This is my first post here on team liquid so be gentle.
( NOTE: @mods, sorry if this thread is in the wrong section of the forums please may you move it to the appropriate section and notify me, thank you.)
I am a silver level player with a question. I have been on the assumption that 24 workers is the max number of workers you need for 1 base. Which appears logical.
However when I view streams of pro players like "White-Ra" etc I notice that when they expand to their natural they will approximately have 16 workers on each base (totaling 32 workers) and proceed not to make any more for a long.
I can understand that the main base will be partially mined out but still why don't they put 24 workers on their natural?
So, when I expand should I do the same and only place 16 workers at each base?
I have read the Maynard worker thread and other related threads but I remain confused.
16 workers on minerals is "optimal return for investment"; it is the largest number of workers before you get diminishing returns. 24 workers on minerals is the "max"; at this point you get very little (or no) benefit from additional workers. If someone like White-Ra has stopped making workers for some reason it is likely because they are intending to exploit some kind of timing attack, or they are waiting for a specific level of tech / army to be able to take their 3rd base safely (so they are choosing to invest that money into other things rather than additional workers that won't help much).
However, as a silver player, I would definitely put this in the realm of "strategies that I should not worry about". I know this gets thrown around a lot, but I would suggest instead concentrating on ALWAYS making workers (only stopping once you reach ~70-80), using your chrono boosts, and keeping your money low. Far from cutting workers, for self-improvement right now you want to be hunting your replays for times that you had GAPS in your worker production and looking to see what it was that was distracting you, so you can try to avoid that distraction in the future.
16 is the most optimal number of workers on a mineral line, 2 workers per mineral node. Basically you get 100% efficiency out of every worker at 16, but every worker above that has slightly less efficiency as it will have to wait every once in awhile to mine. 24 workers being the absolute maximum you can have on one base because a 25th worker won't get you any extra income. To summarize: It is ok to go up to 24 workers on minerals but if you have another base you can put the extra 8 workers at, you will get a higher income.
On March 29 2012 01:07 ThisisRaider wrote: When you box select over your mineral line it should be two and a half rows of scv's, 24.
It's 8 units per row, so 24 scv's is 3 rows or 1 page.
In general, when expanding, I like to drop down my main mineral line to 16 workers on minerals, to maximize income per worker. I transfer additional workers to my natural and rally both CCs to my natural. Once my natural is up to 16 on minerals, I swap my main CC rally back to its own mineral line so both bases saturate at the same pace (assuming no third was taken at this point of course).
2 workers per mineral patch has each working mining at maximum efficiency, (8*2=16) so if you want to do some sort of 2-base allin or very heavy pressure build having your workers perfectly distributed while cutting probes is ideal. If you want to play a more macro game you would keep producing workers and not worry about oversaturation because you would be transfering them as soon as you get your third anyway.
On March 29 2012 01:07 ThisisRaider wrote: When you box select over your mineral line it should be two and a half rows of scv's, 24.
It's 8 units per row, so 24 scv's is 3 rows or 1 page.
No way!.. I have been doing it two and a half rows since forever! xD So bad at maths lol, watch me play LoL, always back to base before I have what I thought I had enough for.
I would try to have optimal saturation (16 per mineral line). Then you can rally all scvs to your new expansion. As Terran and Protoss you can't produce that many workers instantly. So you might go up to 24-27 right before you secure your next expansion, then transfer workers.
22Workers are the ultimate number to have, 6 on gas and 16 on minerals.
If you make more then that you make it so that you can transfer workers over to the other mineral line.
having 17-18 workers gives you close to NOTHING more, if you would just watch your workers mine, you will see that most of them pick the mineral up and have time to drop it off befor the other worker is done with his job.
Only in a few specific maps there the mineral placement is retarded you my actually gain something by having more workers.
On March 29 2012 07:39 Granter wrote: having 17-18 workers gives you close to NOTHING more, if you would just watch your workers mine, you will see that most of them pick the mineral up and have time to drop it off befor the other worker is done with his job.
I have a question on this topic as well. When maynarding to an expansion, lately I've been toying with the idea of sending 16 probes to my natural and rallying both nexii back to my main to prevent it from mining out so quickly, rather than sending half my probes to my nat and rallying each nexus to its own minerals. I realize I lose some mining time (because I'm moving more probes than a normal maynard at this time), but with a fast expand build (ie. fewer total probes) I suppose it won't be a huge amount.
Is there any merit to this? I'm a masters toss if it matters
On March 29 2012 08:17 lolstarz wrote: I have a question on this topic as well. When maynarding to an expansion, lately I've been toying with the idea of sending 16 probes to my natural and rallying both nexii back to my main to prevent it from mining out so quickly, rather than sending half my probes to my nat and rallying each nexus to its own minerals. I realize I lose some mining time (because I'm moving more probes than a normal maynard at this time), but with a fast expand build (ie. fewer total probes) I suppose it won't be a huge amount.
Is there any merit to this? I'm a masters toss if it matters
Interesting idea. Another advantage would be that you can pull more workers if an early attack is coming. But you would have to watch when your main is saturated again, else you are oversaturating and thus negating the effect.
The effect is probably not big and its just easier and more practical to rally to the natural, but it might be worth it especially when you may need to pull workers like mentioned.
On March 29 2012 07:39 Granter wrote: having 17-18 workers gives you close to NOTHING more, if you would just watch your workers mine, you will see that most of them pick the mineral up and have time to drop it off befor the other worker is done with his job.
Wow, I actually didn't anticipate that it's a nearly linear gain going from 2-3 on a far mineral patch as it is going from 1-2 that's pretty crazy thanks for that. I kind of wish it went all the way to 4 workers though just to show the drop off.
If your main has >16 mineral workers, and your natural has less than 16, you will get higher efficiency (more income per worker) by maynarding some to the natural. But for maximum income, 24 workers are needed per base.
The close patches generally don't triple up until the long patches are full, so the additional income from worker #20 to worker #21 is fairly small (less than 20minerals/minute because it's a close patch), #21 to #22 even smaller (since it bounces around a bit looking for the right spot), and #22 to #23 even smaller (since it bounces around even longer). #24 can take a minute or more to find the right spot, so the payback time for him is close to 3 minutes (the first 20 pay for themselves in less than a minute) - worth it if you're stuck on 1-2 bases for that long, but you'd do better to expand.
And that's also why over-saturation is so bad: once you exceed 3 probes per patch, ANY additional workers are just running back and forth looking for a free patch to mine - and there aren't any, so those workers are accomplishing NOTHING. Better to have an additional base which is undersaturated, than to oversaturate anywhere - the soul of maynarding.
I firmly believe in 2 drones per mineral patch, or 16 drones mining minerals per case. The problem is sometimes your drones will triple-mine the far patches and single mine the close patches, so you need to have some decent drone micro and a quick eye to spot that minor inconvenience. You'll notice Stephano rallies his drones to the expansion once he reaches 16-drone saturation, and DesRow always transfers his excess probes (Any more than 16) to his expansion and then rallies there.
Another way of saying this is that the first 20 workers on the mineral line (2 each on the close patches, 3 each on far) pay for themselves in less than a minute. The next 4 don't pay for themselves for 2 1/2 to 3 minutes or so. Any beyond that are a complete waste except for non-mining uses.
On March 30 2012 06:36 TangSC wrote: I firmly believe in 2 drones per mineral patch, or 16 drones mining minerals per case. The problem is sometimes your drones will triple-mine the far patches and single mine the close patches, so you need to have some decent drone micro and a quick eye to spot that minor inconvenience. You'll notice Stephano rallies his drones to the expansion once he reaches 16-drone saturation, and DesRow always transfers his excess probes (Any more than 16) to his expansion and then rallies there.
EPT
![[image loading]](http://i.imgur.com/UGkXZ.jpg)
