|
Update2:
Efficiency per cost
Its far away from being perfect, but you can help me with it.
http://yfrog.com/n5effpercostj
Dps eff. = Dps * sqrt(Range)
chose it, because the factor must be proportional to sqrt, unclear how to value it better
Health eff. = Health + Health * (Armor * c)
c = 0.1 in this chart, it depends on the unit that is attacking yours
Efficiency per cost = (Dps eff. + Health eff. / f) / ( cost / 100)
f = 10 in this chart, tell me if you see a better relation between health eff. and dps eff.
this chart doesnt not punish zerglings more for being range 1 than zealots for example. No idea to improve it so far.
Update:
This chart does not think for you :O If you think its useless, just close it...
But for example you can see that Immortals can be more useful than Stalker even against non-armored units depending on the situation you need them for.
http://yfrog.com/f/n1dpspercostj/
Click on the picture to make it bigger.
I think its too big to include directly :D
Tell me if you find any false data.
Divided cost by 100... looks much better :D
Unsure how to get better data involving thors, tanks and ultralisks splash damage.
Carriers value just shows the dps while interceptors are attacking.
Baneling´s value is more like damage instead of dps.
Kinda surpised how effective rocket turrets are :D
Took an average of each race´s relevant data (1gas = 1,5mins):
excluded each splash, zergling, baneling, PF, carrier (4) etc.
Terran: 7,48 (normal) 8,21 (armored) 8,31 (light)
Protoss: 5,77 (normal) 6,37 (armored) 6,47 (light)
Zerg: 5,64 (normal) 6,01 (armored) 5,64 (light)
Would you say this is compensated by health, buildings and upgrades?
|
Well as terran units are usually squishier (thor and bc aside) it seems somewhat balanced. It's kinda weird that the best DPS unit for protoss is the zealot xD
|
BANELING HAS HIGHEST DPS/COST, ZERG IS OP!!11!!SHIFT+1!!1
In all seriousness, this is pretty cool data.
|
Lings, 28 DPS/c
I wuv my lings :3
|
if you post the direct link with [img] tags it will show up in your OP
but again DPS isn't really a starcraft concept, its more "hits to kill" than damage output
|
Ah, misread what the numbers actually represent. Thought it was minerals/dps!
|
Theorycraft of its finest. Do you think its a good idea to ignore: hitpoints, movement speed, fire rate, fire range, unit size, unit tech, unit abilities, flying/not flying, shoots air/ doesnt shoot air and synergies?
Edit: Just in case: Yes, he did ignore fire rate. I know what dps means, but its not the same if a unit deals 1000dmg each 1000 seconds or 1 dmg each second. Think about it.
|
Does each carrier-interceptor really hit once per second?
|
On September 02 2010 05:21 Goobus wrote:
Why would you ever divide cost by 100? It's just annoying to figure out what the numbers actually represent.
I like to compare 6,4 with 3,4 more than 0,064 with 0,034 :D
|
i think i know what you should do next, put in upgrades!
|
How exactly do you calculate dps for banelings? Wouldn't it be either 0 or infinity? 
|
I'd like to apologize, just because I've not had the best of days is no good excuse for me to being insulting.
User was warned for this post
|
On September 02 2010 05:22 Grummler wrote:
Theorycraft of its finest. Do you think its a good idea to ignore: hitpoints, movement speed, fire rate, fire range, unit size, unit tech, unit abilities, flying/not flying, shoots air/ doesnt shoot air and synergies?
Edit: Just in case: Yes, he did ignore fire rate. I know what dps means, but its not the same if a unit deals 1000dmg each 1000 seconds or 1 dmg each second. Think about it.
Pff, don't try to be the smartguy- we know we know, don't have to "think about it" either lol
Thanks for doing this, upgrades would be interesting too
|
Lings, rines, zlots own the place. But I guess if they were weaker, then a lot of game variety would be lost. As of now it is usually a valid opposition for one side to stay tier 1, while the opponent goes higher tech, and they can compete shoulder-to-shoulder, which is fun.
- Armor plays a big role. Unupgraded lings have 0 dps vs armor upgraded Planetary Fortress (the game counts the damage as 0.5 in those cases though, but that's 1/10th of what's in the table).
- In late game there's one more factor - per food.
- 1 gas really costs around 3 minerals, because the mining rate from a saturated base is around 250:800 per minute. (source)
|
|
|
holy shit @ carriers
twice the dps of fully charged void rays against armored targets, and they do that much damage vs. any target, AND they can kite every ground unit in the game, and they do 60% more dps with 3 weapon upgrades.
At 3 weapon ups, it's 128 dps for carrier vs. about 52 dps for charged VR on armored target.
Like, wow.
I need to start making some viable into-carriers builds.
|
Mothership?
Should be really low for it's attack output ^^;
|
On September 02 2010 05:41 OminouS wrote:
Mothership?
Should be really low for it's attack output ^^;
They do 6 attacks for 6 damage every 2.21 seconds, which is only ~16 dps, if the SC2 wiki tells the truth.
That's crazy low DPS.
|
Aw the first thing I looked for was to compare Zergling vs Adrenal Zerglings but you didn't include it 
|
On September 02 2010 05:42 FabledIntegral wrote:Aw the first thing I looked for was to compare Zergling vs Adrenal Zerglings but you didn't include it
Adrenal should increase dps by 20%.
I think it's bugged or something though, according to some post on TL.
|
On September 02 2010 05:35 Buddhist wrote:
holy shit @ carriers
twice the dps of fully charged void rays against armored targets, and they do that much damage vs. any target, AND they can kite every ground unit in the game, and they do 60% more dps with 3 weapon upgrades.
At 3 weapon ups, it's 128 dps for carrier vs. about 52 dps for charged VR on armored target.
Like, wow.
I need to start making some viable into-carriers builds.
That's absolutely not what you should be pulling from this. SC2 is not a tower defense game where you can win by picking the right units at the right time.
|
On September 02 2010 05:35 Buddhist wrote:
holy shit @ carriers
twice the dps of fully charged void rays against armored targets, and they do that much damage vs. any target, AND they can kite every ground unit in the game, and they do 60% more dps with 3 weapon upgrades.
At 3 weapon ups, it's 128 dps for carrier vs. about 52 dps for charged VR on armored target.
Like, wow.
I need to start making some viable into-carriers builds.
Assuming
1. that the enemy units have no armor (which guts Carriers but does practically nothing for VRs)
2. that Carriers actually attack once per second *per interceptor*. This seems very, very odd. I checked http://starcraft.wikia.com/wiki/Interceptor and it claims the Interceptor shoots twice per 3 seconds (and not twice per second). Can anyone clear this up?
|
On September 02 2010 05:50 Jarmam wrote:Show nested quote +On September 02 2010 05:35 Buddhist wrote:
holy shit @ carriers
twice the dps of fully charged void rays against armored targets, and they do that much damage vs. any target, AND they can kite every ground unit in the game, and they do 60% more dps with 3 weapon upgrades.
At 3 weapon ups, it's 128 dps for carrier vs. about 52 dps for charged VR on armored target.
Like, wow.
I need to start making some viable into-carriers builds. Assuming 1. that the enemy units have no armor (which guts Carriers but does practically nothing for VRs) 2. that Carriers actually attack once per second *per interceptor*. This seems very, very odd. I checked http://starcraft.wikia.com/wiki/Interceptor and it claims the Interceptor shoots twice per 3 seconds (and not twice per second). Can anyone clear this up?
This data was took by the ingame help >.<
|
Starcraft isn't a DPS game. There's way too many different factors like splash damage, unit speed, range, mineral/gas ratios, etc, etc. I guess you could learn some stuff with the DPS but in general, not as useful as it could be in games like WoW or something.
|
On September 02 2010 05:31 ZaaaaaM wrote:Show nested quote +On September 02 2010 05:22 Grummler wrote:
Theorycraft of its finest. Do you think its a good idea to ignore: hitpoints, movement speed, fire rate, fire range, unit size, unit tech, unit abilities, flying/not flying, shoots air/ doesnt shoot air and synergies?
Edit: Just in case: Yes, he did ignore fire rate. I know what dps means, but its not the same if a unit deals 1000dmg each 1000 seconds or 1 dmg each second. Think about it. Pff, don't try to be the smartguy- we know we know, don't have to "think about it" either lol
So, you have read my last sentence. Awesome. Or maybe you read the rest too, but dont want to discuss about it, cause trolling is so much easier.
On September 02 2010 05:53 neobowman wrote:
Starcraft isn't a DPS game. There's way too many different factors like splash damage, unit speed, range, mineral/gas ratios, etc, etc. I guess you could learn some stuff with the DPS but in general, not as useful as it could be in games like WoW or something.
This.
|
On September 02 2010 05:45 Fraidnot wrote:Show nested quote +On September 02 2010 05:35 Buddhist wrote:
holy shit @ carriers
twice the dps of fully charged void rays against armored targets, and they do that much damage vs. any target, AND they can kite every ground unit in the game, and they do 60% more dps with 3 weapon upgrades.
At 3 weapon ups, it's 128 dps for carrier vs. about 52 dps for charged VR on armored target.
Like, wow.
I need to start making some viable into-carriers builds. That's absolutely not what you should be pulling from this. SC2 is not a tower defense game where you can win by picking the right units at the right time.
The roles of carrier and void ray overlap pretty heavily, and carriers are worth more per cost.
That's pretty simple.
Obviously void rays are infinitely more powerful for harassment, but in an army situation, carriers are clearly superior.
|
oh my god. spreadsheets in WOW and now they are in SC2 ! ha...
|
Hyrule19221 Posts
Kind of interesting. How'd you account for upgrade costs (specifically Stim and Siege Mode)?
|
Saying that "this isn't WoW" is a silly argument. No one blindly relied on spread sheets for DPS in WoW for PvP either, for obvious reasons.
It didn't matter what your sustained dps was if you were just going to get kited all day.
It's still good information to have, however.
|
On September 02 2010 06:00 tofucake wrote:
Kind of interesting. How'd you account for upgrade costs (specifically Stim and Siege Mode)?
That would mean to create much more data depending on how many units you have, do you really want that much data? :O
Maybe I´ll do it.
Also it will need to involve movespeed, range etc.
|
On September 02 2010 05:57 berzerger wrote:
oh my god. spreadsheets in WOW and now they are in SC2 ! ha...
Perfect SC2 raid group:
1 Ultralisk tank
2 Medivac healers
10 Zerglings (or as many that can surround the enemy  )
12 stimmed Marines
But joking aside, this is really interesting data. I don't know if it'll effect anyone's decisions on their unit compositions, but it's still nice to know. Nice work Pepe :D
|
For protoss, shields is definitely a compensation.
|
carrier is incorrect, interceptor cooldown is 3, also spine crawler is incorrect as cooldown is .8608, might be more errors me thinks
what is the point of 3 attack calculation?
|
This has been done a few times before but its always worthwhile to look at again or from another perspective. The biggest issue has been coming up with a good valuation of gas. There are 2.4 times as many minerals as gas at a base, and you mine minerals at about 2.8 times the rate of gas. I think gas should be valued higher than your top valuation of 1.5.
|
Nice work here, some stats are interesting. Would be cool to add hp analysis to this as well.
|
Carriers work like this:
Carriers fire Interceptors every 0.5 seconds and Interceptors make an attack every 3 seconds. The Interceptor deals 2 attacks each dealing 5 damage and there can be a maximum of 8 Interceptors. This results to the following:
To fire all 8 Interceptors: 4 seconds.
Attack CD of all 8 Interceptors on their "fire round": 3.5 seconds. (+0.5 second launch time)
The first attack made by the Interceptors in a newly engaged battle, therefore, is 80 (total damage) / 4 (4 is larger than 3.5) = 20.
The following attacks made by the Interceptors after they have been fired (assumes none are killed): 80 / 3 (no need for the additional +0.5 second launch time): 26.6777777r.
Carriers do more damage if they're kept alive than the moment they engage. Alternatively, you can attack some of your own units to launch the Interceptors and then engage, much like the Void Ray can charge it's Prismatic Beam on allied units.
|
On September 02 2010 05:32 figq wrote:Lings, rines, zlots own the place. But I guess if they were weaker, then a lot of game variety would be lost. As of now it is usually a valid opposition for one side to stay tier 1, while the opponent goes higher tech, and they can compete shoulder-to-shoulder, which is fun. - Armor plays a big role. Unupgraded lings have 0 dps vs armor upgraded Planetary Fortress (the game counts the damage as 0.5 in those cases though, but that's 1/10th of what's in the table). - In late game there's one more factor - per food. - 1 gas really costs around 3 minerals, because the mining rate from a saturated base is around 250:800 per minute. ( source)
nothing ever has 0 dps. dmg is minimized at 1.
|
On September 02 2010 06:29 danl9rm wrote:Show nested quote +On September 02 2010 05:32 figq wrote:Lings, rines, zlots own the place. But I guess if they were weaker, then a lot of game variety would be lost. As of now it is usually a valid opposition for one side to stay tier 1, while the opponent goes higher tech, and they can compete shoulder-to-shoulder, which is fun. - Armor plays a big role. Unupgraded lings have 0 dps vs armor upgraded Planetary Fortress (the game counts the damage as 0.5 in those cases though, but that's 1/10th of what's in the table). - In late game there's one more factor - per food. - 1 gas really costs around 3 minerals, because the mining rate from a saturated base is around 250:800 per minute. ( source) nothing ever has 0 dps. dmg is minimized at 1. Damage is minimized at 1/2.
|
On September 02 2010 05:32 figq wrote:- 1 gas really costs around 3 minerals, because the mining rate from a saturated base is around 250:800 per minute. ( source)
Nah, it doesnt... buildings cost far more minerals than gas.
It would just be more than 2 if you are producing mass archons or anything like that.
If you are massing Zealots it would get below 1 and if you are producing mass reapers it would maybe go up to 2.
|
Seems you are missing infested marines
|
On September 02 2010 05:53 Pepe- wrote:
This data was took by the ingame help >.<
The data is wrong according to any source I can find. If Carriers had 2,5 times the DPS of a fully charged voidray against any units without armor, then, no matter how Starcraft works, they would be seen more than every 50th game.
|
On September 02 2010 05:56 Grummler wrote:Show nested quote +On September 02 2010 05:31 ZaaaaaM wrote:On September 02 2010 05:22 Grummler wrote:
Theorycraft of its finest. Do you think its a good idea to ignore: hitpoints, movement speed, fire rate, fire range, unit size, unit tech, unit abilities, flying/not flying, shoots air/ doesnt shoot air and synergies?
Edit: Just in case: Yes, he did ignore fire rate. I know what dps means, but its not the same if a unit deals 1000dmg each 1000 seconds or 1 dmg each second. Think about it. Pff, don't try to be the smartguy- we know we know, don't have to "think about it" either lol So, you have read my last sentence. Awesome. Or maybe you read the rest too, but dont want to discuss about it, cause trolling is so much easier. Show nested quote +On September 02 2010 05:53 neobowman wrote:
Starcraft isn't a DPS game. There's way too many different factors like splash damage, unit speed, range, mineral/gas ratios, etc, etc. I guess you could learn some stuff with the DPS but in general, not as useful as it could be in games like WoW or something. This.
While you are correct in this reasoning (and believe most people here realize that) dps information is still good to have.
Zerglings with adrenal glands may be able to kill shit fast but who cares if one storm anhilates the whole bunch and you lose all of your dps instantly?
There are still a variety of things one can take out of this information.
(Btw your opponent is not always going to have templar ready for the immediate storm)
The statement I am about to make could be made without this fancy chart but I will make it anyway.
It is common knowledge the the immortal is very good versus armored targets. But my opponent is going all broodlords. Even though they are armored it does me no good since immortals cannot attack air. But say I notice he is going very heavy with lings so my gateway/warpgates are going to be pumping out primarily zealots. Well a ranged unit would be good to throw in here as I could shoot over my zealots and add additional dps (kill shit faster). What can I add? Stalkers? No than I have less zealots. What about an immortal? Well to make it simple look at the chart the immortal is exactly double that of a stalker as far as gas and minerals go so I am just going to look at the first dps column and the immortals dps vs light targets is also exactly double that of the stalker. So in short the immortal is the cost and damage of 2 stalkers. (this for the time being excludes that they have about 3x the health and their special shield.) The immortal does not clog my my gateways and I can get ranged support it just as fast. Against heavy zealot numbers an obvious transition for my zerg buddy would be roaches but you already have the immortals to help take care of said roaches. He decides to go air you have all the gateways for stalkers already and your immortals can just sit back for a bit.
Yes there are a million and one factors that go into a strategy game to decide who wins you mentioned a few of them. Cost splash harassment the list goes on. The idea of a dps list not to get too drawn into it and take out of it what information you can. What be relevant for whatever situation you are up against.
|
On September 02 2010 05:45 Buddhist wrote:Show nested quote +On September 02 2010 05:42 FabledIntegral wrote:Aw the first thing I looked for was to compare Zergling vs Adrenal Zerglings but you didn't include it Adrenal should increase dps by 20%. I think it's bugged or something though, according to some post on TL.
Negatory. Adrenal decreases attack time by 20%, which is not the same as increasing damage by 20%.
|
On September 02 2010 07:03 Jarmam wrote:Show nested quote +On September 02 2010 05:53 Pepe- wrote:
This data was took by the ingame help >.<
The data is wrong according to any source I can find. If Carriers had 2,5 times the DPS of a fully charged voidray against any units without armor, then, no matter how Starcraft works, they would be seen more than every 50th game.
Think about it. What other factors could go into this? Possibly the build time of the carrier vs that of the voidray. The larger income neccacery to produce said carriers. Possibly because vikings and corruptors do very well against carriers. Making voids a better investment in more situations.
(and before anyone says this yes I do realize both of these units can combat voids to a degree however, the corruptors and vikings are both better vs carriers)
Also supply might have something to do with it as remember towards endgame supply is the resource that limits you the most and if I feel I can do more damage to my opponent with 6 supply worth of voids then a carrier well.
|
I've already known the Carrier was very powerful from unit tester. The main reason why the BC seems more resourceful has little to do with the race, but more so of the players (TLO, Jinro, etc.). If Blizzard would do as far as to nerf BC because it sees play, then I'm expecting a nerf for the Carrier sometime later, when people use it more.
|
On September 02 2010 07:17 terranghost wrote:
Think about it. What other factors could go into this? Possibly the build time of the carrier vs that of the voidray. The larger income neccacery to produce said carriers. Possibly because vikings and corruptors do very well against carriers. Making voids a better investment in more situations.
(and before anyone says this yes I do realize both of these units can combat voids to a degree however, the corruptors and vikings are both better vs carriers)
Also supply might have something to do with it as remember towards endgame supply is the resource that limits you the most and if I feel I can do more damage to my opponent with 6 supply worth of voids then a carrier well.
A Corrupter has 200 health and 2 armor, meaning the Carrier suffers a 40% damage penalty assuming equal upgrades.
At the stated 80 dps in-game-time (which is 106,6 dps real-time), Carriers would do roughly 64 damage per second real-time to a Corrupter, killing it in ~3 seconds. Corrupters would deal roughly 14 DPS real-time to a Carrier, meaning that 3 of them would require 8,3 seconds to kill a Carrier *if* the Carrier had no armor (it has 2). Carriers would, in other words, deal perfectly well with Corrupters both cost-to-cost and food-to-food.
With 106,6 dps, a Carrier would kill one Viking per roughly 1,17 seconds (which is what a Carrier does to the first Viking it targets if it has the catapult-upgrade) with no upgrade difference between +dmg and +armor. In other words, Carriers would *demolish* Vikings without the slightest concievable fight. There is no way 3 Vikings can kill a Carrier in ~3,4 seconds even if none of them died. Hell, even 5 Vikings would probably lose to a single Carrier that doesnt even have upgrades (which Carriers scale with very well percentagewise).
The data provided in the OP has to be wrong on the Carrier to some extent. I dont know about the Spinecrawlers. I'd be happy to discuss Carrier strategy elsewhere than here - my beef is with the stated dps-value alone.
|
Carriers make In Utter Darkness a walk in the park. The problem is that they are higher up the tech tree compared to VRs and people aren't playing with them in the lower ranks.
|
On September 02 2010 06:29 danl9rm wrote:Show nested quote +On September 02 2010 05:32 figq wrote:Lings, rines, zlots own the place. But I guess if they were weaker, then a lot of game variety would be lost. As of now it is usually a valid opposition for one side to stay tier 1, while the opponent goes higher tech, and they can compete shoulder-to-shoulder, which is fun. - Armor plays a big role. Unupgraded lings have 0 dps vs armor upgraded Planetary Fortress (the game counts the damage as 0.5 in those cases though, but that's 1/10th of what's in the table). - In late game there's one more factor - per food. - 1 gas really costs around 3 minerals, because the mining rate from a saturated base is around 250:800 per minute. ( source) nothing ever has 0 dps. dmg is minimized at 1. 0.5, as quoted.
On September 02 2010 06:40 Pepe- wrote:Show nested quote +On September 02 2010 05:32 figq wrote:- 1 gas really costs around 3 minerals, because the mining rate from a saturated base is around 250:800 per minute. ( source) Nah, it doesnt... buildings cost far more minerals than gas. It would just be more than 2 if you are producing mass archons or anything like that. If you are massing Zealots it would get below 1 and if you are producing mass reapers it would maybe go up to 2. The main limiting factor is the speed at which you gather the resource, and there is precise calculation for it, as quoted. Also, as Zerg you don't depend on Nx buildings to increase the production speed, and as any race at some point this is fixed, and you mainly count the unit cost itself.
|
On September 02 2010 07:38 Jarmam wrote:Show nested quote +On September 02 2010 07:17 terranghost wrote:
Think about it. What other factors could go into this? Possibly the build time of the carrier vs that of the voidray. The larger income neccacery to produce said carriers. Possibly because vikings and corruptors do very well against carriers. Making voids a better investment in more situations.
(and before anyone says this yes I do realize both of these units can combat voids to a degree however, the corruptors and vikings are both better vs carriers)
Also supply might have something to do with it as remember towards endgame supply is the resource that limits you the most and if I feel I can do more damage to my opponent with 6 supply worth of voids then a carrier well. A Corrupter has 200 health and 2 armor, meaning the Carrier suffers a 40% damage penalty assuming equal upgrades. At the stated 80 dps in-game-time (which is 106,6 dps real-time), Carriers would do roughly 64 damage per second real-time to a Corrupter, killing it in ~3 seconds. Corrupters would deal roughly 14 DPS real-time to a Carrier, meaning that 3 of them would require 8,3 seconds to kill a Carrier *if* the Carrier had no armor (it has 2). Carriers would, in other words, deal perfectly well with Corrupters both cost-to-cost and food-to-food. With 106,6 dps, a Carrier would kill one Viking per roughly 1,17 seconds (which is what a Carrier does to the first Viking it targets if it has the catapult-upgrade) with no upgrade difference between +dmg and +armor. In other words, Carriers would *demolish* Vikings without the slightest concievable fight. There is no way 3 Vikings can kill a Carrier in ~3,4 seconds even if none of them died. Hell, even 5 Vikings would probably lose to a single Carrier that doesnt even have upgrades (which Carriers scale with very well percentagewise). The data provided in the OP has to be wrong on the Carrier to some extent. I dont know about the Spinecrawlers. I'd be happy to discuss Carrier strategy elsewhere than here - my beef is with the stated dps-value alone.
Keep in mind that the way a carrier deals damage is through small, 5 damage attacks. Meaning, if the target had 1 armor, the total damage dealt by the carrier would be reduced by 20%. If it had 2 armor (like buildings) it would be reduced by 40%. Corruptors have 2 armor i believe, reducing carrier damage significantly.
|
I really like the compliation of DPS data. But do u have any basis on which u calculated ur aggregation (Any specific reasoning that u chose THAT path out of many)? They seem ab it random to me. Great work though!
|
Terran also has the lowest hp per food units, and has the slowest production time overall in a long macro game.
Useless statistics are useless.
|
On September 02 2010 07:08 MangoTango wrote:Show nested quote +On September 02 2010 05:45 Buddhist wrote:On September 02 2010 05:42 FabledIntegral wrote:Aw the first thing I looked for was to compare Zergling vs Adrenal Zerglings but you didn't include it Adrenal should increase dps by 20%. I think it's bugged or something though, according to some post on TL. Negatory. Adrenal decreases attack time by 20%, which is not the same as increasing damage by 20%.
Oh, then just:
.7*.8 = .56
.7 / .56 = 1.25
25% more dps.
I think it's actually bugged to be .59 delay with adrenal, though. Not sure. That would be only about 19% increase in dps.On September 02 2010 08:09 ZlaSHeR wrote:
Terran also has the lowest hp per food units, and has the slowest production time overall in a long macro game.
Useless statistics are useless.
That's a pretty sensationalist statement.
|
so easy to read. thank you!
|
Updated it... but I think Spinecrawler is fine :O
|
Dominican Republic463 Posts
On September 02 2010 05:35 Buddhist wrote:
holy shit @ carriers
twice the dps of fully charged void rays against armored targets, and they do that much damage vs. any target, AND they can kite every ground unit in the game, and they do 60% more dps with 3 weapon upgrades.
At 3 weapon ups, it's 128 dps for carrier vs. about 52 dps for charged VR on armored target.
Like, wow.
I need to start making some viable into-carriers builds.
Yep I've recently started to incorporate them into PvT earlier than before and I've had great success, def recommend you tinkering around it. I strongly believe its the units that will change PvT meta game in the coming months/years.
|
Whats the reason rocket turrets are so much better than sporecrawlers?
Their armor or life?
|
I'm saddened to see the Spine Crawler pail in comparison to the Photon Cannon when it costs more and only exclusively attacks ground.
And wow that Rocket Turret has a lot of DPS rofl, no wonder my Mutas are getting melted
On September 02 2010 08:09 ZlaSHeR wrote:
Terran also has the lowest hp per food units, and has the slowest production time overall in a long macro game.
Useless statistics are useless.
Yea I'd really like to see where this statement is coming from please, I think someone's shocked that the Terran DPS is so high.
But what's really surprising is Zerglings with Adrenaline Glands having only 1 more point of DPS, that's not even like a 20% increase in DPS.
|
On September 02 2010 08:31 Pepe- wrote:
Updated it... but I think Spinecrawler is fine :O
Nice :-)
Now keep going, I'm curious how things look at gas = 2.5 minerals.
|
1. I feel like banelings cant really have a dps...
2. Am i reading this right , and does an 8 interceptor carrier do less dps than a 4 interceptor carrier?
3. Holy crap turrets are dirty...
|
LOL. I like how Probes out DPS Stalkers and a number of other units against normal armor units.
No Archons!?
|
wow, zlings are really weak... rly need huge dmg-buff...
|
Hmm I've heard banshees are really high dps/cost but maybe I'm readin this wrong as it doesn't seem that high
|
On September 02 2010 14:00 Nub4ever wrote:Hmm I've heard banshees are really high dps/cost but maybe I'm readin this wrong as it doesn't seem that high
well, they are ranged and flying - thats worth a lot.
|
This is one of those statistics like in sports that you can look at all day and it won't actually show that much.
You want to know which unit has the highest dps per food in game? It's the phoenix, guess what, phoenix aren't actually the best unit in game.
|
You guys keep overdramatizing the DPS chart. Did you guys forget about hit points? Or the fact that many units are unique (speed-wise/cast-wise/hp regen-wise)?
Kickinhead: No zerglings do not need a huge damage buff. They have the ability to surround and are very quick. Not to mention using that logic we would have to buff just about every unit as 5 zerglings can easily take down a Zealot upon surround.
The only thing real thing I'm disappointed by in the DPS chart is the spore crawler. Why is it that Terran gets such a great anti-air with high DPS and ability to repair while Zerg gets the measly Spore Crawler? Terran barely has trouble with air early tech with Marines and late tech with Vikings. Zerg, however, can easily get screwed by a single void ray or 2 banshees. No tech switch there, if you couldn't get a scout in because of Terran wall or Protoss wall (with zealot on hold) and have not gotten a den or spire. That's simply stupid in my opinion. At least make Spore Crawler buildable with Spawning Pool.
|
On September 02 2010 05:35 Buddhist wrote:
holy shit @ carriers
twice the dps of fully charged void rays against armored targets, and they do that much damage vs. any target, AND they can kite every ground unit in the game, and they do 60% more dps with 3 weapon upgrades.
At 3 weapon ups, it's 128 dps for carrier vs. about 52 dps for charged VR on armored target.
Like, wow.
I need to start making some viable into-carriers builds.
Socke
|
On September 02 2010 14:00 Nub4ever wrote:Hmm I've heard banshees are really high dps/cost but maybe I'm readin this wrong as it doesn't seem that high
Almost twice as high as Mutalisks and Void Rays so I'd say that pretty good, especially considering they can cloak. Did I miss the sarcasm?
|
What time value did you use for banelings? It's a one-time attack only... not sure how you can choose that and make it consistent.
|
On September 02 2010 14:03 ZlaSHeR wrote:
This is one of those statistics like in sports that you can look at all day and it won't actually show that much.
You want to know which unit has the highest dps per food in game? It's the phoenix, guess what, phoenix aren't actually the best unit in game.
Eh sarcasm noted but it's not like ignoring the statistics completely is any good.
Obviously we can't make a whole lot from this other then some theories, but it's not like theories are jack shit. If anything we should be glad that this chart is up, it might give some people interesting ideas or things to talk about.
Like the most glaring thing to note in this chart to me is the Rocket Turret, why the fuck is the DPS for it so high compared to the Spore Crawler rofl.
|
WHOA...GUYS.....
WHY CARRIERS AREN'T AS RAPE AS THEY SEEM FROM THIS CHART:
![[image loading]](http://imgur.com/vELMT.jpg)
not really. They are not NEARLY as impressive as it looks.
Each carrier has 8 interceptors, each fires 2 shots per attack. You could say they attack in 16 attacks of 5(+upgrades)=128dmg maxed...in 16 attacks....So while their attack gets up to crazy levels, each ENEMY armor upgrade also reduces their attack power by 16.
So maxed carrier vs no armor marine=128damage(per volley)
Maxed carrier vs 3 armor marine =only 80 damage(per volley)
Maxed carrier vs unupped corruptor= 96 damage(per volley)
Maxed carrier vs 3 armor Corruptor= only 48 damage (per volley)
Thats why carriers aren't as rape as they seem
TLDR: Carriers attack in SIXTEEN attacks of 5 damage, each weapon upgrade therefore buffs their damage 16pts, but each ENEMY ARMOR UPGRADE reduces it by 16pts as well.
|
On September 02 2010 05:13 Pepe- wrote:http://yfrog.com/4jdpspercostjClick on the picture to make it bigger. I think its too big to include directly :D Tell me if you find any false data. Divided cost by 100... looks much better :D Unsure how to get better data involving thors, tanks and ultralisks splash damage. Carriers value just shows the dps while interceptors are attacking. Baneling´s value is more like damage instead of dps. Kinda surpised how effective rocket turrets are :D Took an average of each race´s relevant data (1gas = 1,5mins): excluded each splash, zergling, baneling, PF, carrier (4) etc. Terran: 7,48 (normal) 8,21 (armored) 8,31 (light) Protoss: 5,77 (normal) 6,37 (armored) 6,47 (light) Zerg: 5,64 (normal) 6,01 (armored) 5,64 (light) Would you say this is compensated by health, buildings and upgrades?
Totally irrelevant and ridiculous post IMO. You cant calculate dps and then compare it and still have it mean anything. Zealots deal X dps, but they still cant kill a Banshee. Terrain features, having a good unit composition and so on all ... tons of things that have a much bigger impact than statistical data. Sure a Thor is a "hard counter" to Mutas, but in reality that doesnt work anymore.
|
On September 02 2010 16:49 Rabiator wrote:Show nested quote +On September 02 2010 05:13 Pepe- wrote:http://yfrog.com/4jdpspercostjClick on the picture to make it bigger. I think its too big to include directly :D Tell me if you find any false data. Divided cost by 100... looks much better :D Unsure how to get better data involving thors, tanks and ultralisks splash damage. Carriers value just shows the dps while interceptors are attacking. Baneling´s value is more like damage instead of dps. Kinda surpised how effective rocket turrets are :D Took an average of each race´s relevant data (1gas = 1,5mins): excluded each splash, zergling, baneling, PF, carrier (4) etc. Terran: 7,48 (normal) 8,21 (armored) 8,31 (light) Protoss: 5,77 (normal) 6,37 (armored) 6,47 (light) Zerg: 5,64 (normal) 6,01 (armored) 5,64 (light) Would you say this is compensated by health, buildings and upgrades? Totally irrelevant and ridiculous post IMO. You cant calculate dps and then compare it and still have it mean anything.
wtf. its completely valid. rocket turrets do much more damage then spore crawlers. thats a fair and direct comparison yeah ?
|
I don't think this sort of data is truly representative of how Zerg units are usually just running around behind one another and stuck at choke points not doing any damage. SC2 maps are so anti-zerg.
And banelings may do lots of damage when they catch something but without extreme micro you lose like 5 in the same spot lol.
Not to mention Zerg T1 units are have 3 or less range which further reduces the dps per second of nearly 3/4 of them in any given chokepoint/wall.
|
On September 02 2010 05:45 Buddhist wrote:Show nested quote +On September 02 2010 05:42 FabledIntegral wrote:Aw the first thing I looked for was to compare Zergling vs Adrenal Zerglings but you didn't include it Adrenal should increase dps by 20%. I think it's bugged or something though, according to some post on TL.
cracklings arnt bugged in SC2. but.. in BW, the upgrade increase attack speed by 45%, where as in SC2, its only 20% increase. the tooltip description did not lie, just ppl assume the upgrades were the same, since the costs were same.
|
On September 02 2010 14:11 HardcoreBilly wrote:
Kickinhead: No zerglings do not need a huge damage buff. They have the ability to surround and are very quick. Not to mention using that logic we would have to buff just about every unit as 5 zerglings can easily take down a Zealot upon surround.
Next to everyone who has ever played SCBW agrees that Lings do extremely low Damage in SC2 and it's not like their special in any other way:
- They have low health/armor
- They don't have a ranged attack
- They need the Upgrade to even keep up with most of the early Units of the Terran
- Yeah - they can Surround Units, but this is a double-edged sword: Zerglings need surface to attack and because they often are used in large numbers, many of them will simply not be able to hit the target. Also, the opponent can attack in a choke or stand next to a wall so the Lings won't be able to surround the Unit etc.
Lings are totally underpowered in SC2. They may be fast, but in all other aspects, they pretty much suck. This also includes their T3-Upgrade, which won't help too much because they die so fast to the incredibly powerful AoE every race has to hardcounter them.
Just play Zerg for a while, I didn't really notice it by just looking at replays, but after about 150+ games with Zerg at around 1000 Points in Diamond I've noticed that they just don't do any DMG. it's always easy to say that Unit's aren't THAT bad and things are pretty balanced when you didn't actually have to play with those Units or against certain Strategies yourself.
*Edit: Btw. It simply won't get into my head why Turrets do sooo much more DPS per Minerals spent than Spore Crawlers... Okay, you can uproot them, but the difference is HUGE. Besides, Turrets are basically the only Anti-Air Structure that can be used effectively against an Air-Unit, which is the Mutalisk. In every other MU, Turrets/Cannons/Spore Crawlers are just not worth the money spent, but exactly the one Structure you can actually use in a MU is soooo overpowered - WTF? -.-°+
|
attack surface area matters for the melee units too not just cost
ultralisk zealot and thor would be way higher in those charts
|
On September 02 2010 17:16 Jameser wrote:
attack surface area matters for the melee units too not just cost
ultralisk zealot and thor would be way higher in those charts
Exactly that's why Zerglings suck even more...
|
On September 02 2010 17:18 kickinhead wrote:Show nested quote +On September 02 2010 17:16 Jameser wrote:
attack surface area matters for the melee units too not just cost
ultralisk zealot and thor would be way higher in those charts Exactly that's why Zerglings suck even more...
How so? The Zergling is so small, that when he is fully surrounded much less units can attack it simultaneoulsy than an Ultralisk for instance. For all units (even ranged ones) smaller size is better. This being said, Zerglings certainly feel much weaker in SC2.
|
A Corrupter has 200 health and 2 armor, meaning the Carrier suffers a 40% damage penalty assuming equal upgrades.
On September 02 2010 08:07 TSL-Lore wrote:
Keep in mind that the way a carrier deals damage is through small, 5 damage attacks. Meaning, if the target had 1 armor, the total damage dealt by the carrier would be reduced by 20%. If it had 2 armor (like buildings) it would be reduced by 40%. Corruptors have 2 armor i believe, reducing carrier damage significantly.
I am well aware of that. It just underlines my problem with the chart that the Carrier *attacking a 2 armor anti-massive unit* would fare so ridiculously well against it if the chart is correct. The Carrier would be absurdly, absurdly, absurdly overpowered if it dealt almost twice the damage of a charged Void Ray against *any unit* in the game that doesnt have the luxury of having 2 (or 3) armor.
Now I went and tested it, and apart from discovering that my initial game-speed->real-time conversion is slightly wrong, it is very, very evident that Interceptors *do not* shoot once per second *per interceptor*. If this was the case, they'd destroy a pylon in about 5,5 seconds game-time (apart from cluttering the screen in laser) - it takes more than 13 with the interceptors already launched. If the chart is correct, the initial burst of a catapult-Carrier (the deadly part that almost instantkills a Viking if the Carrier has +1 attack) would be dealt every second game-time.
Someone wrote that the Interceptor shoots once (twice for 5 dmg if you will) per 3 seconds game-time, which seems significantly more accurate. Even if this isnt entirely accurate, its definitely much closer to what appears to be happening than the chart is.
|
On September 02 2010 17:25 MiraMax wrote:Show nested quote +On September 02 2010 17:18 kickinhead wrote:On September 02 2010 17:16 Jameser wrote:
attack surface area matters for the melee units too not just cost
ultralisk zealot and thor would be way higher in those charts Exactly that's why Zerglings suck even more... How so? The Zergling is so small, that when he is fully surrounded much less units can attack it simultaneoulsy than an Ultralisk for instance. For all units (even ranged ones) smaller size is better. This being said, Zerglings certainly feel much weaker in SC2.
Its not the zergling size, its the marine ball.
In SC2 its really easy to keep your marines in a tight ball which means lings can't get a surround on any one marine, and many will just scurry around trying to get a hit. This causes a snowball effect because lings will die but marines will get damaged evenly across the group causing the dps difference to grow larger and larger.
|
lol at probes having more DPS / cost than stalkers, haha
|
On September 02 2010 18:05 Nihilnovi wrote:
lol at probes having more DPS / cost than stalkers, haha
if you want to understand this chart, then its important that you compare the dps/cost with the durability and speed of the units also.
protoss units have a shitload of HP+shields. their dps/cost is therefore very low in comparison. the only units that deal higher dps/cost have major downsides, like the slow marauder, the slow voidray that needs to charge and the slow colossus that can be hit by AtA and GtG.
roaches, marauders and hydralisks have letz say medium durability, so they have average dps/cost.
marines and lings have very low durability ->very high dps/cost.
the chart also shows how powerfull stim is. the stimmed numbers really pop out.
|
I want to know what world you live in where marauders are slow and hydras are equivalent to roaches/maras in durability.
|
I hate to jump on the "this is a bad chart" bandwagon, but it kind of is. Even ignoring the difficulty of 1:1 comparisons across races due to a huge number of removed variables, your actual numbers don't make sense at all. Not only are your 4/8 interceptor carrier numbers seemingly backwards, but you would expect something close to double the dps from an 8 interceptor carrier vs a 4 interceptor carrier. I also doubt you are factoring in launch and recovery times as well for these fighters, but if you are, you really should be using the graviton catipoult upgrade. Its so cheap compared to the cost of a single carrier that its always picked up. I haven't studied zerg or terran sides in equal detail, but make sure your numbers are accurate and make sense.
Banelings should not have DPS at all. I don't mean to sound condescending, but dps is damage over time. Depending on how you want to quantify the time in this instance, banelings either have 0.000r1 dps, or infinite dps. In either instance, the resulting number is meaningless in a DPS comparison scenario.
Similarly, multiplying by "hits" for AoE/line attack units is pretty meaningless. A colossus with +2 upgrades that one-shots a line of zerglings didn't do less dps than a colossus with +3 upgrades in the same situation. In fact, in many instances it would be better to have that super-high "3-hit" dps with no AoE factor than distributed damage. This is the factor that makes charged voidrays so scary - high dps with no overkill. For later iterations, I would simply list their dps, with a note of its splash range, if any. You can make up obscene numbers for Colossus, Thors, etc. if you want them to be hitting 20 enemies with each attack, but that doesn't help anybody's comparison, and it has nothing to do with how fast it actually kills a unit.
I would also recommend sorting the units alphabetically and separating by ground, air, and buildings/spawned units. The chart is pretty hard to read with a seemingly random ordering of everything. Getting more colors in there than greyscale would also help. If you wanted to expand this project, adding the impact of +1/2/3 weapon upgrades would also be quite useful.
All that said, there are some definite uses for this. 8x stimmed marauders in a drop do an insane 160dps to a building, which means they could snipe a nexus in ~10 seconds, +2/3s more if you factor in unloading time. Simple calculations like that are quite handy to have a chart to assist with. I hope you continue to work on this project, I think a polished end result would have some excellent uses.
|
Carriers ARE NOT as good as the chart makes them to be. Carriers are about as good as BCs at attack air: quoting myself because people skipped over my post.
Carriers work like this:
Carriers fire Interceptors every 0.5 seconds and Interceptors make an attack every 3 seconds. The Interceptor deals 2 attacks each dealing 5 damage and there can be a maximum of 8 Interceptors. This results to the following:
To fire all 8 Interceptors: 4 seconds.
Attack CD of all 8 Interceptors on their "fire round": 3.5 seconds. (+0.5 second launch time)
The first attack made by the Interceptors in a newly engaged battle, therefore, is 80 (total damage) / 4 (4 is larger than 3.5) = 20.
The following attacks made by the Interceptors after they have been fired (assumes none are killed): 80 / 3 (no need for the additional +0.5 second launch time): 26.6777777r.
Carriers do more damage if they'rekept alive than the moment they engage. Alternatively, you can attack some of your own units to launch the Interceptors and then engage, much like the Void Ray can charge it's Prismatic Beam on allied units.
|
Haha, forget the best looking unit :D
Updated it and corrected Carriers data.
|
Er. Shouldn't the cost also include the unit's cost in supply? A zealot, for example, actually costs 120. Skews the numbers a bit.
|
Archons?
Edit : anyway, what's bothering me is how much terran units have a lot more versatilty (+ damage / ground-air / spash )
|
On September 02 2010 05:56 Grummler wrote:Show nested quote +On September 02 2010 05:31 ZaaaaaM wrote:On September 02 2010 05:22 Grummler wrote:
Theorycraft of its finest. Do you think its a good idea to ignore: hitpoints, movement speed, fire rate, fire range, unit size, unit tech, unit abilities, flying/not flying, shoots air/ doesnt shoot air and synergies?
Edit: Just in case: Yes, he did ignore fire rate. I know what dps means, but its not the same if a unit deals 1000dmg each 1000 seconds or 1 dmg each second. Think about it. Pff, don't try to be the smartguy- we know we know, don't have to "think about it" either lol So, you have read my last sentence. Awesome. Or maybe you read the rest too, but dont want to discuss about it, cause trolling is so much easier. Show nested quote +On September 02 2010 05:53 neobowman wrote:
Starcraft isn't a DPS game. There's way too many different factors like splash damage, unit speed, range, mineral/gas ratios, etc, etc. I guess you could learn some stuff with the DPS but in general, not as useful as it could be in games like WoW or something. This.
You don't say! Just because it isn't the end all be all of analyses doesn't mean that it cant be helpful. I don't recall reading the OP say that this is the only thing you need to look at to completely understand SC2. There's a whole lot more going on, of course, but this analysis can help to put things into perspective. There's a whole lot of unfounded negative sentiment to "theorycraft". How do you think your computer is engineered? How do you think the car which you drive is designed? It starts with a whole lot of theory. Don't take the theory as gospel, realize that there are simplifying assumptions which must be made, but that doesn't invalidate theory completely.
|
OMG, thank you guys for this thread!!
I'm a zerg player, who has been having a lot of difficulty figuring out how to win in economy/army, and failing game after game after game. I could not figure out how to keep my economy competitive, while having an army that can handle both harass and timing pushes. After seeing this chart, I feel a lot more confident, and my game play has actually improved!
My very simple and stupid conclusion was to build cost effective units which have high dps -- that just happens to be zerglings/hydra, and holy shit does it work well.
I open with 14 gas 14 pool, zerg speed + queen, then when queen finishes I lair, add a second gas, max out at 22 drones per base, pump more lings, then transition into hydras. Holy hell am I impressed with this build.. Those hydras absolutely MELT everything toss throws at me, and the zerglings seem to be more cost effective tanks than roaches.
Again, thanks for this thread! I'm so happy :D
/edit Build works fine against terran, as well, except use upgraded banelings + muta instead of hydras :D
|
On September 02 2010 06:19 ShadowReaver wrote:
Nice work here, some stats are interesting. Would be cool to add hp analysis to this as well.
I tried to add hp, range and armor.
So I created the value dps effectiveness, calculated by dps * sqrt(range).
I did this because the difference between 1 and 2 is much bigger than 8 and 9. So this should work approximately.
But this doesnt punish Zerglings more than Zealots for being melee.
In addition I created the value health effectiveness depending on how much an attack is reduced by 1 armor. but Im not sure if I should just add protoss shields to their normal life.
In the end I wanted to add dps effectiveness and health effectiveness to one value like:
dps eff. + (health eff. / x) and divided this by the units cost.
But hard to say which value x should have. Maybe 10?
|
On September 02 2010 18:17 clickrush wrote:Show nested quote +On September 02 2010 18:05 Nihilnovi wrote:
lol at probes having more DPS / cost than stalkers, haha if you want to understand this chart, then its important that you compare the dps/cost with the durability and speed of the units also. protoss units have a shitload of HP+shields. their dps/cost is therefore very low in comparison. the only units that deal higher dps/cost have major downsides, like the slow marauder, the slow voidray that needs to charge and the slow colossus that can be hit by AtA and GtG. roaches, marauders and hydralisks have letz say medium durability, so they have average dps/cost. marines and lings have very low durability ->very high dps/cost. the chart also shows how powerfull stim is. the stimmed numbers really pop out.
SHITLOAD OF HP+SHIELDS!? LOLZ.
Probe is 50/0 for 20/20.
Stalker is 125/50 for 80/80.
4 Probes have the same same HP + Shields as 1 Stalker (80/80).
4 Probes cost 200 Minerals and 0 Gas. 1 Stalker costs 125 Minerals and 50 Gas. This means that at a valuation of 1 Gas as 1.5 Minerals Probes give you the SAME HP and SHIELDS and MORE DAMAGE.
Now, I'm not saying this is proof of Stalkers sucking. Because, hey you can kite those 4 Probes all day long and never take any damage so Stalkers win. But of course, there is something seriously wrong with the fact that I HAVE TO KITE PROBES WITH MY BASIC RANGE UNIT OR ELSE I LOSE.
The truth is Protoss don't have a stable basic range unit. They have a cutesy harass ranged unit (Stalker) and a cutesy spell casting ranged unit (Sentry). And this is fine against ground, because hey just Zealot/Templar it up. It doesn't really matter, because it's designed to be melee heavy, so cutsey stuff for ranged is fine.
But it is a serious issue against air units that Protoss's ground to air unit is almost worse than flying Probes. And it's only worse if there is +2 armor and they are armored.
|
DPS isnt the best way to measure units in SC. What matters is how many hits it takes to kill your target. Also you cant just say that gas = 1.5 minerals bc it doesnt and theres no scientific way of proving it as that.
|
Wow, turrets are about 2x more cost-effective than Spore Crawlers
|
if you factor in unit range / unit durability, terran is so strong
|
On September 03 2010 05:22 whateversclever wrote:Show nested quote +On September 02 2010 18:17 clickrush wrote:On September 02 2010 18:05 Nihilnovi wrote:
lol at probes having more DPS / cost than stalkers, haha if you want to understand this chart, then its important that you compare the dps/cost with the durability and speed of the units also. protoss units have a shitload of HP+shields. their dps/cost is therefore very low in comparison. the only units that deal higher dps/cost have major downsides, like the slow marauder, the slow voidray that needs to charge and the slow colossus that can be hit by AtA and GtG. roaches, marauders and hydralisks have letz say medium durability, so they have average dps/cost. marines and lings have very low durability ->very high dps/cost. the chart also shows how powerfull stim is. the stimmed numbers really pop out. SHITLOAD OF HP+SHIELDS!? LOLZ. Probe is 50/0 for 20/20. Stalker is 125/50 for 80/80. 4 Probes have the same same HP + Shields as 1 Stalker (80/80). 4 Probes cost 200 Minerals and 0 Gas. 1 Stalker costs 125 Minerals and 50 Gas. This means that at a valuation of 1 Gas as 1.5 Minerals Probes give you the SAME HP and SHIELDS and MORE DAMAGE. Now, I'm not saying this is proof of Stalkers sucking. Because, hey you can kite those 4 Probes all day long and never take any damage so Stalkers win. But of course, there is something seriously wrong with the fact that I HAVE TO KITE PROBES WITH MY BASIC RANGE UNIT OR ELSE I LOSE. The truth is Protoss don't have a stable basic range unit. They have a cutesy harass ranged unit (Stalker) and a cutesy spell casting ranged unit (Sentry). And this is fine against ground, because hey just Zealot/Templar it up. It doesn't really matter, because it's designed to be melee heavy, so cutsey stuff for ranged is fine. But it is a serious issue against air units that Protoss's ground to air unit is almost worse than flying Probes. And it's only worse if there is +2 armor and they are armored.
Use your damn Pheonixes. They can hurt a lot, if you don't believe me watch Socke. ( Or was it Nani...)
HuK also transitions to Void Rays with +1 attack in the late game from standard play, which turns out to be a pretty effective strategy as Void Rays do massive damages and become nearly unstable in a mass.
You're complaining that Toss have no stable ranged unit? You're trying to weigh the Stalker by comparing it to probes? I'm sorry that the Stalker is nothing like the marine, I sincerely apologize.
|
rolleyes at your apology. you can keep it. i'm not asking for you goddamn sympathy. i'm trying to point out a freaking issue with the game. if the basic ranged unit is the EQUIVALENT OF A PROBE, then there are some issues here. take your head out of your ass. i sincerely do not apologize.
|
And yes, I do know that Phoenix and Void Rays are awesome. Is there some goddamn point to that statement? Battlecruisers are awesome for Terran... I GUESS THAT MEANS EVERYTHING IS GREAT BECOMES THESE COMPLETELY DIFFERENTS UNITS ARE GREAT. Astounding logic.
|
Comparing probe´s dps per cost with the stalkers isnt good... You always prefer a bigger unit because its dps isnt reduced when a probe dies.
Better compare the Stalker with the Marine and / or Marauder.
|
On September 02 2010 17:12 dybydx wrote:Show nested quote +On September 02 2010 05:45 Buddhist wrote:On September 02 2010 05:42 FabledIntegral wrote:Aw the first thing I looked for was to compare Zergling vs Adrenal Zerglings but you didn't include it Adrenal should increase dps by 20%. I think it's bugged or something though, according to some post on TL. cracklings arnt bugged in SC2. but.. in BW, the upgrade increase attack speed by 45%, where as in SC2, its only 20% increase. the tooltip description did not lie, just ppl assume the upgrades were the same, since the costs were same.
The SC1 upgrade is a 33.(3)% attack speed increase(8 to 6 frames).
The SC2 upgrade is a ~18.6% attack speed increase(0.696 to 0.587 seconds).
|
Would it make sense to update this table with partial supply cost? I mean theoretically 1 thor is about equal to 5 marines comparing the DPS and consumed minerals/gas (don't speculate now about the value of gas) The idea is different supply consumption which make marines little less expensive. Maybe this is not the best example, just want to hear other opinions about it.
|
kim carrier gonna flip out
|
Agreed with a post earlier. While this is a very interesting graph and what not. In starcraft there is so much kiting that DPS really isnt the factor. It will simply come down to amount of hits to kill.
With this information it gives you a little better look at how much you need vs other builds to be viable.. while the DPS is really just another thing to theory craft about.
|
Can we call this A-Move theory? Because that is the only time this matters is a-move versus a-move in an open space...
|
I can't believe some of these "hur hur, OP doesn't take everything into account, hur hur" posts. Come on. For something as complex as Starcraft, it's standard scientific practice to make a bunch of assumptions about the effects of other phenomena on the thing you are trying to analyze, and then proceed with the analysis directly. Usually, you would state these assumptions to make it clear, which he doesn't do here, but still. These numbers *are* accurate for a single unit, already in range of a target, where the target has infinite health, and assuming the damage output is continuous instead of discrete. Sure, these assumptions aren't valid in a real game, but that doesn't change the fact that the information might still be useful...
|
Tank = 14.42
Sieged Tank = 16,67
IMO this looks just stupid, yeah there is splash but still.
|
On October 28 2010 00:58 HwangjaeTerran wrote:
Tank = 14.42
Sieged Tank = 16,67
IMO this looks just stupid, yeah there is splash but still.
It's a well-known fact that a small number of unsieged tanks beats a small equal number of sieged tanks.
|
On September 02 2010 08:09 ZlaSHeR wrote:
Terran also has the lowest hp per food units, and has the slowest production time overall in a long macro game.
Useless statistics are useless.
sorry, off the top of my head i can't think of which T unit has the lowest HP. and Slowest production time. afaik. It's the Hydra with 80 hp for 2 food. Even the infestor has more hp at 90.
ALSO afaik, the Thor is the fastest production of food at 60s for 6 food.
So remind me which unit has the lowest hp for food for T? ty
EDIT: typ
|
On October 28 2010 00:58 HwangjaeTerran wrote:
Tank = 14.42
Sieged Tank = 16,67
IMO this looks just stupid, yeah there is splash but still.
The tank unsieged also attacks much faster, but has shorter range....
And nowadays it is weaker v unarmored while sieged
This chart is pretty out of date... would love to see it updated :D
|
On October 28 2010 01:04 Aquilix wrote:Show nested quote +On October 28 2010 00:58 HwangjaeTerran wrote:
Tank = 14.42
Sieged Tank = 16,67
IMO this looks just stupid, yeah there is splash but still. The tank unsieged also attacks much faster, but has shorter range.... And nowadays it is weaker v unarmored while sieged This chart is pretty out of date... would love to see it updated :D
Yeah I wonder if it would be better to just kite with your army and tanks than sieging them.
|
Ehh... you can't really use DPS to determine anything lol. There's sooo many other factors that go into the game. What about unit size? Since marines are smaller they clump easier and are more susceptible to splash damage. Zealtots with charge will eat all of the splash damage for other units and also shoot it back into your own army...
Sentries aren't actually mean to shoot at things so much as theyr'e meant for the spells, but the forcefields allow the zealot DPS to take full effect as well as to split armies in half, yet they're factored into the average like it doesn't make a different...
Lol.. what about Psi storm dps?
I could make a huge list of this stuff...
|
As a zerg, 4 drones will mine 150 minerals in the same time you can mine 100 gas with an extractor.Not counting extractor build time+cost. But its nothing like 3:1 just because you have 3 times the gas.
Also, you should get DPSxHP/Cost. DPSxHP is basically who wins in a 1v1 fight/ or the general "firepower" of the unit. If you army DPSxHP>his yours will typically win all other things equal.
|
dps in army battles really isnt a factor, its more of how many of X unit shots does it take to kill Y unit and in what amount of time. The term "critical mass" normally refers to a amount of units that once reached will normally start killing units instantly. EX: enough siege tanks that the entire radius of the splash is a kill zone, or enough colo that each time they fire 4 or more units always die.
There are way to many factors in game that will alter dps statistics drastically, esp when u start to factor in overkill and positioning
|
Charged Void Ray DPS vs. armored is still crazy, even post patch. Given that they can fly, hit both air and ground, now do pretty decent damage uncharged vs. armored (almost exactly the same as 2 stalkers, for similar cost), and with the speed upgrade absolutely wreck vikings...
I think its safe to say all the "Voids are useless" QQing was overblown. They aren't an instawin if unscouted anymore, but they are still very, very dangerous
|
On October 28 2010 02:03 NJO wrote:
dps in army battles really isnt a factor, its more of how many of X unit shots does it take to kill Y unit and in what amount of time. The term "critical mass" normally refers to a amount of units that once reached will normally start killing units instantly. EX: enough siege tanks that the entire radius of the splash is a kill zone, or enough colo that each time they fire 4 or more units always die.
There are way to many factors in game that will alter dps statistics drastically, esp when u start to factor in overkill and positioning
Of course its a factor. It's just not as accurate due to overkill. Its still a big factor, but splash and damage per volley does affect it, splash a lot more.
|
Good job!
Can you add the correlation between HP/cost? Maybee you can add the HP-efficency [damage/s/HP]
to your data ? One of them to get the rest would be awesome!
|
Updated and added efficiency per cost that I created some months ago.
|
Good stuff! however I think it is a mistake to not factor supply cost into the equation. 1 supply for each race = 20 mins so its easy to still make cross-race comparisons.
I think this is the part of the reason why low supply units (including workers) come across as so strong in your chart. 1 marine going from 50 minerals to 70 minerals is going to cut efficiency by 29%, but a muta going from 100/100 to 140/100, at 1 min = 1.5 gas, goes from 250 cost to 290 cost, which only cuts efficiency by 14% ... which helps narrow the actual single-target dps between them.
Because terrans tend to have more low supply units compared to, say, Protoss, is a big part of the reason why terrans come out so strong in the overall race v race comparison imo. Plus their low supply units are weighted more; marines & mauraders get 2 entries each, and hellions get 3. I can kinda understand that Zerg should be lower vs p/t, because Zerg don't need to spend additional resources on production buildings, so their per unit costs are higher to compensate.
Note that this adjustment does NOT tell you which units are best to make at max supply, it simply reveals the units true mineral cost.
Oh, and another bug, for the stimmed units, you forgot to reduce to reduce the hp of the unit stimming. so stim marine needs -10 hp and stimmed maurauder needs -20 hp. it would be also nice to have marines both with and without combat shields. =)
|
Sounds well but why 20 mins?
Also its still a difference between losing 1 larvae, time to build depot or just walking to pylons position.
Soon updating values for stimmed units.
|
Oh, I also feel that making the "range dps modifier" just the sqrt of range is a bit iffy. That would imply that a unit with 4 range would have double the effective dps of an identical unit with 1 range. I think ^(1/4) would be more appropiate, to tone it down from 2x to 1.4x.
Alteratively, you could factor in speed as well, which serves the same purpose for melee and clumped units, by multipling dps by the sqrt of speed too. That might be a pain for Zerg units though.
|
err yeah my bad. Not 20 mins, but 100 minerals divided by 8 supply (for just depots/pylons/overlords, ignoring other sources of supply) so 12.5 minerals per unit supply. Its true it doesn't count include opportunity costs like mining and larva, but those are going to be so variable I don't think there is an easy way to quantify them. Plus, then we would need to go down the rabbit hole of valuing opportunity costs per production building, which would be especially problematic for Zerg, and very variable based upon the size of your economy and technology.
|
On October 28 2010 05:49 RoboBob wrote:
Oh, I also feel that making the "range dps modifier" just the sqrt of range is a bit iffy. That would imply that a unit with 4 range would have double the effective dps of an identical unit with 1 range. I think ^(1/4) would be more appropiate, to tone it down from 2x to 1.4x.
You mean having range 9 is just 1.73 times better than range 1? Maybe I shouldnt upgrade thermal lances before I already have 3-4 Colossi :D
If we are talking about only 2 units figting each other you could be right, but when we are trying to get relevant data for a normal match, sqrt fits much better I guess.
|
For the purpose of this chart, yeah it would make sense to delay the range. Sure you can micro 1 range 9 colosses to kill several range 7, and 1 phoenix can kite infinite mutas, however both of those situations depend heavily upon a specific unit matchup. You might as well say Vikings do infinite damage just because colossus can't hit them, and banshees do infinite damage to roaches.
Any chart like this needs to be valued like a unit tester in order to have any relevancy at all. The wonderful thing about this chart is that it can get you approximations without requiring you run a ton of matches via mapeditor for every unit matchup, which is going to have little relevancy anyway because its pretty rare to have a pure unit composition
|
It is balanced by the fact that zerg needs to spend least on technology buildings and has the best mobility while the opposite is true for terran.
|
On October 28 2010 11:31 minithor wrote:
It is balanced by the fact that zerg needs to spend least on technology buildings and has the best mobility while the opposite is true for terran.
About the tech cost, I hear this all of the time but its not really true. It depends on the number of bases, saturation, and unit mix being produced (or perhaps rather desired unit mix). As a general guideline, the fewer the bases and earlier the tech the more expensive it is proportionately for zerg. At least I ran this between zerg and protoss and in some instances your claim was correct, but in others it swung the other way. Thats why all of these racial generalizations (damn I sound so pc) dont really apply to actual game situations.
|
|
|
|
|
|
EPT
