EVE Search - Mathmaticaly Computed Tanks

All Channels

Ships and Modules

Mathmaticaly Computed Tanks

» Click here to find additional results for this topic using Google

Monitor this thread via RSS [?]

Pages: [1] :: one page
Author	Thread Statistics \| Show CCP posts - 0 post(s)
Hegumen	Posted - 2009.02.17 07:30:00 - [1] Hello, I recently wrote for fun a program that given a ship will try every single combination of relevant items and calculate the resulting effect on the ship with the goal of finding the best tanks possible for a given set of parameters. In some instances this means the total number of combinations exceeded 40 billion. N.B. These tanks are not idiot proof. I have extensively tested these setups in EFT and as much as my skills and ISK will allow in game, and my math seems to be in all working order. So now I need your help to see if these tanks really are what they are supposed to be. My challenge to you is to create a tank for a Raven using all lvl V skills, only 1 low slot, meta level 5 or lower modules, and only t1 rigs. Also all active tanks MUST BE CAP STABLE. My computed setups are bellow, your setups must either tank more DPS (in the case of passive and active tanks) or have higher effective HP (in the case of buffer tanks). One more thing: In my calculations I use 2.4 as the multiplier for determining shield and capacitor recharge rate. NOT 2.5. This means my DPS will not be perfectly in line with EFT. You must do the math yourself. Formulae as follow: Shield Recharge Rate = (Shield HP/Shield Recharge Time) * 2.4 Capacitor RR = (Capacitor/Capacitor Recharge Time) * 2.4 Average Shield DPS = ((100 * Shield Recharge Rate) / (25 * (1 - Shield EM Resist) + 25 * (1 - Shield Exp Resist) + 25 * (1 - Shield Kin Resist) + 25 * (1 - Shield Ther Resist))) Effective Shield HP = (Shield / (((1 - Shield EM Resist) * 25 + (1 - Shield Exp Resist) * 25 + (1 - Shield Kin Resist) * 25 + (1 - Shield Ther Resist) * 25) / 100)) Passive Raven Tank: [Med} Large Shield Ext. II x3 Heat Dissipation Amp. II x1 Magnetic Scattering Amp. II x1 Shield Recharger II x1 [Low] Beta Reactor Control: Shield Power Relay I x1 [Rig] Core Defense Field Purger I x3 Shield HP: 19218.8 Effective Shield HP: 37,386 Shield Recharge Time: 620.16 Shield Recharge Rate: 74.376 Shield Average DPS: 144.683 Shield EM Resist: 0.46875 Shield Exp Resist: 0.5 Shield Kin Resist: 0.4 Shield Ther Resist: 0.575 Passive Buffer Tank: [Med} Large Shield Ext. II x4 Heat Dissipation Amp. II x1 Magnetic Scattering Amp. II x1 [Low] Power Diagnostic System II x1 [Rig] Core Defense Field Extender I x3 Shield HP: 35,930.7 Effective Shield HP: 69,896.5 Shield Recharge Time: 1715.63 Shield Recharge Rate: 50.2637 Shield Average DPS: 97.7774 Shield EM Resist: 0.46875 Shield Exp Resist: 0.5 Shield Kin Resist: 0.4 Shield Ther Resist: 0.575 Active Buffer Tank: [Med} Large Shield Ext. II x3 Invuln. Field II x3 [Low] Damage Control II x1 [Rig] Core Defense Field Extender I x3 Capacitor: 6640 Capacitor Recharge Time: 866.16 Capacitor Recharge Rate: 18.3984 Capacitor Drain: 9.6333333 Shield HP: 29,229.3 Effective Shield HP: 107,427 Shield Recharge Time: 1,875 Shield Recharge Rate: 37.4135 Shield Average DPS: 137.506 Shield EM Resist: 0.624709 Shield Exp Resist: 0.812354 Shield Kin Resist: 0.774825 Shield Ther Resist: 0.699767 Active Perma Tank: [Med} Cap Recharger II x1 Shield Booster Amp. II x2 Large Shield Booster II x1 Invuln. Field II x2 [Low] Power Diagnostic System II x1 [Rig] Capacitor Control Circuit I x3 Capacitor: 6,972 Capacitor Recharge Time: 389.373 Capacitor Recharge Rate: 42.9737 Capacitor Drain: 42.4 Shield HP: 9,843.75 Effective Shield HP: 26,237.7 Shield Recharge Time: 1,715.63 Shield Average DPS: 322.253 Shield EM Resist: 0.482515 Shield Exp Resist: 0.741258 Shield Kin Resist: 0.689509 Shield Ther Resist: 0.586012 Thanks for your time. Also I can answer questions and explain any math in more detail.
Terminus Vindictus	Posted - 2009.02.17 08:21:00 - [2] Originally by: Hegumen stuff... Also I can answer questions... What's the point of all this?
LegendInMyOwnMind	Posted - 2009.02.17 08:34:00 - [3] Nice work! I am curious, how long did it take to run the experiments?
LegendInMyOwnMind	Posted - 2009.02.17 08:36:00 - [4] Also, how much grid/cpu did you assume would be taken up by weapons. I would also be curious to see the results for CNR.
shat ghost Caldari	Posted - 2009.02.17 08:44:00 - [5] Originally by: LegendInMyOwnMind Nice work! I am curious, how long did it take to run the experiments? On his computer (read old) it took about 2 hours i think he was saying On my computer (amd 6400+ maxing out one core) it took about 45 seconds, using all lows it takes about 45 min takes up almost no ram
shat ghost Caldari	Posted - 2009.02.17 08:45:00 - [6] Originally by: LegendInMyOwnMind Also, how much grid/cpu did you assume would be taken up by weapons. I would also be curious to see the results for CNR. He does not have a lot of ships data in it but that will be the easy part, and he has a setting where you can tell it how much cpu and power grid is being taken up, so you do not need to worry about that if you are running with missiles.
shat ghost Caldari	Posted - 2009.02.17 08:47:00 - [7] Originally by: Terminus Vindictus Originally by: Hegumen stuff... Also I can answer questions... What's the point of all this? He has about 4 diffrent tanks up there, and it is to maxamise the tank with the spaces and power/cpu requirements you give it.
Norwood Franskly Minmatar Fleet of the Damned	Posted - 2009.02.17 09:23:00 - [8] Originally by: Hegumen My challenge to you is to create a tank for a Raven using all lvl V skills, only 1 low slot, meta level 5 or lower modules, and only t1 rigs. Also all active tanks MUST BE CAP STABLE. My computed setups are bellow, your setups must either tank more DPS (in the case of passive and active tanks) or have higher effective HP (in the case of buffer tanks). [Raven, Silly tank] Damage Control II [empty low slot] [empty low slot] [empty low slot] [empty low slot] X-Large Shield Booster II Shield Boost Amplifier II Shield Boost Amplifier II Invulnerability Field II Invulnerability Field II Heavy Capacitor Booster II, Cap Booster 800 [empty high slot] [empty high slot] [empty high slot] [empty high slot] [empty high slot] [empty high slot] [empty high slot] [empty high slot] Core Defence Capacitor safeguard I Core Defence Capacitor safeguard I Core Defence Operational Solidifier I There you go cap stable 806 dps (1169 overheated) Cap injectors are much more efficient then cap rechargers. However Cap stability is meaningless, even in a mission you don't need to perma run a tank, so why bother. The second you get a neutralizer on you there goes your cap stability anyway. A better use for your program would be to create optimal pvp setups for various ships (meaning mandatory Damage control MWD + 3 damage mods with maximum no of weapons in high slots and largest EHP possible). ~~~ [ 2009.02.05 09:37:43 ] Louis Trenker > - Who's ship is this?- It's a Titan baby.- Who's Titan is this?- BoB's.- Who's BoB?- BoB's dead baby! BoB's dead!
Norwood Franskly Minmatar Fleet of the Damned	Posted - 2009.02.17 09:27:00 - [9] Edited by: Norwood Franskly on 17/02/2009 09:30:22 Heh missed this... [Raven, Extra silly tank] Damage Control II [empty low slot] [empty low slot] [empty low slot] [empty low slot] X-Large Shield Booster II X-Large Shield Booster II Shield Boost Amplifier II Shield Boost Amplifier II Heavy Capacitor Booster II, Cap Booster 800 Heavy Capacitor Booster II, Cap Booster 800 [empty high slot] [empty high slot] [empty high slot] [empty high slot] [empty high slot] [empty high slot] [empty high slot] [empty high slot] Core Defence Capacitor safeguard I Core Defence Capacitor safeguard I Core Defence Operational Solidifier I 814 cap stable If I could use higher meta level and fitting implants you could get a capital shield booster on there (I think). ~~~ [ 2009.02.05 09:37:43 ] Louis Trenker > - Who's ship is this?- It's a Titan baby.- Who's Titan is this?- BoB's.- Who's BoB?- BoB's dead baby! BoB's dead!
Hegumen	Posted - 2009.02.17 10:23:00 - [10] Thank you shat ghost for answering the above questions. I will expand upon them. Originally by: Terminus Vindictus Originally by: Hegumen stuff... Also I can answer questions... What's the point of all this? From my perspective this is a cool application of combinatorics, and no doubt many people will find the application useful in some way. I simply need people who are better at eve than I am to check my work to make sure my application does what I say it does before I decide to release it. Originally by: LegendInMyOwnMind I am curious, how long did it take to run the experiments? Varies from computer to computer and what's running in the background, like shat said it is very cpu intensive and requires little memory. There are filters in place that remove useless items for certain tanks (i.e. shield boosters from passive tanks). So passive/buffer tanks will take between a few seconds and few minutes depending on available slots and what not. As for active tanks I computed an active tank that took more than 10 hours once (on an AMD 1600+ 1.41 GHz). I'm seriously considering adding multi-core support, and potentially getting it to use some of the gpu as well. Originally by: LegendInMyOwnMind Also, how much grid/cpu did you assume would be taken up by weapons. I would also be curious to see the results for CNR. There are settings in the calculator that allow you to specify how much powergrid and cpu you are already using and it will create tanks that work around such limitations. as per request the CNR: Passive CNR Tank: [Med} Large Shield Extender II x2 Heat Dissipation Amp. II x1 Magnetic Scattering Amp. II x1 Shield Recharger II x2 [Low] Beta Reactor Control: Shield Power Relay I x1 [Rig] Core Defense Field Purger I x3 Shield HP: 20,625 Effective Shield HP: 40,121.6 Shield Recharge Time: 527.136 Shield Recharge Rate: 93.9037 Shield Average DPS: 182.67 Shield EM Resist: 0.46875 Shield Exp Resist: 0.5 Shield Kin Resist: 0.4 Shield Ther Resist: 0.575 Note the drop of the third large shield extender for a second shield recharger. This is because the CNR has an already boosted shield HP but the same shield recharge time. Therefore the CNR can passive tank 26.3% better than a Raven given the current limitations. Active Buffer CNR Tank: [Med} Large Shield Ext. II x2 Heat Dissipation Field II x1 Invuln. Field II x2 Photon Scattering Field x1 [Low] Damage Control II x1 [Rig] Core Defense Field Extender I x3 Capacitor: 6640 Capacitor Recharge Time: 866.16 Capacitor Recharge Rate: 18.3984 Capacitor Drain: 10.4333333 Shield HP: 31,368 Effective Shield HP: 134,577 Shield Recharge Time: 1,875 Shield Recharge Rate: 40.1511 Shield Average DPS: 172.259 Shield EM Resist: 0.758741 Shield Exp Resist: 0.7736 Shield Kin Resist: 0.72832 Shield Ther Resist: 0.806993 Note again that because the CNR has an already boosted shield HP that a third shield extender is no long needed. Freeing up a slot. Due to the stacking penalty a 4th invuln field would be useless so one is dropped and two specific hardeners are added. Thus a active buffer tanked CRN is 25.3% better than the Raven, both in DPS and in effective shield HP. I will post an active tank for the CNR when I have nothing better to do with my comp for 8 hours.
Hegumen	Posted - 2009.02.17 10:28:00 - [11] Originally by: Norwood Franskly Cap injectors are much more efficient then cap rechargers. However Cap stability is meaningless, even in a mission you don't need to perma run a tank, so why bother. The second you get a neutralizer on you there goes your cap stability anyway. A better use for your program would be to create optimal pvp setups for various ships (meaning mandatory Damage control MWD + 3 damage mods with maximum no of weapons in high slots and largest EHP possible). Good points. The reason I asked for tanks to be cap stable was for control. I didnt want people to post tanks with all X-large shield boosters that ran for 22s. As for pvp my program can partialy be used for that, just plug in how much cpu and power grid you are already using and how many mid or low slots you have available and it will work around it. One of the reasons I specifically asked for only 1 low slot was to test this. As for cap injectors, while I had thought of them, I dismissed them because they relied on "ammo". But I shall add them to my program and see what new setups it creates.
Arous Drephius Perkone	Posted - 2009.02.17 11:10:00 - [12] Edited by: Arous Drephius on 17/02/2009 11:10:30 Multi-core support + CUDA = massive win. It'll speed it up massively. EDIT: What language did you write it in?
Hegumen	Posted - 2009.02.17 11:26:00 - [13] I wrote it in C++ as thats the language I am most familiar with. Tbh I am not entirely sure atm how to include multi-core support but I am learning.
Norwood Franskly Minmatar Fleet of the Damned	Posted - 2009.02.17 11:51:00 - [14] Originally by: Hegumen I wrote it in C++ as thats the language I am most familiar with. Tbh I am not entirely sure atm how to include multi-core support but I am learning. http://www.advancedlinuxprogramming.com/ Thats a really good textbook, someone at my work pointed it out to me the authors made it freely available online, there's a lot of stuff in there on how to write multi-threaded apps in C (at least for POSIX operating systems), iirc the stuff on socket programming and mmap in there is pretty good too if you ever need to learn that stuff. Not sure how useful it will be for windows programming, I haven't written anything for windows for a few years now, so if you have a nix box sitting somewhere give it a read. I think windows implements at least some of the posix api so may be useful on that platform too, if you are really interested in mathematical/scientific programming having at least one dedicated nix box is probably the way to go. I work in Chemical modeling and CFD and pretty much everything is *nix based. http://en.wikipedia.org/wiki/OpenMP Open MP is something I've heard a lot about for multi processing, never used it myself though. Anyway this is probably widely off topic for ships and modules forum but goodluck with your code. ~~~ [ 2009.02.05 09:37:43 ] Louis Trenker > - Who's ship is this?- It's a Titan baby.- Who's Titan is this?- BoB's.- Who's BoB?- BoB's dead baby! BoB's dead!
Hegumen	Posted - 2009.02.17 12:16:00 - [15] Thank you very much Norwood Franskly I shall check those out. Also here is the Active Tanked Raven my program spits out: Active Raven Tank With Cap Boosters: [Med} Heavy Capacitor Booster II x2 Shield Boost Amp. II x1 X-Large Shield Booster II x1 Invuln. Field II x2 [Low] Damage Control II x1 [Rig] Core Defense Operational Solidifier I x3 Capacitor: 6,640 Capacitor Recharge Time: 866.16 Capacitor Recharge Rate: 132.598 (included cap boosters) Capacitor Drain: 123.673 Shield HP: 9,375 Effective Shield HP: 28,558 Shield Recharge Time: 1,875 Shield Average DPS: 846.06 Shield EM Resist: 0.547201 Shield Exp Resist: 0.7736 Shield Kin Resist: 0.72832 Shield Ther Resist: 0.637761
Altris	Posted - 2009.02.17 15:33:00 - [16] Nice work. I'm not an optimization guy, but my $0.02 on the code side: Open MP and MPI are for distributed computing. Too tedious to use for shared memory multiprocessors like multi-cores. You could try Cilk (http://supertech.csail.mit.edu/cilk/) but will have to un-++ your C. Comet (http://www.comet-online.org/) would probably interest you. Threading is the most direct option, and you then get to experience the joys of debugging parallel programs. Cheers.
Cedric Diggory Perfunctory Oleaginous Laocoon Mugwumps	Posted - 2009.02.17 15:39:00 - [17] Finding holes in your math would be much easier if you furnished us with your program... --- Originally by: 7shining7one7 no.. you're f'ing insane.
Hegumen	Posted - 2009.02.18 01:14:00 - [18] bump for download
Cedric Diggory Perfunctory Oleaginous Laocoon Mugwumps	Posted - 2009.02.18 01:57:00 - [19] Link in the original post shows EFT? --- Originally by: 7shining7one7 no.. you're f'ing insane.
shat ghost Caldari	Posted - 2009.02.18 02:31:00 - [20] Edited by: shat ghost on 18/02/2009 02:31:23 http://depositfiles.com/en/files/rt3v9yjvj
Hegumen	Posted - 2009.02.18 03:24:00 - [21] Sorry my bad, the default forum link that deposit files gave me didn't link properly so I quoted the link for EFT as thats the first place that came to mind, changed the name but forgot to change the link because I was in a rush. It's updated now.

Pages: [1] :: one page
First page \| Previous page \| Next page \| Last page

COPYRIGHT NOTICE
EVE Online, the EVE logo, EVE and all associated logos and designs are the intellectual property of CCP hf. All artwork, screenshots, characters, vehicles, storylines, world facts or other recognizable features of the intellectual property relating to these trademarks are likewise the intellectual property of CCP hf. EVE Online and the EVE logo are the registered trademarks of CCP hf. All rights are reserved worldwide. All other trademarks are the property of their respective owners. CCP hf. has granted permission to EVE-Search.com to use EVE Online and all associated logos and designs for promotional and information purposes on its website but does not endorse, and is not in any way affiliated with, EVE-Search.com. CCP is in no way responsible for the content on or functioning of this website, nor can it be liable for any damage arising from the use of this website.