Interesting CMWW Gear Set (Low APS)

  • #1
    Yesteday saw an old vidposted of some guy killing ghom on MP10 with an effective dps multuplier of around 8 using CMWW build with SA. I'm not sure if he was using diamond shards are not. However, at first I thought that wasn't too big of a deal because there's a couple wizards around the offical forums with similar effective dps multipliers, but then I saw his attack speed: 1.89 or so. I have to say I was very surprised. That, combined with taking the afternoon off, made me curious to try something like that.

    I looked over the gear, and noticed he had 30 APoC, which makes sense because at that low of an attack speed you will need higher APoC to keep WW going. After running a few simulations, I aimed for the 2.00001 APS breakpoint, because it resulted in a huge jump in dps over the one below. The simulation results also indicate higher breakpoints migh result in lower effective dps for various reasons.

    After a bit of time on the AH, I added a couple pieces to my existing Archon gear and ended up with the gear set shown in my profile:
    143k dps unbuffed but with scoundrel
    2.002 APS (26 tics per WW cast)
    41% crit (yes, that's WITH scoundrel)
    1500 LoH (thanks to 600 on wand)
    39k HP
    280k EHP unbuffed

    I ran against Ghom a few times on MP6 and recorded the fights for a few different skill variations. In the end I went with Time Warp for my extra skill and was able to get a 33-34s kill, and he has 36 mil hp, for an effective dps of over 1 mil. That results in an effective multiplier of around 7x, which is about as high as I was ever able to get using a 2.73 APS gear set.

    All isn't perfect with the low APS gear set though. While the effetive dps is quite high for such low APS gear, the survivability is pretty low. MP6 Ghom almost killed me a couple times with the build, though switching from shards to shell on DS mostly elimitated that. The main issue is that even with 1500 LoH my life gains are still lower than I had at 2.73 APS with 1k LoH. I only gain about 5x my LoH per second compared to the 8x expected at 2.5 APS or 9x at 2.73. The benefit of the low APS is that Magic Weapon - Blood Magic becomes very good compared to LoH. For my set up it seems BM is worth about 600 LoH. That means a Life Steal wand with APoC would be great for the setup, but I couldn't find a good price on one with 1000-1100 dps.
    My EHP and DPS Spreadsheet, mostly useful for wizards.
    My wizard, mostly useful for killing demons and collecting loot.
    Wizard CM DPS Simulator, written in Matlab, release version 1.01
    Comprehensive CMWW Guide, including gear and build suggestions, plus Uber strats
    Arcane Mines Guide, detailed guide for the highest dps wizard build, including vids
  • #2
    In the many many discussions people have had about the multiplier in the past the trend of higher multipliers with lower APS seemed to immerge consistantly.

    One thing I keep forgetting to ask is if your simulator shows the same odd behavior regarding crit damage's affect on the dps multiplier...in that even though it increases tooltip dmg, the multiplier is then increased as well.
  • #3
    Quote from BDF2000

    In the many many discussions people have had about the multiplier in the past the trend of higher multipliers with lower APS seemed to immerge consistantly.

    One thing I keep forgetting to ask is if your simulator shows the same odd behavior regarding crit damage's affect on the dps multiplier...in that even though it increases tooltip dmg, the multiplier is then increased as well.


    It doesn't show any CD correlation to multiplier because mathematically it is not included in the simulation. My only thoughts about that are what you're really seeing is a difference in dps based on fight duration, and that's what is changing the multiplier, not the CD, well that or just varience in timed runs.

    CD is already included into your char sheet dps and the multiplier is meant to directly multply your char sheet dps. For example, two people with the same APS, Crit, APoC, latency and skills used should have the same multiplier, even if one only has 10k dps and the other has 100k dps. If one has higher CD, that shouldn't change how many WWs they cast, how much AP gains they get, or anything else about the build itself.

    In short, my opinion is there's a combination of varience from run to run and fight duration that is changing eDPS multiplier.
    My EHP and DPS Spreadsheet, mostly useful for wizards.
    My wizard, mostly useful for killing demons and collecting loot.
    Wizard CM DPS Simulator, written in Matlab, release version 1.01
    Comprehensive CMWW Guide, including gear and build suggestions, plus Uber strats
    Arcane Mines Guide, detailed guide for the highest dps wizard build, including vids
  • #4
    imagine 30 apoc and 3 aps...
    even higher multiplyer.
    but you have to trade in paperr dps from mempo
  • #5
    I don't do a lot of theorycrafting, so maybe I'm a little confused. The "multiplier" you are talking about is how much higher your effective DPS is over character sheet DPS, right? I was under the impression that the damage Energy Twister (and any rune variation) was not affected by attack speed, only the number of ticks and thus crits/LoH.

    So that would mean that boosting attack speed would greatly increase your character sheet dps, but would have no effect on the damage done by WW, creating a smaller multiplier. The opposite is true, when you reduce your attack speed, your character sheet dps goes down, but WW damage stays the same, resulting in a higher multiplier.
  • #6
    Quote from Zeyn

    I don't do a lot of theorycrafting, so maybe I'm a little confused. The "multiplier" you are talking about is how much higher your effective DPS is over character sheet DPS, right? I was under the impression that the damage Energy Twister (and any rune variation) was not affected by attack speed, only the number of ticks and thus crits/LoH.

    So that would mean that boosting attack speed would greatly increase your character sheet dps, but would have no effect on the damage done by WW, creating a smaller multiplier. The opposite is true, when you reduce your attack speed, your character sheet dps goes down, but WW damage stays the same, resulting in a higher multiplier.


    Effective dps is how much damage you actually do. If you record a fight and you do 1 mil dps over 100 seconds, your effective dps is 10k, or 1mil / 100. When I say effective multiplier, that's what you multiply your char sheet dps by to get your effective dps. To continue the example, let's say you have a char sheet dps of 2k dps. Then your effective multiplier is 10k/2k = 5x.

    For WW, you're right in that APS does not change how much damage a WW does. However, it does increase the number of tics the WW performs, which creates more CM, APoC, and SA procs. SA is a huge component of your damage done. More APoC means you can cast more EB while maintaining your WWs. More CM procs means you have a reduced CD on EB and DS so you can cast each more often. That's the gist of why attack speed is useful for CMWW build.

    However, the increase on those values is saturated a bit by the extra AP cost of casting WWs faster and latency, so there can be times where increasing your attack speed results in lower effective dps multiplier, like you mentioned.
    My EHP and DPS Spreadsheet, mostly useful for wizards.
    My wizard, mostly useful for killing demons and collecting loot.
    Wizard CM DPS Simulator, written in Matlab, release version 1.01
    Comprehensive CMWW Guide, including gear and build suggestions, plus Uber strats
    Arcane Mines Guide, detailed guide for the highest dps wizard build, including vids
  • #7
    Quote from BDF2000

    In the many many discussions people have had about the multiplier in the past the trend of higher multipliers with lower APS seemed to immerge consistantly.

    One thing I keep forgetting to ask is if your simulator shows the same odd behavior regarding crit damage's affect on the dps multiplier...in that even though it increases tooltip dmg, the multiplier is then increased as well.


    I had a few minutes to play around with some Ghom runs and decided to put that whole "remove your CD gem" to the test to compare. I did 3 test with the CD gem and 3 without, recording the entire fight. I removed all gear from my scoundrel but I still used him because I am very low on crit%. I'm at 282 CD with gem, 182 without, for reference.

    With CD gem: char sheet DPS = 156739
    Fight durations: 39, 39, 39
    Ghom HP on MP6: 36033267
    Effective multiplier: 5.895x

    Without CD gem: char sheet DPS = 126935
    Fight durations: 50, 48, 48
    Effective multiplier: 5.677, 5.914, 5.914, average of 5.835x

    Also, I looked at how much damage I did for the first 39s of the fight and came up with the following:
    eDPS = 734835, 733555, 734104
    Effective multiplier for first 39s: 5.784x

    My letency was around 120 for all cases, as far as I could tell. The results are that the multipliers seem fairly independent of crit damage. It might just be that SA has less varience at the lower APS though, so who knows. I really wish I had the time to test SA with WW a bit more but I'm just going to be too busy for the next several months.
    My EHP and DPS Spreadsheet, mostly useful for wizards.
    My wizard, mostly useful for killing demons and collecting loot.
    Wizard CM DPS Simulator, written in Matlab, release version 1.01
    Comprehensive CMWW Guide, including gear and build suggestions, plus Uber strats
    Arcane Mines Guide, detailed guide for the highest dps wizard build, including vids
  • #8
    Quote from Jalatiphra

    imagine 30 apoc and 3 aps...
    even higher multiplyer.
    but you have to trade in paperr dps from mempo


    There are some arguments about higher APoC at higher APS. Some say they need it, because if you use teleport wormhole to move around fast, you start fights with around half your AP gone, so you need the higher APoC for the start of the encounters. If you start with near max AP, you will likely see very little benefit from having 30APoC at 2.73 or 3.0 APS over 20 APoC because your AP gains are already saturated. You can only spend AP as fast as you can cast WWs and refresh EB, so any excess AP is wasted.
    My EHP and DPS Spreadsheet, mostly useful for wizards.
    My wizard, mostly useful for killing demons and collecting loot.
    Wizard CM DPS Simulator, written in Matlab, release version 1.01
    Comprehensive CMWW Guide, including gear and build suggestions, plus Uber strats
    Arcane Mines Guide, detailed guide for the highest dps wizard build, including vids
  • #9
    I do the same thing to calculate my true sustained average DPS. I do MP10 Ghom. With a full offensive set up (bone chill, shards, shocking aspect, stretch time to hit 3aps) it took me 2mins 19 seconds. At 138mil hp at mp10 ghom, that works out to 996k DPS. My character stat sheet shows 148k dps with this set up so a multiplier of 6.73, not quite as high some others.

    On a related note, using this method I've found that shards adds about 12% damage for me. Bone chill adds the 15% listed.
  • #10
    What exactly are you trying to show with this topic? High APS is mainly used so you can a ) stunlock (= no gaps for abilities) single targets effectively and b ) use less APoC and thus more DPS oriented gear.

    As for pure DPS:
    (Assuming similar character screen DPS in all cases)

    Shocking Aspect scales with the square of your APS (rounded to certain breakpoints with linear scaling in between). If you're wondering why, just ask yourself what happens when you cast more WW with more ticks each and each tick has a fixed coefficient. I'm not going to post any equations here since if you don't understand this simple correlation, theorycrafting probably isn't for you.

    Energy Twister DPS is heavily tied to freeze uptime in multiple ways. Higher freeze uptime increases benefits from Bone Chill and Cold Blooded but also requires more WWs to be replaced by FN casts. Ideally you'd only want to cast as many FNs as required to lock enemies in place - if you're spamming FN, however, high AS will lead to high overlapping and an unnecessarily high amount of WW casts being replaced by overlapping FN casts. What does this all mean? High AS will at worst increase your freeze uptime at the expense of some DPS. If you're willing to give up some of the gained freeze reliability by using FN in certain intervals, however, you're actually gaining DPS using an APS setup since a lower proportion of your total spells will be spent on non-damaging FN casts (e.g. if you freeze once every 2 seconds, you'd give up 1/4 of your total casts at 2.0 attack speed but only 1/5 of your total casts at 2.5 attack speed).

    The last thing to consider is Diamond Shards/Explosive Blast. While these abilities' frequencies scale with APS through CM, they do not do so in a 1:1 correlation. This is mostly a cause of delays being a bottle neck when it comes to their frequencies (e.g. it doesn't matter quite as much whether you reset EB within 0.05 or 0.1 second when it's 0.2 seconds between a spell being ready to cast and the actual cast command arriving at the server - you'd merely gain a frequency increase of 20%). Thus this is the only case where you're clearly having a higher EDPS multiplier when running a low APS setup.

    In the end a low APS setup only really comes ahead when you're not running Shocking Aspect - and in this case you're already running an inefficient overall gear setup if you want to maximize your DPS without dying. You're much better off using Shocking Aspect in combination with a high APS/EHP setup since you'll have similar DPS to a Prismatic Armor/low APS setup but much better freeze uptime.
  • #11
    Quote from GT4

    What exactly are you trying to show with this topic? High APS is mainly used so you can a ) stunlock (= no gaps for abilities) single targets effectively and b ) use less APoC and thus more DPS oriented gear.


    What I'm trying to show is that contrary to popular believe, you can still get a very high effective dps using the build with a low dps. I have recorded 7x my char sheet dps against a single target with only 2.002 APS. I could barely get that when I had 2.73 APS using the same build. I just found the results very interesting and wanted to share. Besides, with the low APS I can still use DPS oriented weapon (black open socket), helm (SC with crit is decent dps with black weapon), and OH (Triumvirate).

    I know all about how the build works at higher APS. I made a simulator that models it fairly well. For Ubers, you obviously can't beat higher APS but for key or loot farming, lower APS can work reasonably well. Survival is a little lower than higher APS because of the lower LoH returns, but it still works for moderate MP levels. I still think 2.5 APS is the best balance between high dps options and high attack speed for survivability, but I wanted to see how well I could get it to work at a lower breakpoint.

    Also, Shocking Aspect still works pretty well. The armor blast hits for quite a bit because of the higher dps and lower APS. It doesn't quite scale as well as higher APS against multiple targets because of the fewer WW tics, but then again the SA model for WW is really odd and I saw an increase in the relative SA procs at lower APS than high APS, so it might actually even out a bit. I really wish the SA/WW interaction were a lot more straightforward.
    My EHP and DPS Spreadsheet, mostly useful for wizards.
    My wizard, mostly useful for killing demons and collecting loot.
    Wizard CM DPS Simulator, written in Matlab, release version 1.01
    Comprehensive CMWW Guide, including gear and build suggestions, plus Uber strats
    Arcane Mines Guide, detailed guide for the highest dps wizard build, including vids
  • To post a comment, please or register a new account.
Posts Quoted:
Reply
Clear All Quotes