|
Unit battles are determined by effective dps*hp vs effective dps*hp and based on cost.
This is based on equal numbers of units attacking from both sides. With micro or more units on one side, hp can be conserved and dps can be increased by focusing.
Cost is based on 2 saturated bases. (minerals are worth 3.57143 and gas is worth 13.1579 based on time to mine)
Dps is based on a recently posted dps table.
Armour changes change attacking unit dps based on cooldown and can't be included.
Shields and health regen are fixed rates that can be taken off of dps and can't be included.
Attack types, splash and other things like that can be added in yourself, these are ratios.
Some Spells, AOE, and movement can't be accurately included
Melee units have more cost effectiveness since ranged units reduce their group hp before they can do damage.
So speed is an important factor for melee units which is not included.
I'm sure there are many errors. I'm happy to correct them if you post the error.
![[image loading]](http://img692.imageshack.us/img692/934/dpshpc5sc2.jpg)
Why did I post this?
It might help people make builds.
It may be interesting(I find this stuff very interesting).
Someone else might improve upon this, which I support.
If someone knows how to usefully include build times, I'd like to know. I tried and it distorted things too much.
|
Wow, looks like a lot of work went into this. Interesting to see that the Thor has the highest DPS x HP / Cost ratio.
|
lol zealot is really high
|
dps/cost would be an interesting stat as well if you can put that in there 
|
Why does the auto-turret cost 1244 minerals?
|
if i'm correct, it looks like you equalized minerals and gas.... gas > mineral
|
Not sure if I missed it, but is there a reason the colossus isn't added?
|
Interesting... where's the rest? O.o I hope you made a program (or excel macro or something) to do this for you quickly ^_^
|
I gave 1244 minerals to the auto-turret to balance it, since it has no mineral cost. Maybe I should just remove it.
|
what escraper said, but to clarify the compensation, minerals are 4x more abundant than gas geysers (on most maps, iirc there was a base on one that has 3 geysers?) but otherwise this is interesting info to gander at, proves that queens > hydras lol i plan on just making queens anytime i have the thought "hydras might be ok." better range (grooved spines need to just be built in, fer serious), higher dps*cost/hp, creep & transfuse <3
|
Wow, this is pretty cool info. Just wish I was smart/good enough to use it. Good job!
edit: To provide something useful to the discussion... for the people wanting just dps/cost ratio you can divide the current dps*hp/cost number by the hp of the unit.
|
This table holds no real value and should not be referred to. It just seems like you were bored.
|
If you're doing cracklings you could do shield marines, and stimmed marines and marauders ( -hp if you want).
|
@ ventor Why would you say it holds no value? Who cares if he was bored. He has provided more information to the community, that is never a bad thing. At the very least say why you think it holds no value...
|
Battles aren't determined by dps*hp vs dps*hp. They're determined by effective dps*hp vs effective dps*hp (including armor as you note). This means you don't gain much real information from sheets like these. I did one of these when SC2 came out, and I realised that even then all the units that on paper looked like they were the most powerful, were sometimes not used due to entirely different reasons.
A more meaningful approach would be to include all the units secondary stats such as movement speed, surface area etc and make a stat called "usefulness" or something that weighs some of the secondary stats.
|
Unfortunately the problem with theory crafting is that there is so much more than what's on the paper. For example 100 lings vs 100 marines. You may think to look at this table to discover who would have the best damage, cost, HP etc but it doesn't factor in things like marines being able to all fire at once as opposed to only the front line of lings.
|
I added Colossus and removed auto-turret. there is a spreadsheet, but I expect there are some errors, So I'll release it after its more accurate. I Also add DPS/cost.
Yes gas is worth more than minerals, it is based on how long it takes to mine each mineral(seconds per mineral or gas)
I want to add modified dps for stimmed marines.
You can usefully use these numbers if you decrease each units dps in a specific situation by the regen rates, and muliply aoe by average number of units in range.
|
I suppose it's not a bad thing, but like Peleus said, there's 20 other things to consider when theorycrafting with this. I don't see any practical use for this.
It's much faster and more practical to just open up a unit tester, make logical army compositions to fight, and observe how it goes. Taking health regen, different unit compositions, upgrades, melee vs ranged units, build time, tech required, etc, etc is near impossible, and not something you're going to be able to make practical use of during an actual game.
|
Just a suggestion, I think dps*hp/cost/supply will actually tell a different story altogether.
|
For people that dont know math, i would like to remind that the cheap units are actually alot better than in the table because they stack more. For example even thou queen has almost equal dh/c as zergling, 3 zerglings (75 min) still beat 1 queen (150 min) because their dps is three times higer when all lings aare still alive and two times higer when two lings are still around.
|
dps*hp/cost/supply doesn't make any sense, but i added dps*hp/supply for when the restriction isn't cost but supply with maxed armies.
Also added alternating colors.
|
Man, why do they have to go and get rid of the thor in HOTS. 
|
Thors do the most overkill, which is yet another form of intricacy this graph can't capture.
|
On March 07 2012 16:09 Resistentialism wrote:
Thors do the most overkill, which is yet another form of intricacy this graph can't capture.
no that is not it. It is because thors are expensive units and thus stack less as i mentioned in previous post.
|
I thought artosis said carriers have the highest dps in game but according to your chart, the thor does.
|
On March 07 2012 16:03 Ixess wrote:
dps*hp/cost/supply doesn't make any sense, but i added dps*hp/supply for when the restriction isn't cost but supply with maxed armies.
Also added alternating colors.
On March 07 2012 16:02 Sea_Food wrote:
For people that dont know math, i would like to remind that the cheap units are actually alot better than in the table because they stack more. For example even thou queen has almost equal dh/c as zergling, 3 zerglings (75 min) still beat 1 queen (150 min) because their dps is three times higer when all lings aare still alive and two times higer when two lings are still around.
That's why dps*hp/cost/supply might make sense. Tyvm sea_food. On a side note, you left out immortal.
|
Russian Federation164 Posts
Due to how armor works, you kinda need to add something that accounts base attack and base armor. I.e., high-dps units with fast cooldown, but moderate dmg per hit, are extremely vulnerable to armoder units (Marines vs. Ultras)
|
On March 07 2012 16:14 Sea_Food wrote:Show nested quote +On March 07 2012 16:09 Resistentialism wrote:
Thors do the most overkill, which is yet another form of intricacy this graph can't capture. no that is not it. It is because thors are expensive units and thus stack less as i mentioned in previous post.
Honestly, I think the biggest issue with thor dps is that it has high shot damage and slow shots (and thus overkill) combined with the setup time between targeting a unit and firing the first shot. Thors do great dps against large targets but ridiculously bad dps when they have to switch targets multiple times in a fight.
|
Ideas: Secondary dps wich includes the dps of units with bonus dmg when bonus dmg is aplicable.
Adding a cell to set the lvl of attack 0 1 2 3. Shoud be relatively easy if the sheet uses formulas.
|
I think in order for this sheet to be useful, you should factor in the range of each unit. Like a multiplier from each unit of range a unit gets. This way, it will more reflect the usefulness of unit ingame, aka Marines with their range against Speedling for example.
|
Why does lings have 0.5 supply? Don't they have 1 supply?
|
On March 07 2012 18:19 eScaper-tsunami wrote:Show nested quote +On March 07 2012 16:03 Ixess wrote:
dps*hp/cost/supply doesn't make any sense, but i added dps*hp/supply for when the restriction isn't cost but supply with maxed armies.
Also added alternating colors. Show nested quote +On March 07 2012 16:02 Sea_Food wrote:
For people that dont know math, i would like to remind that the cheap units are actually alot better than in the table because they stack more. For example even thou queen has almost equal dh/c as zergling, 3 zerglings (75 min) still beat 1 queen (150 min) because their dps is three times higer when all lings aare still alive and two times higer when two lings are still around. That's why dps*hp/cost/supply might make sense. Tyvm sea_food. On a side note, you left out immortal.
Stacking isn't related to supply, its based on fitting more units in the same space.
supply is not a restriction, and so it does not make sense to divide by it.
|
I didn't find the Sentry in there.
|
rauder vs armored, same with like stalker and other units with differing damage vs armor types
Could also probably include column with dps increase per attack upgrade
|
On March 07 2012 19:22 stingbear wrote:
Why does lings have 0.5 supply? Don't they have 1 supply?
Well, two of them have a combined 1 supply.
|
dps/buildtime would be interesting for me
|
On March 07 2012 15:28 OrchidThief wrote:
Battles aren't determined by dps*hp vs dps*hp. They're determined by effective dps*hp vs effective dps*hp (including armor as you note). This means you don't gain much real information from sheets like these. I did one of these when SC2 came out, and I realised that even then all the units that on paper looked like they were the most powerful, were sometimes not used due to entirely different reasons.
A more meaningful approach would be to include all the units secondary stats such as movement speed, surface area etc and make a stat called "usefulness" or something that weighs some of the secondary stats.
Yep, sadly this is a meaningless spreadsheet.
Micro is by far more determining factor for an engagement. After that it is positioning.
If you're doing 1a v 1a this *might* be useful.
|
Added immortal and sentry, removed dps from meaningless units, and added dps/buildtime, and speed.
Also deleted slightly higher cost versions of archon.
If both players micro at the same skill level this is still fairly accurate(they will have similar total hp*dps and usage skill).
This is also very useful for few units micro in at least 2 ways. You can use percentages to factor in skill difference.
Or you can include all factors, which isn't that much usually, just regen,shields,time for each unit in battle, and focus.
Also you don't have to use it the way I suggest. There are build times, costs, speed, dps, and hp.
With all that you can do many more things. Use your imagination.
|
Where are you getting the numbers for Battlecruisers doing double the DPS than Carriers?
Carriers have 8 interceptors. Each Interceptor has 2 attacks dealing 5(+3) damage each, dealing 80(+48) damage between all 8 per wave, attacking at a fairly fast attack speed(forget the attack speed, someone else may know).
Battlecruisers do 8(+3) damage to ground and 6(+3) damage to air every 0.23 seconds.
|
yeah some units isn't added! O_O
|
Broodlord seems kinda low.
|
This is where I got the dps and stats as well as all other units. http://wiki.teamliquid.net/starcraft2/Carrier
I think I could add in how common regen rates, shields, and medics affect dps in exact numbers. I'll look into that later.
Are any more units missing?
|
|
|
|
|
|
EPT
