Problem: Unreal's Boss Battles Suck.
FPS in general has issues with boss battles, yet Unreal's are notably bad. Do you like recycled boss fights? It's very much acceptable in moderation. But a) Titans are not engaging combatants. and b) they're recycled at minimum four times. Likewise, the final battle was a disapppointment. Here's what I did against the final boss: jumped up to the platform in the middle, strafed in a repetitive circle pattern on this platform over and over while firing away, aaaand victory. That was it. And that's all the bosses of the game. Four mediocre increasingly tired titan battles, and a pitiful end boss.
The solution: make them not suck. You still fight all the same battles, because my faithful design principles demand it. However, through AI enhancements, boss arena design and more, each should be unique from one another, and relatively engaging by comparison to what there was before. The end boss has received the most attention. It should be a real climax like any good end boss.
Problem: Flak Cannon > Eightball
The Flak Cannon is commonly perceived to be a better or more enjoyable weapon than the Eightball Launcher. This may seem unusual if you're familiar with the two. Think about it: the Eightball is the most multi-functional weapon in the game. You can fire rockets OR grenades as the default. You can hold down the fire or alt-fire key to load multiple grenades or rockets to be launched. You can hover over a target to acquire a lock-on for homing rockets. While loading rockets or grenades, you can also hold the opposing fire button to make the projectiles launch in a tight grouping, or don't for a wide grouping. How impressive is that for one weapon? And when is a rocket launcher EVER bad, let alone a rocket and grenade launcher in one... So why is the flak cannon perceived to be the better weapon, personal preference aside? There's two core reasons: first, all enemies that have the ability to dodge have a ONE HUNDRED percent chance of performing a dodge manoeuvre when any and all Eightball projectiles are sent their way, if the conditions for a dodge are met. The Flak Cannon grenade launcher alt fire on the other hand had a ZERO percent chance of ever being dodged. This makes the Flak cannon objectively the better weapon against dodging enemies (All Skarrj types, GasBags, Krall), while with the Eightball you're just wasting rockets in frustration before you finally get the kill. So you learn from this and use the Eightball on slower, non-dodging enemies, specifically the three types of Brute and two types of Titan. Guess what? They have 30% explosive damage resistance, further lending to the accurate perception of the Eightball's inadequacy. This is not a problem for the Flak Cannon of course as its primary fire is not an explosive damage type.
The Solution: I've made eightball projectiles have an 80% chance of being dodged, and the Flak Cannon alt fire have a 60% chance of being dodged (the difference justified by the eightball having better range, the ability to multi-load rockets to ensure at least one hits a dodging enemy, and a lock-on (which is still often dodged, bear in mind). Take note this dodge chance is per projectile, therefore many multi-loaded rockets are still guaranteed to trigger a dodge. Secondly, the dumb slow enemies with 30% explosion resistance now have 20% explosion resistance. It's minor, but it'll help. So there you have it, The Flak Cannon while still best for dealing with those pesky Skaarj will absolutely be dodged here and there rather than the deliverer of easy kills. The Eightball will now be able to shine, rather than either being dodged by everything with 100% frequency, or damage-reduced by everything.
Problem: Weapons can at times be rather unsatisfying to use
You recognise the Eightball, but aren't ALL the weapons just a little underwhelming? It is often said Unreal's gunplay is a little unsatisfactory. A stand-out example would be the Stinger. It's just not that enjoyable to use. This problem is an accumulation of absolutely everything: sound design, animation, projectile effects, impact effects, kickback, perceived usefulness and uniqueness of a fire mode, possible bugs present and so on.
The Solution: See the WEAPONRY category of the changelog. I've faithfully refined and enhanced everything necessary weaponry related. An FPS is no FPS without its guns. You're going to love them.
Problem: Unrewarding Exploration/Unreal's exploration is sub-par
It is. Not the level designs themselves, they're mostly an absolute joy, but rather the supporting design or lack thereof that is supposed to incentivise it. On my first playthrough I skipped over a number of areas. Why is this? Because I was often stocked up on health and ammo on the original hardest "UNREAL" difficulty without having to be completely thorough. This is the primary reason, though I believe longer-term reward is important for such an exploration-heavy game with huge levels, and Unreal has next to none.
The Solution: Resource distribution is tighter on higher difficulty levels. Core issue addressed. As for longer-term rewards for exploration, I've introduced a basic weapon modification system: find upgrade modules in the levels (much like the Dispersion Pistol upgrades) but you can choose one of two weapons to upgrade per module. Which weapons you can choose from and what upgrade it applies depends on the module you find. This is not the sole reason the weapon upgrade system was introduced of course, just one of many, but it would probably be an important reason often overlooked by those unfamiliar with game design.
Problem: Weak Resources Distribution
Unrewarding exploration is not the only downside of weak resource distribution. Challenge of course is an important factor sullied by throwing ammunition and the like at the player. Challenge being one of most important defining factors of a game. So, exploration and finding secrets is discouraged. Strategic management of pickups is not a concern for the player. Engagements can be breezed past with lesser skill. You can spam the same overpowered gun (cough Flak Cannon) without having to diversify your weaponry and tactics so much, and so on.
The Solution: I've left the lower difficulties alone for the most part. Very Hard difficulty and above have tighter resource distribution design. To elaborate, I'm not talking survival horror levels of tight. Think more Doom or Quake. In those games' higher difficulty modes it's always worth being a thorough explorer, managing pickups, fights are that much more intense even though the AI is more primitive than Unreal's. With Unreal Evolution, you'll experience Unreal's great gameplay thoroughly optimised.
Problem: Interior levels are much more detailed than exterior.
Making vast open organic 3D playspaces, especially back in 1998, is a challenge to detail. Unreal was one of the first to even try it. Interior levels were much more suited to 1998 game tech as game engines were not too competent with smooth, rounded surfaces and accurate physics collisions. Furthermore there is a simple common sense rule of the larger your level, the more time and attention it demands. Lastly good old Unreal Engine 1 not to mention typical game dev cycles has its limits...and its priorities. This problem is very obvious in Unreal and all other games of its time that tried it. Compare the level of detail in NyLeve to Terranuix if you'd like an example of exterior vs interior detail.
The Solution: I have the time and the willingness. I have the experience of adding subtle faithful details to an Unreal Engine 1 game with GMDX for Deus Ex. I have the design sense to do it right (again as with GMDX), and lastly I have OldUnreal's updated 227 engine which enables use of advanced graphical features. Don't expect it to look like a modern game or even a mid-2000s game, just expect GMDX all over again: the same levels, just with nice new subtle details that are often confused for vanilla. Interior levels also have new details, but exterior levels typically have more work done to attain a closer level of artistic consistency throughout the game.
Problem: The first third of the game is arguably not as engaging as the second or final
This is a problem with many games, and often turns players away. It comes with the nature in most cases: you only have a couple weapons in your arsenal to do battle with. You've only encountered a couple of enemy types. You're still learning the basic rules of the game. The developers in many cases made the early levels before they had mastered the tools they were working with (not sure if it applies to Unreal, but it's obvious with Deus Ex). And so on.
The Solution: there isn't one per se, as mentioned it is a problem with the nature of the game, but this mod should be a little more engaging from the offset through my conscious design attempts to enhance early game and the general overall quality and vision of the game. Nonetheless, the Sunspire and subsequent levels will still be some of the most memorable and engaging, and I recommend not giving up to would-be dropouts. I will list some enhancements made which should help:
-The Stinger (an early, unsatisfying weapon) is now a much more enjoyable weapon to use...as are all weapons, but it's especially notable with the Stinger.
-Enemy variety and behaviour is more diverse, including an entirely new enemy type introduced early on.
-Levels not only has subtly improved visuals, but you may also find the occasional new secret or challenge.
-The new weapon modification system makes weapons further enjoyable to use, adds fun choice, and makes exploration more rewarding (see above), as does tigher resource distribution.
That's it really. As usual, nothing majorly game changing. I always strive to be relatively faithful to the original vision.
Problem: Sparse Enemy/Encounter Variety
Unreal features a reasonable amount of enemy types and styles, and most of the combat is projectile and melee based. This is good. However, I believe its combat (on the enemy's side) falls a little short in two key ways: you typically only face one-three enemies at a time per encounter, lets say two on average (except late game which are mostly all Krall or Skaarj), and if there are multiple enemies they're usually of all the same class. This means you're fighting a group of enemies with all the same behaviour, where there's less variables you have to account for. On the plus side, Unreal's AI is pretty good, sort of (skaarj at least are notably above most other FPS of the time. Most other enemy types in the game aren't at all on that level however).
The solution: Improved AI behaviour, new enemy placements, and occasional encounters that include multiple enemy class types (monster in-fighting likelihood was added to prevent them all fighting eachother at the most immediate act of friendly fire). Unreal Evolution does not change the game into Doom wherein you fight hordes of varied enemies, especially early game, but it does inject a little variety here and there. This results in more variables you have to account for in combat, more ways in which a battle can turn out, and more challenge.(edited)
For purists, fear not. It's a subtle increase across the board. You will notice it but it won't feel out of place or wrong. It will feel right.
