Starcraft II

[Suji]

I also want to echo what has already been said to Araban. Get well soon. I hope that there isn't any lasting damage.

[ketchup]

I'm not sure if I missed it or not, but anyone know of a re-stream of the GSL finals? I'm thinking of staying awake until then(east coast), but it seems fun to watch "live".

[Solomon]

I'm not sure if I missed it or not, but anyone know of a re-stream of the GSL finals? I'm thinking of staying awake until then(east coast), but it seems fun to watch "live".

Pretty sure it'll be re-streamed here (5am EST): http://www.livestream.com/fishersc2 - There are others of course, but I find his to be the only one with good quality.

You can also watch it through VLC: http://www.reddit.com/r/starcraft/comme ... scriptive/

I'll definitely be watching and cheering for the fruit dealer! Here's an awesome fanpic someone posted on TL:


Also, thanks for the well wishes everyone! I'm still typing awkwardly with 8 fingers, but the finger is healing and the hand surgeon yesterday told me I'm on my way to a full recovery . Now if only I was allowed to take both antibiotic and pain killer pills...

[flOvermind]

One question to the more experienced players:

How do you know that your base is saturated? Or, equivalently, when do you move the rally point of your main to the expansion? Or do you stop producing workers in the main at some point?

I always lose track of how many workers I currently have after starting to produce units. Producing more workers continuously is not the problem, but at what point do I stop? Sure, in the late game, when minerals run out, I can just move the bunch of workers to a new, not yet saturated expansion. Usually I stop making workers at all at this point. But in the middle game, when my main is roughly saturated (but I have no idea about the exact count) and my second expansion is up, I never know whether I should continue producing workers in the main and whether I should rally them to the main or to the expansion.

[schilds]

I just drag a box around the mineral area and make sure there are between 2 and 3 rows.

[Tooveli]

How do you know that your base is saturated?

Two workers per patch is ideal in terms of minerals per worker. With 3 workers, the third worker has to wait around a bit before mining so you are starting to have less return per worker. More than 3 workers per patch and the extra workers don't do anything. In general you want 16 workers (on minerals) at each expansion, and if you have more workers then you're looking to expand. At the beginning of a game it's not always safe to expand so it's ok to just keep pumping workers and transfer them when you finally do. I think this applies more to T/P as Z can have more control about the number of drones at any one time.

As schilds said, you can check by just boxing your workers. When watching streams of high-level players (particularly zerg) they seem do be doing this almost constantly.

when do you move the rally point of your main to the expansion? Or do you stop producing workers in the main at some point?

I'm not sure about this but I think I remember day9 saying that it's bad to rally your main to your expansion, especially on maps where the distance between them is large. You want to transfer roughly half your workers when you expand so that you can make workers at both. I think with zerg you can transfer less if some of the drones at your main aspire to become buildings and with any race you might want to transfer less if the expansion is risky.

[MountainGo]

I did a little test for myself a while ago on mining speed. The first 16 workers mine about 42 minerals/minute each, depending on the exact distance to the mineral patch. After that, each one mines less than the one before. As a point of reference, the 20th worker mines about 20 m/m, and the 22nd only mines about 5 m/m. The game's claim in that one Challenge mission that the ideal is 3/patch is a small exaggeration, as you can see. I'd say 2.5/patch is a better rule. (Fun fact: This means that you want to top out at 26 workers per base, including 3 on each geyser, which means that if you ctrl-click all the workers at a base, you want to see exactly one full selection window of 24, since at any moment two workers will be inside the gas buildings.)

[flOvermind]

You want to transfer roughly half your workers when you expand so that you can make workers at both.

Ah, yes, that makes sense. I might try that next time

It doesn't really make a difference at what base my workers, so I can just try to not saturate any base and build in each one. And when the expansion is risky, my main usually is far away from being saturated anyway

[Aphelion]

How do you know that your base is saturated?

Two workers per patch is ideal in terms of minerals per worker. With 3 workers, the third worker has to wait around a bit before mining so you are starting to have less return per worker. More than 3 workers per patch and the extra workers don't do anything. In general you want 16 workers (on minerals) at each expansion, and if you have more workers then you're looking to expand. At the beginning of a game it's not always safe to expand so it's ok to just keep pumping workers and transfer them when you finally do. I think this applies more to T/P as Z can have more control about the number of drones at any one time.

As schilds said, you can check by just boxing your workers. When watching streams of high-level players (particularly zerg) they seem do be doing this almost constantly.

This is not accurate. While its true that 3 workers / patch (24 workers per base) saturates it, thats beside the point. You never want to stop worker production even at oversaturation, because you want to be ready for when you expand. Therefore when I am at one base, I don't stop at 24 probes. I keep making so that I can saturate my second base immediately when I do expand.

While there are exceptions to this rule (such as when you need to cut troops for a timing push or you are behind and need to do a lot of damage immediately), that is advanced play. Its similar to the "empty triangle" rule in Go. There are situations where cutting workers maybe good, but if you need to ask, don't do it. When the time comes that you can get an edge in your play by cutting workers at specific times, you won't need to ask such a question.

[ketchup]

GSL spoiler, do not read if you have not watched finals.

Very great GSL Finals. Cool/Fruit Dealer was always a step ahead of HopeTorture/Rainbow. Some of his trades were awesome. Game three especially showed off how flexible Fruit Dealer was. Great games to watch overall. Also, it helps he did all of that using Zerg.

[Tooveli]

This is not accurate. While its true that 3 workers / patch (24 workers per base) saturates it, thats beside the point. You never want to stop worker production even at oversaturation, because you want to be ready for when you expand. Therefore when I am at one base, I don't stop at 24 probes. I keep making so that I can saturate my second base immediately when I do expand.

I didn't mean to imply anything about stopping worker production. Maybe I should have said "when" you have more (than 16(+6)) workers you are "even more" looking to expand.

[Aphelion]

Ahh I see. Still, you should always be "looking to expand", so isn't that moot?

[Li Kao]

Ahh I see. Still, you should always be "looking to expand", so isn't that moot?

When you have a few bases your old bases run out of minerals about as fast as you expand to new ones. And you don't want to get too many workers due to the supply limit. So you'll stop around 70 or so.

[Tooveli]

GSL spoiler, do not read if you have not watched finals.

Very great GSL Finals. Cool/Fruit Dealer was always a step ahead of HopeTorture/Rainbow. Some of his trades were awesome. Game three especially showed off how flexible Fruit Dealer was. Great games to watch overall. Also, it helps he did all of that using Zerg.

More Spoiler (Including specific game comments)

So good! Better than any other sport/game I have spectated. Probably all the sweeter that the player I was supporting managed to win. I found it odd that the least recommended game on TL was the Kulas Ravine game which I would have said was by far the best. I thought the roaches were going to move out of sight of the drop ship and fruit dealer would miss it until I realised that they were moving out of its way so it wouldn't see them and they could come in and pincer the drop (which would be at the far back of the ledge due to the spine crawler). As I was being amazed by this, fruit dealer had also prepared a neat way to win the game there and then.

I got the same feeling of him just being a step ahead of his opponent in every game. I wonder if it was because of all the balance discussion. I believe Fruit Dealer had said before (and after about ZvP) that Zerg were UP and maybe this led him to spend more time planning out map-specific strategies. He couldn't just work on his mechanics and feel that it would be enough to bring him to victory.

The Kulas game plan seemed to be - I take the gold and if you rush to tank drop it I have a counter that lets me win and if you don't ... well I have the GOLD! I'm hoping that the game will become a lot more interesting as Terrans learn how to counter these strategies and Zergs in turn need to learn how to adapt to the counters and so on... Also, if Zerg start managing to put more counterattacks into their early game defending (which seems promising given the finals), then the whole matchup dynamics will get more exciting.

[Solomon]

Regarding the GSL Final:

Wow, just wow. I didn't get a chance to watch the whole set (fell asleep one game short), but the games I did watch really tempts me to consider playing Z seriously. Oddly, I'd have to say my favorite game was the one I couldn't watch and had to watch a re-stream of, the 5th game on Scraps. It was just sick that Cool could play a build that's almost insta-loss to proxy reapers and still hold it together, even if building time increased by 5 seconds. There were some cute tactics like using the smoke to bait the reapers that was nice as well. Using the mutas to contain the terran while macroing up like MAD and then unleashing a roflstomp was a sight to behold. A fruit dealer definitely deserved the crown .

For those who missed the final, you can either watch an archived stream of it here or here, or check it out on youtube here.