|
On April 20 2010 16:48 kenhennen wrote: brilliant thinking. no matter if it's gonna be huge or not, great ideas thanks for replay kostja
Thanks. I don't think it's a major thing, it has really limited uses, but I think it's something that in a really close macro war could tip the scales one way or another.
|
Not convinced. When we were promised a minute's difference, I expected all mineral patches in the 2nd gold expansion to live at least a minute past the last one in the 1st one. Instead, for the past 52 seconds (not even a minute), there were only 2 patches left.
I mean it's a good find, but nothing significant that will change the outcome of a game. Have you checked whether longer distance makes gas mining slightly slower? That might be the compensating factor.
|
On April 20 2010 16:58 Sadistx wrote: Not convinced. When we were promised a minute's difference, I expected all mineral patches in the 2nd gold expansion to live at least a minute past the last one in the 1st one. Instead, for the past 52 seconds (not even a minute), there were only 2 patches left.
If you move your Hatchery/Nexus/CC one grid away, you're not going to change the mining rate from the patches, that are farther away. So all you get is mining from the nearest patches faster without side effect on the other patches.
|
On April 20 2010 16:58 Sadistx wrote: Not convinced. When we were promised a minute's difference, I expected all mineral patches in the 2nd gold expansion to live at least a minute past the last one in the 1st one. Instead, for the past 52 seconds (not even a minute), there were only 2 patches left.
I mean it's a good find, but nothing significant that will change the outcome of a game. Have you checked whether longer distance makes gas mining slightly slower? That might be the compensating factor.
Like I said before, I don't think it's a big thing either, but why not get "x" number of extra minerals per minute if you can? It's small, but it's just one of those little things, they all add up, especially in a close game. The reason it went in about 50 some seconds is because it was oversaturated. What I initially said was that with each patch at it's saturation point (no dancing workers, they're operating at 5-10% capacity.) the farther one would mine out over a minute faster, which is true. The idea being that at this point you have enough spare workers that it's 0 cost to add a couple of them to your high yield.for a slight edge in income.
I answered the gas question earlier No noticeable difference.
|
i don't think this will increase mining speed. while it does increase your MPM on the overlay the MPM is highly variable in the first place, with the exact same worker saturation you can be mining 780 one worker cycle and 820 the next. i think by delaying the worker trip time but increasing workers you are returning more minerals per cycle but that doesnt necessarily mean you are mining more. the delay you are introducing + the extra workers just happens to result in more workers returning minerals at the same time per cycle giving you inflated MPM numbers.
|
On April 21 2010 01:58 mahnini wrote: i don't think this will increase mining speed. while it does increase your MPM on the overlay the MPM is highly variable in the first place, with the exact same worker saturation you can be mining 780 one worker cycle and 820 the next. i think by delaying the worker trip time but increasing workers you are returning more minerals per cycle but that doesnt necessarily mean you are mining more. the delay you are introducing + the extra workers just happens to result in more workers returning minerals at the same time per cycle giving you inflated MPM numbers.
Yeah I think this is a valid explanation. If you think of an extreme example, say a mineral field all the way across the map. You have 50 scvs mining it, once they all return it, your mpm will skyrocket. I think this is because mpm is calculated every second or so.
|
Calgary25942 Posts
That doesn't make sense because you're still only going to have 8 returning at a time, even if it's across the map. Someone mentioned the replay shows the 1-space-away expansion mining out first as well, although I can't watch the replay right now.
|
On April 20 2010 16:51 Aether wrote:Show nested quote +On April 20 2010 16:48 kenhennen wrote: brilliant thinking. no matter if it's gonna be huge or not, great ideas thanks for replay kostja Thanks. I don't think it's a major thing, it has really limited uses, but I think it's something that in a really close macro war could tip the scales one way or another.
I think that's pretty major. Placing an expo a square away mines out high yield minerals couple minutes faster makes a big difference. Very good observation imo; when you get a beta key, I want to play you.
|
dude, use bigger words!!
That way Google Korea will translate this into something extremely unreadable.
Just sayin.
|
This concept reminds me when you attack with a large swarm of zerglings and once the target has been completely surrounded, the rest of the zerglings continuously run around the ball until they find a location to attack. I don't know if this is optimal or not, but it's just annoying for me to look at for some reason lol
|
holyshit this is genius ^^.
|
Very nice discovery. It was so obvious that nearer HQ should get more income that I have no idea how you started trying the opposite.
Keep posting more if you find.
|
Quite interesting and not even surprising (I was very familiar with wandering in BW) but I would never have thought about placing a HQ further away to increase mining efficiency.
Did you test the mining rate difference if you have less than 18 workers?
Now as interesting as this may be I won't use it. Since I'm playing zerg I am always lacking gas, especially when I take a high yield. By that time I also don't want too many drones because 150 supply of zerg fighting units isn't even that strong, and every additional drone that does nothing but mine useless minerals hurts my army size even more. Terran probably should focus on MULE efficiency instead (returning 9 times instead of 8 times and timing out after mining the 9th cargo). I think protoss can profit the most from this, especially if they have to play defensively (pvz) and are taking a high yield as their only expansion at the moment.
|
very interesting!!! (wish i had the free time you have!) keep up the good work!
|
Very useful information here, +1
|
Great find indeed. I've always wondered whether or not it would be beneficial to build an extra command center with orbital command at some place other than expansion. It would build up energy for a quick muleing action midgame (giving you about 1500 minerals from high yield for 200 energy paying off with just 2 mules) as well as provide some emergency scans if needed. You might then just lift it to take your 4-th or 5-th base later in the game. Or is the 550 mineral investment early on just not worth it... Has anyone tested it?
|
Ok so let me see if i got this right :
Placing the expansion 1 space away will allow 3 workers on every patch. One mining, one waiting to mine and one bringing minerals. The closest position allows only 2 workers on some patches, one mining one bringing minerals back, and this means you have a tiny loss of mining time. Because even if you put 3 workers there, given the reduced distance, you have 1 mining, 1 waiting, and the third one won't wait but will look for another patch, becoming a dancing worker.
I don't know if that's right, my mind is pretty much fucked now :D And grats on your beta key. In computer science this is a pipeline and of course having a pipeline of 3 is more efficient, even with the slightly increased travel time.
|
I just tested
15 drones with further distance vs 13 drones at closer distance
This was on high yield.
They both finished at the same time.
Not sure if i should have done more drones, but i can link replay if u want.
|
i take back what i said, just tested 18 vs 18 and further mines faster than closer by a few seconds.
with the further cc, the scvs sync'd perfectly 3 per patch no wandering
with the closer cc, the scvs would freak out and wander leaving the closer mineral patches with downtime as they would travel around then just come back to the same mineral.
really neat find.
|
Russian Federation85 Posts
On April 21 2010 03:26 Full wrote: I just tested
15 drones with further distance vs 13 drones at closer distance
This was on high yield.
They both finished at the same time.
Not sure if i should have done more drones, but i can link replay if u want.
If i'm not mistaken high yield has 6 mineral patches. Since in your configuration neither of the high yield is fully saturated (under the assumption of the OP), you would need to calculate the theoritical time for them mineral it off since you have different lengths.
He says that at the max saturation, the distance is important to achive the maximum result with the current AI (i doubt it will change, so).
|
|
|
|