open All Channels
seplocked Ships and Modules
blankseplocked Spreadsheet - Damage Computations (beta version b.26b)
 
This thread is older than 90 days and has been locked due to inactivity.


 
Pages: 1 2 3 4 [5] 6

Author Topic

Ryysa
Mission Fail
Posted - 2007.04.11 08:53:00 - [121]
 

I hope this is not considered necro.

Muninn optimal range bonus seems wrong.

It's set as 25%, as if Muninn would get 5% per level, where in reality it gets 10% optimal per level.

Hoshi
Hedron Industries
Red Dwarf Racketeering Division
Posted - 2007.04.13 20:03:00 - [122]
 

Drone bay of Maelstrom is wrong, set to 75m3 in latest spreadsheet, should be 100m3

Akita T
Caldari Navy Volunteer Task Force
Posted - 2007.05.03 09:23:00 - [123]
 

Edited by: Akita T on 03/05/2007 11:02:11

Originally by: Naughty Boy
* Gun damage: the damage is determined by the tracking formula used in the tracking guide. Each kind of ammo is compared and the optimal ammo is used for every range.
Only hit chance is factored so far, as the formula for hit quality is not yet determined precisely


Well, apparently, hit quality formula has been "reverse engineered" within a reasonable statistical deviation by Kazuo Ishiguro.
Practically speaking, it really makes no difference if the formula has the wrong form, as long as results are very close to actual in-game results, no ? Twisted Evil

Forum thread Linkage

Basically, it's quite simple.
Paraphrasing the thread linked in a very simplified form:

x = random number between 0 and 1
z = To-Hit Chance

If {x<0.01} Then {quality = 3} else {quality = x + 0.49}
If {x<z} Then {damage = quality x base damage}
__

You only deal the "lower quality" hits when hit chance is lower, and the higher quality hits get discarded... so that would imply average DPS decreases a bit sharper as hit chance decreases (i.e. not linearly) in the "relevant" sections.
In other words, switching to a higher range, lower damage ammo is beneficial a bit sooner as previously thought.

At 100% chance to hit, you get 102% of "expected" DPS, not 100% of "expected" DPS, due to 1% chance for triple damage (wrecking shot).

A 50% hit chance would mean 1% wrecking shots and 49% normal hits linearly distributed in the *0.5 to *0.99 (or *1.00?) damage range.
That factors as 1% *3 and 49% with *0.75 damage, for a final DPS of a bit under 39.75% instead of the "expected" ~52% DPS.

A 20% hit chance means 1% wrecking shots and 19% normal hits with *0.5 to aprox *0.7 damage (say *0.6 average hit multiplier), for a final DPS of not ~22%, but instead only ~14.4%.

It also means that at exactly 1% hit chance (and then below), you start ONLY dealing wrecking shots, for a 3% of "expected average" DPS at 1% hit chance.
________
________

In other words, average DPS is (for chance to hit at or ABOVE 1%):

AvgDPS = BaseDPS * [ 0.03 + (ChanceToHit-0.01)*(0.5 + ChanceToHit/2) ]

For chance to hit at or BELOW 1%, AvgDPS = BaseDPS * 3 * ChanceToHit

Kazuo Ishiguro
House of Marbles
Posted - 2007.05.09 10:04:00 - [124]
 

Edited by: Kazuo Ishiguro on 09/05/2007 10:00:46
Marn Prestoc, the current custodian of the spreadsheet, is well aware of this result, but he's a bit busy IRL and I only have OpenOffice, so I can't even use the spreadsheet myself (it does open, but only after about 5 minutes, and it hogs 1GB of ram + vram). I'm sure he'll release a new version when he has time. He's already produced a few graphs, but it may take a bit longer to fix the 'best ammo type' option.

Here's one of the graphs (more in his folder - all the relevant ones have 'newformula' in the filename):

http://go-dl.eve-files.com/media/corp/preston85/newformula_megaVgeddon_50trans.PNG

It's quite interesting comparing the two different models.


Jim McGregor
Posted - 2007.05.16 15:20:00 - [125]
 


Looks like version b23v6 came up today. I was just going to play around with the spreadsheet and noticed the new version. Can be downloaded from here.

It looks different... will be fun to play around with.

Marn Prestoc
Minmatar
Maelstrom Crew
Paradigm Alliance
Posted - 2007.05.17 02:25:00 - [126]
 

I'm aware of the hit quality formula, I produced some example graphs to show the difference when using the hit quality formula while the reverse engineering of the formula was going on which were linked above. They were a pure hax job to show the previously formula and the modified formula including the hit quality formula as features that look quite simple such as Best T1 Ammo have to be modified to. I won’t be implementing the hit quality formula fully until after the next update has been released and tested.

Current plan is to add the hit quality as a true/false option but will have to see as the next update (rather than a fix like v6) will contain rigs. The rigs update is a huge one as its doubling the number of module variables. Then making sure the stacking penalty is correctly applied to them. Its purely a time issue of going through all the cells and updating them rather than difficulty, and my time is short until mid-June.

I leaked a early screenshot here which shows the modified layout to allow the setups your entering to have a chance of fitting onto the screen.

Rigs so far hooked up in that SS are DMG, ROF and Falloff with tracking and optimal having the stats there but not modified the formula's to use the stats yet.

Originally by: Jim McGregor

Looks like version b23v6 came up today. I was just going to play around with the spreadsheet and noticed the new version. Can be downloaded from here.

It looks different... will be fun to play around with.


Not sure how it looks different... I just fixed the Muninn and Maelstrom because someone pointed it out to me elsewhere since I rarely read E-O so didn't spot it mentioned here. The Muninn's bonus has been 25% instead of 50% since at least version b.17, Maelstrom must of changed from the early stats when I first add/updated the tier 3 BS and didn't notice/forgot to update it to 100m3. Embarassed

Kazuo Ishiguro
House of Marbles
Posted - 2007.05.17 18:45:00 - [127]
 

Edited by: Kazuo Ishiguro on 18/05/2007 08:42:45
I really should have checked this thread first Laughing

I tracked down a university PC today and modified b23v5 to include an option to use the variable hit quality formula. This involved reworking most of the 'best named t1' parts of the spreadsheet, but I'm reasonably confident that everything is now functioning correctly; all the lookups were left intact, and in fact 95% of the spreadsheet is unchanged. The only annoying problem was with half-absolute references to cells on the same sheet dotted all over the chance to hit formula in about 20,000 cells Evil or Very Mad

There was a spare cell under the 'miscellaneous' settings in the input-output sheet, so I've used that for setting the hit quality option. If used, it applies to any damage-based variable. It does not apply to damage from drones (which are typically very accurate and do not suffer greatly from low hit quality).

This is by no means an official release, but it'll do until the rigs update is put out. Please post here if you find any problems with it:

//edit: link removed in favour of Marn's newer version below, which follows the correct naming convention and includes a few things from b23v6 that weren't in v5.

Marn Prestoc
Minmatar
Maelstrom Crew
Paradigm Alliance
Posted - 2007.05.18 03:31:00 - [128]
 

Edited by: Marn Prestoc on 18/05/2007 03:32:37
I've had a quick look and test and looks to be correct so i'm happy to roll with it and let others just report any problems they might find.

I've renamed it v7, included the Muninn and Maelstrom corrections, edited the graph version number to v7 and the graph title to show if the hit quality formula is being used or not.

beta_23v7

Be aware that the link provided by Kazuo only renamed the zip file. The excel file inside it and the name displayed on the graph still says v5 so any screenshots taken of the graph could be mistaken as a previous version with the old errors.

Many thanks to Kazuo for adding his formula to the spreadsheet.

Darth Patches


ISD YARR
Interstellar Services Department
Posted - 2007.05.19 18:41:00 - [129]
 

After much requesting, and after personally checking this spreadsheet for consistency I am proud to announce. Stickification is complete.

-Darth Patches

Naughty Boy
Chronics of ordinary hate
Posted - 2007.05.20 21:28:00 - [130]
 

Thanks for the sticky, and thanks for the hard work Kazuo and Marn. The link in the first post and the thread title are both updated. Very Happy

NB.

Jim McGregor
Posted - 2007.05.26 07:10:00 - [131]
 


Have you guys considered adding faction ships? I know I would love to compare those to eachother.

Kazuo Ishiguro
House of Marbles
Posted - 2007.05.27 22:33:00 - [132]
 

Edited by: Kazuo Ishiguro on 28/05/2007 21:03:22
The spreadsheet doesn't use named ranges, so this involved doing a find & replace job over the entire workbook- in openoffice- to make everything work with the expanded ship table. I think it was using almost 2GB of ram at one point (as opposed to the usual 1GB under normal circumstances, and 400MB in Excel).
Shockedugh

Still, it seems to be working now, so do have fun with it:
http://dl.eve-files.com/media/corp/KzIg/beta23v7a_kali.zip

I hope I've made the appropriate choices where ships have more than one possible turret/launcher combination. If not, it's easy enough to change them on the ships sheet.

Some of the drone bays might be a bit out on the state issue battleships, but they can all still field 5 heavy drones, so it doesn't affect the dps calculations. Let me know if there are any problems - Marn is on holiday from EVE at the moment.




//edit: It appears that after I performed open heart surgery on the spreadsheet to enable it to use the hit quality formula, it developed a minor complication: when using 'best t1' ammo for turrets, figures for all turret ships were being calculated using the damage/tracking/range mods entered for the first ship in the list, rather than their respective modules.

Marn has received the updated version with faction ships and, due to a combination of injury and bad weather, has spent his bank holiday working on the rigs update, which will correct the above error. Accordingly, I'll wait for him to upload it before I think about doing anything else.

Kazuo Ishiguro
House of Marbles
Posted - 2007.05.29 09:41:00 - [133]
 

Marn is currently unable to post here, so here's the latest update from SHC:

Originally by: Marn Prestoc
beta24

Going to follow the traditional naming as I feel this update is a lot more than just updating the database stats.

Through the various 23v1 -> 23v7 versions during the REV expansion the stats have been added and updated as per patches. Finally, the last addition from REV has been completed by adding RIGS to the sheet, ending the 23v naming as it was orginally only meant to be while REV was in testing as changes to stats were happening weekly. So I give you beta24. The following is a general summary from the last NB version beta23 to this new version.

Whats changed from 23 to 24?
- Rev ships added (tier 2 BC and tier 3 BS)
- Added Heavy Assault Missiles + Ammo
- Nighthawk Bonus changed
- T2 ammo updated
- AC's updated.
- Arty updated.
- Muninn optimal bonus corrected.
- Hit Quality Formula implemented by Kazuo Ishiguro. Can be activated/deactivated through cell N10 on the INPUT OUTPUT sheet.
- Faction frigates, cruisers and battleships added by Kazuo Ishiguro.
- RIGS ADDED.
- Updated Layout to accommodate increased ship setup options due to rigs.

Please report bugs either at SHC Topic or Eve-O Topic. Both are stickied so easy for me to see. Every page has been modified so I would be very suprised if there are no bugs at all in either the formula's, stacking penalties or database.

Naughty Boy
Chronics of ordinary hate
Posted - 2007.05.31 18:18:00 - [134]
 

Thanks a lot Kazuo and Marn. First post and thread title are both updated.

NB.

Baudolino
Gallente
Royal Crimson Lancers
Posted - 2007.06.05 14:08:00 - [135]
 

Edited by: Baudolino on 05/06/2007 14:12:08
question solved



Marn Prestoc
Minmatar
Maelstrom Crew
Paradigm Alliance
Posted - 2007.06.20 16:00:00 - [136]
 

Edited by: Marn Prestoc on 20/06/2007 16:04:37
Rev 2 Update + Fix

beta25

Updated Pulse Laser tracking by 25%.
Added t2 Sentries
Fixed major error in ROF lookup referencing.


To do: Add stealth bomber changes. (this was a quick fix for the error and threw in the other quick changes) Will probably throw motherships in to and add the hit quality formula to drones thanks to Kazuo Ishiguro's testing.

J'Mkarr Soban
Posted - 2007.06.22 19:10:00 - [137]
 

Edited by: J''Mkarr Soban on 22/06/2007 19:20:20
How exactly are skills used in this? What's assumed? Because I'm getting different ranges between QuickFit and this spreadsheet, and I don't know which to believe.

Naughty Boy
Chronics of ordinary hate
Posted - 2007.06.23 08:59:00 - [138]
 

Edited by: Naughty Boy on 23/06/2007 08:58:05
Originally by: J'Mkarr Soban
How exactly are skills used in this? What's assumed? Because I'm getting different ranges between QuickFit and this spreadsheet, and I don't know which to believe.
The spreadsheet computes everything with maxed skills. If you see something that is not consistent with the values you are observing in game, you can report it here.

NB.

J'Mkarr Soban
Posted - 2007.06.23 13:03:00 - [139]
 

Originally by: Naughty Boy
Edited by: Naughty Boy on 23/06/2007 08:58:05
Originally by: J'Mkarr Soban
How exactly are skills used in this? What's assumed? Because I'm getting different ranges between QuickFit and this spreadsheet, and I don't know which to believe.
The spreadsheet computes everything with maxed skills. If you see something that is not consistent with the values you are observing in game, you can report it here.

NB.


Ah, it's ok, I found it. Best named T1 gear is far above the baseline listed. It's ok, I adjusted the values manually and get the correct values now.

Jim McGregor
Posted - 2007.06.24 10:21:00 - [140]
 


b25 link doesnt work currently.

Marn Prestoc
Minmatar
Maelstrom Crew
Paradigm Alliance
Posted - 2007.06.25 11:56:00 - [141]
 

Works for me, eve-files must of had a hic-up.

Skarned
Inroads
Posted - 2007.07.19 19:39:00 - [142]
 

Edited by: Skarned on 20/07/2007 02:54:45
EDIT: I'm an idiot, figured out why I was seeing what I thought I was seeing.

inflCMNDR VoO
Posted - 2007.07.23 20:23:00 - [143]
 

Edited by: inflCMNDR VoO on 23/07/2007 20:28:23

Nambu
Caldari
Nova Prospekt Initiative
Posted - 2007.07.24 20:47:00 - [144]
 

Dont know if people are still updating this, but the missile damage for ships that have specific damage type buffs is not calculating correctly (ie Carc). The data on the ship tab seems correct, but in the <Launchers> tab the damage calculations are all the same instead, of the bonus that kinetic missiles get.

If I took it apart right, it looks like the vlookup against the Ship table is off. The column stays at col 25 for each damage type rather than iterating to the other damage bonus columns. This is normal Excel behaviour as it the column specifier in vlookup is not a refrence so it dosen't update as you copy across columns. So in summary all missile damage is being calculated as if it were EM damage, which is fine for most ships, but the Carc, Condor, and Kessie (others?).

Marn Prestoc
Minmatar
Maelstrom Crew
Paradigm Alliance
Posted - 2007.07.24 22:07:00 - [145]
 

Thanks, I noticed this error while doing the version 26 update so is fixed in that version, which should be released with a week.

Salvator Dali
Dracula Order
Bang Bang You're Dead
Posted - 2007.09.14 12:23:00 - [146]
 

Interesting

Marn Prestoc
Minmatar
Maelstrom Crew
Paradigm Alliance
Posted - 2007.10.16 20:24:00 - [147]
 

beta27

- Boosted the Curator sentry drone, max Range, falloff and tracking improved.
- Added 20% tracking to Tech II sentry drones.
- The flight time on Assault Missiles has been increased.

HAM missiles flight time increase:
t1: 3 sec -> 4 sec
jav: 8 sec -> 10.5 sec
rage: 4 sec -> 5 sec

- Increased turrets of Sac Mk2 to 1 from 0 as they were given turret hardpoints back (with 6 highs can only fit 1 turret with 5 launcher hence 1 turret set).

- Added check for specific missile bonuses:
Sac only gets bonus to heavy assault
Stealth bombers to only cruise
Malediction and Vengeance to only rockets.

Think thats everything upto date for TQ stuff. Will do a sisi version when I get time and I get screenshots of the stuff on there.

<3 Chribba.

Kazuo Ishiguro
House of Marbles
Posted - 2007.10.16 22:48:00 - [148]
 

Looks good - I'll see if I can put in hit quality for sentry drones at some point without breaking anything (I'll be using Excel, don't worry Wink). It would also be possible to add it in for normal drones, assuming that they orbit at their normal orbit speeds and optimal ranges, but it's only really noticeble for sentries.

Ath Amon
Posted - 2007.12.08 09:18:00 - [149]
 

firsto lot of love to NB, Marn and Kazuo

ther's any version around with trinity stats?

i tried the 27trinity on evefiles but seem quite buggy (sig radiust can't be changed and some data seem not right)

thx :)

Kazuo Ishiguro
House of Marbles
Posted - 2008.02.04 23:47:00 - [150]
 

New release!
http://go-dl.eve-files.com/media/corp/KzIg/beta28bTrinity.zip

It's been a while since the last one, so here's what's happened over the intervening versions:

Originally by: Marn Prestoc on SHC
27bTrinity

- Interceptors changed:
Removed bonus taken by the WD range (Malediction, Raptor, Ares and Stilleto).

- Torps Mk2 added:
Changes to launcher, t1, rage and Javelin torps.

- Changed order of ships listed to largest to smallest then faction largest to smallest.

-Added T2 ships (with temporary hull name for quick reference):
Electronic Attack Ships: None have damage bonus but there are hardpoints and drone bays + bandwidth.
Heavy Interdictors
Black Ops (indicated by BO)
Marauders (indicated by M)


IMPORTANT:
- Drone Bandwidth, I haven't read anymore info but saw the stats in the dev blog so I have assumed that 5Mbit/s bandwidth = 5m3. So the new Amarr EW frig has max 20Mbit/s out of 60m3. So in the spreadsheet i've used 20m3 as the spreadsheet doesn't need to know about spare drones.

- EW bonuses have never been included, so obviously haven't for the new ships.

- Only used info from dev blog, haven't checked SISI for bandwidth info or any other ship changes except what i've stated.

p.s. Made the graph auto the axis scales again which KzIg had disabled in 27b.


Originally by: Kazuo Ishiguro on SHC
beta28bTrinity

* Updated to test server ship stats for the most recent patch
* Added scripts for tracking computers & tracking disruptors, including the new falloff effect of the the optimal range script.
* Fixed target painters & sig radius calculation in general
* Fixed a bug in flight time calculation for T1 torps


At the moment there is no option to use TC/TDs without scripts, as in almost all cases it's better to use a script - at close range, tracking/falloff is crucial and optimal is irrelevant, and vice versa at long range. Also, it'd be a *lot* more work to add that option.

This was quite a major change, so I'd be grateful if people would let me know of any bugs they find.


Pages: 1 2 3 4 [5] 6

This thread is older than 90 days and has been locked due to inactivity.


 


The new forums are live

Please adjust your bookmarks to https://forums.eveonline.com

These forums are archived and read-only