I was nervous before the first playtest. My worst-case scenario could be summed up in two words: “This sucks.” I’ve played on countless community maps where that was the sentiment during the game as broken spawns, clunky sight lines, epic stalemates and generally broken design made for a hellish – if often hilarious – experience. If that was the sentiment about Crossroads, it would take a major overhaul to remove the suckage.

Going into the test I had a list of specific issues to watch for. My top concern was whether new players would be able to grasp the layout and concept of the map quickly. I try to build maps where the structure of them becomes immediately apparent, rather than bury the design in layers of complexity. The best BTB maps – Valhalla, Avalanche, Standoff, Headlong, Relic, etc. – all had a central concept that was easy to grasp from just one look at the map. Right alongside that was whether the map was detailed such that the layout worked: were players plunging over the sides in vehicles, were they getting caught out in the open too often, were the bases too easy or impossible to defend? And finally, did the respawn scheme work, in that it booted players away from combat zones and into safety, but not so far away that they were too removed from combat to counter attack. Other issues such as vehicle and weapon timers, balance and placements were easy enough to adjust.

The map has been play tested three times now, with a handful of tweaks and adjustments between each. In the interest of brevity, I’ll just treat them all as one big playtest and work through the issues that each identified, and the updates I’ve made in response – so far.

Feedback fell into a wide range, but can be grouped into three categories. The first is specific observations, where a concrete issue is identified as being a problem (usually while they were having it). Next up is general feedback, summing up impressions at a high-level. And finally, there’s information that goes unsaid but which I gleaned from observing player behavior, both during the game and by reviewing the saved films.

What I found was, somewhat paradoxically, the less specific the feedback, the more it helped me reshape the map, because the general and observed issues spoke to larger problems that individual players either didn’t pick up on or didn’t articulate.

The good news is that I never heard those two dreaded words, and there was a lot that seemed to work well. The spawns did their job, as players killed at the base kicked out to the weak spawn zones and into safety; I could find no instances of players spawning into lines of fire or death traps. The sniper rifle didn’t dominate, and the overall weapon and vehicle mix seemed pretty good; no team was able to pile up vehicles or power weapons and dominate the other team.

The first reaction I heard in the first test game – CTF – was a mix of praise and general commentary. Namely, about the size of the map. Actual quote: “Dude, this map is huge.” Yes, yes it is. The upper man cannons were a hit, and it was less than a minute in before people were abusing them delightfully.

The consensus appeared to be that the map was fun to drive around on, which was a huge relief, but a number of specific quibbles came into focus quickly. There were several complaints that the sniper rifle was too far away from the base, and that players felt too exposed when they tracked it down. A lot of people noticed that the “welding joints” I’d put down to close the gaps in the road due to the tilted Coliseum pieces were too bumpy, for both infantry and vehicles. And then there was this:

There is a physics bug with the Revenant relating to how it interacts with Forged surfaces. I don’t know the how’s or why’s, but it’s enough to break their usage entirely. One or two launches took place on every game, and once is too many: a small chance the vehicle will murder you for driving it still too big. That last instance was particularly devastating, as the Revenant’s launch cost that team a flag cap. It was a deal breaker.

The Revenant is no more, sadly, replaced by the far less interesting Ghost. I liked the Revenant because it supplied an additional passenger vehicle, and honestly, because of the raw chaos it creates wherever it goes. But the physics bug is out of my control. There have been no launching incidents with the Ghost – so far.

To address the feedback on the sniper rifle, I could either more it closer, make it more visible from the base, or do both. I settled on doing both, and have moved the sniper rifle up to just outside the base, leaning up against the archway to the t-ramp. That also solved an unspoken problem I realized: with the sniper rifle so far down on the map, no one could tell when it had respawned without making a significant time investment to check.

The fixes for the joints in the road were simple: I pulled the Coliseum pieces closer together, and hid the edges they formed under other objects. Specifically, behind the rocks and the draw bridge. So while the road around the rocks looked even, underneath it wasn’t.

The refinement made the map smoother and got me some much needed objects and budget back. I had a mere $30 in reserve, which didn’t leave enough for any major upgrades without other cuts. Between the surface cuts and replacing the Revenant with the Ghost, I got a couple hundred dollars in budget back.

One comment I heard during the second test game stuck with me when I turned to the film. Someone, while trekking out in the open, exclaimed, “Where am I getting shot from?!” shortly before his demise. I knew I’d left the center of the map open, partly by design and partly because I simply ran out of budget for guard rails and cover. But if it was to the point where people were getting nailed from all directions, then it was too exposed. I didn’t have enough to go on from yet, so I went to the films to find out what to do about it – which is also where I noticed a number of other problems.

Base Design

I began my review of the first game by focusing on behavior around the base. The CTF game ended with a 2-1 score, which I was pleased with because it meant there was a decent balance between a defensible and assailable base. But in watching the film, attackers were able to repel a large number of assaults from a distance; attackers only got close in vehicles or from over the top man cannons. Also around the base, the grav lift chute spent a good long while with a Mongoose (or two!) bobbing up and down it; players were charging the base and when they jumped out of the Mongoose, it dropped into the grav lift well. That effectively blocked that route of attack for a while.

On several occasions, players in vehicles bumped into the ramps between the side man cannon and the ramps up to the base, which is a fairly narrow gap on a corner.

Someone mentioned to me that another player had observed how Bungie “would never get away with” the 45 degree ramps attached to the base. And he was right; I’d used them to save space, but they were too steep for something like the base. The reason was simple: a 45 degree grade makes it too hard to see what’s at the top of the ramp. The standard 30 degree ramps do not have the same issue.

With all that in mind, I retooled the base in a number of ways. To make the base a touch easier to assault, I removed the machine gun turret (budget savings: $150). To clear up the corner, I swapped out the two steep ramps in front for one normal incline, which cleaned up the corner (savings: $20). And I used some of the extra budget I’d acquired to block off the grav lift chute on all four sides, so Mongeese wouldn’t tumble into it with ease. (Cost: $200.)

And then there was the problem of people plunging to their deaths. I was expecting to see a few incidents, as there were some edges I didn’t cover as well as I wanted, and I knew people would drive off as they learned the map. I didn’t expect the post-game stats to show 18 falling deaths, which was over 9% of total player deaths. No wonder people were complaining the vehicles kept going missing. (I upped the Mongoose respawn timer to 15 seconds, to be safe.) The kicker: one of those was me: I drove a Warthog, gunner and all, right over the edge. When the guy who built the map is driving others to their demise, there’s a problem.

But I didn’t, as yet, have enough budget to deal with it. The problem area was the center of the map and below the man cannon towers. Only two of four corners were protected, while the ramp to the lower level had, literally, nothing on it at all. Those were where most deaths took place.

I knew that if I was able to add the guard rails, I’d probably be able to address the issue of how open the center of the map was in one fell swoop. But I needed to dig up some budget to do it.

And as it happens, I had four of the most expensive objects available out in the center of the map: four covered bunkers at $150 each. They were intended to serve as safe havens for players moving across the center, and also to provide covered firing positions. But in all the games I watched, they went almost completely unused. The reason, I think, is simple enough: the rest of the center is so open, people didn’t want to hole up in the center. They wanted to get the heck out of it. To the extent that they were useful, they provided some cover for from the bases as players approached them. But anything can do that – it didn’t have to be an expensive little hut.

So out they came, replaced with four tall columns (net budget savings: $560). And with the new pile of cash, I was able to block off the remaining uncovered sides on the map. That still didn’t address all the issues in the area, but it got close.

To provide at touch of cover to players on the bridges in the center of the map – where anyone on them could be fired from, well, just about anywhere – I swapped out the flat bridge for a covered one (a steep ramp, flipped).

Capping Flags

There were three primary ways to snag a flag on Crossroads: make a quick get away in a vehicle, take the man cannon next to the base, and pop up the grav lift jack-in-the-box style and drop back down to a waiting vehicle. But I watching the films, two big problems emerged.

The first was that no one was using the side man cannon to put some quick distance between them and the base. It was only a five second hike or so, people weren’t using it, opting to take the on foot route along the side.

But that route was far too long and open to make it far.

I watched this happen twice during the first game, and decided to do a little in-game testing of my own. I made a run at the flag using the side man cannon to see if it was a viable option, and the cap below was the result.

It played out exactly as I had hoped: a I got some quick distance from the base, and then some tug of war as the flag was inched away, and then a vehicular rescue. So why was no one else trying it?

At first I dismissed it in the first game to players not knowing the map. But the same thing happened on every game, even when people knew the map. I think the issue had to do with the placement of the man cannon: when people snag a flag, they are expecting to be rewarded with the chance to move it some distance right away. And when there wasn’t anything available, they just took off in the best looking direction. On every good CTF map, snagging the flag and moving it even a few feet results in a big opportunity for the offense. Tossing the flag to the ground on Headlong or Relic, leaping off the man cannon (or teleporter) on Avalanche and Valhalla, getting the flag out the door on Standoff. But on Crossroads, you go the flag and then had to hike in the open for the escape.

A similar problem was on the lower level: if a vehicle wasn’t waiting, the flag wasn’t going anywhere because the area was too open. Infantry needed some cover in order to make it a more flexible flag cap route.

Back down to $30 again, I decided to check to see if anything extra was on the map; $30 struck me as an odd amount of budget to be left over. So I punched into Excel the number of objects I had left in each category, how much they cost and what the cap was on the category. The result was an inventory list showing how much I’d spent on what. And it said I had two Revenants and three Warthogs on the map, somewhere. I’m guessing they were placed by people who had joined my Forge games at some point and put them down after launching the originals off the map; they were probably on the island somewhere.

Deleting them (removing all objects of that type) got me another $350 in budget, and with that I added some tunnels to the center of the map for infantry to use. It sealed off one side, thereby breaking a sight line in the middle, and gives players a way to move across the middle while sheltered. In the last play test, I successfully snagged a flag using them, entirely on foot.

That left the man cannon next to the base. The (currently untested) solution was simple: I moved it to onto the base itself, so players snagging a flag can launch away similar to the launches from the bases on on Valhalla.

Sprinkled among the major changes are a couple dozen smaller, but still important, ones: lowering the central man cannon towers so they weren’t such a long hike up, adjusting ramps to make more floor space, ironing out all the bumps in the road, shifts in border placements to better protect vehicles, and so on. Lots of little nips and tucks to smooth the map out.

Progress Report

The play testing has been a godsend, because there was simply no way to find all of the issues that emerged without subjecting the map – and players – to some actual games on it. Some of the issues jumped out right away, while others took a couple weeks and several viewings of the saved films to identify. It’s been a learning experience in understanding player behavior, and how small decisions can have large impacts on the flow of the map.

On that first CTF game, there were 18 deaths due to players falling off the map. In the most recent, there were just six. Vehicles and players are moving more smoothly and swiftly over the map, and areas that had been death zones have become viable play spaces. But there’s still a ways to go: I’m mid-way through a base rebuild, removing the steep ramps in back, working in the man cannon, and streamlining the foundation while making it more compartmentalized and more compact. The layout of the map overall has seen a few nips and tucks, and I’m working in a few new routes for infantry, so that the map is more flexible for them. And finally, improving the border security for vehicles further.

I’m planning to give Crossroads two more rounds of play testing and polishing, before shipping off to Bungie’s Forgetacular contest, and moving more fully on to map #2. I’m not sure Crossroads will end up a great map, but  I think it might end up pretty decent, and I learned so much from building it that – hopefully – map #2 is stronger for it.

GhaleonEB

Click here to return to the Forging Ahead Index