|
Introduction: My philosophy has always been that even the smallest macro advantages add up over the course of a game and I’ve always been an extremely strong macro Zerg as a result. Every little bit counts over the course of a game with exponential growth. In lieu of that line of thinking I performed a detailed analysis of the position of the mineral fields relative to a Zerg player’s first hatchery and larvae spawn position. Here are the results.
Hypothesis: The larvae-mineral distance of a base has a notable, although not necessarily game defining, impact on the player’s economy.
Procedure: Two tests will be performed to determine the impact of a base’s larvae-mineral distance.
(a) The time it takes in game seconds(gs) for a newly formed drone to reach a mineral patch will be measured from the close larvae-mineral position, minerals below, and from the distant position, minerals above. Using the drone mining rate of 40 minerals/gs we can get a concrete amount of minerals lost per worker positioned, and extrapolate to find the loss of the initial 6 workers and beyond. This will then be compared to a known tactic for gaining minerals early on, splitting drones 3:3 instead of putting them all on a single patch.
(b) A series of test games will be done using a single build performed to a high degree of accuracy. The results will be used to determine the relative time lost between each larvae-mineral position on Shattered Temple.
The two most relevant numbers, the best case and the average case, will be extracted and compared. The average case is the performance you could normally expect from a game and the best case is the performance closest to “perfect”. The defining moment will be the time it takes for the third Hydralisk to begin production to 1gs accuracy. We can then project losses and gains based on that advantage.
Results:
(a) Through a series of tests it was determined it takes roughly 2gs for a drone to reach the mineral patch from a close position and 4.5gs from a distant position. Which mineral patch being mined has a negligible impact on relative distance, less than .25gs.
At a mining rate of 40 minerals/60gs and 2.5s extra travel time we can determine that each drone created gives a loss of 5/3 minerals at a distant larvae-mineral location. The initial 6 workers position gives a loss of 10 minerals which is comparable to the 10-15 minerals gained by splitting workers 3:3 compared to putting them all on a single patch. This number rises to 35 minerals at 21 drones, extractor pool saturated gas and two rows of workers. Even if you round down to 2gs difference it becomes an 8 and 28 mineral difference at 6 and 21 drones, respectively.
(b) Results based on time hydralisk #3 begins production, 1gs accuracy.
Top position results
+ Show Spoiler +
Average: 6:29
Best: 6:26
Bottom position results
+ Show Spoiler +
Average: 6:23 + 1/3
Best: 6:20
Note that the distance between the average and best times are both roughly 6 seconds. At a mining rate of 960 minerals/minute that’s almost 100 minerals, enough to buy two drones or 8.33% of your economy.
If about half of the worse position’s income goes to army, tech, and upgrades that means you can maintain the same army/tech/upgrade level as before while the minerals availible for economic growth increases by about 16%. Add in the exponential growth of economy and the benefit of even 6 seconds explodes rapidly over a prolonged game.
Conclusion:
(a) While the difference between a close larvae-mineral position and a distant one won’t cause a great player to lose to a terrible one it can still have a marked effect on a game between two players of equal skill. The potential difference with the first 6 workers alone is almost as large as the gain from a 3:3 split and the advantage grows even more as more drones are produced.
(b) In this particular build; spawn position created a 6 second difference on shattered temple. That equates to enough extra minerals for two free drones, +8% mineral income, and results in a significantly larger potential bonus to economic growth. Note that economy then grows exponentially from that point.
The difference in larvae-mineral position is large enough to be significant in some, plausibly most or almost all, situations.
Edit: I've done more samples for part B which means a larger data set.
+ Show Spoiler +It would be too difficult to upload every replay I have but I do have them saved. I guess i could email the replays to people if they really wanted it. and gave a good reason in PM. Here is the new data. Combined new & old data Close mineral-larvae position + Show Spoiler +Average: 6:23.5 Best: 6:20 Far position + Show Spoiler +
6:26
6:29
6:32
6:25
6:27
6:28
6:30
6:31
Average: 6:28.5 Best: 6:25
|
This is really cool and it's nice to see hard data, but I think most people have realized for some time that many spawns on many maps are not equal, though it seems doubtful that blizzard will change this anytime soon.
|
Always thought it would be negligible. I´m sad now...
|
wow, we knew there was a difference but there is SUCH a difference?
this is like a creep tumor on scrap station, clearly 1 side has advantage/disadvantage
|
Are you just testing this for academic purposes or do you advocate some change be made?
It seems like you're just doing this for academic purposes which I like because I was kinda curious about this myself.
|
seems like something that should be fixed.
|
I always get annoyed when spawning with my drones/larvae on the far side of the hatchery and away from the mineral fields.
At least in BW you could micro your larvae a bit so that the drones spawned as close as possible to the minerals, to somewhat make up for the difference. All the zerg pros did it and that should be saying a lot.
Good write up, always wondered how large the difference was.
|
On November 26 2011 04:32 Klondikebar wrote:
Are you just testing this for academic purposes or do you advocate some change be made?
It seems like you're just doing this for academic purposes which I like because I was kinda curious about this myself.
I did this mostly because I could and it was interesting. It also supports my postion of "Tiny changes can make or break a game for a good macro Zerg".
|
On November 26 2011 04:34 Duban wrote:Show nested quote +On November 26 2011 04:32 Klondikebar wrote:
Are you just testing this for academic purposes or do you advocate some change be made?
It seems like you're just doing this for academic purposes which I like because I was kinda curious about this myself. I did this mostly because I could. It also supports my postion of "Tiny changes can make or break a game for a good macro Zerg"
I would tend to agree that interfering with timings for something as trivial as larva positioning is kinda silly. But I would hold off on actually calling for tweaks until Nestea says "I lost that game because my larva were too far from the minerals."
|
How about Terran production buildings & addons? Depending on the spawn position (ex. 6 o'clock on Shattered Temple), If I wall off with a Barracks I can't put an addon on it because of the position of the ramp.
Every race got some drawbacks, in my opinion it's all part of what makes this game a Starcraft game.
And it's clearly not as game breaking as the creep tumors on Scrap station, I don't even understand why people brought that up. Scrap station is clearly a MAP design issue.
|
your Country52798 Posts
On November 26 2011 04:34 Duban wrote:Show nested quote +On November 26 2011 04:32 Klondikebar wrote:
Are you just testing this for academic purposes or do you advocate some change be made?
It seems like you're just doing this for academic purposes which I like because I was kinda curious about this myself. I did this mostly because I could. It also supports my postion of "Tiny changes can make or break a game of a good macro Zerg"
Well, thank you for doing so.
Crazy results...
|
That is interesting, but I might add that those map differences do not only affect zerg. Terrans have their add-on placement that changes if they want to properly wall off. I'm pretty sure it does not affect protoss walloffs, because basically you can make units spawn from either side of a building.
I agree that this difference might be significant in the case of a game agaiinst the AI where you hit everything dead on an no micro is required. In a real game, however, even a 2 second delay on a larvae inject will nullify any "advantage" you get by a good spawn position, so while I agree the exponential nature zerg economy grants the spawns a bigger and bigger advantage (or disadvantage), I just don't see it really being a cause for loss/win in even pro games today.
|
THATS how you discuss. With evidence and data. Cheers for testing it!
|
On November 26 2011 04:37 Ghoststrikes wrote:
That is interesting, but I might add that those map differences do not only affect zerg. Terrans have their add-on placement that changes if they want to properly wall off. I'm pretty sure it does not affect protoss walloffs, because basically you can make units spawn from either side of a building.
I agree that this difference might be significant in the case of a game agaiinst the AI where you hit everything dead on an no micro is required. In a real game, however, even a 2 second delay on a larvae inject will nullify any "advantage" you get by a good spawn position, so while I agree the exponential nature zerg economy grants the spawns a bigger and bigger advantage (or disadvantage), I just don't see it really being a cause for loss/win in even pro games today.
6 second difference could mean your zergling speed is 6 seconds later, which means your 4-6 zerglings get caught by their speedlings with similar openings.
That's quite important.
|
terran has a similar disadvantage depending on starting positions with which way the tech lab/reactor will go. it's not minerals but time that is wasted in lifting off to make the add on with a wall
|
Yeah, I always thought that it would made a different, especially if you're having a wall at your gas (your pool at your gas to help vs Banelings or w/e.). Then they have to go around the building. I dont think you win games cause of it but still, no one should have an advantage because you happened to spawn at 12 instead of 6.
|
This is why I always 6-pool when I get top spawn on shakuras, drones start on right side and lings pop on right side. Makes rush 0.5s more effective.
|
now larvae must spawn closer to minerals for all hatcheries.
That or add a rally for larvae(limit distance to very close to hatchery). While you are at it add a rally for overlords. And let addons be placed with mouse instead of being locked on a position.
Game fixed.
|
On November 26 2011 04:43 Zephirdd wrote:
now larvae must spawn closer to minerals for all hatcheries.
That or add a rally for larvae(limit distance to very close to hatchery). While you are at it add a rally for overlords. And let addons be placed with mouse instead of being locked on a position.
Game fixed.
You assume it's broken...it could very well be that this is just a tiny way that maps create diversity of play.
Difference does not necessarily mean problem.
|
On November 26 2011 04:45 Klondikebar wrote:Show nested quote +On November 26 2011 04:43 Zephirdd wrote:
now larvae must spawn closer to minerals for all hatcheries.
That or add a rally for larvae(limit distance to very close to hatchery). While you are at it add a rally for overlords. And let addons be placed with mouse instead of being locked on a position.
Game fixed. You assume it's broken...it could very well be that this is just a tiny way that maps create diversity of play. Difference does not necessarily mean problem.
Someone is ALWAYS behind the other player by default of spawn positions. That is an unfair, undue advantage.
|
|
|
|
|
|
EPT
