EPTWhy is this a problem? Defender's advantage. The oft-quoted, frequently misunderstood byword, mentioned in hushed voices on the strategy forums, and in screeching, incoherent rants by BW elitists on LR threads. It is often mentioned that BW had a more thorough and more elegant application of this holy grail of strategy theories. 30% chance of missing a shot to the high ground, huge pathing issues associated with ascending a ramp, stronger Siege Tanks, the existence of Lurkers, Reavers, etc... Why wouldn't these work in SC2?
Do you remember Ohana? Cloud Kingdom? Entombed Valley? A very strong defender's advantage in SC2 usually leads to 3-base turtle fests, hour long stand-offs, and worst of all, Protoss victories! No one likes watching those. How did Brood War avoid these problems? Brilliantly designed units? Flawless balance? Self-important fans? Terrible pathing? Mature metagame and mapmaking scene?
No, no, no, sort of, and no. It was due to the benefits that map control gave in BW. Back in the days of 360p spectator gaming, base saturation worked very differently than it does now. Simply put, 6 bases with 8 workers each would give a player more income than a player with 3 bases with 16 workers each. At an additional cost to supply (and therefore, a smaller maxed-out army) a turtling player (usually a Terran, as it tended to be) could go up to full saturation on their 2 or 3 bases, which was usually thought of as between 2.5 and 3.1 workers per patch; usually it was on the higher end of that number if they were planning on expanding. This enabled them to make a sacrifice of running out of bases sooner in a macro game, having a less supply-heavy army, and having minerals lines more vulnerable to splash damage harassment. This was in exchange for a temporarily roughly equivalent income to an opponent on more bases, maintaining a base architecture and strategic positioning that was more resistant to counterattacks, and investing fewer resources into expensive CCs/Hatheries/Nexii and the supporting defensive structures necessary to defend them, thus allowing more minerals to be dumped into attacking units.
There is a famous picture that has floated among the forums that explains this in a clean visual way that it difficult to explain with words and numbers: mining scalability. Some posters have theorized methods of ameliorating this issue with test maps, such as this one: Breaking 3 base.
While they have toyed with the notion of an elegant solution quite admirably, they are misguided in these efforts. An awkward problem demands an awkward solution. Nearly all of the solutions so far have had to do with either A. mining time per mineral patch, B. worker acceleration, deceleration speed, or C. repositioning minerals + Show Spoiler +
credit to SluggyDeezy
If you've read this far, you probably know what's going to come next. That's right, yet another suggestion for altering mineral collection so that it makes large numbers of bases on medium supply of workers equal the collection rate of medium numbers of bases with a large supply of workers.
What is needed
1. a noticeable curve between the mining rate per worker of two "unsaturated" values
2. maintaining the marginal decrease in efficiency between 16 workers' and 24 workers' worth of saturation.
3. similar total mining rates during the early and early-mid game, so as to avoid completely ruining any semblance of balance.
My proposal
A awkward, inelegant piecewise function! Yes, it sounds clumsy. But clumsy, yet fun to watch is a much better situation than elegant, yet turtly and boring. I propose giving mineral patches that have just been mined a 7 in-game second delay before delivering their full 5 minerals per trip again. If they are mined during this time, they will only cough up 4 minerals during said harvesting. This means that after the 5.4 second time to harvest minerals, there will be a 1.6 second period where any workers harvesting a recently-mined patch will return with 20% fewer minerals than normal. At 12 workers, the minerals per minute rate will be roughly 462, rather than the 495 it would be normally. At 16 workers, it would be 594, instead of 660. At 24, it gets even more noticeable in absolute numbers, yet not much more marginally significant, with ~718 minerals per minute, as opposed to 816 normally. This is about 88% as much as normal mining rates, compared with 90% of normal mining rates for 16 workers and 100% of normal mining rates for 8 workers.
The overall slightly slower mining rates also discourage players somewhat from waiting until 200 supply to move out. Why? Because they're likely to take about 10% longer to max out, thanks to these mining rates. In a 15-minute max game, that's equal to about one and a half more minutes. That's one and a half more minutes that an aggressive opponent has to put pressure on you while you refuse to play the game of Starcraft.
Don't like the idea of slowing the game down? An alternative proposal is to make the default mineral cargo 6, but keep the cargo of 4 during the 7-second harvesting cooldown. This makes the mineral efficiency gap between 8 and 16 workers even more severe, thus encouraging expansion over increasing saturation even more. On the other hand, it also encourages early game rushes. Here is some of the math, for comparison.
8 workers: 396 minerals per minute
12 workers: 528 minerals per minute
16 workers: 660 minerals per minute, aka the same as it is in vanilla SC2
24 workers: 785 minerals per minute
Here is a chart detailing this, although I couldn't figure out how to cleanly demonstrate space between 8, 12, 16, and 24 on that program: Minerals per minute.
As you can see, the absolute rates of even such an extreme proposal as this are not even quite as high as the game's normal rates. However, the early game will progress faster, as unsaturated mining will be more lucrative, and very fast expansions, if they can be held, will be even more rewarding than they already are.
