Request to Roserio - calculate DPS for units

Discussion in 'Battle Strategies' started by Arr, Dec 18, 2014.

  1. Arr

    Arr Captain

    Hey Roserio,

    You've been posting great videos and conducting some pretty in-depth tactical analyses. Have you thought about calculating DPS for certain units that you have?

    Since you have the timestamps on your videos, you can get a good estimate of damage per second by knowing the damage of the particular attacking unit, the strength of the building being attacked, and the time elapsed.

    For example:
    Gunner, level 21, damage 54 ... versus Voodoo Hut, level 3, 3600 hit points
    Run a video and send a solo gunner against a lone Voodoo Hut and check exactly how long (seconds) it takes to bring it down. It should take the gunner 67 shots. Again, you have the time stamps on the video so you know start and end times - better for a more accurate DPS estimate.

    I'm starting do to this. That same dmg 54 gunner takes ~10 seconds to drop a 800 point outhouse. DPS ~80. I tried it against a level 2 voodoo hut, hit points 2400, it took 32 seconds and 45 shots. DPS ~75, . More shots, more accurate estimate of DPS.

    Edit: Actually better way - figure out shots per second, then multiply by the damage per shot.
    (damage of unit) * (number of shots / time to destroy unit)
    Gunner: 45 shots / 32 secs = 1.4 shots/sec
    54 damage/shot * (45 shots /32 secs) ~= 76 DPS


    You should confirm. No saying whether DPS changes when the unit upgrades. But would be interesting to know other units, e.g., heavy gunner.


    Same methodology could be applied to defensive units. Example:
    2x Canons, level 12, damage 38 ... Versus Brute, level 18, toughness 1500
    The two cannons would take 20 shots to take the Brute down, but how long? This will be a bit difficult to find a good control for, but it's possible, especially for your analysis techniques.

    Edit: Again, just count how many shots the canon gets off in a certain time period, assuming that it is constantly firing without stopping or turning. this would work great with a bunker trained against a single brute or jugg.

    These are the numbers for shots per second that I have so far, just using a timer and quick counting (N=1 ... science people will catch my drift):
    • Mortar: 10 1 shots / 4.0 sec = 0.25 shots / sec
    • Canon: 17 shots / 20.6 sec = 0.83 shots / sec
    • Bunker: 100 shots / 19.8 sec = 5.05 shots / sec

    With the above shots/sec numbers, we can estimate damage per second:
    • Mortar, level 9: 55 dmg/shot * 0.25 shots/sec = 14 DPS
    • Canon, level 12: 38 dmg/shot * 0.83 shots/sec = 32 DPS
    • Bunker, level 2: 12 dmg/shot * 5.05 shots/sec = 61 DPS

    It'd be great if you could look into this. With these data, we can mate it with the cost of upgrading per level in the academy. It would play a lot into battle strategy and troop upgrade paths, i.e., is a 5% in DPS and 2m grog worth the 20% increase in recruitment time. We know that academy 6 upgrades are expensive and almost not worth it - with these numbers, we can apply some science to it.
     
    Last edited: Feb 12, 2015
  2. I second that Arr, great idea, although, anyone could do this with his videos, although I"m sure he has more footage to work with then any of us would. It would be at least nice if midoki somewhere, even the forum could provide us with the shots per second so at least we could calculate easily.
     
  3. I do think however you can get a pretty good idea just by watchiung though. Like gun towers are a little slower than bunkers, but have 3 guns instead of 1. Mortars are big hits but slow, and cannons are the middle ground.
     
  4. Arr

    Arr Captain

    Midoki seems to coy about the shots per second, but we can make a good estimate by exactly what you said - just watching and counting.

    Once we figure this out, we can figure the economics of upgrades. We'll take a look at the academy upgrade numbers - cost and time to upgrade in academy, as well as cost and time in the tavern once upgraded. Everyone suspects that many academy 6 upgrades aren't worth it - this way, we can add some numbers to support that theory.
     
  5. Roserio

    Roserio First Mate

    oops sorry for the late reply,
    wow that's quite scientific and logical :)
    i also was curious about this because when we had starcraft tv channel i remember that pro gamers and the BJ was aware of these number stuffs to develop more accurate attack and defence.
    but i bet this will take a lot of time for me haha

    i'll first ask Midoki first to provide the info.
     

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice