Reply To: Version 11: Q&A, discussion, and report bugs here

Written by  on  

MTrek forums mtrek forum Version 11: Q&A, discussion, and report bugs here Reply To: Version 11: Q&A, discussion, and report bugs here

#7967
obitobit
Keymaster

In short, yes.

There have absolutely been hidden (or I prefer the term foundational) changes. With only a few exceptions, there were balance adjustments throughout the game, to include shield/hull strength, class-specific gold earning rates, and there was another less-obvious on the surface, but perhaps much more significant change, in the power/speed differentials among all classes. These changes were made in consideration of, and complementary to- a slower rate of bot spawning, a wider expanse of space to battle in, and a tightened gold economy.

The game balance theory for Ver 11 can be simplified like this (and I promise, the incredible wall of text that follows really is the shortest version of it):

Pretending warp speeds are constant across all ship types, let’s say we take each ship class and assign it a generic “attack” value, based on the general effectiveness of weapons due to ranges/mechanics, damage-per-second potential, and how long that rate of damage can be sustained. Then we give it a generic “defense” value based on countermeasures like cloak duration, shield/hull, and xwarp. Weapons such as teleporter phasers, plasma used to cover your escape, and even the deterrent benefit of having really strong torps at a certain range- will also add to that “defense” value.

We combine those attack/defense values to come up with a sort of “power level” for each class. It’s true that this power level is somewhat subjective, and the weight of each factor used to determine powerfulness varies greatly in the actual game, and it depends on meta/environmental contributors like player skill, strategies & goals, starting energy levels, how many bots/enemies are in the area, distance to bases, resources, & terrain like nebulae, ‘roids, black holes etc. etc. etc.

It’s easy to miss the forest in so many trees, so for the sake of brevity, we’ll assume each ship has a quantifiable amount of power, and each ship also has very definite speed (remember we first determined power in the absence of any speed differential). But ship speed truly functions as a multiplier on ship power, so to balance the classes, we need to account for speed. Speed shortens or lengthens the effective distance to resources, terrain, and safety. It grants the faster ship initiative in combat, and it can make all the difference in the world when it’s time to escape or destroy a target. (There’s a diminishing return on speed multiplication once you get above warp 15 or so, and similarly, a diminishing return on having more power when your speed is below warp 11… but that’s for a more in-depth analysis. We’re keeping it simple here, right?)

In a perfectly balanced game, we could graph speed (max cruise warp) along an x-axis and power (as above) along a y-axis. As we add data points (ships by power/speed) the result is a sunken curve from the upper left to the lower right. We’ll have slower more powerful ships at the top left, faster less powerful ships at the bottom right, and mid-power mid-speed ships filling in the line between the extremes. This line can be referred to as the “frontier”.

We don’t have a perfectly balanced set of ships to pick from. If we did, it would be boring. Instead, we have this frontier of ship balance that accounts for a core majority of ships available in the game, but we also have outliers. These outliers are ships below and to the left of the frontier line, which have an objectively worse speed/power ratio than those along the frontier. We also have ships above and to the right of that frontier line, and these outliers can typically be described as being overpowered.

Since we don’t have a set of perfectly speed/power balanced ships, the outliers are instead accounted for in the meta game. In some cases, the rock-paper-scissors effect comes into play, and we’ll acknowledge that we’re either sacrificing something because our ship is overpowered, or there’s a very specific benefit to be had from using the typically underpowered ship. So identifying where everything falls on the speed/power curve is our real starting point when it comes to balancing the ships, which is essentially where I started.

Methodology:

I began by graphing out the 1990s “core-14” ship classes that have been a staple of mtrek basically forever. With their original specs, these ships give us the beginnings of a speed/power frontier line. We already have our first outliers too (for example, rbop & br5 are overpowered, while cda180 & br1k are distinctly underpowered). Next, I added the stock JavaTrek ships to the graph, and what I noticed is that our frontier line had shifted a little up and to the right. That is to say, the pool of ships as a whole became slightly faster and more powerful than before. Unfortunately that also means “underpowered” ships got even worse with the addition of jtrek ships. R.I.P. the CDA-180 and BR-1000…

Then I added the “new mtrek.com ships” from Ver 10 of this server to the chart. As you might imagine, that shifted the frontier line even further up and to the right, and much more drastically so. That was basically the starting point for Ver 10’s ship balance formula. Bots were also cranked up to the max, and the map layout was basically condensed. The meta environment and speed/power curve would have made bone stock classic/jtrek ships practically unplayable at that point. Even when we add in the power increases that were applied to classic/jtrek ships, they still largely fell short of the frontier. You can see the results today by looking at the scoreboard. There’re exactly 2 regular “jtrek” ships in the top 20 and they’re both KBOPs. There’s also a D-10, which was technically a “pre-core 14” ship that CLP removed from the game in the early 1990s, and the Valdore/Vorcha which were added to jtrek late in the cycle as upgrade-classes and those were overpowered by design. There are zero standard mtrek ships. If you expand the scoreboard out to the top 100 ships, there are fewer than 20 ships of any kind from before the new mtrek.com generation.

Only 3 of the 48 ship classes (Cube, SV-1, and Vorcha) account for nearly 50% of all the gold in the game, even counting all the dead ships. The top 10 most popular ship classes have earned something like 90% of all gold. None of this is to suggest that Ver 10’s ship balance was “broken” per se, but that it could be better. Most of the 48 Ver 10 ships were actually viable to some degree, but unless you flew one of the small handful of overpowered ships, you weren’t playing with a full deck either.

If we want to achieve better balance and keep all of the 48 ship classes (I think we do), our options are (1) to boost the speed/power ratio of our lower-left ships, (2) reduce the speed/power ratio of our upper-right ships, or (3) condense the entire speed/power curve into a narrower band so everything gets a little closer to the frontier. I opted for a combination of all 3.

Since we can move the speed/power curve whichever way we like, it was important to first decide where we want it to be. The standard 90s-mtrek ships + stock jtrek ships would be my anchor. That’s about half the ships balanced, locked in, and done. Counterintuitively, I also decided to *increase* the multiplication effects of speed in our speed/power ratios by dialing back the pace of play and spreading the bots out more evenly across a larger galaxy. Doing so gives a much higher degree of fidelity when adjusting the curve with our added newer ship classes.

You’ll notice that the the most obvious balance adjustments were to ships’ max cruising warp speeds. Across the entire pool we now have:

4 warp 10 ships
10 warp 11 ships
10 warp 12 ships
10 warp 13 ships
10 warp 14 ships
4 warp 15+ ships

Which gives us a 50/50 distribution of ships faster/slower than warp 12.5

Compare that to what we had in Ver 10:

3 warp 10 ships
5 warp 11 ships
9 warp 12 ships
14 warp 13 ships
12 warp 14 ships
5 warp 15+ ships

Giving us a 65/35 distribution of the same.

If you’re curious, regular 90s mtrek had:

1 warp 10 ship
3 warp 11 ships
2 warp 12 ships
3 warp 13 ships
4 warp 14 ships
1 warp 16 ship

Or a 57/43 distribution (pre-1994 mtrek had only 13 classes and a 53/47 distribution). That’s faster than Ver 11, but still way slower than Ver 10. if you consider a speed/power curve for the core-14 ships, larson, kpb, br1k, cda180, and cda120 were all low-left underpowered outliers. Aside from sui-larsons which were sometimes used to scum kills around earth/vulcan/sb3/sb4, and then typically discarded, none of those were ever popular or viable enough to regularly show up on the top 20 scoreboards of old. Personal recollection and the few scraps of scoreboard we have on USENET’s alt.games.mtrek strongly suggest the real meta game of the time was actually a lot closer to a 50/50 fast/slow distribution.

Whether “warp 12 slow, warp 13 fast” is ideal or not can be debated. But since we’re using mtrek+jtrek’s speed/power curve as our anchor, that’s roughly where all the Ver 11 ships are balanced.

By the way, did I mention that it’s basically impossible to achieve a perfect state of balance with so many ships? Because it is. I can’t imagine a way to pull it off without making every single class homogeneous and bland. We need lower-left outliers to challenge ourselves, to accomplish specific goals, and to appreciate how powerful our frontier+ ships are. We also need upper-right outliers to feel like badasses and to appreciate how much more difficult the frontier-and-below can be to operate in. Best case scenario, we make sure the low-lefts are as viable as possible- or at least make sure they have *something* of value to add to the game. We also do what we can to reel in the overpowered up-rights so they don’t eclipse every other option in the game.

At this point, there’s little else that can feasibly be done to improve balance. It was as much a thematic or stylistic choice as it was anything else to keep the mtrek/jtrek ships as bone-stock as possible. Now we have our galaxy, we have our bots, we have our arsenal of ships. The only thing left to adjust programmatically, is the economy.

The Economy:

I’ll preface this part by saying that in hindsight, it was probably an enormous mistake to not reset the scoreboard. That option is still very much on the table, but I hesitate to employ it because you guys have so many freaking hours invested in building up your fleets.

Aside from the annoying “database artifacts” carried over from Ver 10 to Ver 11 causing inconsistencies between existing/newly-launched ships, there’s an issue with fairness and player expectations. Fairness issues, because a ship launched in Ver 11 will struggle to catch up with one that’s spent most of its existence running in Ver 10. Expectation issues, because players will inevitably compare their Ver 11 ships’ capabilities with those of their Ver 10 predecessors. Most Ver 10-popular ships have moved down a notch or three on the food chain, and that could lead to frustration or ultimately less satisfaction with the game.

As for other economic issues and how they relate to overall game balance, I mentioned earlier that the pace of play has been slowed down by a fair amount. The bots spawn with zero gold just like a human ship would and they’re fewer and farther between, so we’ll have to hunt for our food a little harder than before. Fights are now a little more meaningful in terms of gold, so this becomes another tool for balancing the game.

Basically, our low-lefts now earn an increased amount of gold depending on just how much lower and to the left they are than our speed/balance curve. Similarly, up-rights earn comparatively less gold-per-DmgGvn than a majority of other ships. And that’s it. Simple.

Sorry if I rambled on a tad bit there. I hope I answered your question? Lol

  • This reply was modified 1 month, 2 weeks ago by obitobit.

Category :