Regardless, that is like what, 10 seconds of level 1 protection until it is up to level 2? Again, unless you are being actively attacked I say the dispenser is what I save.
Let's re-assess this in real terms, because this debate is at a standstill. I see the dispenser's side of things, certainly. You get set up faster again, assuming you aren't under any sort of attack.
My side of things works under the assumption that you WILL be under attack, and frequently. Therefore, the sentry is going to make more sense to save.
In order to come to consensus, we need to look specifically at this scenario, because while 9 times out of 10, I do believe that the sentry is the correct answer, each scenario is different and requires different assessment.
You're on the roof in badwater. There is a 100 metal spawn beside you, and you're overlooking the point that the cart has to pass through. That last point is the most important to our debate.
Usually, in my experience, once a spy saps your buildings, it means that the other team knows about you. Typically, it means that the other team (or at least, those with voice) have come to the conclusion that your set-up needs to go away. This usually leads to a spy on the other team going for a sap. This is a number of assumptions, but I hope we can agree that
a spy sapping your buildings typically means the other team is aware of both your location, and the fact that your equipment is down.
The next set of assumptions we have to make is around the location of the cart. we can likely assume that the cart is past the first point, and around the corner of the second point. There is little reason for a spy to run 2 points ahead and destroy any equipment on it, especially when he could be sapping equipment on the point in immediate contention, or just backstabbing the next point's defenders. Why worry about the end when you haven't gotten past the beginning yet? Therefore, for this to happen,
there's a good chance that the cart is past the first point, but not past the second yet.
So, let's put the two together. The other team thinks the point is clean and the cart is likely right around the corner from it. The obvious next step is to get a large number of people escorting the cart to the second point. Before this step, however, likely, a scout or an eager pyro might step up and try to solo-cap it. So, based on the previous 2 assumptions, we can assume that
the point is about to be under heavy contention.
Now, back to the engineering scenario, we have to choose something to save. Originally, I argued sentry first, then teleporter, and dispenser last in terms of priority, and I maintain that. according to these assumptions, the point WILL be under attack RIGHT NOW. This means saving a dispenser and having a level 1 may just lose you a point. Your gun is going to come under fire no matter what happens, and it's going to happen almost immediately after you're sapped, because the spy was likely coordinating with his team.
In this case, I would absolutely save the sentry first. a dispenser and teleporter will leave 120 metal behind between the 2 of them, which is enough to immediately get a new dispenser. You then run to the metal spawn to get the 100 metal back, and whack the dispenser to get it up in 15-20 seconds. If you need to interrupt this process with shotgunning people or wrenching your gun, that's just fine, because your dispenser is not nearly as critical to your success in defending the point
IN THE IMMEDIATE FUTURE.
After the dispenser is up, you presumably have 120 metal still, so you hit it up a bit, and get the metal spawn (which has respawned by now) and upgrade it to level 2. At this point, you start to slowly rebuild.
The important thing in this case is that your gun is maintained, because during this time when you're setting up your dispenser, there's a good chance the other team is trying to push the point. Your gun has 216 health, and deals great knockback and flat damage. There's a good chance that since the spy informed the other team of your demise, you will catch a good number of people off-guard. This includes those scouts and pyros I talked about before that want to be the hero. If you only had a level 1 up, yes, perhaps you could tank it so it wouldn't die, but people would take a bullet or two and immediately know what was happening. This carries a few issues.
First, no kills. Even a sandman scout could probably get away in time. Second, that scout would inform the team that you're rebuilding. This carries the assumption that you aren't well established, which would be accurate. you would then likely be dealt with by stronger classes to a magnitude that not even tanking could fix.
Why is having a level 3 different? Well, the scout wouldn't live, first of all, which means a minimum of about 10 seconds before he can communicate his cause of death to the team. This potentially leads to more 'walk in' deaths before the team knows again. Secondly, he will communicate not that you're setting up again, but that you're still there. This connotes that you are still well established and that more resources, like an uber, need to be diverted to fix you. This will delay the enemy on a psychological as well as a physical level, and in a map style where time is very key,
this will delay the opposing team longer than saving your dispenser.
I say delay because neither set up is permanent; in either case, you'll be taken out again quickly. However, in one case, a soldier will peek out and kill you with no damage to the team, where in the other case, you'll probably get a few unexpected kills and potentially waste an uber. Ultimately, keeping the sentry maintains the pressure and keeps the area protected with no gaps, whereas the dispenser will cause a leak in defense that not only shows weakness on a psychological level, but will potentially cause you to lose the point without diverting any significant amount of the other team's resources.
