Rokrin writes a guide: Engineering.

This is copied directly from Rokrin's post on hkcentral.net.

Let me preface this by saying I am, by no means, the best engineer on the server. However, it is my most played class, and consequently I've picked up many of the niches. As this is not the first engineering guide posted, I will not cover the basics of engineering; the other guides can teach to to swing the wrench and be a perfectly good engineer. What I will do is cover the less known, or practiced, niches f the class; go elsewhere to learn the class, come here to fine-tune it. I'll probably add to this guide, I'm kind of just writing up what I was thinking about today.

1) Patience, young Padawan
Learning to wait.

Engineer is probably the most boring class in TF2, especially when setting up. Everything you can do takes time, and nasty people don't leave you alone while you build. Spies will sap you, Demomen will sticky you, Soldiers will spam you, Heavies will brute force you, Pyros will burn you, and medics will build ubers ready to destroy you later. 6 of the 9 classes have a way to piss you off, and they will use them, and you will become pissed. Grab a snack; the best things take time. Perseverance will make it that much sweeter when your level 3 gun earns you dominates, and your teleporter wins your team the game. While you're building, eat your snack, and observe. Watch the kill updates so you can see what's coming to get you. The advantage you have as an engineer is the biggest annoyance of the class; you have time to plan. How do you increase your patience? The best thing you can do is continue eating and re-build. The more you do it, the less pissed off you'll be.

2) Higher Math
Know your pantry

You are an engineer. You are the problem-solver. You are the jack of all trades handyman. And just like any good handyman, you should know where your tools are. Handymen do not have some fancy little thing to tell them where the hammer is versus the screwdriver; they know where they're suppoed to be. So ditch your scripts. You will be far better at the class if you learn it the right way. Here's a rhyme! Everyone loves rhymes.

4 plus 1 makes you a gun
4 plus 2, dispenser for you
4 plus 3, teleporter entry
4 plus 4 is your exit door

Congrats, in 4 lines you just learned the engineer build menu, in order of what keys to push. Wipe your config.

3) Construct Additional Pylons
Managing your resources

Buildings cost metal (rundown of costs below, if you're reading this you probably don't need it anyway). Metal is a resource, and while it's semi-renewable, incorrect management means your buildings are going to get rolled. Hard. Like, grandma's-makin'-cookies-out-of-rocks hard. There are two things that affect how you manage the metal, and consequently what you build and when you build it: availability, and current situation.

Building costs:
Sentry Gun: 130 Metal
Dispenser: 100 Metal
Teleporter Entrance/Exit: 125 metal each.

Availability: Several things give you metal in addition to your dispenser.
- Ammo Kits: Small/Medium/Large feed you 41/100/200 respectively
- Weapons: Dropped weapons are 100 metal, every time. No exceptions.
- Gibbed buildings: These seem to vary but it's rare you get very much metal, don't rely on these unless you absolutely have to.

Availability is crucial to your build order. If you're building in an area where there are no ammo kits around, the first thing you get up needs to be a dispenser. The dispenser will feed you 25 immediately after it finishes, and will give you 40 shortly thereafter. That's enough for you to throw a gun up and start banging on it while your dispenser continues to stock with metal. Inversely, if you are building in an area where you're right beside a large ammo kit, you have more options. This is where we get into metal management, which is how we explain WHY you should be throwing a gun up before a dispenser. It's actually just general math.

Code:
You are sitting, ready to build next to a large ammo container. You have two options sitting at your disposal (since teleporter exit should almost never be the first thing to build, with the exception of defense on certain maps). One is your Sentry Gun (s), one is your Dispenser (d). Which do you build first?
Most engineers would be quick to say Dispenser, as it's important to have a supply of Metal. But you have an instant 200 metal crate sitting near you.
d= 100 (metal)
sg= 130 (metal)

You start each life, as an engineer, with 200 metal. So building either building means:
d= 200 - 100 = 100 (metal)
sg= 200 - 130 = 70 (metal)
The dispenser leaves you with more metal, so it's the better option here, right? Wrong. If you have a large ammo crate ready to provide you with 200 metal, and you pick it up after building a dispenser with 100 metal left, you're only getting 100 metal from that large crate. And it doesn't instantly respawn. If you build a sentry, you're picking up 130 of that metal, which then leaves you with enough metal to quickly throw a dispenser up and still have 100 metal leftover to use before that crate respawns. Had you done the reverse order, you would only have 70.

That 30 metal often means the difference between a scout killing your mid-build gun, and it getting up and shooting the scout to death.

Obviously pure math cannot lead an engineer. If your team is on the frontlines, trying to survive, what good is a gun going to do them? Wouldn't healing be better? That comes to our second point.

Current Situation:

As an engineer, not only are you the jack of all trades, but you are the fallback. People will rely on you to save the intel when the team is dead, to protect the point, and to be the scouting offense when everyone else is dealing with the current threat. You must be able to step back and analyze what your team needs, versus what you want to do. So in addition to being able to manage your metal, you have to be able to realize what your team needs. Sure the Sentry Gun might be the economic choice, but if your team needs to continue assault on a forward point, your teleporter is going to save them. On Badwater basin, the first thing (I believe) to be done is to have at least one defensive engineer build a teleporter. Get it up FIRST THING during setup, because the people that die to that initial onslaught of kritzkriegs and ubers (normally offensive classes dying, like soldiers/heavies/demos and medics) need to be able to get back fast. Your teammates will, in turn, be able to hold off the enemy while you finish your stuff. This is a situation that applies rarely anywhere else on the map, but because it's what your team needs, you need to be able to do it, and you need to do it efficiently.

So for building your weapons and tactics, two things must enter your mind above all else:
- What does my team need
- What is best for me to build based on the available resources?

One final note on this: people constantly harp that you need to be hitting buildables right when you put them up in order to speed up the process. In certain situations, this is true; the offensive engineer is an expert at throwing level 1's up to do quick guerilla damage, and thus needs to hit that gun to get it up ASAP. However, when any engineer (red or blue, offense or defense) is setting up anything more than one thing, you should utilize the time it takes for one thing to build to get something else up. Chefs time their meals so that the chicken breast is ready the same time as the noodles; you should do the same. If you throw a dispenser up, and rather than hitting it continuously, throw a gun up, turn around and hit the dispenser, your dispenser and gun will finish roughly at the same time (gun might be slightly after. Suddenly you've got a level 1 gun that you can upgrade and keep intact, versus the engineer that suddenly has a dispenser and no defense against the scout who just dropped down.

4) Patent Pending
Innovation is the key to success

Look. Everyone knows there's going to be a sentry guarding the courtyard of 2fort. Everyone knows you're going to build on the roof of the second cap on dustbowl. Everyone knows there's going to be a sentry in the intel room on Well. Everyone knows there's a teleporter in their damn sewers. Because people know these are there, and have seen them before, people know how to deal with them. So as an engineer, not only do you have to solve problems, but you need to do it in ways people don't expect. There's not much I can teach in the way of this; it's simply through trial and error that you discover fantastic spots for things. And there will be a lot of error. The best advice I can give is to spend time with your maps. Waiting to join a game on #4? You know the maps that are on 4, there's only 5 in rotation (I think). So load up Dustbowl, join as an engineer, and run around with your build menu out. Find out where you can build, look at the area it covers, and look at the area it does NOT cover. Know the points strengths, know the weaknesses. If the benefit outweighs the risk, try it when you join the game. You'll learn how good it is, make a judgement, and maybe use the place again.

The only danger with this is that good innovation, like anything, will be replicated. If someone sees you're fantastic new teleporter spot, you can bet your life they're going to try it when they play engineer. Soon it's well-known, people find ways to deal with it, and then you're back to the drawing board. An example of this is the teleporters on granary; before, nobody put them on the upper tunnel leading to the drop hole. It was simply unheard of. But it is a phenomenal teleporter spot, and it has been replicated to the point where spies now actually go out of their way to check that area. So back you go, and hopefully you find a new place. Occasionally, you strike upon a place that wins games. I've only found one of those to date, and I will not be sharing it, because frankly it is the best teleporter spot I have ever seen and I'll be damned if anyone else starts using it on a consistent basis. Selfish? Maybe, but that way, it'll work longer.

5) And now for something entirely different
Screw strategy, let's surprise them

When in doubt, try shit out. This is going to be the shortest section; in fact, I don't quite know why it's here. But when you feel like all hope is lost, do something stupid. Do something drunk. You'd be amazed the things you can come up with when you're desperate. The example I have here? Enemies can't move through your buildings. This presents two very nice opportunities; blocking doors, and creating detours. Being chased by someone? Spin around, drop down a dispenser or a sentry gun, and start shooting. They can't move through it, which means they either have to take time to jump over it or go around it. Running and know you're going to die? Turn and try and throw down a gun somewhere. Anywhere. Then keep running. One of two things will happen: either a) they'll stop to kill the building gun and you can escape, or b) they'll continue chasing you, gun goes up, and you'll probably get at least one kill out of it before the thing goes down.

Experiment. Be spontaneous. Rape a cactus.

Don't be a builder.

Be an Engineer.

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License