Transcripts (of Day[9]'s Podcasts) - Page 2

Hello everyone, this is Sean Plott, a.k.a Day[9], presenting today: "A New Look at Build Orders." Naturally, before I begin anything, I must apologize for the extended delay for this podcast. I just had a whole bunch of end of the year stuff going on, with finishing up all my school projects, and my finals, and now I have an internship starting up, and I had to do my laundry, and there's this really good taco I ate. But finally, I have some free time. Rather than ask for your forgiveness though, I figured it would be significantly more kick ass if I presented on what I feel is the most important component of Starcraft strategy, and that is how you look at build orders, and how you think about them in the proper way. I find build orders to be such an important topic of discussion because the way people discuss them in the forums is actually quite bad, and I would say flat out WRONG in most circumstances. So I’m going to divide this talk into the following four sections. First, I wanna ask the most basic questions about build orders, and I’m gonna answer them in this incorrect way I just mentioned, and I’m going to point out all the flaws in that line of reasoning. In part two, I’m gonna ask those same basic questions again, but I’m gonna answer them properly, avoiding the mistakes I discussed in part one, and more important, highlighting the strength and versatility of the proper line of reasoning. In part three, I’m going to take a very simple example of the proper thought process and build orders and apply it to a simple situation. As a result, I can go super-duper in depth into a build order that most people are very familiar with and isn’t too tricky. Then, in part four, I’m going to wrap up with a personal example of how I constructed a build order from scratch using this appropriate line of reasoning, and I’ll also include, uh, how I could’ve gotten side tracked or made a whole bunch of mistakes if I thought about it incorrectly. So, without further ado, let’s begin part one.

The two questions that I wanna ask that are really the most simple questions that Iyoucan ask are the following: “What is a build order?” and “How do we talk about it?” The reason I think it’s important to ask those two different questions is that, surprisingly, most people think that the answer to these questions is the same. They will say ok, the food is your unit count in the top right corner of the screen, and a build order is a list of instructions based on that food. So a build order, for a 2 gate opening with protoss would be 8 pylon, 10 gate, 12 gate, etc. and then when they answer the question how do you talk about a build. Well you talk about it using that unit count in the top right corner. We always discuss builds in regards to this food number, obviously there are good bits of truth in this answer because in the game staracft when you’re trying to figure out the appropriate time to build something you don’t have a clock in the game so using food or unit count as a notation is very very convenient and very accurate. Here is the big problem I have with the answer I just gave, if you think about it, this food based instructions set is extremely detailed, that’s a huge level of accuracy and precision in terms of telling you when to build things. And when you have something this precise, there had to have been some step before hand that led to that instruction set. For example, let’s take this recent shift to five hatch hydralisk into mutalisk in Zerg vs Protoss. A player didn’t wake up one day and magically have this perfect build order. He sort of shaped and sculpted a whole bunch of various ideas together and smoothed it down and refined it until he had a very clean smooth 5 hatch hydra into mutalisk buld order. That instructions set that we all use nowadays was the final end product of a very long process. Moreover, the concepts, all the ideas, the tests, the failures, the successes of this build when you just look at this list of instructions based on your food. The final product, this instructions set, is tremendously problematic for players who are on forums discussing, sharing, and learning new build orders. The two biggest problems, is that players can’t deal with variations in game play and they’ll also have no understanding of their builds. So for instance a player will learn this 5 hatch hydra build but what happens if suddenly his opponent gets a whole bunch of gateways early on and pressures him with zealots. That player will feel totally lost because now his instruction set can’t be used appropriately. If he tries to make drones here, when he gets attacked he’ll die so he’ll feel pressure to feel hydralisks. But now he doesn’t have enough money later to make mutalisks later and it’s just a big mess. Moreover, their understanding of the all the underlying of the build are lost, they might not even want to do this build on a new map, but they just do it anyways. And they have a whole bunch of failures and they just don’t know why. Or they will be doing the build perfectly making all the units they need to and getting all the timings right but they won’t be using the units properly, they won’t be positioning the unit properly. All these little subtle details laong the way. With all that in mind, lets move on to part two.

I’m gonna ask the same questions again, but there’s one crucial idea I want to keep in mind. Remember how I said there was some step before this detailed instruction set, I want to make sure to focus on that step. That is the key to answering any question about build orders. So again, let’s ask the question “What is a build order?” A build order is an optimization of an idea. That is it’s the best possible way to do some thing. Now, I do mean idea in as broad a sense as possible. An idea could be something as simple as, “Hey, if I have +1 upgrade for my zealots for attack they can kill a zergling in two hits.” And we have a lot of builds that now revolve around getting that upgrade very quickly. Or a terran player might say, “Wow, zergs really like mutalisks and lurkers against me. Maybe I can try to kill them before even gets his lair tech going.” Now all of a sudden we have bunker rushes and fast tank pushes being motivated from this idea. Your idea could be something tactical, like “Wow, this bridge is very narrow, I bet if I controlled this my opponent would have a very hard time breaking out this front.” This is a great motivator for a huge number of builds. And although it sounds a little vague a build is an optimization of an idea, I want to use that definition because it is very important that I keep referring back to my initial idea. And make sure that I’m not contradicting that or violating it in some way. Now it’s time for the second question. “How do we talk about build orders?” I want you to know that this is the most important segment of this entire presentation. Alright, get ready! A build order is divided into several key components, these key components that are the tings that are critical for accomplishing your idea. A build order therefore is how you arrange these key components leading up to the execution of your idea. The most important aspect of this notion of build orders is that you can rearrange these key components and you still will en dup accomplishing the same big idea you had in mind if you do it right. Way way way too many starcraft players think of strategies as this huge branching tree that you open strategy X. And if you see your opponent do this, you move down this deviations on the left. And if he does that, you move down this deviation on the right. What a clever player does, is he says, “I wanna go down this middle path. “And my responses are not gonna be going down the different deviations on the left and right. Rather I’m gonna rearrange the key elements of my build in different fashions to make sure that I’m always going down that middle path so that everything is the same in the end anyways. After I have figured out all the rearrangements based upon the various responces, that is when I fleshing them out and getting them down very precisely until at the very very end, I have an instruction set based upon food that I can execute every single time. Notice that the food instruction set that most people would describe as the “build order” was the last final thing we did when we were trying to make a build. And if all this sound super abstract and theoretical, don’t worry because I’m gonna move on to part 3 right now. A super simple example that will demonstrate these ideas in a real game.

So suppose you’re a terran player, and you’ve been trouble against zergs. Here is a good idea for a build order, you would say to yourself, “Hm, I’m having a lot of trouble against mutas and lurkers. Maybe I can kill that zerg player before he even gets them. I’ll need some tanks to break down his sunkens at the front, and I’ll need some medic marine to defend my tanks. And to run in once those sunkens are broken and hopefully if I do this right my zerg player will be dead before he even gets his lair tech off the ground. Great! Our big idea is killing zerg veofre lair tech with medic marine tank push. Now it’s time to identify the key opponents to this build. What are all the things rthat are necessay to make this build work. I came up with a list of 6 key elements, which might seem like a lot, but it’s a hella of a lot less than remembering every little instruction of some detailed build order that you find in the forum. So with that said, number one. I need to get two barracks up so I have enough medic marine to defend my tanks. Two, I need a factory so I can get tanks with siege mode. Three, I need to get an academy so I can research stim and get my uhhh firebats and my medics. I need stim for my marines and firebats because they’re really not very good without it. And potentially 6, for maybe an engineering bay. Again I want to list it because I might need it to build turrets to help bolster my push up front because he might have mutalisks up. But there’s a chance that my build is so good that it’ll crush the zerg before he gets mutalisks. So I might not need the engineering bay at all. One I need a second barracks, two I need a factory, three I need an academy, four and five are siege and stim, and six is potential this engineering bay. Alright, sweet. Right now note that we haven’t played a game yet. You’re just a terran player sitting alone in your room still devising a strategy, and it’s important that before you pick up and play that first game that you have some arrangement, some logical arrangement, of these ekey lements in mind. Most logical thing would first to start with the easy part. I think I’m gonna build my engineering bay last. Because again, I can’t conceive of needing it early because you can’t get mutalisks that fast. I mean, you know, the whole point of the build is to kill him before he gets mutalisks. And the only thing to get it early would be to get some sort of attack and defense upgrade but I don’t think that would be done in time for the tank push, so e-bay would probably come last. Siege mode would obviously come second to last because I would want to start researching it so that it would finish right when the tanks get to the front of his base. Stim would come next before that so I would have stim by the time I get to his base perhaps along the way. So right now my last three elements in order are stim, siege and engineering bay. Now I have the three perhaps most important components of this build which are the two factories, the barracks, and the academy. So continuing in this reverse the academy seems like the last of those three that I’ll get because I don’t need the firebats or the medics too early because I can just hold my ramp with marines early on. And again, I’m not rushing to get stim early so right now the only thrings we have left are the two barracks and the factory. And the last four elements in roder are going to be the academy, stim, siege, and engineering bay. And there’s a little bit of debate about whether I should make that second barracks first or that factory first. If I make the two barracks first I will have a lot of units early on and I’ll have a lot more medic marine for when my tank push comes out, but it’ll be a little bit slower. Likewise, if I get the factory first my tank push will come out earlier. But I’ll might not have as many units as before.

Most important thing when you're faced with this pro con analysis and you're not sure, is just to pick one. So right now I am just going to pick two barracks first, followed by a factory, then my academy, then stim, siege mode, and an engineering bay. Now it's time for me as the terran player to start playing the game. And what I'm gonna do now, is think about how to make these key elements relative to one another. I have an ordering for them but I might say, "Start siege mode when stim pack is half done" or something like that. Or, I might wait till my second barracks is finished before I even consider building my factory, or I might be able to build my factory while my second barracks is making. All these sorts of things are what you flesh out through playing your build order several times in a row. Anytime you're doing a build in practice and you have these key elements all set up and everything seems to be going well you'll still have a handful of big questions that you'll need to answer. So for example with this fast tank push build an excellent question is "should I be cutting marine production at anytime? Should I stop making marines at the start so I can get that factory up really fast and then go right back to pumping out marines from my two barracks?" If you're unsure again, just pick something and test it out. So for example, lets say I'm my terran player and I start out I get my two barracks which is my first key elements and I make five marines before I get my factory started and then I do everything as appropriate. I get my academy a little while afterwards and then I get stim, and siege mode and move out with medic marine army once I have a tank or two. I might get to his base and say to myself, "Wow, I'm a little bit late with this tank push. He already has his lair units and I have trouble getting this tank push started. Moreover, I seem to have way too many medic marine." Well, ok, duh, early on I clearly made too many marines because I seem to have an excess, and that also appears to have delayed my tank push. So then I might cut back and say, "Ok, well let me make only one marine." So lets say I've now done this build 10 times or something like that, and everything is feeling really really good, I have all my timings set out, I have this ordering of my key components, and when I build them is based on the other key elements. I don't say something like, make my academy at 30 fo0od. I would say something like, make the academy when the factory is halfway done. As a result I'll have a much easier timing set should I encounter any sort of variation. Now with this in mind, let’s think about the true power of using this key component build order set. Let's say for some reason that in those 10 games I practiced where everything was feeling really great I was only playing against super standard zergs who did something like 3 hatch opening and they gassed on 16 and went straight for a lair for mutalisks. Suppose in my eleventh game doing this build I encounter some zerg player who's going 3 hatch mass zergling. He seems to have spent his first 100 gas on zergling speed and I'm seeing very few drones being made from those hatchers. What do I do now? Well the key idea of my build was to try to kill my opponent right before he got his lair units up and running, right before those mutalisks and lurkers came around. So, the appropriate way to respond then is not to discard this whole fast tank push strategy, it's to rearrange the key elements to respond to these zerglings. So in an appropriate response would be to get the academy before the factory, as a result I can get more firebats and bigger medic marine army and later tanks. but again, this is ok, it's ok that my fast tank push is a little bit later because his lair is a little later as well. So again, I still get my stim and my siege mode and my tanks coming out with the same relative timing that I was using before. I just got that academy before the factory, that way I could get more anti zergling units so my later push could deal with his zergling army, but still crush him before he got those mutalisks out. Suppose in my twelfth game using this strategy I encounter an opponent who's going fast 9 pool speed zerglings. I can't now suddenly just get one marine and go straight for my factory as I did in my original build. I should be getting a lot more marines early on to hold off that zergling number. Moreover, this again doesn't screw me up the idea of my tank push suddenly isn't invalidated because he got zerglings early. He still almost certainly will be expanding and getting mutalisks. So it's ok for me to make those extra marines and then proceed as normal, get my factory, my academy, my stim-siege, and possibly my engineering bay to wrap things up. Again, this is the key idea that I want to hammer home again, and again, and again. A strong player with a good idea for a build doesn't not look for ways to deviate from his build based on what his opponent is doing, rather he looks for ways to rearrange all the crucial elements of his build so that he can continue down the same path that he wants every single game. It's all about rearranging key elements, not about deviations and changing your game plan. With all this in mind, I want to move on to part four, which is a personal anecdote of how I created a zerg vs terran build on Gaia using this build order theory I've just described. I started with a general idea, identified the key components, and then figured out how to respond to all the different possible terran openings to make sure my build order worked the exact way I wanted every single time.

Before I say anything, let me briefly note that this build is not for vertical positioning on Gaia. That is, it is not for top right vs bottom right or top left vs bottom left. Those games play out totally differently and really weird so I'm not even gonna go into it. So, lets suppose my opponent and I are on opposite sides of the map. like I'm top right and he's top left, or I'm bottom left and he's top right, something like that. The big question I always ask myself whenever I'm developing a new build or a new style of play on a map is "what do I want my mid game to look like?" The answer to that question is my big idea of my build order. That has a lot to do with the fact that mid game tends to converge in Starcraft, which is whole other discussion, but again, just to sort of breeze right through it, I just ask myself what do I want in my mid game to look like? On Gaia, I note that in the center is a big, fat, wide open lane. it's not quite an open circle like on Lost temple or on Luna, it's more of a wide path. So when I look at that, in the mid game, I want a really big lurker-hydra army to crush terran's first push. That is my big idea. Is to secure the mid game with hydra-lurker. Yeah, sure, I might be going for Guardians, and defilers, and Ultra's later on in the game and I might open up with lurker ling or something like that, but in the mid game I want to have hydras and lurkers as my main defense. So, what are the key components to this build? Obviously I need the three upgrades, hydralisk speed, range, and lurkers. I also want two evolution chambers upgrading +1/+1. I want two extra gas so I have four gas total. I want those expansions to be my side expansion and my natural at the other main. I don't want to take the other main on my side because my big hydra lurk army is a little slow so if that main got dropped it's kinda tough to get all my units up there. However, if I just took the natural of that main my hydra lurk can just mosey right over there and defend it much more easily. I need a spire so I can get scourge to kill drop ships. Again my hydra lurk isn't too mobile, if I have a big hydra lurk in the middle to hold this push, and he drops in my main, it's pretty tough for me to get all the way back into my main, so those are pretty deadly. I also want overlord speed because it's really nice to be able to spot his army completely. I'm not gonna be doing any cute harassment stuff with mutalisks, so I won't get the vision potential given by those mutalisk. I need overlord position in really nice place. Again those overlord with speed are also instrumental in, in holding off dropships with the scourge. And the last component I want is Queen's Nest. I'm not going into extensive detail of what I'm doing in the late game, all I need to know is that I must transition to that at some point, so I may as well throw the Queen's Nest and the Hive in there. So my short list of key components is: Hydra speed, hydra range, lurkers, two evo chambers upgrading, two expansions, a spire with scourge, overlord speed, and one thing I didn't mention earlier is that I might need ling speed in some of these variations. Zerglings are very good early on. Let’s discuss now all the openings that terran can do and how I rearrange the key elements of this hydra lurker build so I have a big hydra lurk army in the middle. In virtually all circumstances I'm gonna be opening with the standard three hatch sort of build, I'm gonna be going 12-hatch, 11-pool, 13-hatch, totally normal opening, gets me a lot of drones, awesome. Let’s talk about when my opponent does a fast tank push, the same tank push I've described earlier in this talk. First of all, think about what zerg actually sees, zerg sends his scouting drone out sees two barracks and refinery go up kinda quickly. But then that drone has to leave, fortunately I do have the following information, my terran opponent is pretty much not really expanding, doing some sort of one base play against me. So he's gonna be having more units early on and he's gonna be getting his starport faster with drop ships. So in my base I'm gonna open up with my standard three hatch tech, go for lair, and get lurkers and zergling speed but I'm gonna cut drone production and make sure I have enough units. I'm also gonna be getting my spire along side that lurker ling combination because I know that for example if he did some sort of fast tank push a lot of players will follow that up with a drop ship. or he might even go straight for drop ship and it's really important that as I push out more to the middle of the map that I have the scourge in time to kill any drop ship harass he's doing. A potential concern would be that if I'm getting that spire and I'm getting a bunch of lurker ling early on my economy would be wounded, not only do I have a lower drone count, but I will delay taking those two expansions that's part of my whole build plan to begin with. This is ok however, because although my economy is delayed, it's delayed proportional to my terran opponent. My opponent is one basing, so his economy is behind as well, so it's ok that I don't have nearly as many drones. It's more important for me to just not die. So in some circumstances if I know I won't be able to get out of my base, so let’s say he's really aggressively pushing, he has a bunch of bunkers and turrets outside of my base, I'm totally comfortable getting my evolution chambers early on. I might not get two evolution chambers to get the +1 range attack and +1 carapace but I'll throw down one to get the carapace upgrade early on. And again, if I'm still being held in my base and I can't take those two expansions, I'll get overlord speed. Once I do end up breaking out, taking those two expansions, and powering just briefly maybe getting 6 or 7 drones, that's when I'll get hydra speed and hydra range as well as another evolution chamber for that ranged attack upgrade. As a result, in the end, I'll have that same mid game plan that I wanted. I'll have all my expansions, my big hydra lurk army in the middle, but it will have been significantly delayed because of this early pressure my terran opponent put on, and again this is totally fine because my terran opponent equivalently delayed his expansion. As a matter of fact I'd play pretty similarly if my opponent does some sort of two barracks fast academy type of play. I'd just make sure to get enough lurker ling to push him out and gain center control. And then I go right back to the same general style aiming to get my expansions, fairly quickly, followed by my +1/+1 upgrades, and if there's any delay I in getting those expansions I will just get my evolution chamber a little earlier and my overlord speed a little earlier. In short, against one base play from terran that seems to be medic marine based, I simply delay getting my two gas expansions and focus more on getting all those critical upgrades and buildings first like the spire, the evo chamber, and the overlord speed upgrade. In some circumstances the terran player won't even expand for a while after his one base opening, he'll add on a third or maybe even a fourth barracks and start pushing you aggressively with one factory, and one starport making vessels, and 4 rax making marines. Since that's pretty difficult to deal with without making a lot of units, I will just happily delay my third and fourth expansion. I'll generally take my third gas right when I bust out of whatever early pressure he has, but I'm totally comfortable waiting a little while to take that fourth gas to make ensure that I am in fact making enough units to hold off that aggression. In all cases, the end result is that, in the mid game, I have a big hydra lurk army, I have four gas, two evo chambers, a spire, and overlord speed upgrade. And after I've done all that and I'm feeling pretty content, that's when I throw down the Queen's Nest and in all these circumstances. Let's consider something totally different. Let's say I scout into his base and I see my terran opponent has one barracks and a refinery. Immediately I note he's doing some sort of fast tech opening. Briefly I'm going to ignore cases where he goes all out mech. Like does some sort of goliath, tank, vulture composition for the remainder of the game. I'm gonna stick to these medic marine army and well address that in a little bit. But so right now based upon medic marine openings he could be going for fast drop ship, he could be going for two port wraith, he could going for one factory vulture harass into expand and then get a bunch of medic marine. All these sorts of techy annoying options. If I see this, then I do again with three hatch but I will get a hydralisk den and then I'll get speed [sic] really early on. As a result I can handle some vultures running into my main, I can handle wraiths, I can fend those off pretty easily. I also can deal with even if he does weird stuff like goes one rax fast academy and comes out with super fast firebats or something like that. Cause I have these hydras with range [sic], I can hold all of that off effectively. And then, once I get my lair up, note that I'm getting lair before range for hydras [sic]. Once I get that lair up, I research lurker spines, and again, I move out into the middle, burrow a bunch of lurkers there, and I take my third and fourth gas. And everything's pretty much the same from here on out, I opened up with hydralisks and delayed my lair and made sure to get speed [sic] to hold off his early aggression stuff. But after that, I pretty much went for lurkers, took control of the middle, and now I'm expanding and powering drones and I'm going back and getting my +1/+1 upgrade. I'll probably will already have gotten speed for overlords, sorry, there's a huge truck passing by, I'll let that one go by, ok. I'll probably have gotten speed overlords already if I see him doing something like fast wraiths, if he's not doing that I may well skip it and get a spire followed by that overlord speed upgrade to sort of wrap things up. And again, my mid game is identical in this circumstance. And then lets finally consider the most common opening for terran, he opens up with a one rax and early expand or even two rax fast command center. In this circumstance I'm more than happy to make a few extra drones, I'm still gonna be pushing out with lurker ling early, but instead of maybe 36 to 40 lings, if he was doing some sort of super aggressive opening, I might have somewhere like 18 to 24 lings. Just enough with my lurkers for me to push out to mid and secure that wide alleyway so I can take my two expansions. And then I'm very comfortable powering drones for a little bit and then I'll get my +1/+1 upgrades before I get range and speed. Cause again, I only need range and speed right before his big attack in the middle. And after I take these two expansions and I get my +1/+1 upgrades that's when I generally get my spire for the scourge and drop ship control. An interesting thing to note is that I didn't mention overlord speed in this discussion against early expanding. What I discovered when I did this hydra lurk opening is that there were a handful of circumstances where I did have to deviate. One of those circumstances is when my terran opponent would early expand and get two factories very very quickly. When his middle game push came out, although it came out a minute later, around the 10 second, or 10 minute 30 second park as opposed to usual the 9 minute 30 second mark for a early push. Even though his push was a little bit later, it just had so many tanks that I had a huge amount of difficulty dealing with it despite the fact that I had this oversized hydra lurk army. As a result against an early expanding player I tend to get my overlord speed upgrade very quickly to see if he is getting that two factory playing. In circumstances where he is doing that, I generally just skip hydras altogether and I go straight for a Hive. I just generally do the lurker ling, and then go straight to Guardians, and Ultras, and defilers and all that good stuff. And also you'll note I skipped talking about mech. Hydra lurk is quite bad against mech play so I also have a transition out from that. Remember that I said against a one basing, gassing terran player, I generally open up with hydras with range upgrade [sic] and then go straight for lair. But if it appears he's going mech, I won't get lurkers I'll just throw done a spire. And that way he's forced to make some goliaths and he's forced to make some tanks early on to deal with my hydras. This allows me to take two more expansions as usual and instead of going a hydra lurker army I got a hydra muta army.

Lemme sort of recap cause I've done a whole bunch of rambling about all sorts of variations. In essence what I'm doing is against one basing medic marine players, I'm delaying my expansions and getting more upgrades and the spire and the scourge and all that critical upgrade stuff first. Against a one basing mech player, I open up with hydras. If he's continuing going medic marine I transition into that same lurker hydra play and if he's going more mech, I transition smoothly into a muta hydra with lots of expansions play. If he's early expanding I just push out into the middle as normal and expand a little bit more and make a few more drones and then go back and get all those critical upgrades and buildings like the spire, the speed, the +1/+1 upgrade. And if he is in fact going for this two factory, really fast, I just smoothly transition away from this hive type play. In all circumstances, I tried as hard as I could to make my big idea work, to try to get this hydra lurk army to work in the middle. And you'll note for the most part I got that to work in every case except for two, cases where he mechs and cases where he goes early expand two factory. All the rest of the time I did an excellent job of forcing the game to go down the exact path that I wanted to. So again, even if all this just seem like blabber the important thing to note is to force your gameplay down one path by rearranging the key components of your big build order. So slight apologies for having a little bit long audio, this thing just ran over 30 minutes but hopefully this was really helpful and I'm actually I glad I took this long to talk about this concept because, really, thinking about build orders the right is so key to your learning. If you try to focus on all these food instruction sets, you'll get lost in all the details and miss out on the big pictures. So start on the big pictures and then work your way down as an end to that food instruction set for all these deviations I just described, I do have a loose notion of what I'm doing based on food. But all of that is based upon my foundational logic that I began at the outset of everything. So, good luck everyone, cheers.

Hey what's up everyone, this is Sean Plott, a.k.a Day[9], and today, "Relative Timings." But before I can go on about that, we just have to stop and jam out to this song for a bit, ugh, god, it's so good.

That is one of my all time favorite songs. It's called "Snowflake" by Trent Moller (sp) on his album "The Last Resort", which you should all get by the way, because it will totally change your life. So speaking of changing lives, lets talk a little bit about the concept of relative timings in Starcraft. So in my last podcast, which was "A New Look at Build Orders" I mentioned the notion of relative timings here and there and I wanted to spend some time going into the concept in depth. So the idea behind relative timing is fairly simple, relative timing is the idea is making the unit or building or upgrade based on the progress of some other unit or building or upgrade. Not using the food in the top right corner. The classic example of this is lurkers. If I want to get lurkers as quickly as possible, I should get my hydralisk den and my lair to finish at the exact same time. This timing's really well known, you just make your hydralisk den when your lair is exactly halfway finished and both will finish simultaneously so you can start that lurker upgrade right away. Another classic example is with reavers. You start building your robotics support bay when your shuttle is half way finished. That way the shuttle finishes right when the robo bay finishes and you can start both simultaneously. So now that we know what we're talking about the most important question is why is this concept useful? And the idea is we always in Starcraft want to squeeze out as much efficiency as possible. Technically it doesn't really matter if your robo bay finishes a little bit early because you'll still be making the reaver right when the shuttle finishes, but in game like Starcraft where it's so intense and how you use resources is so important, why not figure out these timings because if you make that robo bay you might end up with a slightly later pylon or you may not have been to squeeze in an extra dragoon or probe. And not being able to do a little things like that, throughout an entire game, adds up tremendously. In general, there are three situations which relative timings are useful. The first is if you get thrown off early, or thrown off your build. The second one is for decision you'll be making late game. And the third one is for organization, for things like macro. So lets go through each of those with some examples and stuff.

So lets begin with getting thrown off your build order. Suppose for a moment that you're playing a zerg vs terran and you want to be going lurkers in the mid game. If you get bunker rushed early on you're gonna be using a lot drones to hold that off. You might lose a lot or a little or you might have to make a bunch of extra lings and you'll end up in this sort of unfamiliar circumstance. But if you remember your standard lurker timing you simply make your den when you're lair is halfway finished, both finishes at the same time, and given the circumstances, you're still getting lurkers as fast as possible. So this example might seem might seem pretty obvious, because you start the hydralisk den at such a close time to each other that it's kinda hard to mess up. However, this example highlights the key idea of relative timing, which is that you want to do one thing that requires multiple other things. So relative timing is figuring out how to get all of those things to line up. So a great example of relative timing and its power is in the old school Protoss vs Zerg style, and in old school Protoss vs Zerg, Protosses would early expand, and rather than go for corsairs, they would spend their first 100 gas getting +1 attack upgrade and then they would go for zealot with leg speed and they would gate 4 gateways and they would come out right when +1 finished and zealot leg speed finished, they both finish at the same time, and they have about 15 to 20 zealots, and that was a huge issue for a lot of zergs to deal with. But the problem is that a lot of zergs were really aggressive early on, some of them would try to do 2 or 3 hatch ling all ins, some of them would 9 pool, and a lot of these protoss players that were doing this build order would get screwed up and I would watch a lot of replays against my opponents and see that sort of thing happen where I would make a whole bunch of zerglings early and then once they got on their feet, they would start their +1 attack upgrade too early so it would finish and they wouldn't have leg speed done or they wouldn't have that many zealots. A really funny example: one guy got the zealot leg speed really fast and +1 attack upgrade really fast but he only had 4 zealots when he came out to attack and was totally not a problem to deal with. So in that example the Protoss was going for zealot leg speed and +1 attack upgrade and having a lot of zealots, those are the three key pieces of that attack. So regardless of what happened early game, whether he got to early expand normally, or he got 9 pooled or something like that, he should know when to start the cybernetics core relative to when he started the +1 attack upgrade so that he can get zealot leg speed and whole bunch of gateways and have everything finish right on time.

Relative timing is also very useful in the late game, when there is some action you'll be doing very often in a lot of your games but when you do that action is high variable. So a great example is getting arbiters in Protoss vs Terran. A lot of times it's up to the Protoss' judgment when he feels comfortable getting arbiters. So we're not going to talk about when it's right or wrong to do that, but what's important is that once you have decided to get arbiters how do you go about getting it in the most efficient way possible. Surprisingly not many protosses have this number stored in their head, they just generally make a stargate and a templar archive at the same time. And then when both are finished they start a arbiter tribunal. Technically you should start a templar archive when a stargate is at about 1/5 or 1/4 completion, that's when they'll finish at about the same time. All this really means is that the Protoss should start making the Stargate before the Templar Archives by just a little bit. And what's very useful is especially late game is that you'll be able to get 1 or 2 extra units in by making sure you line those up in the timing I just said. And again stuff like that might seem inconsequential but it definitely adds up over time if you know a lot of little relative timings. So still in regards to late game, relative timing is great for little tactics such as storm droppings. So suppose you're a Protoss player you have the robotics facility finished, you have your templar archives done, but you don't have templar, storm, or a shuttle. When should you research those three things relative to one another so that you can storm drop in the most efficient way possible? So theoretically that means when your shuttle unloads your templar or shortly before that is right when your storm finishes. Now we're talking about something that is a little bit more clearly useful. It's really great if you're a Protoss player if you don't have to waste the 200 gas upgrading storm if you can delay it as long as possible to build other units that you will need for defense.

The third really powerful application of relative timing is organization. Let's say you're a terran player and you're macroing out of a whole bunch of factories. You optimally want all your factories finishing units at the same time. So lets say you have 4 factories making vultures, you want all the vultures to pop at the same time you only have to go back once to those factories and hit v four times to build those vultures. A lot of terran players will have factories that kinda have a stuttered production, and they'll end up with units cued and that's really really bad to have units queued at your factory. You're essentially wasting money that you could've been using elsewhere. So they might have three factories where the vultures finish at the same time and one factory that's a little bit delayed so it has a vulture cued at all points in time. So how do we use relative time to avoid that? Naturally we're gonna be doing it the exact way we did before. Suppose I already have two factories and I want to make three more, when should I start factories 3, 4, and 5 so that they finish right when factories 1 and 2 have completed making whatever units they were making. So now I have 5 factories making nothing and I can start units in all those 5 factories at exactly the same time. Now I know a lot of you are thinking, "Oh, well that's a handy thing to think about." But I need to emphasize how cruuucial it is that your organization of your gateways and factories is really really good, because if it's not, you'll be, one you'll be shooting yourself in the foot, you'll at least be shooting yourself in the face for the whole game because there's a lot of players out there who are really good at saying, "Hmmm, now seems like a good time to add on factories and gateways" and they'll just immediately start building 'em without any thought as to whether appropriately timed. As a result, when all the gateways and factories finish, they have this stuttered, slightly off production in all their unit making structures and as a result they'll have maybe 3, 4 units queued up all over the place, and if you think about that's probably 4 to 500 maybe, even 600 minerals that are just sitting queued doing nothing. That's like an expansion, that's 4 cannons, that's a big deal. So definitely you get those relative timings set up right so you can have really smooth macro that will not only save you money but make macroing way way easier so that will naturally improve as well. So in your own gameplay, try to think of as many situations where you could try to get things line up in a relative manner so you don't have to rely on the food in the top right corner. Which allow you to play dynamic variable games a lot more effectively than you used to be able to.

So let's just finish by asking a whole bunch of questions, like a bunch whole of little brain-teasers for each race as motivation to figure out these sorts of timings. Lets begin with zerg, suppose you're playing a zerg vs terran and you're wondering how you can get defilers out before your opponent's first push gets to your base. The question you'd be asking is, "When should I start my Queen's Nest, to get Hive and defilers, relative to when my terran opponents starts his factory?" which will allow him to get science vessel. Answering that question will allow you to get defilers before your terran's opponent's first push. Lets say you're playing a zerg vs protoss, you want to have your spire with scourge out before your protoss opponent has corsairs killing all your overlords. When should you start upgrading to a lair relative to the progress of your opponent's cybernetics core, or even relative to the progress of his gateway so you can have your spire out to block his corsair. Lets say you're playing zerg vs. zerg and you're really tired of losing overlords all the time when your opponent's spire finishes and he gets mutalisks. So here's a question you to ask yourself, "When my opponent starts his spire, how many hit points does the spire gain relative to how far my overlord moves?" That is, if you figure something out like oh, if your overlord moves this far in the amount of time that it takes for a spire to build, I know around the time his spire's finishing based upon how far my overlord has run from his base. Lets say we're playing terran vs zerg, we want to make sure we push out before our opponent gets defilers, we want to hit him at this crucial time when he feels weak. Again, in the reverse sense, how far along should your factory be relative to your opponent getting Hive so that you know your first push doesn't have to deal with defilers. This would allow a terran player to compose his army appropriately if he doesn't sees queen's nest anywhere in sight, he knows the zerg is gonna be trying using a mass number of units to kill off his army as opposed to some cute defiler tech. Suppose you're a terran player against a protoss player and on destination you want to early expand with some tanks but then you immediately you want to get a drop ship with vultures that have speed and mines. You need to make sure you have enough tanks so that you're safe from whatever he's doing early on but you need to start making the vultures and that starport really fast so you can have that really early harass as well. So you should say, "When do I start making the starport and getting all those vultures relative to when I stop making tanks?" So by answering this question we are both safe and harassing as quickly as possible. So, last, suppose you're a protoss player against terran and you're having trouble against the timing of a fast +2/+1 push. So naturally you'll be able to get an observer into his base early so you will know that this is coming pretty far in advance but you really want to know exactly when his first attack is gonna come. Well you should ask yourself "How many minerals do I mine in my main relative to when he starts his first +1 attack upgrade?" So for instance, if you find out it takes 1200 minerals mined for him to get +2/+1 from the start of his upgrade, you'll be able to plan accordingly. You'll be able to click on the minerals in your main and constantly see how close you are to him getting +2/+1. So that about wraps up all the different sorts of examples, so hopefully this is gonna be a easy enough thing to incorporate into your own play so you can squeeze out as much efficiency as possible. Best of luck, everyone, cheers.

      Hey everyone, this is Sean Plott, AKA Day[9], presenting today, a simple Goon Reaver push for Protoss. So you hear me say time and time again that the most important component of having strong strategy is having really good transitions. It's linking together all the parts of your game in a seamless fashion.
      For instance, let's take very standard Terran vs Zerg. The Terran opens up with a Medic Marine Tank Vessel push and when Zerg starts approaching Hive tech, Terran gets a second Starport and starts going Vessel. At this point Terrans tend to stop making tanks and they start making Vultures with mines because then they can save all the gas that they would have spent on tanks on those extra Science Vessels and the +1, +1 upgrades, or I guess +2, +whatever upgrades, at that point, for the Medic Marine army. And this simple transition is great because it allows the Terran player to move from a tank-centric mid-game, that involves a three to four Tank push and one Vessel, to a Vessel-centric late game, where the Terran has a full control group of Vessels and has Mines planted everywhere, and it's much more focused on mobility in the late game as opposed to some really strong, one-punch style push.
      With that in mind, let's stop and look at UpMaGiC vs EffOrt, a Terran vs Zerg on Outsider. Upmagic opened up with a 3-Port Valkyrie-Wraith opening. It was really aggressive and really, really unusual, and it sort of threw EffOrt off kilter. I mean if you saw the live reports or the chats during that game, everyone was saying, "Awww, I love UpMaGiC, he always does these awesome, aggressive openings. Go UpMaGiC!" Personally, I like this style of play, but when I saw it, my immediate concern was, "How on Earth is Upmagic going to transition out of this? He doesn't have any important tech structures; he only has one Factory, he only has one Barracks. What is his stage two and how is going to get there without dying?" And if you watch the game, a lot of players were shocked when Upmagic lost pretty easily in the later stages of the game. EffOrt pretty much looked like he was losing the entire time, and then just won all of a sudden. And that is a direct result of the fact that Upmagic's play has a very difficult transition OUT of that 3 Starport aggressive play.
      So with these two examples in mind, I figured that instead of doing some high level, long podcast that has lots of theory, which is a lot of my other podcasts, I instead would focus on one really simple, easy build that we could look at as an example to thinking about things in terms of transitions. The build we're going to learn today is a very easy Goon-Reaver push for Protoss. I specifically chose this build because pretty much anyone can do it. If you're a D level Protoss, and you're struggling in Protoss vs Zerg, I strongly recommend this build, because it's just so easy to pull off. So first we're going to do a really fast run through of the build and talk about why the transitions in it are so good, and then I'm just going to redo the build a little more in depth, so that way you'll have a very solid starting point to start fiddling around and crushing faces on iCCup.
      So here's the basic play: Protoss opens up with a one Gate, fast gas opening, and goes straight for Corsairs. Instead of getting Dark Templar and expanding, which is the typical play, the Protoss instead gets a second Gateway and goes straight for Reavers. And when the Protoss moves out with his first push of Dragoons and Reavers, he generally wins, or he'll cripple the Zerg and win with the second Goon-Reaver push. And a lot of you are thinking that sounds way too trivial for me to make a big deal about it. However, all the components of this build work together so well. So the basic underlying idea is that the Protoss screws up the Zerg's economy enough early on, so that the Zerg can not have enough units to hold off that Goon-Reaver push. And this is a subtle point that's really important to make.
      A lot of players think, "Oh, I'm just gonna screw up his economy, I'm gonna storm drop him and harass him because that's - that's good." But you should be doing that harassment and putting him at a setback so you can take advantage of it in some fashion. With this Goon-Reaver push, the Corsair comes out and the Zerg now has to spend all this money dealing with anti-air, and he has to pull all his Overlords back, so he's lost information and he's lost all that money trying to keep his existing Overlords alive. But right when Zerg is feeling comfortable holding off that Corsair, Protoss is coming out with two Gateways worth of Dragoons and Zealots, so now the Zerg is spending all this money on Sunken Colonies at his front. In fact, a lot of times when you watch this build in practice, if the Zerg goes 3 Hatch, one of those Hatcheries tends to start building up larvae, because the Zerg just doesn't have enough money to build stuff from it. At last, when Zerg is feeling confident that he can hold off all these Zealots and Dragoons at his front, and hold off those Corsairs running around, a Reaver comes along, and pushes right up the middle of his base to those Sunken Colonies and then kills them. Moreover, this build is so awesome because it's so logical and easy to do. Pretty much anything that the Zerg does, this Goon-Reaver push can transition to a counter really easily. If the Zerg is making a ton of Zerglings early on, then the Protoss doesn't need to move out with his initial Zealot-Goon army, he just waits until he has a Reaver and deal with it. If the Zerg is going for really fast Mutalisks, well, Protoss already has two Gates making Dragoons, and he can immediately start making Corsairs out of that Stargate. If the Zerg starts going Lurkers, you can just get an Observatory because you already have a Robotics Bay (Facility). And all this is so easy because you got that Corsair early on, so you see exactly what the Zerg is doing, and you know exactly what position he's in. Plus, since you're one basing for so long, you don't need to worry about some sort of elaborate defense for your first expansion. You have a ramp with a whole bunch of units on top; you're fine. So we have all these easy transitions and most importantly, everything we do has purpose. We're not harassing for the sake of harassing, or messing up his economy for the sake of messing up his economy. We're doing all those things specifically because it magnifies the strength of our Goon-Reaver push.
      So let's go through the build once more in detail and wrap things up. The Protoss will open with an 8 Pylon, 10 Gate, and in general, I like to keep those pretty close to my Nexus, just in case the Zerg tries to do some early Zergling stuff, there is no need for it to be all the way at the ramp, in the case I was 2 Gating, or something like that. Protoss gets 11 Gas, 13 Core, still hasn't scouted yet - I like to keep a probe mining in my base instead of scouting just because the extra minerals makes things just a little bit more smooth, and plus, you're not going to be doings any major deviations early, so scouting him ultra-fast, like after your Gateway or your Pylon just is not necessary. At this stage, I strongly prefer to get the Stargate before the first Zealot because as a Zerg player, it really messes me up much more when that Corsair comes out unbelievably quickly. In fact, a lot of times, I don't even have enough gas to start my Hydralisk Den at the time that the first Corsair pops out, so I always prefer the faster Corsair because of the horrible position it puts the Zerg in. And I know a lot of people really like getting that first Zealot early on, because they think they need it for the 9 Pool. You really, really don't. In fact, of all openings the Zerg can do, then 9 Pool is the most important to get the fast Corsair against, because the 9 Pool has NO defense against it; What anti-air is he gonna get? With just 8, 9 drones, or however much he'll have at the early game. That - the fast Corsair will screw over 9 Pool the most, and the way that you can hold off the 9 Pool, the actual Zerglings that get to your base is to make sure the second unit out of your gateway is a Dragoon. A lot of players, when they go fast Corsair, go Zealot, Zealot, Dragoon. You always, ALWAYS, no matter what the Zerg is doing, you wanna go Zealot, Goon, Zealot. Getting that Dragoon really, really, early, is absolutely key. It will allow you to hold off Zerglings much more easily and any Overlord that tries to sort of sneak in your base and see what's up, you can kill it without having to bring your Corsair all the way back.
      So, at this point, my Corsair is getting ready to start running around his base, I have my Dragoon and my Zealot, and I'll be making a third Zealot - Or excuse me - my second Zealot, my third Gateway unit, and at this point, I have two choices: Based upon what my Corsair sees, I can either get my second Gateway, and then get Range, and then my Robotics Facility, or I can get my Robotics Facility, then my second Gateway, and then get Range later. In the first case, if I get my second Gateway really early on, I can put a little more pressure on the Zerg with just Dragoons and Zealots. Naturally, the disadvantage is that my Reaver is more delayed. In the second case where I get my Robo Bay, then my second Gateway and Range, this means that my Dragoon Reaver push is a little faster. I can still do pressure with Dragoons and Zealots in the early game, but it might as threatening, it might not force my Zerg opponent to as much as I want him to, and a really important timing to note: If you start the Range upgrade RIGHT when the Robotics Facility finishes, then the Range upgrade will finish right as your Reaver does, and you can move out with about 12 Dragoons with Range and Reavers. Personally, I prefer the second choice, I prefer to get my Reaver out a little faster and have fewer units, but make sure my push happens very, very quickly, definitely feel free to experiment, just wanted to make sure those two options were very clear.
      Whichever one you choose to do, however, it's very important to note that around 40 food, or maybe even as low as 36, 37, you can start cutting Probes. You do not need nonstop Probe production when you're doing this build. Nonstop Probe production, SCV production is used to make sure when you're expansions finish, you can transfer Probes and SCVs to them to make sure they're operational at max capacity immediately. However we're not going to be expanding any time soon, so they're really no point in getting that many probes early on. Given the structure of this build, it's much more important to have all the units out, in maximum numbers, as quickly as possible. At this stage in the game, you have so many options. You're first push, a lot of the times, will kill him. If that doesn't work, and it looks like it did a lot of damage, you can choose to add some more gateways, and start really pounding him down with Dragoons and Reavers, and if you're low on gas, you can throw some Zealots into the mix. If your opponent seems to be turtling really hard and and going for Mutalisks, you can get a whole bunch of Corsairs and Dragoons, and you can do something cute like an elevator drop with your Shuttle. Or if you just think the game is looking pretty even, you can take your expansion. Again, the basic structure is just Corsairs into ground units into Reavers, and it's so fun, because you seize control right from the outset with the early Corsair, and you never let up, all game long, and you have so many options, no matter what the Zerg is doing, that it's just a very, very solid way to open.
      And just a few other tidbits to note before I wrap things up, this build works great on maps that have very close starting positions. So, for instance, Requiem, is one of the best maps that I can think of right now , that I can think of right now. It would be much more difficult on a map like Katrina, where the starting positions are very, very far away, and the Zerg can be in a really good position defensively, with that fat ramp at his front, and the ability to make more Sunken Colonies there. So just keep that in mind whenever you're using this, that on maps where things look like they favor aggressive play, this is the build you want to be using.
      So I hope all of you who listened feel like you have an awesome Protoss vs Zerg build to test out and most importantly, I sincerely hope that you can use this as an example to improve your own analysis of your own play to try and really flush out those transitions and make them really easy and really solid. Uh, I guess that wraps it up. Thanks everyone, cheers!

Hey, what’s up everyone? This is Sean Plott a.k.a Day[9] and today I want to present to you, Mechanics Part 1- The Basics. In this audio, I want to describe what the essentials are for having strong mechanics. I want to begin by talking a little bit about how to hold your keyboard hand and then how to use hotkeys and the F keys effectively in your play. I’d then have to talk about the mouse hand, how to holding your mouse hand correctly, tips for improving accuracy, and general advice on how to control your units the way you want too. Naturally, I will be discussion mechanics with the ideal of improving micro and macro so I will be trying to give lots of examples relating back to that. Great, let’s start talking about the keyboard.

So I keep my keyboard pretty close to the end of my table. Now if you take your left hand and you face it towards yourself, in the bottom right corner of your left hand you’ll a feel a really solid bone if you push there with your thumb. That is the bone that you generally want to have as your pivot point. So I generally keep my hand rested on that bone. Now something that is really really important that you do is that you keep your pinky at your left hand of the keyboard – or let me rephrase that - you make sure that your left pinky does not rest beyond the left end of the keyboard. Essentially what I am addressing is a problem that a lot of players develop when they first started to use hotkeys. Often times when player’s first learn about hotkeys and want to hit Ctrl-1, they hold the Ctrl key with their left thumb and press the ‘1’ key with their index finger. Obviously this isn’t a big deal if you only play Starcraft twice a year at a LAN party or sometimes but there is a lot of players that want to improve that get stuck with this sort of bad habit from the get go. I will go to Tournaments and see players who hit control 123 with their thumb on the control key. As a result when they want to do 1a2a3a, they tend to hit ‘1’ with their middle finger and ‘a’ with their index figure, and then ‘2’ with their middle finger and ‘a’ with their index finger, and ‘3’ with their middle finger and ‘a’ with their index finger. Obviously this is grossly inefficient because you are essentially bending your arm around the left side of the keyboard and hitting the keyboard sideways; moreover your fingers are incredibly far away from the keys that are used to build workers and build units and buildings and stuff.

You want to be holding your hand in the middle of the keyboard so that the CTRL key is hit by your pinky and then the 12345 keys are hit by your index finger or your middle finger or even your thumb for a lot of the higher keys, like if you wanted to hit CTRL + 8 or something. Moreover, don’t be afraid to have your hand move around the keyboard a lot. For example, when I want to do CTRL 0 and CTRL 9, I take my left hand and I move it to the right CTRL and then I hold CTRL with my thumb and press 0 or 9 with my index or ring finger. As a result, I don’t have to do something that is ergonomically inefficient like hitting CTRL with my left pinky and swinging my thumb all the way up to hit 9 or 0. It’s just as easy to move my hand a little bit to the right to take care of that.

In essence, I just don’t want any of you to think that you should ever keep your hand still. I’ve provided a sort of good default that you should rest at the left side of the keyboard but make sure that you’re constantly bouncing around. Not only does bouncing and moving around a lot allow you to do better things in Starcaft, but also it increases the movement and hence the blood flow to your hands so since you’re getting a lot of blood flow toy your wrists and your hands that will help to prevent a lot of symptoms of carpal tunnel and exhaustion and what not. Alright Awesome! So now everyone has an excellent positioning for their left hand.

Let’s talk about hotkeys a little bit. First of all, allow me to say a rule that you should never ever break. Here it is: Do not every click on something that you could build or do by pressing the key on the keyboard. I’m going to say that again: Do not click things if you can use the keyboard. Don’t do it. Don’t be one of the Protoss players who says the ‘p’ key is all the way on the right side of the keyboard. I guess that means I will just click on the probe icon whenever I want to build a probe. That is horrendously newbie. Do not ever ever do that. I don’t care if you’re playing BGH and doing a 2v2v2v2 and going mass scouts. You better be pressing ‘g’ if you want to upgrade scout speed. Now, if you don’t know the hotkey, you should mouse over it, look at what the hotkey is and then press it. Then you will start training your brain to move your hand to the right position on the keyboard. Do not fall into this trap where you just start clicking on things that are somewhat inconvenient. The keyboard is just as powerful a tool as the mouse. You should use it.

I want to spend some time now discussing hotkey setups. When I set up hotkey setups I mean what you do you have CTRL 1-2-3 and all the way up to 0 set to but before I do that I want to talk about the ‘F’ keys. F2, F3, F4. A lot of players do not use the ‘F’ keys because it feels a little inconvenient on the fingers to reach that high however they are absolutely essential for strong play so I am going to talk about them first. So that way when I do discuss hotkey setups like the 1234 all the way up to 0, I can also talk about how to incorporate the ‘F’ keys into that. So let’s begin with the ‘F’ hotkeys.

Let me briefly explain what the ‘F’ keys are for anyone who is potentially unfamiliar with them. So let’s say I select 12 Zerglings and I hit CTRL 2, every time I press 2 it selects those Zerglings. The ‘F’ keys work in the same fashion, however instead of selecting a unit or a building, it centers over a particular screen. So let’s say we are playing on Lost Temple and you are looking on your ramp, I can hold shift and then F2 and go do something else and then every time I press F2 my screen centers at my ramp. Nothing gets selected. It doesn’t matter if I have vision of that area or not. It just moves the screen there. The most typical use for the ‘F’ hotkeys is just to have one screen for each expansion. So I would have my main hot keyed at F2, my expo to F3, my second expo to F4. As a result if I get reaver dropped or storm dropped or vulture dropped, then I can hit F2, make a box around my workers, hit F3 and click on the minerals there and my workers start moving. F2 select the rest of the workers, F3 send them away. Using the ‘F” keys is clearly better than going to your main and making a box and then scrolling to your expansion, clicking on your minerals, and then scrolling back to your main. That takes too much time and all your workers will be dead by then. Another great use for the ‘F’ keys is for unit producing structures. Late game, a lot of Protoss and Terran players will have so many units that they want to use more hotkeys for those units. Moreover, since they have so many unit producing structures, there is no way that they could hotkey each structure individual. So these players will do is hotkey the screen of all their unit producing structures to F2. So they will hit F2 and go back make a bunch of units, and hit 11 and then go back to battle.

Another extremely powerful use of the ‘F’ keys is resetting rally points and a lot of people seem not to even know about this, let alone use it. Here’s the idea. Suppose I am a Terran player and I have 6 barracks and they are all rallied right to my natural. Suppose then I take another expansion and I want to re-rally all my barracks so the rally point is a defensive position for my next expansion. An example of this would be a Terran player on Andromeda who takes one of the left or right expansions. He wants to rally one of the marines up there so they can defend appropriately Here is a great way to use the ‘F’ ketys to re-rally all those barracks in a accurate, clean and click manner. Go to the barracks in your main and hotkey that screen as ‘F2’ and then go to where you want to rally all of them and hotkey that screen to ‘F4’. You hit ‘F2’ click on a barracks, F4 right click on the ground, F2 click on another barracks, F4 right click on the ground and continue this process to all 6 barracks are re-rallied. You can end up doing this very very quickly. Using the ‘F” keys is way better than all the alternatives. A lot of players will go to barracks and they will try to rally using the minimap and the problem with this is that if they are off by just a few pixels then they can 1 barracks that is being rallied to a totally random location way off to the side or something. Or some players will try to click on the barracks and scroll to the area and click and then do this back and forth. The problem is that it takes so much time that a lot of players end up missing a barracks or 2 and then they will have 5 barracks going to the right spot and 2 going again to some totally random location.