cap recharger or CCC would help me recognize how much benefit I'm getting from that module
> A Notes panel underneath the fit like EFT does where I can drop notes on the item; what this fit is for, how to use it, strengths, weaknesses, other chatter.

Nice work on the program.

Is there any way to delete old fittings (e.g. "delete all fittings that have not been modified for X days") other than deleting saveddata.db as a whole? I don't want to lose my settings and fresh fittings, just get rid of pre-tiericide stuff as painlessly as possible.

All,

Real life has reared it's ugly head and I'll be extremely busy for the next few weeks, which means I'll be a bit slower with development / checking the thread. In the mean time, if you have ideas / bug reports, please remember to post on the github issues page noted in the first post of this topic. I tend to check it much more frequently ;)


Scroll all the way to the bottom, there should be a pyfa-1.22.2-yc_118.6-1.0-win.exe link. use that to install.



Unfortunately not. We don't track modified date, only created date. I know it sucks, especially for the past 2-3 years now that CCP is doing balance passes every update and nullifying old fits. =/