|
On December 02 2010 05:17 AllNight wrote: I love the financial approach. But I think it's more complex than that and common sense is just in order. Just always mule when you don't need scan.
You need more than common sense in order to use MULEs/Scans effectively at the highest echelons of play.
- - - - - - - - - - -
A MULE has a fixed cost of 50 energy and 270 minerals over 90 seconds. It's important to note that it's never "270 minerals" and is always "270 minerals over 90 seconds", because a MULE does not create more minerals than what are already on the map. A MULE is an injection of income, not an injection of resources. The real value of a MULE is it allows you to do more things sooner; what it does not do is give you the ability to do things you never would have normally been able to do in game (IE creating minerals from thin air).
|
I'm pretty sure a Scan actually costs 8 supply.
|
When used correctly, the scan really doesn't cost 270 minerals at all. It's only a waste when you use it and have gained no information, or use it at an inappropriate time (the first 50 energy of the OC). When used successfully, the scan is often able to reveal game changing information, which is much more valuable than 270 minerals. Most of the time when your econ has stabilized, I'd say scan is more useful than Mule. It's better to constantly know the location of your opponents army than use the mule to gain more minerals that you don't immediately need. But when you're down on minerals, and you really need that extra boost, using a scan instead of a mule would definitely be the equivalent of losing 270 minerals.
|
i think this is pointless.
scan = 270m/90s = 8 supply
50 energy = 270m/50s = 8 supply = scan it all depends on the situation on what you need. if you're supply blocked and have 1000 minerals, you're not gonna use mule, you'll drop a supply. if you have dt in your base, you're not gonna drop a mule, you'll scan. if you mess up, say mule then a dt comes into your base and unprepared, you're done.
in the end, its just 50 energy and what you do with it are all equally valid. you're not sacrificing the others for one, you're picking one advantage to have(at particular situations). same with chrono boost, you're not "delaying" wg for probes or the other way around, you're gaining an advantage in one area.
|
On November 30 2010 21:21 eluv wrote:Show nested quote +On November 30 2010 20:53 Nycaloth wrote: maybe T players rely on the mule too much as an income source and therefore get less workers then the other races and need the mule to stay even and that is where the 'cost' mentality stems from. It's not really a choice - Terrans can't actually produce workers as fast as either of the other races on equal bases. Terrans can build more CCs than they have bases, reasonably, because they can float them into position later. Protoss can't. (Of course, this is expensive, and it will take a while for the extra workers/OCs to pay for it... and if Terran expands exactly like Protoss and does not use OCs, they'll be in a shitty position.)
Zerg will always have the fastest potential rate of workers since each Hatch + Queen can produce ~twice as fast as a Nexus that's dedicated to Chronoboosting itself. Hence the general understanding that Zergs need to be pressured so they won't just mass drones (or drone/queen).
|
I see horrible logic. And dead people.
First of all, how you use the scan has no bearing on what it costs. The cost is irrelevant to how it works. You exchange something for a mule, whether or not you gain anything from the something has no relation to your investment. Whether the drop supply or scan helps you is how you measure how good was the investment and if it was worth it, not how big the investment was.
Now, on to the cost. Lets make some assumptions: * The time value of money holds. This is basic finance theory, meaning it is better to get an x amount of money now than get them some time in the future. This means you cannot equate a 90s stream of income to the exact notional value of the total income. The value of your investment will always be lower than the minerals you receive in the future, so the cost will never be 270.
I have actually thought about the concept in SC2 before, and it would actually be interesting to get a rough idea of how much less minerals are worth one minute from now. Imagine you are given on option to spend 100 minerals 5 minutes into the game. How many minerals would you want back at the 6th minute? The more you want in exchange, the less the scan will actually cost.
* Since the mule is such an awesome thing, your opponent will want to kill it as fast as possible, valuing killing it over other workers. This means you cannot expect to always get the full amount of minerals. So you have to factor in the 270 minerals by some sort of probability that it will stay alive for 10 seconds, 20 seconds, the whole 90 seconds, etc. Lets say on average, the mule gets killed 45 seconds into mining one out of every four times you cast it. This means that a scan, on average, will net you 236 minerals.
Disclaimer: all views derived from traditional finance theory. I think it describes what happens in SC2 pretty well.
I think a reasonable approximation might be that a scan instantly costs around 200 minerals or a bit less, on average. When there is a banshee harass going on and the mule would be instantly killed anyway, it might cost 0 minerals to scan, or, alternatively, when you don't care if you get any minerals a minute from now (for instance, because you're going to lose/win even if you get the mule), it might also cost 0 minerals to scan.
Obviously the 200 or whatever minerals are worth way more early game than late game when they represent a smaller percentage of your income/resources, so the investment decision is more for mules early game and less late game.
|
Think of it in terms of opportunity cost: If you feel that you know what is going on and there is no real threat of cloaked units scan is better. If you do not have sufficient knowledge of the opponents strategy or you're facing cloaked units scan is a better choice.
I think the energy on the OC is very similar to the larva mechanic, you have to make a conscious choice when deciding whether MULEing (droning) or scanning (making troops) is more beneficial.
|
A scan doesn't cost 270 minerals. In a game that goes to where the entire map is mined out, you are not down 270 minerals. Either way, you will eventually get that 270 minerals, you only got the minerals faster. However, it does cost you 100 minerals if you do not use the supply drop. Simple logic guys.
|
11 pages of a dumb semantics debate?
i've seen more sense on the battlenet forum.
|
well, if a scan is 270 minerals, Every zerg buildings cost infinite minerals? Those drones could of stayed mining for ever =/
|
On December 02 2010 08:58 facefart wrote: well, if a scan is 270 minerals, Every zerg buildings cost infinite minerals? Those drones could of stayed mining for ever =/
Get ready to be flamed by people that think that you could just rebuild that drone and not have n+1 drones.
|
It is all semantics and it depends on the in-game context. As someone pointed out 50 energy=1scan=270min/Xsec=8supply. Saying that scan costs 270 minerals is like trying to determine cost of the forcefield in hallucinations. Of course you can do it, but does it make sense or is useful terminology ? Sometimes it can, but as a blanket stytement it is useless and baseless.
|
I have actually thought about the concept in SC2 before, and it would actually be interesting to get a rough idea of how much less minerals are worth one minute from now. Imagine you are given on option to spend 100 minerals 5 minutes into the game. How many minerals would you want back at the 6th minute? The more you want in exchange, the less the scan will actually cost.
Frankly, this weighing this is a good portion of the strategy in Starcraft. Like...the entire game. Seriously.
---------
This is a pretty pointless thread that really just exists for people to continue to stream really bad comments nobody else cares about. So let me piss my 2c into the pot.
Forget about minerals in total. You have a rate per minute. A mule will add 180m/minute.
MR = Mule Rate (Per Minute) = 180m ET = Minutes to get 50 Energy / 1.5 (we divide by 1.5 because a mule lasts 1.5 minutes)
The Mule Income Bonus then is: MIB = MR / ET
I'm not sure what the real value of ET is. Let's say its 90 seconds exactly, that means that mules provide terran with a constant income stream of 180m/min.
I don't know the exact income rate a fully saturated base has - but I think its around 800m/minute (100 m per patch per minute), meaning that mules provide an income bonus of 22.5%/ET -- if it takes 90 seconds to get 50 energy (ET = 1) - then terran recieves 22.5% more income than any other race. Thats tremendous. Seriously - even if ET is 1.5 its still a 15% bonus to income (in bursts - which gives you that "money now > money later" bonus as well). Pretty insane, and if neither base is saturated the percentage increase in income is greater. Amazing.
Now ... the word Cost. If scanning "costs" you money or not depends on how you mentally categorize the Mule Income Bonus. The word cost is somewhat subjective here.
If you think of MIB as something that you are always entitled to when playing terran, then it yes, there is a 'cost' in income to use a scan.
If instead you think of MIB as a free bonus - then it doesn't 'cost' you anything but energy to opt for the scan.
So you can think of it as costing you something or not. Just depends on how you like to categorize it - as a free bonus (no cost in income), or as a given (yes cost in income).
For what its worth, because MIB is constant (assuming there is a CC next to a mineral patch) I personally think that you should count it as a cost. Further, I think that people who consider it an 'opportunity cost' are somewhat technically correct, but because MIB is constant I think it is more accurate to be considered a part of base income.
And in conclusion....holy shit, I had no idea mules were so imba - even if ET is 1.5 - 120 m/minute is 15% extra income which is insane. Even if ET hovers around 2 - thats around 90m/minute extra. In a 10 minute game thats 900 minerals. Thats insane, and I think its probably closer to ~1,200 mineral advantage in a 10 minute game -- amazing. I need to figure out how to get a hell of a lot more value out of chronoboost... 900 minerals is 6 extra gateways -- I really doubt that I'm getting 6 extra gateways worth of value out of chronoboost.
Also -- please -- let this thread die, or if people continue pissing random opinions just keep linking to this post in response. If you took the time to actually understand the above it'll actually teach people something about starcraft (as well as how to like...think... in real life ;D)
EDIT: I can't get over how valuable mules are...does anyone know how long it takes to get 50 energy?? I looked over the logic above a few times and it looks right...which ... i sorta can't believe - did blizzard not do the math here?
EDIT #2: I think that this might very well be the real reason that terrans early game is completely out of control. Someone should look over this (someone smart) and see if I messed something up. I'm really shocked.
|
You can argue that MULES cost ~270 minerals and I wouldn't complain.
However, defining scans as a cost is a little more murky.
Scans COST 50 energy. They don't cost 270 minerals. You can compare that 50 energy is worth 270 minerals, which is why people make the analogy that scans also cost 270 minerals.
How do you define energy cost? Well, there are three outputs for OC energy, and I will examine all three of them with respect to min/energy.
MULES are perhaps the simplest to define. Optimally on a blue mineral patch, they will harvest 270 minerals within 90 game seconds. Safe to say that the 50 energy used will return 270 minerals. However, you must also account for little discrepancies such as the common debate "money now is worth more than money later" (which I agree with wholeheartedly) and the fact that the amount of minerals in a patch will deplete more quickly. All in all though, it's safe to say for the MULE ability, 50 energy ~= 270 minerals.
Supply drops get a little more ambiguous. Simply put, they can be worth 100 minerals for the cost of another supply depot. Definitely not worth it compared to a mule, right? Which is why nobody uses them. However, you must also take in the fact that it delivers 100 minerals INSTANTLY in the form of supply. If you were supply blocked and tried to build another supply depot in time for your units to come up, but he's already pushing up the ramp, wouldn't it be better to instantly redeem that 50 energy for 8 supply? Also other little discrepancies such as building space.
EDIT: People get a little confused with my supply drop analysis. Let's say you have 100 minerals and 50 energy. You need to create a supply depot and a bunker. You can either wait for the mule + scvs to mine that amount (which will probably take 12ish seconds, give or take), or you can supply drop and create a bunker simultaneously. It's the idea of instantaneous redemption of mineral/energy rather than improved income.
Now we come to the bulk of the discussion: The cost of scans. Like I stated before, the basis that scans cost 270 minerals isn't accurate, but rather we should look at in terms of OC energy.
The thing about scans is that their worth is very varied. They give absolutely no real material gain that can be recorded by the player. They have one purpose, they convert that 50 energy into information. Whether it be scanning to destroy banemines, scouting opponent tech or destroying cloaked banshees in your base, assigning a mineral 'cost' to scans is inaccurate in my opinion. In this regard, I disagree with OP.
However, scan is nowhere close to being free. We have already seen that 50 energy can be easily redeemed for 270 minerals, which is why you should make your ENERGY count.
If you so desire a formula for the mineral worth of a scan, then try using this pseudo formula. Worth of Scan = x + y + z Where x is equal to the mineral cost of units that have been saved by your scan (ie. moving up with your force, scanning ahead to seeing a massive amount of siege tanks covering your approach and falling back), where y is equal to the mineral cost of enemy units that have been killed by your scan (ie. cloaked banshees or dark templars), and where z is equal to the mineral cost of information that has been gained from that scan. (LOLWUT?)
Can you see why giving scans a material worth can be so hard to accomplish?
tldr; Scans don't cost 270 minerals, they cost 50 energy. 50 energy can be worth 270 minerals (in the case of MULEs), 100 minerals (Supply drop), or an undefined amount of minerals (Scans).
|
ok but, not using the scan and letting your base die to dts is even dumber so I just dont get all this mule talk.
|
On December 02 2010 10:45 TheGiftedApe wrote: ok but, not using the scan and letting your base die to dts is even dumber so I just dont get all this mule talk.
Hence why in my formula, Worth = x + y + z, X would be a very larger number (the total mineral sum of your base structures) and Y would be the cost of the DTs, which obviously shows that the scan is worth much more than the mule at that point in the game
|
is MULE really free or does it cost anywhere from
-extra supply -valuable, valuable, valuable scouting information -valuable, valuable, valuable detection from DTs
are SCVs merely 50 minerals or do they cost 2 marines, one for the cost and one for the supply?
i also don't like how you phrased it. the words "free" and "cost" implies dangerous assumptions. did you HAVE to get that mule, and because you didn't, it "cost" you 270? as a protoss, if i chrono anything but a nexus, am i costing myself mining time? what if chronoing something else is more valuable than mining time or minerals? terran has the advantage of flexibility, so saying scan COSTS 270 is like saying you HAVE to get the mule and NEVER anything else. what if a scan is worth more at the moment?
|
if a scan is worth 270 minerals then any terran building besides the cc costs infinite minerals cuz the cc can be turned into an OC which can drop infinite mules
also, you say the drone can be remade? yeah well you can sit there and wait for your energy to hit 50 again too
|
On December 02 2010 07:04 jinorazi wrote: i think this is pointless.
scan = 270m/90s = 8 supply
This is correct
When your orbital has 50 energy you have 3 choices, Scan, Mule, Supply. When you choose one you LOSE/Sacrifice/whatever you want to call it, the other two options instantly until you get another 50 energy. That is opportunity cost.
Its basic econ 101.
I think people are getting confused because they think when you can scan you have 270 minerals subtracted from your current mineral count which of course is wrong, but over 90 seconds you have 270 less minerals than you would of had if you had muled, so in that sense you "lost" 270 minerals in those 90 seconds.
Also, no one is saying you should mule with every single 50 energy, early in the game you should mule, almost always with the first 100 energy but as the game goes on, the relative value of 270 minerals decreases and the value of information (scans) increases.
Its also understood that you aren't getting 270 minerals out of thin air and that each expo has a limit which will be mined out but in what games is the entire map mined out? Basically none, so that is 270 minerals you wouldn't of otherwise had since once you mine out one expo you go to the next one, the Mule will always have 270 more minerals in a given time period over not muling (until you mine out the entire map which almost never happens)
And as for zerg drones costing infinite minerals...well first off there aren't infinite minerals on the map, you are trading that mining time/minerals would of been accumulated *faster* for a building that has a value to you. In effect, slowing your economy to produce a building, kind of like a scv losing mining time while building, and the time it takes for a probe to reach the position of the building and warp it in and back to the mineral line. Obviously in this regard, zerg has the highest cost to produce a building than any of the other races, this is partially balanced out by the fact that they have to make the fewest buildings of all the races as they don't need unit producing structures since the units spawn from larva.
|
Frankly, this weighing this is a good portion of the strategy in Starcraft. Like...the entire game. Seriously.
Hi, Captain Obvious? I was talking about a numerical estimate, how many minerals would 1 mineral 1 minute later be worth now. 0,9? 0,8? 0,7? Etc. This would also help estimate how much not using a mule "costs", among other things.
Anyway, since everyone just spouts out whatever they "think is right" without thinking about previous arguments, this thread has indeed turned into a dead end.
|
|
|
|