Oblivion

I've put about 40 hours into Oblivion at this point. During all the pre-release hype, I never had the intention of buying it. I skimmed a few previews and saw the huge anticipatory buzz in the Games forum, but I hated Morrowind, so its sequel didn't appeal to me. I've been fooled enough times by sequels, or the promises of an author/director/studio's latest work, to know that if I didn't like the first title in a series I should tread extremely carefully when considering its successor. Morrowind was a really dry, empty game. You spent most of it walking across desolate, unpopulated wastelands. When you came to a town, the majority of inhabitants were robotic, all sharing the exact same dialogue trees and standing in place. I can't remember a quest I took on that wasn't a fedex. I'd come across a random cave in the landscape, go inside, and be killed instantly by level 30 ghosts. My Morrowind experience ended relatively quickly.

Oblivion is quite the opposite. The world is lush and densely packed with towns, dungeons, inns, countryside manors, abandoned daedric temples, and the Oblivion gates. Besides the limited number of Rumor dialogues in each area, every NPC has their own unique set of dialogue, all of it voiced aloud. Enemies in any given location scale to your level, providing a challenge that is evenly proportionate across the entire game. Navigation is no longer entirely in real-time, as you can travel instantly from any major location to another, and the location of your next quest goal is always marked on your map. And the quests are now the best part of the game. Since interminable traveling or wandering is no longer a valid delaying action as it was in Morrowind, the length of each quest is extended by adding twist after twist to the main objective. Something as simple as "retrieve the book that so-and-so has" can be twisted into a five- or ten-step quest involving questioning people around town, infiltrating a castle prison, finding a secret passage, fighting vampires, aiding the escape of a convict, fleeing the city, then sneaking back in to finally retrieve your goal. Almost all of the quests are like this, turning into something you never could have predicted from the initial description, and it makes them greatly engaging, a series of fun little stories to play out. It's the best questing I think I've ever done in an RPG.

I'm playing as a thief and assassin, which makes security one of my primary skills. Breaking, entering and stealing can be really satisfying, but the functional game systems they have in place to play out lockpicking and pickpocketing mar the experience to some extent. Essentially, they encourage the player to save immediately before each sneaking/stealing attempt and jam on quickload upon failure.This is lame in an of itself, but from a logical standpoint it's also bad-- anything that occurs between quicksave and quickload never happened in the gameworld, meaning that all those failed attempts don't have any impact on the player or the gameworld. When a level 1 thief quickloads 30 times while trying to steal an elven broadsword off the count of Cheydenhal before succeeding, from the gameworld's point of view, it's as if he succeeded on the first try. Functionally, a level 1 thief can clear any obstacle in the game with no repercussions if they're persistent enough on the F9 button. This is a broken system.

How pickpocketing (PP) works in Morrowind: The player enters stealth mode, approaches an NPC, and right-clicks to enter the NPC's inventory. When clicking on an item therein, the player either successfully receives the item, or the inventory window closes, the NPC goes aggro, and the player's bounty amount rises. This is a binary success/failure system, which the player can instantly reset by quickloading. Since the penalty for failure is much more trouble than it's worth (especially since a thief loses all stolen items in his inventory when nabbed by the guards) the player will always quickload.

What needs to change: PP shouldn't be binary. Instead: if a player is unseen by any NPC during their PP attempt, failure never results in being caught. Instead, failing simply closes the NPC's inventory window, and PP can't be attempted for five seconds. Chance of success is a percentage based on the following factors: PC's security skill, mass of item being stolen, and relative gulf between PC and NPC's character levels. Hereby, a level 1 player with 20 security skill attempting to steal the elven broadsword off the level 25 count of Cheydenhal might have a 1.2% chance of success. Since the player would have to wait five seconds between each attempt, the penalty is the player's own time investment, which could be major. Most likely, the player will give up after his 20th try, functionally making the overall PP attempt a failure without directly punishing the player (and prompting a quickload.) Of course, with enough persistence/luck the level 1 player could eventually be successful, but this success would be earned through the player's (and thereby the PC's) time investment.

Also, as no failed attempt would be detected when the PC was unseen by NPCs, any failed attempt would be detected when the PC was seen by NPCs. Because of the five-second wait between each opening of the inventory window, the player with a low success rate would have to monitor whether they had been spotted between each attempt before trying again. On the other hand, no successful PP attempt will ever be detected, whether they PC is in an NPC's sight or not. So while an unskilled thief would have to follow their target to an isolated area and be sure no one would be around for a while to guarantee a successful theft, a top-level thief could walk down a crowded street stealing people's coinpurses and jewelry without worry, knowing that they had an extremely high chance of every one of their attempts being successful. This entire system would work better from the standpoint of game logic, and be more satisfying for the player from a play standpoint.

How lockpicking (LP) works currently: The player buys lockpicks from a thieves guild or assassins guild rep. Upon encountering a locked door, they enter a lockpicking minigame, in which the player much manipulate each tumbler of the lock, then click it into place at just the right moment. Once all tumblers in the lock (more or less depending on difficulty) are raised, the door opens. The game is paused the entire time this minigame is played. If the player is seen when initially attempting to pick the lock, they will be pursued by a guard. When the player fails to click a tumbler in at the proper moment, their pick breaks, drawing from their total supply of picks. When the player runs out of picks, they cannot proceed.

Since picks are a limited commodity, this setup encourages the player to quicksave before approaching a lock, then quickload if they waste too many lockpicks. If the player is not seen before approaching the lock, they can spend as much time as they want in the LP minigame, meaning that the speed of cracking the lock has no value, counter to what one might expect.

To this end, picks should not be expendable. A lockpick is a single tool which has durability like any other piece of equipment. The player can buy or find better quality (more effective) lockpick kits over the course of the game. The player's LP skill is determined by the following factors: his security skill, the quality of LP kit he's using, the condition (current durability) of his pick, and the relative difficulty level assigned to the lock in question. The LP skill, then, functionally determines the window of opportunity that the player has to click each tumbler into place. For instance, a level 1 player with a level 1 pick against a level 10 lock may have .1 second to successfully click a tumbler into place. If there are five tumblers, and missing your click on any given tumbler can undo another tumbler you've already successfully raised, the likelihood of success, and likely time investment required by the level 1 player to successfully raise all tumblers consecutively, becomes daunting, while still possible. On the other hand, a level 40 player with a level 10 pick against a level 2 lock might have one or two seconds in which to click the tumbler in, making the opening of this lock an elementary endeavor that could take less than a second or two to pass.

Another factor here is that the game does not pause while the LP interface is overlaid. So, if the locked door is in a patrolled zone, opening it as quickly as possible is to the player's advantage, lest the patrolling guard return to the area and spot the player in mid-LP. This would be another functional advantage for the skilled player, and another factor discouraging the low-level player from attempting a lock that's out of his league, which would greatly increase the likelihood of being caught by the guards. One situation that would be greatly satisfying to the player under this system: say the front door to a jewelry shop sports a level 20 lock with five tumblers, and a guard walks from one end of the block to the other, leaving only five seconds for the player to manipulate the lock unseen. At first approach as level 1, the player has almost no chance of successfully entering the shop, since breach such a disproportionately leveled lock will almost certainly take much longer than five seconds. However, upon returning to the lock after significantly leveling up, the door becomes passable in only a couple of seconds, allowing the player easy access to territory they were earlier denied. Very cool, and makes the player feel like a badass.

In both of these cases, the proposed game system more accurately simulates its real world counterpart, and completely invalidates the quickload button as a valid strategy. These changes would make the thief career track much more satisfying for the player, and allow greater opportunities for obstacle design and balance for the developer.

THANKS!