|If you write a little expalanition of how it works in UMS mode it would be very usefull.|
|Ok, I uploaded this map again. I guess it got deleted due to the hacker.|
Though it might as well have been deleted because it is no melee map. Well, I made it follow the naming conventions and such, so that it hopefully is ok to put it here. If not, I'd appreciate if it was put under the files section since some people might want to have that map :)
I included some changes to the version I uploaded the last time. They are not great, but you might want to download the new version if you already got the old one.
The gas issue has only been tested for 8 special positions out of the 50 possible ones, and there are differences between the races, which have not been tested yet. Since there sometimes are major harvest effectivity differences for minor position differences, this map is designed to test all the possible positions.
- Start the game ;)
- do any preparations such as addons, lair/hive...
- send all the required peons to harvest (of course you can limit your test to just a few positions)
- select one of the "test run X minutes" beacons with your civilian
All geysirs will be set to 5000 gas and the timer will be initiated.
You just have to wait for the timer to elapse, all peons will stop harvesting automatically.
- calculate how much gas has been harvested from each position and evaluate your test run
This procedure can be repeated as often as you want.
There are the beacons "remove addons" (should be clear) and "reset hatcheries" (resets all lairs/hives to hatches, takes a few seconds)
I placed a Wraith hero with 2500 attack power that you can use to destroy buildings and such (for example when testing with supply depots), and an overlord hero for dropping things around (dunno if you really would need it)
The "reset gas" beacon sets all geysirs to 5000 and your gas count to 0. That way you can do tests that last any custom time that is lower than 15 minutes, though there probably is not much use in it, because the amount of gas you harvested increases linear.
If any problems/bugs appear, tell me, and also any suggestions, of course.
|Hehe, well, LGI, you simply were too fast ;)|
|aren't the two articles here on the gas enough enough?|
|Now, due to that damn hacker I have to repeat everything...|
I don't force you to download and test.
It's just that there are more aspects to the gas issue than discussed in those two articles. There is, for example, a position for protoss that harvests as fast as the left and top positions, but terran and zerg are remarkably slower. I am using this for balance issues on a natural expand.
There certainly are other effects that one MIGHT want to test, and I delivered the instrument to do so.
Btw: Aren't you the one who always cries out "gas issue"? ;)
|yeah, but I thought the two articles on this site about the gas issue are enough, I didn't realize that race made a difference. My understanding of the gas issue is solely based on those two articles.|
|Make on test run over 5 minutes and click on all the geysirs. You'll get a good overview by just this ;)|
|U cant fix that problem realy, BUT u can use FOUR Probes to minimalize that Problem.|
i tryed it:
Pos 2 (right)reached depleted at first
after 4 sec
reached pos 1 (up) + 3 (down) depleted
after 1 sec
reached pos 4 (left) depleted
|How wise you are... Oo|
|I like how the zerg bases looks on the pic;just like pizzas.Mmmm,tasty...|
|lol wtf I see nothing that resembles a pizza I would eat...|
Not to mention that comment is really useless, just about as useless as this comment, which is pretty useless -_-
|I made a testrun over 15 minutes and evaluated the results:|
Blue means best mining speed, red is worst.
The positions are numbered like this: The position where the geysir is diagonally to the top left is labeled as position 1. Position 2 is one tile to the right. Position 8 is directly above the main building; position 33 directly below.
To include a position into the blue coloured ones, I accepted a difference of at most 5% in mining speed; those 4 position where all 3 races have that speed are to be preferred.
Of course these are just rough values, though the test run over 15 minutes (approximately 10:05 real minutes) evened out a lot. Reasons for differences in a real game can be: buildings (I didn't use any comsats or whatever), terrain, some random stuff like from where your units came to the geysir. The biggest effect would certainly come from buildings and terrain, though.
modified by spinesheath
modified by spinesheath
|wow nice work :)|
|yes very nice. very informative|
|6-12 is center top then?|
looks pro though, we should make a 3rd article for gas issue including the information from the 1st two articles along with this chart thing.
|you should have made it like with a circle around the CC. so 0° is 12h, 90° is 3h etc.|
|I would get very stupid values then, and you would have to do some calculations to get out which position is actually meant...|
|you should place the tangent of the angle as the values!!1|
|Ya, and an incredible value for 90°|
modified by spinesheath
Okay, I know this may seem like a bad idea, but.....
One of the problems I have with current maps is some of the imbalances caused by having the gas in the same positions, even though the maps themselves are mirrored/rotated.
What I have been experimenting with is placing the gas on a diagonal, causing players to have to use 4 workers for maximum efficiency.
The problem with this is that when there is less than 4 on the geysers, the vespene gathered can vary a lot, even on the diagonals. They can also vary dramatically in different matchups.
I came across this map by accident, but the graph above shows the exact same thing that I found, through trial and error.
Basically, the 4 positions I am talking about are diagonal, but they are 2 squares towards being centred top/bottom of the hatch/cc/nexus. On the picture above, the geyser positions are "3, 13, 28, 38".
Having the geysers on these positions force the players to use 4 workers for maximum efficiency, and if the players choose to use 3 or less, the rates are close, and fairly even for each race.
They aren't perfect for 90 degree rotated maps or x or y mirrored maps, but they are good. They are very good for x/y mirrored maps or 180 degree rotated maps.
Shoot me down if you think I'm nuts, but I would rather see geysers at these positions than always at the 9/12 o'clock positions.
|Blue means good or bad xD|
|Learn to read:|
"Blue means best mining speed, red is worst."
I once have spotted those 4 locations too, and used them in expansions, where imo the geysir placement is more important than the income rate (as long as the difference is not too huge). But for main bases, I definatly would stick to the 12/9 positions for most maps. Usually you can place the resources so that there is no huge difference in harrassability because most mains have quite some space around the resources. There certainly are exceptions.
On and by the way: Afaik nobody has went through the effort of testing all the possible combinations of geysir placement and supply depots etc.
On 4-worker positions those CAN make a huge difference, sometimes even achieving perfect mining rate with only 3 workers. So you would have to test all of this for your new geysir placements to guarantee that there are no imbalances.
12/9 does't have this problem since there is no way to mine perfectly with 2 workers ;)
modified by spinesheath
modified by LasTCursE
modified by LasTCursE
|Totally outdated (only one race, only 8 positions and so on...). Also no need to bring up a picture that already is in one of the gas issue articles.|
Of course no disrespect to those who started the gas issue research (Starparty?). Helped positional balance a lot.
|What about if you put unbuildable tiles around the geysers? You would prevent buildings from being built near the geyser. It wouldn't be a big problem for your buildings in general, but you could prevent a player from using buildings to speed up mining rates.|
You would get better symmetry in maps.
|Well, definately possible, but highly unorthodox. Protosses will be totally perplexed in pvz when they can build all their stuff in those big clumps around their nexus (which makes muta so damn powerful since you can easily outmicro archons ;) ).|
Also, terrans with minerals to the right of the CC usually want to build a supply depot below the CC to make their workers spawn to the right of it.
Zerg loses lots of building space (imagine 2n2 when the zerg has to turtle and can't place his sunks properly anymore) and there certainly are several other aspects...
|I don't mean most of the base being unbuildable - I mean just a bit near the geyser, literally a few tiles, to prevent buildings from increasing the mining rates.|
That way, the gas could be placed symmetrically, and there would be only a minute imbalance when less than 4 workers are on gas.
Yes, Zerg loses a bit of building space..... That's one downside to it that I also thought of, but it shouldn't be a big deal. There should still be tons of space for sunkens if needed, they just can't be placed directly in contact with the geyser.
|Hm...interesting...did you try this when Terran has Comsat on?|
|no. Certainly would be interesting, but shouldn't matter (that much) when geysirs are at the top or to the left.|
|You'd be surprised. Sc mining is one of the most complex, ridiculous things I've ever seen.|
|No actually I wouldn't be surprised even if adding a scanner at python's top main would speed gas mining at the bottom natural. I've learned too much about SC's behaviour to be surprised about that.|
It certainly is not all that complex, though. Afaik SC uses a variation of the A* Algorithm for pathfinding (and gathering is mainly about pathfinding, though of course different than normal unit's pathfinding) which is quite a simple one.
(Certainly simpler than the ISOM Algorithm, that's for sure TT)
|What if the geyser is on the right and you build a scanner, isn't going to slow down more?|
I was playing computers on Python yesterday and I was at 6, I expanded and built a comsat. I noticed the SCVs were being weird and going under the comsat taking a rather much longer time then a regular geyser in the right.
|try it out. That's what I uploaded this map for.|
|I can't :( I'm at my cousin's house where he has no SC|