Embedded Powerpoints!

Props to Microsoft for this feature. I’m thinking of ideas for a PowerPoint post already. 🙂

Edit: Nice! If you modify the PowerPoint, it will automatically update on the blog and you will not need to repost it!

Edit 2: This feature isn’t too polished yet; for one thing, it takes forever to load sometimes.

Edit 3 (11/29/10): Aww, it stopped working! 😦

Social Networking as News: Examining the 2010 UT Gunman Incident

Yesterday, at the University of Texas at Austin, a gunman armed with an assault rifle committed suicide after firing several shots in the Perry-Castañeda Library. Thankfully, no one else was killed or injured.

I was informed of this in real time when numerous Facebook statuses from my UT friends began appearing on my news feed. (I’m from Austin, so I know plenty of people there, and I am deeply glad that y’all are safe.) Even without seeing video footage or any official report or news article about the event, I knew instantly about something happening 1400 miles away. Certainly instant communication isn’t a new technology. But widespread use of social networking is. As I began to wonder, I realized Facebook was giving something much different from what the conventional mass media would.

Facebook and UT

This is what I realized. I wasn’t seeing news articles on my news feed. I was seeing personal reactions from people at the scene, things that might be considered unprofessional to air on TV, such as how several people said they were happy that classes would be canceled that day (and, to be clear, they were happy for that fact alone, not for the cause behind it). It was also different because I was receiving information from people I know very well, instead of from a bureaucracy.

Granted, there were plenty of news articles appearing online within an hour, but the real question is, What role did online social networks play in yesterday’s incident? For me, I felt a very close, personal connection with the distant event because of whom I received the news from.

There are other factors that I felt connected too, e.g., that I’ve studied in the Perry-Castañeda Library a few times and that UT is my hometown college, but even so, this incident seemed so different from anything before, especially in how it was presented to me. If I had merely watched news coverage of this on TV or read an article, it would be different as I would have been one step more removed from the event. Social networking is not only decreasing the physical distance between us (allowing us to communicate via “broadcast” across large distances), it is decreasing the bureaucratization of media. Perhaps we are more connected, not just statistically, but genuinely.

Will the personal social-networked media, e.g. Facebook and Twitter, replace more “official” ones such as news networks and (online) newspapers? Probably not soon, but we may see a gradual transition. Already, online news sites are increasingly personal. With Google News you can select whatever topics you want to read articles on. The New York Times and CNN are increasing blog-like: they have lists for most recent and most popular articles, corresponding to most recent and most popular posts.

Is networked news necessarily better? Would I rather get news from Facebook or the Associated Press? I’d say official news, but then again, Facebook is less than a decade old, and it has plenty of room for evolution.

I’ve neglected the role of social networking within the incident. Therefore, a question to UT kids and all other Austinites: By what method did you first find out about the gunman? Some possible answers: Facebook/Twitter, email, text, call, TV, word of mouth, sight (you saw the police), personal (the police told you something), etc.

My Writing Sins

There is no way of writing well and also of writing easily.

—Anthony Trollope

Originally I planned this post to be about my writing pet peeves, things I don’t like in other people’s writings. But it would be inappropriate to criticize other people’s work as I’m not a good writer myself, so instead, I decided to write this post introspectively as to talk about my own writing and what sins I commit in it, and what we can learn from them.

This blog contains plenty of examples of my writing. Some posts are written better than others, some worse. The better ones are usually the ones that I spent the most time on, and the worse ones the least. This is because the first draft of any writing, more often than not, will suck. It is a fundamental law of writing. An interviewer once asked Ernest Hemingway about revision, and the great writer replied that he rewrote the last page of A Fairwell to Arms thirty-nine times. What took so long? Hemingway’s response: “Getting the words right.”

But why does “getting the words right” take so much time? Because there are so many factors to consider for each paragraph, for each sentence, for each word. We need to consider tone, word choice, sentence length, flow, and logic. Writing is like putting together a puzzle: you can set most of the pieces in place, and a passerby will usually figure out what the picture is, and be satisfied. Satisfaction is, however, not enough—we want the the passerby to come and take a look at the puzzle, to muse and study it like a painting; and for this to occur, all the pieces must compellingly fit together.

A writing sin is a part of the puzzle that is incomplete.


Sometimes science and cold logic take the better of me. When writing about technical ideas, such as velocity and acceleration, we must be cautious with our terms. We cannot replace “velocity” with “speed,” even if they are synonymous in everyday speech (actually one is a vector, the other a scalar, which could be very important). So, to be correct we end up writing the word “velocity” or “speed” many times without regard to artistic effect.

While mandatory for technical writing, this habit can dull a normal piece of writing. I find I use repetition frequently, even when another word might both sound better and have a more accurate meaning (this shall be the topic of the next sin). The lesson is therefore to avoid unnecessary repetition.

That said, you should not just use a thesaurus and look up synonyms to replace repeated words. If you know precisely what the synonyms mean, and are sure you are using them correctly, then it may be acceptable, but it is in general not advisable. This is because a thesaurus doesn’t give words that mean exactly the same thing. They give words that have similar definitions. For example, if you look up “odor” on thesaurus.com, you will find that one synonym for it is “fragrance.” Surely you can’t use them interchangeably! “An odor swept across the room” conveys a much different meaning than “A fragrance swept across the room.”


This is perhaps my most committed sin. Looking at the comments on an essay handed back to me for my English class, I observe that roughly half are on words or phrases that don’t quite mean what I was trying to say. For instance, I used the word “notice” instead of “observe,” “tease” instead of “taunt,” and even “rather than” instead of “instead of.” (Yes, there is a difference in the last two. “Instead of” involves replacement whereas “rather than” involves preference.)

As this essay was essentially a first draft, this led me to wonder, why is my first draft so filled with imprecision? And furthermore, I wondered, since most of my blog posts are first drafts, how much imprecision lurks in them? It turns out to be a substantial amount for some of them. On the science- and logic-oriented ones, this sin is committed less, while on more free style posts, the sin is committed more. Lesson: when trying to produce good-quality, precise writing, don’t turn in a first draft. This will apply to my blogging. Starting this post I shall put focus on the writing itself instead of just on the content.


I tend to liberally use the phrases “seem to” and “appear to.” Even in that sentence, I used the phrase “tend to,” as opposed to “I liberally use…” I also use the words “normally,” “usually,” and “generally” a lot. Sometimes I try to be so careful in creating a supported argument that I over-qualify it with these phrases, so I cannot be pointed out to be wrong. If someone finds a counterexample, I can just make the excuse that I said “usually,” and the counterexample is just an unusual case. Because of this, I often don’t make strong, emphatic points.

Of course, we shouldn’t go to the other end of the spectrum with absolute words (“all,” “every,” “never”), as that would create false statements. There needs to be a balance: enough qualification so that the argument is supported and can be defended, but not so much as to make the argument worthless. Lesson: Revise.

Dropping Arguments/Not Following Through with an Argument

My instructor for the same English essay also pointed out a few of my good arguments that I left off without explaining or capitalizing on them. For dramatic effect, I’ll leave this section at that. (Edit: Yep, I dropped the puzzle analogy.)

Lack of Flow

Regarding the same essay still, my instructor pointed out several sentences and passages that simply didn’t flow from the previous one or to the next. In one case, I introduced a quotation, blabbered for one sentence, and then made an argument using the quotation!

For flow, we should have each sentence follow the one before it. A passage should not jump from topic to topic. A well-written argument should convince, not confuse, the reader. Lesson: Revise.

Awkward Phrasing

Lesson: Revise.


I actually felt bad for seeing this comment on a couple places in my paper. I had read Strunk and White’s The Elements of Style, and its central message is to “omit needless words,” i.e., to be clear and concise, and have tried to apply this to my writing. It appears to be effective, but there are still some parts I write that are excessively wordy. Lesson: Re-read The Elements of Style, and revise.


If the word “Conclusion” in bold confuses you, because it might either suggest that this passage is the conclusion of this post or that this is one of my writing sins, do not panic: it is actually both. I am horrible at writing conclusions. Writing so many “formulaic” essays in high school killed whatever I had that ability.

I’m certainly trying to improve this aspect of my writing. Meanwhile, enjoy either the lack of conclusions or the really cheesy conclusions in my posts.

Gumption Traps and Writing

This is the first time in over a month that I’ve blogged on two consecutive days. In contrast, I posted for 42 days in a row between late May and early July. I simply lost blogging interest for the last couple months. And it occurred to me that the best term to describe this decline in interest is “gumption trap,” from Robert Pirsig’s Zen and the Art of Motorcycle Maintenance. I haven’t brought the book here with me, so I won’t be able to cite anything (you’re probably glad for this anyway ;)), so I’ll see how much the Wikipedia page jogs my memory.

A gumption trap is what makes you lose motivation and interest in something. It could be that whatever you were working on burst into flames, or you found you could have done it in a much better way, or you simply got bored.

Carl Larsson - The Lazy Corner

It would go too far to say that life is a series of gumption traps, but I have observed it a lot in my life. I’m a chronic non-finisher: I’ll start a project with a complete, well-formulated idea, and I never finish it, or even continue it after a certain point. I suppose that makes me chronically gumption trapped.

When I talk about projects, I don’t mean those for school. I am referring mostly to writing or programming projects, where I try to create something original. My last two writing projects, a play and a novel, have both been gumption trapped, one whose first draft is 50% done and the other 25%.

Incidentally, I planned out a short story as I thought I might actually finish one of those—however, instead of working on that I’m writing this blog post. This is where I will tie in my last post, on multitasking. The vast majority of my novel-writing happened this summer in China, where I could sit down and write without distraction. Right now, as I am typing this blog post, I am in the Olin Library, and I find I am much more productive here than in my dorm room. Writing in general is hard.

As fiction author Holly Lisle warns in Mugging the Muse on full-time writing:

Writing is a business for self-starters. It’s hard. If you don’t know beforehand that you can sit down and make yourself produce pages in spite of weariness, boredom, lack of inspiration, or over-abundance of distractions, don’t kid yourself that writing full-time will magically cure this.

I don’t remember where it is—maybe it was Trimble’s Writing with Style—some author observed that when you are writing, your body will automatically try to distract you from it, even if it just means your body is trying to make you feel like taking a break. Either way, writing is an art full of gumption traps.

In light of this, I’m making a plan to improve my ability to resist gumption traps in writing. This is first going to involve regular blogging again, perhaps not on a daily schedule, but at least three times a week. That’ll be my goal for the remainder of the year.

*Just a side note: Outside of writing, I’m not as bad in falling into gumption traps. There was one project that I had virtually stopped, and then picked back up one-and-a-half years later, and continued it for the next two years. Since both of my large fiction-writing attempts were recent (in the last five months), I hope I’ll be able to find the motivation to continue those in the future.

Memory and Multitasking

How many times in this era of “information overload” have you forgotten something, whether it be somebody’s name, a random fact, or a website’s name, not five seconds after you learned it? Just now, for example, I thought of something that I needed to look up, so I pressed control-T (new tab) on my browser, typed in “www” for the beginning of the url, and suddenly realized that I had no idea what it was that I was searching for.

It turns out I was attempting to go to the Weather Channel and look up tomorrow’s weather.

In my Intro to Sociology class today (err, make that yesterday, as this post took longer than expected to type up), we watched parts of a documentary about the effects of recent information technology upon mainly the American youth. By recent, it means in the last decade. And the video’s main point is, recent technology makes us multitask, and the physical effects of this are not are not wholly good.

I would consider myself a multitasker, but not as much as many others are. This is because I use my cell phone very seldom—my main forms of communication are in-person and the Internet. And the Internet, excluding mobile use in this case, is not yet an all-pervasive technology (i.e., everywhere), which means I can usually engage with another person in conversation undistracted. In contrast, the documentary showed an MIT (or was it Stanford) student at a desk with a laptop and an iPhone out—he was in the middle of an email, IMing someone else, and texting still others. And probably supposed to be doing homework.

And yet, I have somehow forgotten the name of this video.*

Even this blog post so far has taken me at least a couple hours. I’ve been looking at other sites and doing other things.

More concretely, one report has shown that multitaskers are bad at multitasking. Another says that we can’t multitask at all: we instead switch single tasks quickly. This second result is of pivotal importance. It means we cannot consciously focus on more than one thing at a time, and that if we try to do a second action, we lose attention of the first.

It would seem, then, that this is one way to “multitask” well: We perform some task to repetition so that it becomes ingrained into our subconsciousness rather than our consciousness. For example, our hearts are beating and lungs are beating at the same time. But we don’t have to think about either of those.

If you teach a baby to walk and to speak at the same time, it might not work. But walking eventually becomes part of our subconsciousness (muscle memory), and at some point, even talking becomes natural. When such a process is put into subconscious memory like this, it no longer needs our attention. Therefore, some things are possible to multitask: walking and talking, for example. While solving a Rubik’s Cube.

Let us try to apply this practically. In a situation where we have to multitask, we would want to train one of the tasks, or all of them, as much in muscle memory as possible. Unfortunately, some actions are currently impossible to train so. You can’t put a conversation, for example, into muscle memory, because conversations are different each time, and you have to consciously think about what you’re going to say.

The ability to train other actions may be limited by evolution. While we are evolved to walk—and at the same time watch out for predators—we are not quite evolved to make sense of the environment moving 20, 40, 60 miles per hour around us, as in driving. An experienced driver would thus need less attention to the driving process than does the new one.

What if we compare the rates of accidents for texting while driving by experience (age)? That is, do older people who text have a lower accident rate than younger people who text, simply because they are better at driving and can multi-task at it more easily? I have not yet found data sheets collected this way; if anyone could point one out, that would be appreciated.

Now let’s see if you remember what this article said about memory.

*Edit: The video I was referring to is Digital Nation.

Starcraft II: Wings of Liberty Walkthrough, Difficulty Guide

StarCraft II


Mission Name

Difficulty: A/B/C/D (out of a 1-5 scale, 1 being the easiest, 5 being the hardest; A on casual, B on normal, C on hard, and D on brutal)

Achievements: E/F (also a difficulty on a 1-5 scale; E for the achievement on normal mode, F for the achievement on hard mode)

Then tips, advice, and other info about the mission.

I. Mar Sara Missions

1. Liberation Day

Difficulty: 1/1/1/2

Achivements: 1/1

As the first mission, this is very easy: you control Raynor and a few marines, and you need to destroy the Dominion headquarters, killing Dominion units and Mengsk statues along the way. Brutal is somewhat difficult because you lose marines pretty fast if you don’t control them well. Even so, you can win the mission with just attack-moving: both times I tried it, I lost all my marines by the time I got to the end, winning with only Raynor + the rebels with molotov cocktails.

2. The Outlaws

Difficulty: 1/1/1/3

Achievements: 1/3

This mission is also not very hard—on the first three modes I just build a tech lab, a medic, and then push out right away. But on brutal, this will not work. I would suggest massing a large clump of marines and medics before moving out. Why? The Dominion units are upgraded +1/+1, and you have no engineering bay and so cannot upgrade your units. Thus, you must outnumber the enemy significantly to do a sustained push. Once you reach about 30-40 marines you’ll be unstoppable.

The normal achievement, picking up mineral/gas pellets, is easy, but beating the mission on hard in under 10 minutes is a bit more intense. After getting a medic and a second barracks, you should push right away, rescuing the other rebel base as quickly as possible. This adds another barracks so you’ll have 3 in total. Just keep pumping marines from all three of them, and push. Against the Hellions, just take them out quickly so they don’t get many attacks on you, and the rest is easy.

3. Zero Hour

Difficulty: 1/1/2/4

Achievements: 1/5

This one is a throwback to mission 3 in the original Starcraft game, where you had to defend a base with bunkers for 25 minutes. In SC2, it’s again pretty easy on the casual and normal settings. Hard mode is a bit more tough because the Zerg start out with more upgrades, and thus can sustain longer attacks and do more damage to your bunkers. In brutal, the Zerg get more numerous and more dangerous attack waves, with more upgrades, and they can also take over your outside bunker zone easily. I had to keep a mass of marines and medics in my inner base for this one.

The normal achievement is a piece of cake–just make sure you have SCV’s repairing those bunkers, and a sufficient number of marines to defend. But the hard achievement is very difficult: you must kill 4 Zerg hatcheries, which requires offense. It took me at least 5 tries to do it, and it was by far the hardest achievement to get in all of the first few missions. The key is just building a large clump of marines and medics—about 50 marines and 10 medics is what I had, and just attack with the entire force sticking together.

II. Colonist Missions

1. The Evacuation

Difficulty: 1/1/1/2

Achivements: 1/2

A very easy mission, especially if you have reapers. Speaking of which, my preferred order after the Zero Hour mission is Smash and Grab, which unlocks the marauder, and then The Devil’s Playground, which unlocks the reaper. I also get the U-238 weapon upgrade on the reaper immediately after unlocking it—this single upgrade alone makes some of the early missions much easier.

You basically just escort the colonists to the spaceport. On hard and brutal the Zerg occasionally send attacks from multiple directions immediately following each other, so that your army might be preoccupied with one Zerg wave and not the other which is attacking the convoy. Even so, it’s not too bad, and reapers again make moving from place to place very easy, as they’re very fast units. A mass of marines, marauders, and reapers should make this mission easy.

For the achievements, just defend the convoy well, and defend the bunkers well, and you’ve got them. The hard achievement for this mission seemed to be one of the easiest hard achievements to get.

2. Outbreak

Difficulty: 1/1/2/2

Achievements: 2/2

Survive zombie attacks by night, and kill as many infested structures as you can by day. Again, reapers are perfect for this mission—I found them more effective at clearing infested units and structures than hellions, which are introduced here. Case in point: reapers do 30 damage to buildings, hellions 8. Also, the hellion takes some time between acquiring a target and actually firing, whereas the reaper attacks instantly. Strategy: Fill one bunker with marines at each of the three entrances, place a couple of firebats and medics in front of each bunker (and hold position on them), and then just mass an army of reapers and hellions. (You would optimally go all reapers, but you’ll have bonus minerals which you should use on hellions anyway.) At night, you can use this reaper/hellion army to jump from bunker to bunker to defend, and at day, you just attack like heck.

The normal achievement requires winning the mission in 28 minutes (reference to zombie movies 28 Days Later and 28 Weeks Later), and this is super easy with reapers. I haven’t tried it with just hellions though. The hard achievement is to kill 15 structures at night. Well, infested structures spawn infested units upon being attacked during night, so this is really stirring up the hornet’s nest, as Tychus says. With a large number of reapers though, this should be easy since their instant attacks will kill the spawned units almost immediately.

3a. Safe Haven

Difficulty: 1/2/2/3

Achievements: 2/2

You have to kill 3 nexuses and then a very powerful Mothership. Strategy: build SCV’s and a bunker in the beginning, and fill that bunker with marines. That’s about all the static defense you’ll need. Concentrate the rest of you resources on building vikings and getting air upgrades. All 3 nexuses are killable (even on brutal) with vikings alone.

The achievements are basically time runs. Both are easy if you just mass vikings. To kill the first protoss nexus (top-right of the map), keep the vikings in air mode and snipe the warp prisms powering the photon cannons. Once that happens, land your vikings on the high ground and hit the nexus from there. For the second nexus you can also just snipe warp prisms and kill it from the high ground. But for the third nexus, it is surrounded by water and the edge of the map, and has archons/sentries/high templar. To beat this, I would again snipe the warp prisms powering the cannons, and then land a mass of vikings outside of the base, and march in. You should be able to take out archons and sentries without sustaining too much damage. For the high templar, just send one viking to attack it, and it will be chased away. Finally, for the mothership, just send all your vikings at it, and try to keep at max range. It goes down pretty easily if you have enough vikings.

3b. Haven’s Fall*

Difficulty: 1/2/3/3

Achievements: 2/2

This mission is ridiculously easy if you just mass vikings. the very first thing you should do is build a reactor on the starport. I was able to beat the mission on brutal by building only 1 bunker of marines and 2 siege tanks for defense, and about 20+ vikings for offense. To take out the new infestations, use the vikings in air mode to take out mutalisk and brood lords, and then transform them into ground mode to kill the virophage and other units around it. To kill the main Zerg bases, again use air mode to kill the mutalisks and brood lords, then switch to ground for a frontal assault. If you have a mass of vikings it will be easy to attack. For the bottom-most base I had to keep about 5 vikings in air mode to deal with newly spawned mutalisks.

The achievements are also super easy if you mass vikings. Again, kill the mutalisks and brood lords, switch to ground, and take out the ground forces and the virophage. This mission, however, is overall harder than Safe Haven because you constantly have to worry about Zerg infestations.

*On the choice missions, I will mark the path I choose with an asterisk (*).

III. Artifact Missions

1. Smash and Grab

Difficulty: 1/1/2/4

Achivements: 2/3

This mission is pretty easy on the first three difficulties as all you need to win are a mass of marauders and medics (in addition to a bunker of marines in you main base). But in brutal, the Protoss get void rays! I didn’t know about this, so the first time I tried brutal, I led with a mass of marauders and medics, and pretty much quit the mission once I saw a void ray.

The first achievement is to kill the stone zealots without losing a unit—this is easy if you have all marauders and medics in your army. If you have marines you should move them off to the side. The second is to beat hard in less than 15 minutes, and this is not too bad if you maintain a constant push into the Protoss base.

2. The Dig

Difficulty: 1/2/3/4

Achievements: 1/5

This mission scales in difficulty quite well—it gets genuinely tough on higher difficulties, where swarms of Protoss units attack. The “aerial assaults” become especially dreadful. My strategy is to maintain a force of siege tanks in siege mode (most important unit), bunkers, and goliaths. A backup army of marines, marauders, and medics is helpful.

The normal achievement is to kill 20 units with the laser, which is pretty easy. But the hard achievement, to kill 50 structures with it, was impossibly hard. The problem was that every time I tried pushing into the Protoss base, they would just pump replacement units and structures so that I was always pushed back. You really just have to be patient for this one.

3. The Moebius Factor

Difficulty: 2/2/3/4

Achievements: 2/2

A fairly fast-paced and mobile mission, and quite fun. I found that having the Hercules research helps here, because it is much more sturdy than a medivac. Basically, just follow where Dr. Narud tells you to go, and build marines and War Pig mercenaries, and the infantry weapon upgrade definitely helps. The mission is beatable by building only a relatively small number of units—if you locate the Moebius survivors, it becomes much easier.

That is the first achievement, and the second achievement, to win before Kerrigan accesses 6 data cores, is also easy: she had only accessed 3 data cores on my first attempt at the achievement. Just move around a lot, and you’ll beat the mission quickly.

4. Supernova

Difficulty: 2/3/4/5

Achievements: 1/2

This is one of the hardest missions in the game, especially on higher levels. On brutal the wall of fire advances so quickly that by the time you’ve settled into one base, you need to clear the next one so that you have a place to where you can relocate. While there is a cheesy way to win (hide banshees/vikings in the bottom-right corner, wait for the fire to kill most of the Protoss defense, and then attack the artifact vault), the conventional attack becomes progressively more difficult with each difficulty level. You’ll be fighting virtually everything in the Protoss army. This is also the only mission where I had to restart on casual mode (I underestimated the last Protoss base and my army got slaughtered in the final push).

Achievements for this mission, however, are easy. It is nearly impossible to lose a unit to the wall of fire, and killing 75 units with banshees is also easy, because there are so many ground units on the map.

5. Maw of the Void

Difficulty: 2/2/3/4

Achievements: 1/4

One of the funner missions in the game, as it basically prompts you to mass battlecruisers. It’s really not that hard of a mission, but perhaps time-consuming.

To get the first achievement you need to destroy all of the rip-field generators, and you’ll already have the vast majority destroyed anyways to win. The second achievement requires patience and some saving/loading of the game.

IV. Covert Missions

1. The Devil’s Playground

Difficulty: 1/1/2/4

Achievements: 2/3

Lava floods the map every few minutes, and to make it worse, the Zerg send in consistent attack waves. On brutal I found myself spending as much on defense as I was mining, so I just launched an all-out attack on the Zerg bases and wiped out most of them. It is a very fun mission with reapers. You’ll want to beat this mission as early in the campaign as possible to unlock reapers, which really help in the first two Colonist missions.

This is the first mission with a Brutalisk, and a fun way to kill it is to lure it to the low ground when it becomes flooded with lava. I personally prefer reaper kiting, but the lava is necessary for the achievement. The second achievement is to locate all of Tosh’s miners, and this will simply take a while to do.

2. Welcome to the Jungle

Difficulty: 2/3/4/5

Achievements: 1/3

I found this to be a fairly hard mission, just because the Protoss have strong units and lots of air units, especially void rays. (Though on my hard-mode run through the campaign, I did an order in which I had vikings available for this mission; they made it a lot easier.) Just create a large army of marines, marauders, medics, and goliaths. On brutal, try to prevent the Protoss from sealing the first two or three terrazine altars, but don’t worry about the rest—just try to collect the gas faster than the Protoss can seal them.

To not lose an SCV, just be careful with your SCVs. Stopping the Protoss from capping a single terrazine altar is a bit more tricky. My advice is to create a backup group of 4-5 hellions. If you can’t beat the defensive force guarding the probe, just send the hellions to snipe the probe and then rub back. Hellions have more hp than reapers, and so, are more suited for this task.

3a. Breakout

Difficulty: 1/1/2/2

Achievements: 1/3

A very fun mission where you control only Tosh and engage in an AoS-type gameplay. Anyone who has played Warcraft III custom maps (*cough*, DotA), should love this mission. It’s also very easy because you can focus all your attention on one unit.

The first achievement requires that Tosh not go below 100 health; this is easy because there are plenty of allied medics around. The second achievement is a time run and is more difficult, but still not too hard—just advance as quickly as possible and use your nukes when they become available.

3b. Ghost of a Chance*

Difficulty: 1/1/2/2

Achievements: 1/3

The most fun mission in the game, in my opinion. It’s a stealth mission with Nova and a handful of other units. You also don’t start out with a base, so you can focus all your effort on infiltration. It is not too hard.

Killing 15 units with dominated units is the first achievements, and this is incredibly easy: the siege tanks alone should have 15 kills. The Maelstrom Rounds upgrade on the siege tank (the one that adds +40 damage to the primary target) is very useful: it allows you to wipe out the units on the middle bridge of the third part of the mission with just a dominated tank; without Maelstrom Rounds, the tank gets killed first. The hard achievement is to kill every last unit, and if you’re a perfectionist, then you’ll be doing this already. Don’t worry about the miscellaneous missile turrets—they don’t count.

V. Rebellion

1. The Great Train Robbery

Difficulty: 1/2/3/5

Achievements: 2/2

This mission is really fun on easier modes but is truly brutal on brutal. I could not beat it the first time around, and had to play the mission that unlocked siege tanks first, so I could kill the bunkers and the two marauder strike teams from afar. Once you get siege tanks it’s a lot easier.

The normal achievement is to kill the marauder strike team—fairly easy. You get the achievement during the game, so you could save the game, kill the strike team (getting the achievement but for example lose half your army), the reload and play on without killing it. Siege tanks in siege mode crush the marauders.  The hard achievement is to not miss a train, and this is quite easy.

2. Cutthroat

Difficulty: 1/2/3/3

Achievements: 3/2

Here is again a very fun mission, where you must gather scrap minerals around the map and deny the opponent from gathering minerals. Once you gather the minerals you can just do a siege tank push or a fun ghost/spectre push.

The normal achievement is to kill 25 units with vulture spider mines, and this is actually not easy to do—I had to go to melee range of a unit, lay a mine, then run back, to get enough kills. Just laying mines at base entrances doesn’t seem to get enough kills.

3. Engine of Destruction

Difficulty: 1/1/2/5

Achievements: 3/2

Very easy on the first three modes: even on hard, you can just send 1-3 SCV’s repairing the Odin and create a small wraith fleet (6-8 of them) and win, with no ground support. On brutal, the opponents target your SCV’s! On several of the base pushes, I had to send ahead an army of siege tanks/marauders and basically suicide them so that Tychus in the Odin wouldn’t die. For the final push, just bring all your SCV’s off the mineral line and have them repair Tychus. Have some wraiths to hit the battlecruisers, and it’s do-able from there.

Find and destroy the Loki is the normal achievement. Find the what? It turns out the Loki is this battlecruiser in the back of the fourth base. It’s pretty easy to kill with cloaked wraiths: you just kill the detectors and own the Loki with impunity. The hard achievement is to not let the Odin go below 30% hp, and this is super easy if you have SCV’s repairing it.

4. Media Blitz

Difficulty: 1/2/2/3

Achievements: 1/3

Now you actually get to control the Odin, in which Tychus acts as a one-man army. This makes the mission very fun as well. The best base to destroy is the one with the factories and heavy armor—in normal and hard you can afford to destroy this base and some starports in the air base; in brutal, you only have enough time and hp to destroy this one base. It’s still an easy mission though, with the Odin being as strong as he is.

Killing an enemy factory, barracks, and starport is super easy—you just do what the achievement says. Beating the mission in less than 20 minutes, though, is somewhat tough—you’ll need to wipe out as much as you can with the Odin first.

5. Piercing the Shroud

Difficulty: 2/2/2/2

Achievements: 2/2

Being the secret mission, this is pretty easy no matter which difficulty you’re on, but not incredibly easy. The only trouble is the invincible hybrid, so even casual mode is not a true walk in the park.

Normal achievement is to kill the Brutalisk without losing a unit—easy if you use plasma rounds. Next is to locate all the weapon pickups—also quite easy if you know where they are.

VI. Prophecy Missions

1. Whispers of Doom

Difficulty: 1/2/2/3

Achievements: 2/3

The first Protoss mission, fairly easy since Zeratul is so powerful. On later levels you’ll have to think about how to do some of the puzzles, though some can certainly be cheesed. For example, on the baneling part, I just run back and eventually the banelings will give up chase and roll back, whereupon you can take free shots at them.

Surviving with 3 or more Stalkers is not too hard if you blink properly. Not getting Zeratul damaged is a bit tougher, and you’ll also need to be patient for his shield to fully regenerate after each fight.

2. A Sinister Turn

Difficulty: 2/3/4/5

Achievements: 2/5

This is one truly tough mission, where the hybrid Maar consistently attacks and resurrects every time you kill it. He casts very powerful abilities too, becoming more powerful over time. On brutal, I won by sneaking into the pillars with a few immortals, stalkers, zealots, and dark templar, damaging each preserver pillar so that it only has 50-ish hp left, then kill them all quickly. This is because when you kill a pillar, Maar teleports back and begins to slaughter your army, so by damaging all the pillars first, you can quickly kill the remaining two pillars after Maar’s teleportation.

The first achievement is to kill all of the Protoss, which isn’t too bad on normal. On hard and brutal, they have void rays and rebuild their bases, which makes it much tougher. The win in 25 minutes on hard achievement was impossible for me until I did the sneaking in strategy.

3. Echoes of the Future

Difficulty: 1/2/3/3

Achievements: 1/4

You need to get Zeratul to four Overmind tendrils, and you meet a lot of Zerg resistance and counterattacks along the way. Easiest strategy is to mass colossi and stalkers. The colossi own ground units, and the stalkers will take out mutalisks and occasional brood lords.

The first achievement is to kill 50 Zerg units with Zeratul; if Zeratul is in your army at all, it is almost impossible to not get this achievement. The second achievement is much tougher: win in 20 minutes. It’s not impossible though, but you’ll need to act quickly. Securing both warp gates early on and powering up the robotics bay helps.

4. In Utter Darkness

Difficulty: 1/3/4/4

Achievements: 3/5

An overall very tough mission. You have to kill 1500 enemy units on casual and normal, 2000 on hard, and 2500 on brutal. On casual, enemy units have only 50% hp, including hybrids, so it is pretty easy if you have lots of colossi/phoenix/carriers. Starting normal mode, the mission is a defense against a relentless attack. Massing colossi is definitely the way to go. (There is also the dark templar wall strategy which I have tested, and it is very efficient.)

This mission contains perhaps the hardest achievement in the campaign. The first is to kill 250 additional units on normal mode, and the second is to kill 750 additional units, also on normal mode. This second one is nearly impossible. After several tries, I was able to kill 2252 units before my last unit died—that’s only 2 kills above the needed amount. My strategy was massing phoenix and then carriers as soon as they were available. When you lose your base, you just hide on a corner that only air units can access, then just kill mutalisk/corruptors for score.

VII. Final Missions

1. Gates of Hell

Difficulty: 1/2/3/3

Achievements: 1/3

It’s one of the last missions but a fairly easy one. You rescue Dominion drop pods and then make a push against the Zerg to kill three nydus worms.

The first achievement is to kill all the spore cannons, which is very easy because the spore cannons don’t do much. The second is to rescue all 10 of the drop pods, which is somewhat hard to do if you’re using your ground army to rescue them. To be cheap, you can call down a mule to rescue a drop pod if you really need to.

2a. Belly of the Beast

Difficulty: 2/3/3/4

Achievements: 2/5

This is a mission where you get to kill a TON of Zerg. And you use the four biggies from the Hyperion crew: Raynor, Tychus, Swann, and Stetman. Each person has a different unique ability, so it almost feels like an RPG dungeon crawl. Just use your abilities often, and you’ll win.

Normal achievement: don’t let any hero fall incapacitated. This is not hard at all—just watch the queen at the very end. The hard achievement is to kill 50 units in one shot of Raynor’s Penetrator Round. I must have spent 20 tries on this before I got it to work. I used the beginning of the second part of the mission, where you meet an infestor and a ton of infested units in a narrow alleyway.

2b. Shatter the Sky*

Difficulty: 2/3/3/4

Achievements: 1/4

A mission that requires some thinking. I built many different types of units for this mission, including vikings, battlecruisers, marines, medics, goliaths, tanks, and ghosts. Some good upgrades to have: both goliath upgrades make the goliath a very good unit for assaulting the lower-right platform, as it can attack both ground and air at once; ghost’s infinite cloaking as well as range upgrade, as they can easily kill mutalisks, zerglings, and hydralisks; the viking’s aoe attack and range upgrade, so it can destroy clusters of air units from far away; and the science vessel, so you can repair vikings and goliaths. Ghosts were the fun part though, I built 8 of them and had them all perma-cloaked, and just went around attacking and sniping stuff. At the end all 8 of them had at least 20 kills each.

Not losing a unit to platform explosion is super easy: after you destroy a cooling tower, just retreat. Beating the mission in less than 25 minutes is tough though, and you don’t have enough time to mass a sufficient number battlecruisers (plus, they’re slow units). This is why I went more for vikings, ghosts, and goliaths.

3. All In

Difficulty: 3/4/5/6(!)

Achievements: 3/3

Oh my god. This mission is way harder than any mission before it—the difficulty ramp is huge. You have to defend the artifact for some amount of time, and in the meanwhile you are attacked by massive waves of Zerg. In all four of my runs I killed the air in Shatter the Sky, so I was up against Nydus Worms in this mission. Banshees, siege tanks, and marauders will be the most important units. Use Banshees to take out Nydus worms. When the end approaches, one way to gain some extra time is to send all your mining SCV’s up to the artifact’s high ground and just build a ton of stuff. Each building will both delay the Zerg final push and stop Nydus worms from popping up at that location. That is how I beat brutal after many frustrating, failed attempts.

The normal achievement requires religious use of Energy Nova, while the hard achievement requires very little usage of it. Both are fairly tough, as I could not manage to get the sufficient number of kills on normal at first, and I could not beat hard without using it more than once. But after a while, it becomes easier to do and both achievements just take some practice.


Protoss Tech:

  • Ultra Capacitors (instead of Vanadium Plating): I prefer bonus DPS over bonus survivability. When your units are in a large clump anyways you have strength in numbers.
  • Orbital Depots (instead of Micro-Filtering): This is just really convenient.
  • Automated Refinery (instead of Command Center Reactor): A passive bonus; and queuing two SCV’s at a time still makes the SCV’s the same cost.
  • Science Vessel (instead of Raven): Heals mechanical units.
  • Tech Reactor (instead of Orbital Strike): Building two high-tech units at a time is much more amazing than sending infantry anywhere.

Zerg Tech:

  • Fortified Bunkers (instead of Shrike Turrets): I went for the other bunker upgrades as well, and this research fits the upgrades better.
  • Planetary Fortress (instead of Perdition Turrets): I ended up not really using either one.
  • Hercules (instead of Predator): Only useful in The Moebius Factor and Supernova, but better than Predator, which was not useful in any mission.
  • Regenerative Biosteel (instead of Cellular Reactor): Auto-healing mechanical units is an amazing ability. Cellular Reactor would only benefit units like Ghosts/Spectres, Wraiths, Banshees, and Battlecruisers; and their rare use is not nearly as much as the usual massing of mechanical units.
  • Psi Disruptor (instead of Hive Mind Emulator): Passive effect requires less focus than an active one.


Base Upgrades:

  • Bunker: Both upgrades. Both bonus range and bonus slots are useful for defense.
  • Missile Turrets: None. (Turrets are already strong enough as they are.)
  • SCV’s: Double repair upgrade. I thought the multiple construction upgrade was useless, especially if you have Orbital Depots.
  • Buildings: Both. Fire suppression is very good, as is Orbital Command.

Infantry Upgrades:

  • Marine: Both upgrades (stim and shield). Come on, name one mission where you don’t use marines.
  • Medic: Stabilizer medpacks, because is amazing. The ability to train without tech lab is useless later on with tech reactors.
  • Firebat: None.
  • Marauder: Both. Concussive Shells (slowing) and Kinetic Foam (+hp) are both very helpful.
  • Reaper: U-238 Rounds (bonus range, and damage to light armor). It makes some of the earlier missions much easier. The G-4 Clusterbomb takes too much effort to use.

Vehicle Upgrades:

  • Hellion: None.
  • Vulture: None.
  • Goliath: Both. Two attacks at once? And super range? Awesome.
  • Diamondback: None.
  • Siege Tank: Both. Bonus target damage and reduced friendly splash damage are really good.

Starship Upgrades:

  • Medivac: None.
  • Wraith: None.
  • Viking: Both. (Makes Vikings excellent as anti-air throughout the campaign.)
  • Banshee: Both. (Useful for last mission.)
  • Battlecruiser: Both. (This made Shatter the Sky easy.)

Dominion Upgrades:

  • Ghost: Both. (They become really fun to use with infinite cloaking, and bonus range makes them really good at sniping, both with normal attack and with ability.)
  • Thor: Immortality Protocol. (Rebuilding a dead Thor is awesome.)
  • [Spectre: Both. (Lash is awesome, as is infinite cloaking.)]

This is not a Starcraft 2 campaign walkthrough in the normal sense; I just want to give an overview of the mission difficulties and any tricks I found to completing them. And I have certainly played each mission on all four difficulty levels at some point or another.

By the writing of this post I have amassed 1570 of the 1590 achievement points for the campaign (missing only the Lost Viking gold and the normal mode speedrun) and a total of 3000 achievement points.

Any questions, criticism, advice? Just make a reply!