|
Hi all,
Recently I was thinking about how I can train my mechanics and try and improve my macro outside of the typical "play a heap of games". I find that the mass games approach isn't always the most effective way of practising, although it certainly has it's place.
Instead of trying some other challenges which is simply to max out asap I wanted to do something with my queens as injects are a big issue for me at the moment, and this is what I came up with. Feel free to give it a go, I'll add my replay at the end if you like for my attempt. I also like playing against a 'hard' opponent, not because it's difficult but because it adds a small element of randomness to it where you can't blindly focus 100% on macro.
Map: Entombed Valley
Race / Opponent: Zerg against Terran on Hard AI
Objectives to win:
By the 20 minute mark:
- You must have a maxed ling / baneling / mutalisk army, with 3/3 upgrades and adrenal glands on the lings.
- You must have every base on the map, apart from the main, natural and 3rd of your opponent.
- You must have a queen at every base
You lose if:
- Your money goes over 600 minerals or 600 gas in the first 10 minutes
- Your money goes over 1000 minerals or 1000 gas before you're at 185 supply
- Any queen on the map gains 50 energy or more at any stage
You cannot:
- Drop creep tumours with your queen past 1 with your first two queens (or kill something and transfuse i.e. dump energy)
- Attack the opponents base or harass his workers in any way.
I found it pretty hard, didn't quite get there (completed about ~21 minutes) but it's definitely achievable. I also think 1-2 of my queens may have hit 50-60 energy at one point.
Here is my first attempt - http://drop.sc/93364
If you master it and you want more of a challenge, here are some things you can try to increase difficulty.
- Not looking at your base bar to select a worker to build something, then looking away for the first 5 minutes (aka working blind).
- Having 1 extra queen spreading creep who's energy cannot go over 40 for the entire game (and spreading the already laid tumours).
- Time trial, achieve all the objectives in the shortest time possible.
- Increase difficulty of opponent to very hard, or insane
Would love to see some replays of some Zergs smashing this out, hopefully it helps some people think outside the box a little for mechanics training. I'm sure there could be similar scenario's / restrictions for protoss / terran, if someone wants to write them up that would be cool (i.e. energy can't pass 60 on any orbital or nexus throughout the game, etc).
|
So is this a custom map or something? I searched for Macro Challenge for Zerg and did not find anything. Would love to give it a go when I am bored
|
On January 17 2012 17:12 CP-Jun wrote:
So is this a custom map or something? I searched for Macro Challenge for Zerg and did not find anything. Would love to give it a go when I am bored
On January 17 2012 16:58 Peleus wrote:
Map: Entombed Valley
Race / Opponent: Zerg against Terran on Hard AI
that hard to find?
|
O.o I will give this a go tomorrow definitely.
This could probably be refined as a way of practicing and made into an actual custom map :D
P.S. Something like that probably already exists, so don't rage at my ignorance ^.^
|
Protoss:
Map: Entombed Valley
Race / Opponent: Protoss against Terran on Hard AI
Objectives to win:
By the 20 minute mark:
- You must have a maxed zealot / stalker / sentry army, with 3/3/3 upgrades and charge/blink/hallucination on the gateway units. (+3 shields and charge/blink/hallucination accomodates for having to get 3/3 ground and air for ling/bling/muta and also hive tech for adrenal glands. if anyone could give a suggestion as to what other upgrades we could add to make it more fair, please edit this.)
- You must have every base on the map, apart from the main, natural and 3rd of your opponent.
You lose if:
- Your money goes over 600 minerals or 600 gas in the first 10 minutes
- Your money goes over 1000 minerals or 1000 gas before you're at 185 supply
- Any nexus on the map gains 50 energy or more at any stage
You cannot:
- Attack the opponents base or harass his workers in any way.
(I feel that dumping chrono boost is necessary because you cant constantly be producing off of forges, cyber core etc. and having that many nexuses means a ton of chrono boosts)
Feel free to suggest ways to make this better for protoss players :D
|
This sounds very interresting, I'll be sure to try this tonight.
Turning this into a custom for zerg macro as CP-Jun stated might be a very good idea.
Let's see how my Silver macro does 
[ Replay will be posted here later ]
|
On January 17 2012 17:19 ElusoryX wrote:Show nested quote +On January 17 2012 17:12 CP-Jun wrote:
So is this a custom map or something? I searched for Macro Challenge for Zerg and did not find anything. Would love to give it a go when I am bored Show nested quote +On January 17 2012 16:58 Peleus wrote:
Map: Entombed Valley
Race / Opponent: Zerg against Terran on Hard AI
that hard to find?
Well sorry Mr. I Can't Talk without being Sarcastic. I thought he made a custom map with these objectives with the map Entombed Valley.
|
On January 17 2012 17:36 CP-Jun wrote:Show nested quote +On January 17 2012 17:19 ElusoryX wrote:On January 17 2012 17:12 CP-Jun wrote:
So is this a custom map or something? I searched for Macro Challenge for Zerg and did not find anything. Would love to give it a go when I am bored On January 17 2012 16:58 Peleus wrote:
Map: Entombed Valley
Race / Opponent: Zerg against Terran on Hard AI
that hard to find? Well sorry Mr. I Can't Talk without being Sarcastic. I thought he made a custom map with these objectives with the map Entombed Valley.
gotta be honest though, i chuckled with his response
Nice idea, though. (about the macro test!)
|
Sounds like good training, but I agree that it really should be a custom map which forces these constraints. It's bad enough having to focus on doing all of those things well, you also have to focus on making sure you're following those rules. A custom map like Multitask Trainer would be awesome, could even have difficulty settings such as whether or not you're warned as your minerals/energy climb etc.
|
I'm sure it would be easy enough to turn into a custom map if someone wanted to, but I'm sorry I have zero experience with the sc2 mapmaker.
Saying that, you can easily check for yourself when reviewing the replay. Just because at 8:32 you break a 'rule' doesn't mean the next 12 minutes is useless. It gives you more training when you have more bases, as long as you are always trying to achieve the rule it's doing it's job and gives you something to work towards.
|
Hmm I actually really like this! I always found it abit silly with those who just try to max out as fast as possible and that's it, but this challenge is really well thought out!
I wish there was an actual custom map for this. It kinda reminds me of the Multitask Trainer, but the setup of that map is abit artificial on that map.
|
OP, you need to try the [Macro or die] series of maps. Instead of losing when your macro falls apart, your mineral line gets nuked, which is equally as awesome.
Oh and if you're not going to try it, I think you may need an addition to your rule..because it's a common occurrence to start saving gas up for mutas at some point.
I also don't see the significance of having "every base on the map" (also excluding the ones you stated) if you're not mining off them. Seems more like a mineral sink that way...
|
In my opinion, the best way would be to just try your best, setting up a buzzer for the time limit.
Then, watch the replay (on like 2x) and see how well you did.
This is still a pain and a custom map would indeed be far better, but its better then trying to monitor yourself at the time.
|
I played around with this for a bit, here are some thoughts on optimization:
3/3 + adrenal glands:
- Double evo takes 35 seconds to make, and costs 150 minerals
- +1/+1 takes 160 seconds to research, costs 250/250. At 50% completion start your lair
- +2/+2 takes 190 seconds to research, costs 375/375. At 40/190 start infestation pit, then hive immediately
- +3/+3 takes 220 seconds to research, costs 500/500.
- Adrenal glands take 130 seconds to research, costs 200/200
So overall the ling upgrade path takes just over 10 minutes to complete. You could go for the spire at lair, but keep in mind that you have to spend 375 gas right away on upgrades, 150 gas 40 seconds later on the infestation pit and 150 more gas for hive 50 seconds after that. At that point you have 100 seconds to amass 500 gas for the +3 upgrades. The thing is, you are also expanding to your 3rd and 4th bases at this point after defeating the first AI attack. So realistically, massing muta at this point is counterproductive to the goal.
The other goal of taking every expansion was a huge pain, as it is rather unlike any situation you would come across in a normal game. Basically from just 4 gases (and 4 bases) I was banking tons of gas as I made 8 hatches one after another in quick succession, then queens at each expansion. Injecting is a pain if your queens end up coming out staggered, or even cycling through and hotkeying the queens is a bit tiresome. The real issue is that all the units you're making cost as many minerals as gas, and at this point you need to spend pure minerals on getting some drones for your 3rd and 4th, you need to spend pure minerals on lings to defend the AI attacks, and of course pure minerals on the mass expansions and queens. In terms of optimization it would be best to get as little gas needed to support the fast hive + upgrades, get a drone pump for your 3rd and 4th, switch to ling production while mass expanding then finally taking a lot of gases to make the mutas. Ideally you'll time the 8 last queens to come out at around the same time that you max out, so you don't have to deal with them, and also have it occur at the same time as your upgrades.
Anyhow, in my unrefined execution I was able to max out with the upgrades at 18:30, though I'm sure it can be several minutes faster with better optimization. I was also banking tons of gas in the second 10 minutes, because I was trying to expand and therefore had no minerals to spend on mutas at the same time.The first 10 minutes were pretty crisp, with money dropping quite low after each production round, though in terms of optimizing for this exercise you would want to get even earlier gas to start the path of upgrades sooner. I went 16h 17p 17g, kept 3 in gas then went metabolic boost, double evo at 40 and then lair. The timings worked out quite well like this, but in a race situation the evos could go down earlier thus requiring an earlier/double gas.
|
On January 17 2012 16:58 Peleus wrote:
You lose if:
- Your money goes over 600 minerals or 600 gas in the first 10 minutes
- Your money goes over 1000 minerals or 1000 gas before you're at 185 supply
I always find these restrictions very uncharacteristic of Zerg challenges.
Half of teching to something is banking resources to churn out a bunch at once as soon as the structure finishes. In many cases, there's no way for Zerg to spend all his gas before the Spire is finished. If you say to simply not mine gas, then you're losing out on quite possibly a dozen potential Mutalisks emerging a few minutes earlier than they otherwise would by trying to be 'efficient'.
|
On January 17 2012 19:12 Proof. wrote:
I also don't see the significance of having "every base on the map" (also excluding the ones you stated) if you're not mining off them. Seems more like a mineral sink that way...
Basically one or two reasons.
Firstly, Hard AI is hardly a challenge to actually beat, so there must be a difficulty increase as you progress through the game, just the same as what there is in a typical game.
Also, in a typical game as more stuff happens people's macro tends to fall apart, along with the fact you typically have more bases.
Having a requirement that you take every base on the map helps add a little challenge, while also dramatically increasing the amount you need to macro since you have so many queens / injects required. It means that you're constantly being reminded to macro harder as the games progress, which ideally you should be doing in typical ladder games.
Also, I'd like to point out that I've never theory crafted this and only thought of it on the way home this afternoon. It's not designed to be optimised and beaten as efficiently as possible. By all means benchmark yourself to see how you are improving, but for all I know the optimal time may be '13 minutes' or it may be '19 minutes 45 seconds'. If however you're focusing on purely how to get the time down more and more or trying to skip bases or whatever, you're kinda missing the point. What you're doing is exactly like focusing on winning a ladder game, looking at the end result rather than the journey.
20 minutes was arbitrarily picked because it's a typical time to transition into T3 as a zerg and about when you want all your upgrades done by, so it's trying to make it 'realistic' in some senses, while I completely acknowledge you're never going to take every base on the map in a real game. That's an artificial constraint to try and actually test you with the bonus of specifically testing your mechanics, which is unfortunately not likely to occur in any real game.
|
On January 17 2012 20:23 Tropical Bob wrote:Show nested quote +On January 17 2012 16:58 Peleus wrote:
You lose if:
- Your money goes over 600 minerals or 600 gas in the first 10 minutes
- Your money goes over 1000 minerals or 1000 gas before you're at 185 supply
I always find these restrictions very uncharacteristic of Zerg challenges. Half of teching to something is banking resources to churn out a bunch at once as soon as the structure finishes. In many cases, there's no way for Zerg to spend all his gas before the Spire is finished. If you say to simply not mine gas, then you're losing out on quite possibly a dozen potential Mutalisks emerging a few minutes earlier than they otherwise would by trying to be 'efficient'.
There are always exceptions to rules and yes banking for muta's is one in an actual game. This is not an actual game, this is to try and improve mechanics. You don't take all the bases in a real game either, but once more it's not the point. I'd much rather get efficiency drilled into my brain and have to make a one off exception for an in game scenario, especially when only doing this to learn macro / mechanics.
|
On January 17 2012 20:23 Peleus wrote:Show nested quote +On January 17 2012 19:12 Proof. wrote:
I also don't see the significance of having "every base on the map" (also excluding the ones you stated) if you're not mining off them. Seems more like a mineral sink that way... Basically one or two reasons. Firstly, Hard AI is hardly a challenge to actually beat, so there must be a difficulty increase as you progress through the game, just the same as what there is in a typical game. Also, in a typical game as more stuff happens people's macro tends to fall apart, along with the fact you typically have more bases. Having a requirement that you take every base on the map helps add a little challenge, while also dramatically increasing the amount you need to macro since you have so many queens / injects required. It means that you're constantly being reminded to macro harder as the games progress, which ideally you should be doing in typical ladder games. Also, I'd like to point out that I've never theory crafted this and only thought of it on the way home this afternoon. It's not designed to be optimised and beaten as efficiently as possible. By all means benchmark yourself to see how you are improving, but for all I know the optimal time may be '13 minutes' or it may be '19 minutes 45 seconds'. If however you're focusing on purely how to get the time down more and more or trying to skip bases or whatever, you're kinda missing the point. What you're doing is exactly like focusing on winning a ladder game, looking at the end result rather than the journey. 20 minutes was arbitrarily picked because it's a typical time to transition into T3 as a zerg and about when you want all your upgrades done by, so it's trying to make it 'realistic' in some senses, while I completely acknowledge you're never going to take every base on the map in a real game. That's an artificial constraint to try and actually test you with the bonus of specifically testing your mechanics, which is unfortunately not likely to occur in any real game.
Ah kk, but my thinking would be that having more bases != better exercise for having to remember injects. In fact, with such a high number of hatches, it would be a little tedious and nonhelpful to the overall macro exercise. At that point, it just becomes an exercise of how "fast" you can inject each base, which just requires a good hotkey set up.
As for your point about making it challenging, you should really try out that map series I suggested! There's many versions of the map (xel naga, TDA, shattered, etc.) all equipped with green tea ai, which almost always performs a different build every time you play it (unlike the extremely predictable blizzard ai).
|
The restrictions aren't realistic and pretty arbitrary IMO. Quuen at every base? Seriously? 3/3 and cracklings when going ling bling muta? Not realistic, people don't tech to hive for upgrades alone.
This challenge should be to max out on ling bling muta with 2-2 ground and 2-0 flyer, in 16 minutes with a maximum of 4 queens, with creep reaching the opponents tower. By disallowing mineral sinking with additional hatches and queens to make it seem like you have 'good macro', you make it much more skillful.
|
close to pointless doing this.
|
|
|
|
|
|
EPT
