|
I have noticed that most pros produce more workers than their bases can support. While I understand that it is to transfer when another base is done, I have three questions.
1) Why is it better to produce workers before the CC/Nexus/Hatch instead of cutting workers, building the building, then restart making workers?
2) Why is it better to oversaturate rather than long distance mine? From what I understand, having more than 17-18 workers on minerals give less minerals than having those excess long distance mine.
3) Why is it better to split the workers evenly instead of just transferring the excess (above 16-18) and rallying new workers to the new base?
Any advise would be appreciated.
|
acctualy it takes 3 workers to saturate a miner patch, which brings it to 21 plus 6 in gas to saturate an expansion.
|
Okay, but I am not talking about the numbers, but rather the order of building,training,moving,rallying etc. And I think it only takes 3 for the further minerals? I understand that theres a tiny gap in between the workers if there's only 2 workers on the patch, but I think long distance mining gives more than using the drone to cover the tiny gap. I remember a thread where someone did the experiments and had it graphed out and all, but I can't find it >.<
But even if its 3, just change the numbers in the original post to whatever number you think full saturation is at.
|
1) Sometimes they do cut workers for a bit if they can't get up a new base very easily, but ideally you want to be able to fully saturate your new base as soon as possible. You generally can't do that if you cut workers for a bit and resume worker production when your third comes up. The answer to this question is a bit more ambiguous with Zerg since you have the ability to produce so many more workers at once.
2) Bases are saturated 24 workers on minerals and 6 on gas. Distance mining will only result in increased income if you have over 60 workers on two bases. You also have to take into account the fact that you now have to defend those workers. If you don't have a third up then it should be because you are not yet prepared to defend it, so sending workers out to distance mine puts them in danger.
3) You don't want your main base to mine out too fast. If you mine out your mine too quickly then you are put in an awkward situation where you need to obtain a fourth before the other player to stay on even income. This spreads you out even further and leaves you vulnerable to attacks since you still have to defend your main even though it is not giving you any income. The less area you have to defend, the better.
|
1. Because then you'll have extra workers faster. Zerg usually don't oversaturate as much because they can make all the drones at once later, which is safer.
2. It's dangerous
3. No real reason actually, in SC2 it's more of a preference thing. It can be risky to constantly have workers moving between your bases though when they become more spread out.
|
On November 30 2011 02:22 Imbaman wrote:
I have noticed that most pros produce more workers than their bases can support. While I understand that it is to transfer when another base is done, I have three questions.
1) Why is it better to produce workers before the CC/Nexus/Hatch instead of cutting workers, building the building, then restart making workers?
2) Why is it better to oversaturate rather than long distance mine? From what I understand, having more than 17-18 workers on minerals give less minerals than having those excess long distance mine.
3) Why is it better to split the workers evenly instead of just transferring the excess (above 16-18) and rallying new workers to the new base?
Any advise would be appreciated.
1. Unless you're on 3 command structures, you can't build enough workers to saturate a new base after you've started building it. Because you are weakest after expanding but before mining from that expansion, when you are expanding you want to be as close to saturation as possible when the expansion finishes to minimize the window between building it and hitting your economic advantage. So, you need to continue building workers.
2. If you aren't risking the loss of workers, having anything over 2 per close patch and 3 per far patch sent to long distance mine makes sense. In-game situations tend to make this dangerous and the risk of losing those workers outweighs the small benefits of long-distance mining.
3. Mining near patches at the new base is better than mining far patches at the old base. Other than that, I don't see a reason to do it. I do it because I've seen pros do it, but now that I'm thinking about it you're right and I'm going to stop doing it. I'll drag excess or enough off far patches to hit the near patches at the new base and then rally over.
|
Thanks for all your points. I want to seek some clarifications though.
1) Is it possible that the advantage of having the building finishing earlier outweighs the fact that it is less saturated when finished i.e. only have 70% saturation when finished, but finishes a minute earlier or something? This new structure can then start producing workers earlier too.
2) Other than terran who can build the CC and float it down, wouldn't the player have already set up defences to defend the structure being built?
3) Wouldn't having less mining bases be less area to defend rather than more?
In case of any wrong perception, I am not trying to say that I am right and the pros are wrong. I just want to understand why do what they do even though it seems unintuitive to me.
|
The reasoning I see behind it is two-fold.
1) Not-so-hypothetical situation: Say you have 21/24 workers on minerals (saturated), and a ling run-by happens; or a drop in the mineral line; or phoenix harass. Suppose they kill 10 workers before you kill / drive them off. You're now down to 11/14 workers (roughly 60% saturation).
Now you have 31/34 workers. They kill the same 10. You're still fully saturated; no income loss.
2) Say you don't get harassed and you safely expand. You now have a bank of workers, at the previous base, that you can transfer for no income loss while the transfer occurs. Or, better yet, you can transfer a set amount to establish the 2nd base that much faster. I'm sure I'm not unique in transferring 1/2 of my probes + 6 when expanding to 2nd base, and a 1/3 + 3 to 3rd base from each of my previous 2 (extra 6 (or 3x2) for gas claim).
Unfortunately, myself being an MMR-unstable toss (somewhere between low diamond and high gold), I rarely get my 4th before the 1st is mined out.
Anywho, hope that helps,
Rasera
|
I think long distance mining when you're oversaturated on one base is a good idea. I end up doing this after I lose my main or if I'm 1 basing really hard like in PvP.
|
I think these people are all playing victim to the status-quo bias. I think you're right and you're not the first person to mention this. Some players will long-distance mine when they're over saturated (which is >30 workers at one base). For the most part they don't do this for 2 reasons:
1. It's not safe to have workers wandering around past your natural
2. They really should expand, but the same mistakes or issues that are preventing them also make it impossible to either spend them time or risk the workers to long-distance mine.
|
On November 30 2011 02:56 Imbaman wrote:
Thanks for all your points. I want to seek some clarifications though.
1) Is it possible that the advantage of having the building finishing earlier outweighs the fact that it is less saturated when finished i.e. only have 70% saturation when finished, but finishes a minute earlier or something? This new structure can then start producing workers earlier too.
2) Other than terran who can build the CC and float it down, wouldn't the player have already set up defences to defend the structure being built?
3) Wouldn't having less mining bases be less area to defend rather than more?
In case of any wrong perception, I am not trying to say that I am right and the pros are wrong. I just want to understand why do what they do even though it seems unintuitive to me.
1) Yes, and it's frequently done. Fast Expand builds are popular for that very reason, but you cut unit producing structures and/or tech, instead of workers. The idea being that as your 1st base income is increasing, there will be a point that every unit structure you have building, as well as workers building, and you still floating 300(Hatch)/400(CC/Nexus) and then drop your expo.
2) In theory, yes. However, throwing down static defense instead of keeping your army in position will delay your expo further. It becomes a balance between how much defense you need in order to ensure that your expansion is safe, and how much you want to delay your expo so you can have that defense.
3) The more bases you have, the more you need to defend. The more bases you have mining, the more income you have available to defend with. Example: toss can support 6 gate, robo (5 gate if still building probes) on 2 base. They can support 10 gate, robo; or 7 gate, 2 robo; or 7 gate, robo, stargate; on 3 base. While you have to defend more bases, you can build that much more.
|
When I'm one basing and not pulling workers, I'll usually send 10 workers at the natural to distance mine, since I like to keep up SCV production no matter what. I'm usually not afraid of losing them because my one base play is (hopefully) forcing my opponent to be on the backfoot on his side of the map.
|
On November 30 2011 02:22 Imbaman wrote:
I have noticed that most pros produce more workers than their bases can support. While I understand that it is to transfer when another base is done, I have three questions.
1) Why is it better to produce workers before the CC/Nexus/Hatch instead of cutting workers, building the building, then restart making workers?
2) Why is it better to oversaturate rather than long distance mine? From what I understand, having more than 17-18 workers on minerals give less minerals than having those excess long distance mine.
3) Why is it better to split the workers evenly instead of just transferring the excess (above 16-18) and rallying new workers to the new base?
Any advise would be appreciated.
1) Well, I don't think it's necessarily better to produce workers before CC/Nexus/Hatch but most people do (particularly zerg/toss) because they have minerals to spend and do not have an early timing planned so you might as well invest the spend in something that will give you an advantage a little later. Having instant saturation once the CC/Nexus/Hatch is done will make up for the lost minerals very quickly. Terrans don't do this as much since building CCs give them mules ne ways.
2. It's not better to oversaturate economically, but it's just safer generally to not have your drones walking across the map and it has really has a lot to do with the answer to #1.
3. Again, it's not actually better, but it's just safer. In theory, rallying new drones to a new base and keeping full saturation on existing bases is slightly more efficient economically but having drones walking across the map constantly can be a bad thing. Not to mention you just lose track of your rally sometimes and will get crazy oversaturation on a single base.
|
On November 30 2011 03:32 ChanmanV wrote:Show nested quote +On November 30 2011 02:22 Imbaman wrote:
I have noticed that most pros produce more workers than their bases can support. While I understand that it is to transfer when another base is done, I have three questions.
1) Why is it better to produce workers before the CC/Nexus/Hatch instead of cutting workers, building the building, then restart making workers?
2) Why is it better to oversaturate rather than long distance mine? From what I understand, having more than 17-18 workers on minerals give less minerals than having those excess long distance mine.
3) Why is it better to split the workers evenly instead of just transferring the excess (above 16-18) and rallying new workers to the new base?
Any advise would be appreciated. 1) Well, I don't think it's necessarily better to produce workers before CC/Nexus/Hatch but most people do (particularly zerg/toss) because they have minerals to spend and do not have an early timing planned so you might as well invest the spend in something that will give you an advantage a little later. Having instant saturation once the CC/Nexus/Hatch is done will make up for the lost minerals very quickly. Terrans don't do this as much since building CCs give them mules ne ways.
Unless you're cutting for a really deep all in, Terran should almost never stop building SCVs. You use excess minerals to invest in the Macro CC if you already have spent your money into the SCVs and whatever extra production facilities you have. Late game, sure go for it, but it's a bit ridiculous to claim that OC's will cover the investment fast enough (550 minerals, early mid game, how will you survive/keep up existing production?!).
|
On November 30 2011 03:39 Gamegene wrote:Show nested quote +On November 30 2011 03:32 ChanmanV wrote:On November 30 2011 02:22 Imbaman wrote:
I have noticed that most pros produce more workers than their bases can support. While I understand that it is to transfer when another base is done, I have three questions.
1) Why is it better to produce workers before the CC/Nexus/Hatch instead of cutting workers, building the building, then restart making workers?
2) Why is it better to oversaturate rather than long distance mine? From what I understand, having more than 17-18 workers on minerals give less minerals than having those excess long distance mine.
3) Why is it better to split the workers evenly instead of just transferring the excess (above 16-18) and rallying new workers to the new base?
Any advise would be appreciated. 1) Well, I don't think it's necessarily better to produce workers before CC/Nexus/Hatch but most people do (particularly zerg/toss) because they have minerals to spend and do not have an early timing planned so you might as well invest the spend in something that will give you an advantage a little later. Having instant saturation once the CC/Nexus/Hatch is done will make up for the lost minerals very quickly. Terrans don't do this as much since building CCs give them mules ne ways. Unless you're cutting for a really deep all in, Terran should almost never stop building SCVs. You use excess minerals to invest in the Macro CC if you already have spent your money into the SCVs and whatever extra production facilities you have. Late game, sure go for it, but it's a bit ridiculous to claim that OC's will cover the investment fast enough (550 minerals, early mid game, how will you survive/keep up existing production?!).
There were a bunch of really good Terrans going for fast 3rd Orbital during Dreamhack this Weekend. And it did not look even a bit vulnerable.
|
On November 30 2011 03:46 Xanatoss wrote:Show nested quote +On November 30 2011 03:39 Gamegene wrote:On November 30 2011 03:32 ChanmanV wrote:On November 30 2011 02:22 Imbaman wrote:
I have noticed that most pros produce more workers than their bases can support. While I understand that it is to transfer when another base is done, I have three questions.
1) Why is it better to produce workers before the CC/Nexus/Hatch instead of cutting workers, building the building, then restart making workers?
2) Why is it better to oversaturate rather than long distance mine? From what I understand, having more than 17-18 workers on minerals give less minerals than having those excess long distance mine.
3) Why is it better to split the workers evenly instead of just transferring the excess (above 16-18) and rallying new workers to the new base?
Any advise would be appreciated. 1) Well, I don't think it's necessarily better to produce workers before CC/Nexus/Hatch but most people do (particularly zerg/toss) because they have minerals to spend and do not have an early timing planned so you might as well invest the spend in something that will give you an advantage a little later. Having instant saturation once the CC/Nexus/Hatch is done will make up for the lost minerals very quickly. Terrans don't do this as much since building CCs give them mules ne ways. Unless you're cutting for a really deep all in, Terran should almost never stop building SCVs. You use excess minerals to invest in the Macro CC if you already have spent your money into the SCVs and whatever extra production facilities you have. Late game, sure go for it, but it's a bit ridiculous to claim that OC's will cover the investment fast enough (550 minerals, early mid game, how will you survive/keep up existing production?!). There were a bunch of high Terrans going for fast 3rd Orbital during Dreamhack this Weekend. And it did not look even a bit vulnerable.
Yeah, but they didn't cut workers ==; I'm not saying "DON'T BUILD MACRO CC'S" I'm saying don't cut workers!
What's going to happen is that you'll get the OC faster, but you'll be behind on income, and won't be able to fully produce or invest. It's just more efficient to keep up the SCV production, keep getting better income, and invest in the OC a little later.
|
takes wayyyyyyy too long to make workers for protoss/terran, better to constantly produce until you reach 60-80 workers and cut army productions for 15 seconds to get that extra base out sooner.
It is better to long distance mine after you have 30 workers in your main, but it is opportunity cost as they could easily be picked off until the base is established.
The split depends on worker count again, I don't know the critical mass but sending half while keeping 18 ish in the main(on minerals) will ensure more immediate minerals instead of the walking delay during the transfer, long term it is better to send half and rally to the base with more mins so you keep your income for as long as possible and don't put all youreggs in one basket late game. They are after all, a resource and you need to manage it effectively at the risk of getting a planetary baneling busted and having no mins to fall back on. This is also why over muling is bad. I should write a blog about this over xmas 
|
The good answer to #3 has been given by noel:
On November 30 2011 02:35 noelsusman wrote:
3) You don't want your main base to mine out too fast. If you mine out your mine too quickly then you are put in an awkward situation where you need to obtain a fourth before the other player to stay on even income. This spreads you out even further and leaves you vulnerable to attacks since you still have to defend your main even though it is not giving you any income. The less area you have to defend, the better.
With some reasonable points added by other folks. It even makes sense to leave the old base undersaturated in some cases - if you have expanded fast enough that you don't have enough workers to saturate all bases, leave the old ones with less workers, that way the bases take longer to dry out and later on you'll be mining from more patches.
I would add that it usually is a good idea not to transfer all the needed workers at once if that is going to significantly reduce the income from your old base. In order to have more of a constant overall income I tend to transfer half the workers first, and once they're mining from the new base, I'll send the other half.
|
Since CC/Nexus only costs 400 minerals, which is 8 workers, wouldn't building the CC/Nexus be better as long as they are able to produce 8 or more workers within the building time of the building (which is possible for protoss/zerg on one base and terran on 2 base)? I think we may have overestimated the the number of workers we are cutting in order to get the building out faster.
Also, while I understand the point on having the main base last longer, transferring less to the natural initially would also mean that the newer bases last longer, and therefore while we may have to get a faster fourth, we can get a slower fifth and sixth, so doesn't that compensate for the faster fourth?
On November 30 2011 03:39 Gamegene wrote:
Unless you're cutting for a really deep all in, Terran should almost never stop building SCVs. You use excess minerals to invest in the Macro CC if you already have spent your money into the SCVs and whatever extra production facilities you have. Late game, sure go for it, but it's a bit ridiculous to claim that OC's will cover the investment fast enough (550 minerals, early mid game, how will you survive/keep up existing production?!).
I agree with that when the bases are not yet saturated, but if bases are already saturated, cutting SCVs doesn't reduce income, because any new SCVs will not provide any income anyway (unless you long distance mine, which even so provides quite little).
On November 30 2011 04:54 Gaspa wrote:
I would add that it usually is a good idea not to transfer all the needed workers at once if that is going to significantly reduce the income from your old base. In order to have more of a constant overall income I tend to transfer half the workers first, and once they're mining from the new base, I'll send the other half.
This is one of the reasons why its better to rally over new workers too.
|
On November 30 2011 02:22 Imbaman wrote:
I have noticed that most pros produce more workers than their bases can support. While I understand that it is to transfer when another base is done, I have three questions.
1) Why is it better to produce workers before the CC/Nexus/Hatch instead of cutting workers, building the building, then restart making workers?
2) Why is it better to oversaturate rather than long distance mine? From what I understand, having more than 17-18 workers on minerals give less minerals than having those excess long distance mine.
3) Why is it better to split the workers evenly instead of just transferring the excess (above 16-18) and rallying new workers to the new base?
Any advise would be appreciated.
1) the sooner you get more workers on a new Nexus, the sooner that it starts paying itself back, and the less vulnerable you are.
2) This one I don't have an answer for - I think its mostly because in a lot of cases long distance mining is more risk than it is worth.
3) Some pros do only transfer the excess workers, but in general its a good idea to have even worker counts on both bases and rally each Nexus to its own minerals to cut down on the time new probes spend running around before they get to where they need to be.
|
On November 30 2011 02:22 Imbaman wrote:
I have noticed that most pros produce more workers than their bases can support. While I understand that it is to transfer when another base is done, I have three questions.
1) Why is it better to produce workers before the CC/Nexus/Hatch instead of cutting workers, building the building, then restart making workers?
2) Why is it better to oversaturate rather than long distance mine? From what I understand, having more than 17-18 workers on minerals give less minerals than having those excess long distance mine.
3) Why is it better to split the workers evenly instead of just transferring the excess (above 16-18) and rallying new workers to the new base?
Any advise would be appreciated.
1. is a very hard question that involves a bunch of ideas. First, you have to take a huge opportunity cost to make a Nexus, so you need to make sure you are safe enough to do that because starting a Nex and cancelling it 30 seconds later, clearing an army and remaking, is way worse than waiting 30 seconds originally to make it. Also you want to make up the opportunity cost of the Nexus as fast as possible, so having enough workers to start getting a good income instantly is important. Zerg is really completely separate from Protoss and Terren because their Nexus is their entire production. So yea, most importantly its opportunity cost
2. It is beneficial to long distance mine if you are exceeding 24 probes on minerals for a base(over 24 workers you get 0 additional income so any LDM amount will be greater than 0). Reasons people don't this are 1: its apm/mental energy they don't have to spare. 2: it is not always safe to have workers long distance mining so losing theoretical income in the short term until you can secure a base to properly mine from may be the correct decision in a lot of situations.
3. It is not better it is preference/style. Doing a large maynard of probes when a new base comes online (large=enough so that the original base is no longer optimally saturated) puts all the lost income from travel time up front. Setting up optimal saturation in your main then simply re-rallying spreads out the lost income over a longer period of time. So, for example, if you were in a situation where you need every single mineral you can get NOW you would probably want to re-rally. But, conversely, if you want to saturate 2 bases as quickly as possible while teching then doing a large maynard will get you saturation faster. (after the initial bump in income that is)
Low Master P on NA if its anything.
|
On November 30 2011 10:54 Imbaman wrote:
I agree with that when the bases are not yet saturated, but if bases are already saturated, cutting SCVs doesn't reduce income, because any new SCVs will not provide any income anyway (unless you long distance mine, which even so provides quite little).
Here's the thing:
You're making a third command center because in all likelihood you're going to take a third expansion. You're not making a third command center just for mules, because if you're staying on two base it's more than likely you're going to want those extra minerals in whatever timing you have, and the mules are going to take a while to pay for themselves, and you're going to deplete those 2 bases faster. So if you are planning on taking the third expansion, you want the SCVs to saturate that third base faster, which is why you keep producing SCVs.
|
On November 30 2011 13:28 Gamegene wrote:
Here's the thing:
You're making a third command center because in all likelihood you're going to take a third expansion. You're not making a third command center just for mules, because if you're staying on two base it's more than likely you're going to want those extra minerals in whatever timing you have, and the mules are going to take a while to pay for themselves, and you're going to deplete those 2 bases faster. So if you are planning on taking the third expansion, you want the SCVs to saturate that third base faster, which is why you keep producing SCVs.
Yes I understand, and my question was pertaining to taking new bases, but not necessarily just the third. But what I was thinking was, upon reaching saturation at existing bases, cutting worker production in order to get the new base up quicker, and then restarting worker production would allow the new workers to start mining faster, as well as use the new CC to start producing workers faster. From my simulations in the build order calculator:
Constant worker production, at building CC at 400 minerals
Cutting workers at saturation, building CC, then continuing worker production
As you can see, in both cases, there are 3 raxes continuously pumping out marines. I made a lot of excess supply depots to reduce the mineral count to show to effect of cutting workers, because otherwise with just 3 raxes and no add ons, you would reach 400 minerals before saturating the first base. The minerals spent on supply depots can easily be used to build refineries, factories, starports etc. I just build supply depot to simplify the simulation and remove the need to build gas.
Even with only 3 marines being produced at a time, not cutting workers got the CC started at 45 supply, while cutting workers got the CC up at started supply. As you can see, the cutting workers BO also ended up with more workers and more minerals mined at around 8:50 than the constant production BO.
I understand that this simulation is over simplified, but in both cases, barracks were built at the same time, marines were in constant production, and both were never supply blocked, both have same number of supply depots, so I think it is fair to say that they are on a level field.
|
Russian Federation164 Posts
Hey, if you have 30+ workers at each base, it's ovesaturation etc, but usually it's not safe to pull long-distance mining, as mentioned. How much minerals per minute do you expect to get eve with 10 mining workers? 100/minute? that's just not worth it, 2 kills - and you're even.
But what if you go for 1 geyser? With like 6-7 workers. It's pretty easy to rescue them since they're gathering in 1 line, and the extra 50-70 gas/minute may help to squeeze an extra upgrade out.
What we also often see is players don't use long-distance mining when Nexus/Hatch is morphing, while the main is oversat, although most the time it's pretty safe (I mean, you don't expand when there's a huge threat, right?) - that could be just a habit.
|
|
|
|
|
|
EPT
