EPTI did my own calculations a little while back, and most of my numbers are pretty consistent with the chart, but a few standout.
The biggest one of note is the roach. My calculations were 1.8 seconds, not 2, and that was after 5 trial runs.
DPS chart for units - Page 3
| Forum Index > SC2 General |
|
Karas
United States230 Posts
I did my own calculations a little while back, and most of my numbers are pretty consistent with the chart, but a few standout. The biggest one of note is the roach. My calculations were 1.8 seconds, not 2, and that was after 5 trial runs. | ||
|
hzhao
China75 Posts
On March 10 2010 16:47 Karas wrote: Curious how the numbers were calculated? I did my own calculations a little while back, and most of my numbers are pretty consistent with the chart, but a few standout. The biggest one of note is the roach. My calculations were 1.8 seconds, not 2, and that was after 5 trial runs. as far as I know, from mpq | ||
|
starfries
Canada3508 Posts
| ||
|
Railz
United States1449 Posts
On March 10 2010 17:47 starfries wrote: Oh this is cool. Battlecruisers don't feel that strong in the game, interesting They get hurt my armor more with such a fast attack speed, but yes, they are way more powerful - problem is, finding a time to get any amount that becomes feasible is unlikely. Getting 4 or so in late game is doable I find but you have to protect yourself with cannon fodder. | ||
|
Tamerlane
Canada424 Posts
On March 10 2010 13:11 Pillars wrote: Might be useful to factor in unit cost as well. Indeed, that's why I did it : http://spreadsheets.google.com/ccc?key=0AvtPWMS53wIgdEFfLVFWNllJYl90YUFmODRTQ05jQUE&hl=fr Although, a few comments are in order : I wanted to have a relative idea of the burden a unit represented and not only the raw cost of producing this unit, gas is worth much more than minerals and that had to show in the data. At first, I thought of using a relative scarcity of resources (there are 12.5k minerals and 5k gas on each resource spot), but in the end, what counts in a game (cost-wise) is how fast can you produce a unit depending on your gas income, what are your alternatives to these units and how do they compare with other units of other races. In order to do so, I defined the time it took a worker to get in minerals and gas. A worker gets in 5 minerals or 4 gas per trip (that's a "worker-hour" <= of course it has nothing to do with the time unit, but I'm an economist so screw you) and given the availability of minerals patches and gas deposits, you can get 40 mineral and 8 gas units per worker cycle (that's a worker-day, I thought about the better "worker cycle" term by typing this post and I'm too lazy to edit, so screw you again), that is : all stocks are harvested once and the resources are sent to the main. Then you get a neat cost unit in minerals/gas worker-days, which you simply add to get the total cost of a unit in terms of resource gathering cycles. Of course, this is not a measure of the real cost of a unit, but rather its timely economic burden to produce. If you don't think this is a good calculation, I would tend to say "screw you" again, but I'm opened to look at YOUR calculations. NB. You can find the detailed results in the sheet labeled "Total cost calculations" A bit of acronyms : - UCPD = Total Unit Cost per Damage per Second (DPS) - UCPD bonus = UCPD with bonus damage - UCPD full = UCPD with full upgrades - UCPD full bonus = trivial The index is simply a normalization of the values with a reference, I picked marines and marauders becase we all belong to the homo sapiens sapiens specie and should agree to this ethnocentrist bias. I also added all mutalisk' bounces to simplify display, just keep in mind that mutalisks will have higher UCPD against units with 1 to 4 armor. Finally, I removed the Void rays because, well, it's obvious. Oh, and remember to compare boobies with boobies!! (or melee with melee if you will) | ||
|
rackdude
United States882 Posts
| ||
|
spinesheath
Germany8679 Posts
So less UCPD should be better, right? Nice to see that zerglings still are on top  | ||
|
Tamerlane
Canada424 Posts
On March 11 2010 01:58 spinesheath wrote: Small issue: Archons can cost 1 of 3 values, because you can merge HTs and DTs in any combination. ah, I thought archons would get only merged from 2 HTs - I obviously don't play protoss! that explains the "various" in OP's spreadsheet I'll update this soon with the 3 different values. So less UCPD should be better, right? Nice to see that zerglings still are on top correct, the lower the UCPD, the least strain the unit has on the economy for every damage unit it deals | ||
|
Disastorm
United States922 Posts
| ||
|
fearus
China2164 Posts
On March 11 2010 11:10 Disastorm wrote: spine crawlers look so bad compared to photon cannons and bunkers (assuming a bunker is 4X marine damage). Why do spine crawlers do half the damage of a photon cannon and they can only attack ground?? Cause you don't need to invest in a forge early on You can move them By using a drone as an initial investment, you have more flexibility. | ||
|
Makatak
United States10 Posts
That's like those Street Fighter 4 newbs that talk about frame data for certain character moves. I don't care how far you have to stick your **** through the data files to figure that **** out. It doesn't make you a better player. | ||
|
hzhao
China75 Posts
On March 11 2010 11:27 Makatak wrote: And this is useful how? This is Starcraft, not ArithmeticCraft. That's like those Street Fighter 4 newbs that talk about frame data for certain character moves. I don't care how far you have to stick your **** through the data files to figure that **** out. It doesn't make you a better player. who claims he is a better player in this thread? | ||
|
Karas
United States230 Posts
On March 11 2010 11:27 Makatak wrote: I don't care how far you have to stick your **** through the data files to figure that **** out. It doesn't make you a better player. It can in certain ways. For example, did you know that it takes 3 roach hits to kill a marine? Meaning 3 roaches attacking together will kill a marine every hit. If I have 9 roaches, and micro them in groups of 3, for the first few seconds I kill marines 3x as fast as if I just told them all to focus on the same guy. Knowing the numbers can tell what counters are useful, when to micro and when to focus more on macro, etc. While its not certainly the sole of gameplay, it can be a useful addition. | ||
|
jimbobdwayne
United States53 Posts
| ||
|
Brett
Australia3822 Posts
On March 11 2010 11:27 Makatak wrote: And this is useful how? This is Starcraft, not ArithmeticCraft. That's like those Street Fighter 4 newbs that talk about frame data for certain character moves. I don't care how far you have to stick your **** through the data files to figure that **** out. It doesn't make you a better player. I make no comment about the usefulness of this data given its infancy, but if you think frame data is useless in a fighting game such as SF4 then you're completely clueless. | ||
|
-orb-
United States5770 Posts
| ||
|
phexac
United States186 Posts
| ||
|
DeCoup
Australia1933 Posts
On March 11 2010 12:04 Karas wrote: It can in certain ways. For example, did you know that it takes 3 roach hits to kill a marine? Meaning 3 roaches attacking together will kill a marine every hit. If I have 9 roaches, and micro them in groups of 3, for the first few seconds I kill marines 3x as fast as if I just told them all to focus on the same guy. Knowing the numbers can tell what counters are useful, when to micro and when to focus more on macro, etc. While its not certainly the sole of gameplay, it can be a useful addition. No you wouldn't. There is no overkill in SC2. Assuming that all your roaches are in range of all the marines, waypoing attack targeting every marine with all 9 of your roaches selected would kill them in the same amount of time as using 3 groups, because only 3 roaches will fire at each marine. | ||
|
Tamerlane
Canada424 Posts
On March 11 2010 18:49 DeCoup wrote:No you wouldn't. There is no overkill in SC2. Assuming that all your roaches are in range of all the marines, waypoing attack targeting every marine with all 9 of your roaches selected would kill them in the same amount of time as using 3 groups, because only 3 roaches will fire at each marine. If there were only 3 marines as targets, that would be true, but if there are more : your roaches will not "1 shot" 3 marines at a time, instead they will attack the closest unit to them, therefore spreading the total damage over multiple units instead of focusing on a few. | ||
|
Rickilicious
United States220 Posts
On March 10 2010 12:18 majesty.k)seRapH wrote: wait, so does this mean that hellions are weaker than DRONES? at least when not upgraded... It has that follow throug-ish type of attack, and anyway, hopefully you realize that they are only 50 more minerals than drones/probes/scvs and what not and you get a lot more utility out of them for the 50 minerals. I've always thought these types of charts and graphs etc were very misleading, it's still all about how you use your units and how they counter and what not. DPS is for nerds who want to make perfect calculations on nothing and when they lose complain that their units don't do the correct dps or whatever. No shit the BC and thor do the most damage but if you have two of them and you're opponent is ready, you're screwed. | ||
| ||
