It's very similar to mine. A couple issues worth noting with using it or my simulator are CM and SA procs. We really don't know the model for CM and SA procs with WW so correction factors are used instead. In his simulator it's defaulted to something like 2.5 for SA and 2.0 for CM, meaning SA procs 2.5x as often as you would expect and you get about 2x the CM procs as expected. Those values agree reasonably well with in game testing but it means your results could be off by a bit. Also the additional procs seem to depend somewhat on APS so a constant value might not work as well for a range of attack speeds.
EDIT: Also, it doesn't support IE, so I haven't used it much to compare my results to his.
^^ My first thought is that he used your file to create his simulator, but I saw the code (the source is on the website) and it was coded in Lisp, while yours was in MatLab. I don't think he ported your code.
I used/tested with chrome, successfully.
So, off-topic, I could port your simulator to JavaScript or C#, but you would have to explain the formulas to me, as my MatLab-fu is close to null.
GNU Octave is an open source Matlab alternative. I've had good luck with running unmodified Matlab code in Octave. Many linux distros should have octave in their repositories.
Matlab itself is pretty easy but there are few things like matrix operators that are semi-unique to these math scripting languages. Unless I'm mistaken, my biggest pet peeve is that Matlab uses 1-based indexing for arrays/matrices instead of zero-based for C, java, etc.
couple questions on this simulator... is there a way to get explosive blast without chain reaction? (lol should I be using chain reaction? I'm using time bomb)
Main question, how does this compare to DPS from other characters? I have sheet dps of 160 (unbuffed) and this simulation lists me closer to 700k but how would that compare to a hota smash barb? A certainly dont seem to be outperforming my 300k dps buddy
couple questions on this simulator... is there a way to get explosive blast without chain reaction? (lol should I be using chain reaction? I'm using time bomb)
I like chain reaction, because on my view it can proc more than other runes. But Loroese is the head theorycrafter about CM wizards, he can answer that better than me.
You can uncheck the option Chain Reaction on the simulator by clicking on it, but there's no alternative rune for explosive blast on the simulator.
Main question, how does this compare to DPS from other characters? I have sheet dps of 160 (unbuffed) and this simulation lists me closer to 700k but how would that compare to a hota smash barb? A certainly dont seem to be outperforming my 300k dps buddy
Probably your 300k dps buddy has more effective DPS than you... The best test I've seem on the forums so far is to do timed kill on Azmodam or Ghom.
Do one run with your CM wizard and another run with you buddy on the same MP level, and compare the time. Than you can see who is performing best solo over time.
But remember one thing: CM wizards do stand-out on a crowd. If you have 20+ enemies, I think nothing can beat a CM wizard.
I've used that calc, its good except for one thing - with 18apoc and 51 crit chance, it says I should be slowly losing arcane as I'm getting back slightly less then Im using on average. This doesn't seem to be the case as I dont' run out of arcane.
Regarding similar dps numbers among other classes and true damage. I do a lot of timed kills against mp10 ghom to get true damage. To match a 300dps HOTA barb in damage output you need to be at around 300kdps stat sheet and run SNS set up. Of course thesea re rough estimates and depends on that 300kdps barb whether thats unbuffed or not.
Regarding similar dps numbers among other classes and true damage. I do a lot of timed kills against mp10 ghom to get true damage. To match a 300dps HOTA barb in damage output you need to be at around 300kdps stat sheet and run SNS set up. Of course thesea re rough estimates and depends on that 300kdps barb whether thats unbuffed or not.
My WW barb does like 1.5-1.6m DPS against Ghom. With 92k DPS unbuffed (without passives). Replacing my SoJ with my hellfire ring obviously hurts the multiplier, because it makes both my effective DPS fall and my unbuffed DPS rise. But in any case, even if you include passives, it's a multiplier of at least 10, up to 17 or something like that.
And HotA should be even better against single targets.
Granted, CM has crowd control. But in terms of raw damage, it doesn't come close to a barb.
Not sure where you get 17 multiplier from. My 316k stat sheet unbuffed barb (with soj) can down MP10 ghom in 43 seconds for 3.22mil true dps. That's a multiplier of around 10. SNS wizards get around a 7-8 true multiplier.
couple questions on this simulator... is there a way to get explosive blast without chain reaction? (lol should I be using chain reaction? I'm using time bomb)
Time bomb is fine. I recommend it for Ubers but I prefer CR for solo farming because I tend to kill trash mobs fast so time bomb can be wasted frequently.
If you want to know how time bomb compares, just multiply the resulting CR damage by 315/(3*97). The proc rate is the same overall for both so it should even out. The main differences are the damage done and the delay, but in a long fight the longer delay isn't as much of an issue.
As far as porting my code to another language, you're more then welcome to if you want, and I'd be happy to help explain it all. Mostly all I have in my code is a bunch of matrix notation. The math is pretty simple and I just track everything in matrices. I haven't used anything besides MATLAB for years or I would have done it myself. There is an "import to C" option in matlab, so maybe I could take a look at that sometime and see what it does. I don't think I have a C++ compiler though. Like I said, it's been ages since I used C/C++.
Not sure where you get 17 multiplier from. My 316k stat sheet unbuffed barb (with soj) can down MP10 ghom in 43 seconds for 3.22mil true dps. That's a multiplier of around 10. SNS wizards get around a 7-8 true multiplier.
I just dug up a recent video from the recycle bin. With Scoundrel (Buriza, no damage, just freeze and CC bonus) I took 80 seconds for MP10 Ghom. That's 138425408 / 80 = 1730317 DPS
My DPS without passives, with follower: 102164 *
With passives, with follower: 138772
Depending on your definition of "unbuffed" (one could argue that passives are buffs - after all, Cold Blooded also counts towards the multiplier, not unbuffed DPS), that leaves me with a multiplier of either
1730317 / 102164 = 16.94
or
1730317 / 138772 = 12.47
Again, that's regular WW farming spec with Overpower/Killing Spree. Nothing I don't use in regular gameplay.
And last time I tried, switching to a HotA spec gave me a considerable bump in (single target) effective DPS, despite me having to swap out my generic 30% SoJ for a 23% HotA SoJ.
So I think it's a fair assumption that if you really try, you can achieve even higher multipliers.
* I said 92k earlier, but that was the diabloprogress value. Which I believe doesn't count paragon levels either. The values above are taken from ssdrawr's DPS calc.
Diabloprogress does count paragon levels but it doesn't count buffs or follower. It also sometimes has to be updated a couple times to properly update your paragon level.
I think it's generally accepted that wizards have the lowest dps multiplier out of the classes, or at least lower than monks and barbs. The SNS build has the highest multiplier of all wizards but we lack the extra buffs that other classes can easily stack.
Diabloprogress does count paragon levels but it doesn't count buffs or follower. It also sometimes has to be updated a couple times to properly update your paragon level.
I think it's generally accepted that wizards have the lowest dps multiplier out of the classes, or at least lower than monks and barbs. The SNS build has the highest multiplier of all wizards but we lack the extra buffs that other classes can easily stack.
Thanks, you were right. diabloprogress was stuck at plvl 15, while I'm actually 52. Thus the difference.
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
I'd like to share this calculator with you guys, found it last week, and I'm pretty happy with the results so far:
http://d3cmww.com/cmww
Disclaimer: I did not make the calculator nor I know who did. I just found the link and I'm passing it by.
I'm only posting because I saw no mention of it or a similar tool here on our forums or on battle.net forums.
Loroese's CMWW calculator only works on MatLab, and since I don't own a copy of MatLab I had to search for something online.
Hope this helps someone as helped me.
Wizard | Demon Hunter
EDIT: Also, it doesn't support IE, so I haven't used it much to compare my results to his.
Crusader DPS and EHP Spreadsheet, meant for Crusaders
My Wizard
I used/tested with chrome, successfully.
So, off-topic, I could port your simulator to JavaScript or C#, but you would have to explain the formulas to me, as my MatLab-fu is close to null.
Wizard | Demon Hunter
Matlab itself is pretty easy but there are few things like matrix operators that are semi-unique to these math scripting languages. Unless I'm mistaken, my biggest pet peeve is that Matlab uses 1-based indexing for arrays/matrices instead of zero-based for C, java, etc.
Main question, how does this compare to DPS from other characters? I have sheet dps of 160 (unbuffed) and this simulation lists me closer to 700k but how would that compare to a hota smash barb? A certainly dont seem to be outperforming my 300k dps buddy
You can uncheck the option Chain Reaction on the simulator by clicking on it, but there's no alternative rune for explosive blast on the simulator.
Probably your 300k dps buddy has more effective DPS than you... The best test I've seem on the forums so far is to do timed kill on Azmodam or Ghom.
Do one run with your CM wizard and another run with you buddy on the same MP level, and compare the time. Than you can see who is performing best solo over time.
But remember one thing: CM wizards do stand-out on a crowd. If you have 20+ enemies, I think nothing can beat a CM wizard.
Wizard | Demon Hunter
Regarding similar dps numbers among other classes and true damage. I do a lot of timed kills against mp10 ghom to get true damage. To match a 300dps HOTA barb in damage output you need to be at around 300kdps stat sheet and run SNS set up. Of course thesea re rough estimates and depends on that 300kdps barb whether thats unbuffed or not.
And HotA should be even better against single targets.
Granted, CM has crowd control. But in terms of raw damage, it doesn't come close to a barb.
Time bomb is fine. I recommend it for Ubers but I prefer CR for solo farming because I tend to kill trash mobs fast so time bomb can be wasted frequently.
If you want to know how time bomb compares, just multiply the resulting CR damage by 315/(3*97). The proc rate is the same overall for both so it should even out. The main differences are the damage done and the delay, but in a long fight the longer delay isn't as much of an issue.
As far as porting my code to another language, you're more then welcome to if you want, and I'd be happy to help explain it all. Mostly all I have in my code is a bunch of matrix notation. The math is pretty simple and I just track everything in matrices. I haven't used anything besides MATLAB for years or I would have done it myself. There is an "import to C" option in matlab, so maybe I could take a look at that sometime and see what it does. I don't think I have a C++ compiler though. Like I said, it's been ages since I used C/C++.
Crusader DPS and EHP Spreadsheet, meant for Crusaders
My Wizard
My DPS without passives, with follower: 102164 *
With passives, with follower: 138772
Depending on your definition of "unbuffed" (one could argue that passives are buffs - after all, Cold Blooded also counts towards the multiplier, not unbuffed DPS), that leaves me with a multiplier of either
1730317 / 102164 = 16.94
or
1730317 / 138772 = 12.47
Again, that's regular WW farming spec with Overpower/Killing Spree. Nothing I don't use in regular gameplay.
And last time I tried, switching to a HotA spec gave me a considerable bump in (single target) effective DPS, despite me having to swap out my generic 30% SoJ for a 23% HotA SoJ.
So I think it's a fair assumption that if you really try, you can achieve even higher multipliers.
* I said 92k earlier, but that was the diabloprogress value. Which I believe doesn't count paragon levels either. The values above are taken from ssdrawr's DPS calc.
I think it's generally accepted that wizards have the lowest dps multiplier out of the classes, or at least lower than monks and barbs. The SNS build has the highest multiplier of all wizards but we lack the extra buffs that other classes can easily stack.
Crusader DPS and EHP Spreadsheet, meant for Crusaders
My Wizard