Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 24, 2014
arrowPress Releases
October 24, 2014
PR Newswire
View All

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

Analysis: Video Games - Downgrade Complete?
Analysis: Video Games - Downgrade Complete?
July 8, 2009 | By Simon Parkin

July 8, 2009 | By Simon Parkin
More: Console/PC

[In this critical examination, British games journalist and Flash game producer Simon Parkin takes a look at a satirical game that reveals much about the state of game design.]

The evidence that video games may yet emerge from their period of extended adolescence comes not from the dizzying realism of the next Forza, nor from the unrivaled spectacle of the forthcoming God of War, nor even the news that Lara's improbable cleavage is scheduled for a sober reduction in the next Tomb Raider.

Rather, it's in the emergence of a new breed of satirical web game, one most famously exemplified by last year's Achievement Unlocked, which poked fun at gamers' obsessional pursuit of Xbox Achievement points and PlayStation trophies.

These snappy experiences parody not the grim cliches of gaming's stories, settings, or visuals, but rather the more subtle underlying systems that drive them or, in the case of Achievement Unlocked, surround them. They compel us to play via the very same hooks that big budget titles employ, but their exaggerated presentation and irreverent context encourage us to evaluate the worth of these mechanics and, in doing so, question the very reasons why we find them so irresistible.

Upgrade Complete is the latest such satirical game in this vein. It begins by presenting players with the bare bones of a shoot 'em up; a blocky, silent retro game whose music, graphics, menus and even developer logos must be bought and upgraded one by one with in-game currency.

At first glance Upgrade Complete appears to be making fun of downloadable content, those upgrades -- new costumes, weapons, characters and levels -- released by a developer for a modest fee after a game's initial release. After all, until you purchase a humble loading bar you can't even start this game (the developer "lends" you $1000 to make this initial purchase).

But as you play on, the game's target is revealed to be a more substantial and pervasive one: that of the in-game upgrades that furnish our characters with better weapons and abilities, a feature found in almost all contemporary video games from Fallout 3 to Call of Duty 4.

In Upgrade Complete, your slow-moving ship starts out with a single front-mounted gun. But as you shoot down enemy ships and collect the coins they drop, you can pay to upgrade and "customize" its maneuverability, weapons and effectiveness, revealing, piece by piece, the final, ideal iteration of the ship originally designed by the game's creator.

This mechanic originated in traditional RPGs, where defeating monsters earns experience points and coins used to improve your character's abilities and purchase new spells, armor, and weapons. The system is compelling because it offers a clear way in which your time spent with a game results in demonstrable progress, while also strengthening the idea of your character undergoing a journey through which they're growing and developing.

But it's also a fiercely linear sort of progression: the more you play, the more your character realizes its predestined potential, one that has little to do with your own choices or successes.

In recent years more diverse games have begun to adopt this unlock trajectory, using it as a way to hook their player in and to artificially pace the game’s progression. Recent open-world titles Infamous and Prototype each feature characters who start off as pale reflections of the avatars they eventually become, the pacing of each game set by the character abilities that unlock in a steady trickle.

So it is with Upgrade Complete, a game that only begins to resemble its final, finished state as you play and invest in it. It's compelling because the economy of cause and effect is immediate, overt and frequent. But it is also a cheap trick that highlights a fundamental change in the way designers are constructing their games.

In the beginning, non-RPGs rarely employed this kind of system. In Pac-Man, for example, players enjoy the full range of their avatars' abilities right from the off. The focus isn't on playing the game for a long time to develop (or, more accurately, complete) their characters, but on perfecting technique with a defined, immovable skill set. The better you are at Pac-Man, the further into the game you are able to progress, its rewards structured around perfecting skill rather than merely investing time.

Likewise, the full breadth of Street Fighter IV's content is unlocked very quickly. And yet tens of thousands of players are still heavily invested in the game, not because there is more content to "purchase," or better moves with which to upgrade Ryu and Chun-Li, but rather because they are on a quest to play the game more effectively and beautifully, to perfect their technique and to better manipulate its systems.

There is no in-game reward structure in Street Fighter IV for a player who invests more time than another player, other than the probability that they will grow to become better at the game. As such the payoff for the player is in learning and improving a skill, not in purchasing and accruing in-game items or upgrades, a crucial distinction that bucks current consumer-based gaming systems.

In part the use of sequential character upgrades in games is a byproduct of the pursuit of linear narratives. If a game is telling the story of a character's journey, then one of the easiest ways to communicate a sense of that journey through the game system is by adding to the character's abilities. But this approach, while effective, encourages sloppy design.

Imagine if Super Mario's jump distance extended as you worked through Super Mario World (as it does in Prototype). The need to craft thoughtful levels and puzzles and construct a balanced difficulty curve through them would diminish as the pacing of the game is instead dictated by withholding abilities, not finding ever more inventive ways of challenging a player to apply what they've always had.

Good games encourage players to literally better themselves, rather than simply toiling away to unlock game features or character upgrades that should have been present from the start. And yet, this most simple of reward schemes can prove irresistible, locking us in to patterns of play that distract without enriching.

We should be glad of satirical games such as Upgrade Complete which challenge us to identify what it is we find attractive about a particular game, and, as with choosing a partner, discern whether that attraction will be good for us in the long run. Therein lies maturity.

Related Jobs

Giant Sparrow
Giant Sparrow — Playa Vista, California, United States

Lead Artist
Digital Extremes
Digital Extremes — LONDON, Ontario, Canada

Digital Extremes
Digital Extremes — London, Ontario, Canada

Generalist Programmers
Petroglyph Games
Petroglyph Games — Las Vegas, Nevada, United States

Unity Engineer


John Mason
profile image
Fantastic piece Simon, and not just b/c I have another game to go check out (seriously, when I read the description I started laughing so hard b/c it's sadly such a great parody, from what it sounds like. Is is a browser game or freeware download?). Personally I don't own any of the next gen consoles (sadly), but a friend of mine has all three, and thankfully he isn't *quite* the type to lust so strongly for trophies and achievements. I say "quite" b/c when it comes to certain games-like SFIV-he does have a tendency to show off w/e titles he's unlocked, which is okay w/ me but I just hope I never have to get an intervention for him b/c he obsesses over achievements/trophies 24/7 xD.

Now as far as what you've said about the different types of progression used for many game design, I completely agree w/ you there too. Like you said, one of the main reasons games have taken to the route of encouraging play through unlocking abilities rather than giving those abilities at start and challenging them w/ thinking of more creative ways to use what they've got, is b/c so many games these days are going after the "narrative" angle (and many of them having not-so-good gameplay as a result, imho) and one of the best ways to encourage that type of angle is through limiting a character's starting ability and have them improve over time; personally I feel it all really started w/ the PSX, when developers began releasing games like Resident Evil and Metal Gear Solid, both of which (in addition to many others) went after more cinematic atmospheres. Of course, two things those games have that many games this gen (in my opinion) don't have is raw challenge, and that sort of leads me to some of the other reasons why I think developers take the "upgrade" route:

1: Less intimidation: Now that so many developers want to make their stuff "accessible", they don't want to scare people off by giving them all their abilities right off the back. Some people tend to get frightened off at that. Some games still give you all the bare assets you'll need right off the back, but the majority of games these days don't. Personally I'm not intimidated when a game gives you everything you need right from the start, which is why I play so many old-school games, shmups, fighting games, etc. But some people don't necessarily want to be challenged through skill/creative skill, and just want to play a good game and be taken for a journey, which is cool too, so long as I can still my fix the other way around.

2: Lengthier games: IGN did an article on Sonic, and in the last part one of the developers of Unleashed say something along the lines of Sonic games these days needing so many 'non-Sonic' gameplay sections mainly b/c the developers felt that gamers thought a 'core Sonic game' wouldn't be enough to justify a full-priced retail release. For me, it's pretty easy to see how that might apply to developers these days that take the 'upgrade' design mentality approach; they might think that many gamers today feel as though a game isn't worth it if it starts by giving the player everything they need, b/c so many players might interpret that as a mediocre game. Not my personal opinion, just sayin'.

3: Different audiences: To be totally honest, I don't think the majority of gamers these days even play their games more than once, not w/o incentives like Achievements or Trophies anyway. Given the length of games these days (which seems a bit shorter compared to last gen anyway) and how many hit the market, a good # of gamers probably don't have the time to play through games again. But if they *did*, and there were no incentives like achievements/trophies/whatever,...would they play the games over again, and again?!? Unlike back even 8 or so years ago, I doubt it. The teens have the texting, parties and whatnot, the college students have, well, college, and parents have their families to deal with. There's exceptions to all those groups but that's just being general. You'd think that'd be a reason to encourage the 'replay' aspect of design via like older games, but I think the drive for narrative in games these days is too strong for most developers to take that path.

There's more factors at hand than that, but that's all I have the time to speak up on. Like I said before, I favor the 'replay' aspect of design and can easily see a way for exceptional developers (perhaps ones in the future, moreso than any of the big cats out there today) to meld narrative operatics (if they want that) w/ gameplay that encourages the player to better their skill at the game and try perfecting the systems there. It'll happen, someday, but until then we gotta make the most of what's thre. Great article piece Simon; looking forward to more. Later.

Mark Brendan
profile image
I disagree with the premise that games with levelling up or unlocking (or a combination thereof) style progressions all start with a predetermined ideal endpoint that you approach in a deterministic fashion. Certainly in the more RPG-oriented ones like Fallout 3, they're a way to customise your character as you improve. It is therefore a branching progression and not a linear one, and one that allows the player to personalise the game experience and thus achieve greater investment and immersion in the game world. I think it's a bone fide evolution of gameplay (and one that pen and paper and tabletop games mastered way before video games) that will continue to devlop along its own branch--it's not the cul-de-sac it's being painted as here. That said, I don't want to spoil the punchline--after all Progress Quest was ace ;-)

Steven Conway
profile image
Great article. I think Mark is correct though in pointing out that we should not blindly assert that the majority of games these days are linear a-b-c pathways to progression; many are rhizomatic, branching in numerous ways.

Joshua Sterns
profile image
"Good games encourage players to literally better themselves, rather than simply toiling away to unlock game features or character upgrades that should have been present from the start."

I disagree. Good games come in all shapes and sizes. Bioshock would have failed if you started out with all the upgrades. RPGs, as Mark has mentioned, allow players to customize their character in a variety of ways enhancing the overall experience. Ninja Gaiden and The Force Unleashed could have started out with everything available, but the unlockable weapons/powers help with the learning curve and add something besides a story to look forward to. Also both those games allow for a second play through allowing players too experience a fully powered up character from the start.

On the "skill" side there are also great games. BlazBlue is a blast, and the depth of each character encourages me to practice/play more. Geometry Wars and Tetris are too of my favorite games, and they only reward is a high score. Halo is a great fps, and I play it often. This doesn't mean, however, that I am trying to get the 50 rank in mp.

Good games can be skill and/or progression based. When it comes down to the bare bones it is about the fun factor. If a game is fun, then people will play it. Some will have fun "bettering themselves" (although I doubt my mother would think I am bettering myself via gaming) and others will enjoy upgrading and collecting their way to perfection.

I did enjoy this article. It reminded me of the Simpsons game released by EA last year. The story progresses to the point where the modern video game Simpsons met with their 8 bit counterparts. The conversation that ensues is hilarious. 8bit Homer says, "No highscore! Well then what the hell are you playing this game for." Modern Lisa, "It's not about the score. Experiencing the game from start to finish is the reward." 8bit Bart, "That's lame." I am paraphrasing here, but you should get the idea.

Joshua McDonald
profile image
"Imagine if Super Mario's jump distance extended as you worked through Super Mario World"

Of course when you take a game that was designed a certain way and throw in a trait that it never intended to have, that trait sounds bad. "Imagine Tie Fighter if you could double in size by eating a mushroom."

Joshua Sterns said it best: "Good games come in all shapes and sizes." There are very few things that are absolutely bad or absolutely good in games. It always depends on implementation.

RPG's expect you to, at least some degree, personify your character, and acquiring better items, abilities, etc. mimics what people seek to do in real life (better yourself, better your circumstances, etc.). In addition, the "rags to riches" type scenario is extremely appealing to most people: The idea that a semi-average person rises to greatness.

Finally, the accessibility thing has been mentioned above, but I think it could use more emphasis. Many games in their later parts would be ridiculously inaccessible, even to hardcore gamers. Take somebody who's never played Eve Online and ask them to properly outfit a huge battleship and send it into battle. They'll quit before they ever reach the battle.

Simply put, I enjoy many types of games, some with more of a leveling focus, and some with more of a skill focus (and many with a lot of both, such as Tactics Ogre). To claim that one is good and one is bad is completely wrong. Both have good implementations, and both have some pretty awful implementations. I could make a counter-satire piece where a game required you to implement a 5 button combination with perfect timing to make a simple jump, but it would not prove that games that require exteme levels of skill are somehow inferior.

Evan Combs
profile image
It all depends on what the game is going for. I prefer skill based games, but there are upgrade games that I enjoy greatly as well. Each is good in their own merits, but I do think that upgrades shouldn't just be thrown in just to make it more accessible or to increase replayability like a lot of games seem to do it anymore. The game needs to be designed from the ground up with the upgrades in mind.

Ben Taber
profile image
"I could make a counter-satire piece where a game required you to implement a 5 button combination with perfect timing to make a simple jump"

Sorry but Sega already made Virtua Fighter.

Okay that was a throwaway joke, I'm actually posting to point out a couple of things I thought were worth mentioning that I noticed about playing Upgrade Complete;

a) As you upgrade the graphics, the playfield gets more cluttered and confusing. It gets more and more difficult to actually pick out the important information from all of the shiny explosions.

b) As you upgrade your ship, the amount of work you actually have to put in as a player diminishes, to the point where you can begin clearing entire waves by sitting there holding space and all the coins come to you.

c) The final upgrade of the menu items are mostly simple clean vector art while the mid-tier upgrades are unnecessarily busy photo-textures. ALSO the mid-level upgrades are in no way superior to each other, just different photo textures. Particularly amusingly, the final logo is exactly like the one before it, except 10% bigger.

All in all, it was an interesting commentary on how upgrades can actually work to undermine the gameplay instead of support it. Honestly, though? I still thought it was a fun game.

An Dang
profile image
I don't agree with the conclusion.

This is a slap in the face to all RPG developers and RPG fans--or anyone who would implement RPG elements into their game.

Travis Johnson
profile image
I think "slap in the face" is a tad bit harsh

All he's really saying in the article is be mindful of your design decisions--before you include (or design a game around) a certain mechanic, ask yourself if it really adds to the experience you are trying to create. Linear or non-linear, story-driven or character-driven, RPG, FPS, whatever, make sure you are devoting all the resources at your disposal (time, talent, money, etc.) toward the creation of a game devoid of cheap tricks or tired (even useless) conventions.

And if a developer says to him/herself, "Psh, it doesn't really matter--it makes the game work, that all that matters", well that's one's already lost.

Jon Waholic
profile image
There are some differences between RPG level ups and unlocking skills/actions/powers.

In RPG, selecting a class, skills or stats upgrade means player has given up another skill. Example: becoming thief, you have given up warrior's heavy armor proficiency or priest healing spells. Leveling up the stealth skill, you will be better in hiding but less successful in picking lock or disarming trap. The process of choosing is interesting and challenging.

However, creating a unreachable ledge, then forcing the player to replay whole stage after unlocking double jump much later part of the game, is bad waste of time. Tasking player with non-challenging errands under the guise of 'replayability'. If there is a high ledge and player has not learn double jump, put some narrow moving platforms and make it a more challenging task. Casual players who forgo the challenge can come back later after learning the skill. Same for more difficult monsters. They are difficult but not impossible. Weaker players can come back to challenge them after powering up. Designing more challenging task means more effort. "Locked door in stage 3, put the card key in stage 8, force player to replay stage 3", is simpler to implement.

Replaying game stages with avatar who has powered up, player can breeze through the level and easily trash previously tough bosses. Challenge is gone. Things get boring quick. It is a cheap effort to lengthen the life of the game while disregarding player's gaming experience.

Jon Waholic
profile image
BTW, Great article, Simon. A good refreshing reminder for me to put in more careful thoughts about upgrade feature in games. Thanks.

Alexander Bruce
profile image
Jon, you've basically just said that Metroidvania games are a bad waste of time. I'm not sure if you're aware of that conclusion, but I'd hardly call games that follow the Metroidvania structure "lazy". Their level designs would take far more effort than linear games because they're integral to the experience.

Bart Stewart
profile image
My impression is that the effect Simon describes -- of designing a game so that the complete progression sequence of a player character is predetermined -- tends to be justified by designers in one of two ways.

1. The developer wants to tell a specific story. Predefining the progression path allows moments of gamerule-based progress and key plot points of the story to heighten each other's effect.

2. The developer believes that any moment of gameplay not consciously defined is one that may be sub-optimal. The only way to guarantee maximum fun is to tightly control every possible gameplay-gameworld interaction. Player choices must be tightly constrained at all times to insure that they can only do what they're supposed to be doing at that moment in the game.

I could be wrong, but I feel like I'm seeing more and more games being designed under that second assumption. That doesn't mean I think all games should be completely unscripted, but I do wonder whether some aspects of fun are being lost as developers try harder (to minimize financial risks) to control every possible outcome of every moment of gameplay.

How hard is it becoming to persuade the money people to fund the development of games that allow more than an illusion of player choice?

How much room is left for games in which "advancement" comes not from a mechanical increase in the numbers defining the character's power, but from the player developing a better understanding of how the gameworld works? And to what extent can the latter type of game allow the player to enjoy a good story?

Francis Page
profile image
Metroid, Zelda and Metal Gear Solid are all games are all games based on this. Those are also the games I enjoy the most. I prefer linear games, I feel lost in games with too many options. I really don't think that this kind of design should disapear.

James Hofmann
profile image
I think upgrades are considerably more complicated than the article states.

The case of an explorable world with implicitly unlocked areas - as with Zelda and Metroid, etc. - has its own set of implications. First of all, seeing unreachable areas early on ignites player curiosity. Second, when the player backtracks to explore the unlocked segments, it acts to extend the reward of the upgrade.

The case where player powers increase is likely to modify the nature of the gameplay; a player character that moves three times as fast at the end of the game as at the start isn't uniformly better - the character now requires more dexterous handling to be controlled optimally. Similarly, with new abilities, different strategies have to be adopted to incorporate the abilities. In the ideal case this not only lets player skill progression match game skill progression, it lends a different tone and style to each segment of the game. It requires a _lot_ of thought and balancing to make this kind of progression live up to the ideal, though. Close to zero games really make any headway into this level of balance.

It's the situations where upgrades are simply increments of abstract statistics or flashier versions of "keys to the door" that tend to be the least appealing or rewarding, but even here, most games can adequately flesh out their statistics or "door keys" so that they don't feel quite as abstract.

The essential problem is that upgrades get abused as a stand-in for other forms of content, rather than a structure on which to build more fleshed-out designs. If you throw in upgrades arbitrarily, of course they are going to suck. (I say this having tried using simple incremental upgrades and increases in enemy power in an attempt to simplify one part of the tower defense genre in exchange for more emphasis on a positional game. It backfired; the majority of TD's strategy emerges from giving both the enemy waves and the upgrades bold, well-defined properties, and the positional game alone was underwhelming.)

Michael Martin
profile image
There is one core thing that prevents gamers from lapsing into boredom in video games, and that is presenting new and interesting experiences for the player. If a game promises nothing but more of exactly the same activity without any variety, a player will quickly lose interest and find something more entertaining. Tetris, although the gameplay consists exclusively of dropping blocks into a play area and attempting to form solid lines, is constantly challenging the player - either by increasing the speed, or by presenting new situations in the layout of the dropped blocks.

It's for this exact reason that backtracking in a game is generally considered a Bad Thing.

A good Metroidvania game, although it will invariably include some backtracking, will typically allow the player to either take new paths or use new techniques to rapidly travel through old territory. Castlevania: SotN, for example, has the Wing Smash in bat form or Speed of the Wolf in wolf form, both designed to allow the player to rapidly travel through long areas unhindered. Super Metroid, on the other hand, either designs areas to be rapidly travelled through with Speed Dash, or breaks up old areas with new sections that can be explored with the Spring Ball, Jump Boots, Ice Beam, or whatever, meaning the backtracking is frequently broken up by new areas and new experiences. Or, the backtracking is made shorter by means of a shortcut.

Simply providing the player with the means to massacre enemies that were previously genuine dangers is, in and of itself, a new experience, and it can work well - for a time. Before long, it's not a new experience any more. If the player isn't challenged, the game quickly becomes 'Press X to win'.

Running the upgrade route is all well and good, but only if the upgrades are substantial enough to cause a noticeable change in gameplay. This is one of the areas in which the classic Metroidvanias excel - nearly every upgrade permits you to play the game in a new way. My least favorite upgrades in Super Metroid are such things as the Gravity Suit, which permits you to move normally underwater. It opens up new areas to explore, but it doesn't change the gameplay at all like the Charge Beam or the Jump Boots, and there aren't enough underwater areas to make it feel substantial. As James says above, it's like a flashy key (although it's a flashy key that lets me soak a lot more damage, which redeems it somewhat).

Of course, then there's games like Mario 64, which gives the player nearly all the tools at the beginning of the game; only a few extra goodies, that are used sparingly, need to be unlocked. However, not only is the game a joy to play through the first time, it also has a high replay value, because Mario's abilities - though limited - are usable in a variety of situations, and as the player grows more skilled, the challenge changes from simply completing the stages to doing so in increasingly speedy and challenging ways by using those abilities in places they weren't meant to be. The sideflip and wallkick are wonderfully broken in this sense, utterly breaking the level designs when executed properly. The challenge, then, is executing them properly, which requires very precise maneuvering and timing. This is helped significantly by the compact levels and the lack of forced pathing in the level design. If it were possible to tear through the levels in Mario Galaxy in the same way, I'd have a lot more fun playing through it again as Luigi.

The flipside of that is Shadow of the Colossus. It's an excellent game, and I'm glad I purchased it, but I played through it once and that was it. In essence, there is one way and only one way to take on each colossus, and no amount of skill will change that. It's possible to complete the game much faster on the second time around, as the experimentation and discovery is no longer required, but at the core of it, you're performing exactly the same actions as the first play-through, and merely leaving out the actions that didn't come to fruition. There's no new experience. Fortunately, the single play-through of Shadow of the Colossus is such a moving experience that it stands on its own without the need for a second time around - and that seems to be how Team Ico meant for it to be.

Now, if you want to talk about extreme cases of the upgrade treadmill, take a look at MMORPGs. They're absolutely grind-tastic, and yet Blizzard is raking in money hand over fist. So what is it that appeals so much about the minute-but-measurable character progress that exists all over in the genre? I still think the key is new experiences. The players know that there's dungeons out there that they'll be able to run through if they only increase their character's level a bit more, collect that bit of armor they're drooling over, get that new spell that means they can crowd-control more effectively. And the fact that it's a shared experience means it never plays quite the same twice, even if the only difference is the chatter coming across the speakers.

Michael Martin
profile image
Sweet mercy. I practically wrote an entire essay. Sorry if it wound up a TL;DR.

Jon Waholic
profile image
Thanks Michael. You have explained everything ;-)

Castlevania and metroid open world games are structured in a way that whole castle/alien planet is a stage that allows free exploration. Many games breaks things up in sequential chapters and levels instead. I am talking about the latter. Sorry for not stating the differences earlier.

Michael Martin
profile image

I'll admit that in the context of linear games, the comparison doesn't hold up entirely... Let me see if I can clarify a bit with your first example, because I think a lot of the points are still valid.

Whether or not you are a skilled player, you shouldn't be required to replay a significant portion of -any- stage, because again, the goal of entertaining the player is done by presenting new and interesting experiences to them. What you described, with the moving platforms leading up to a double-jumpable ledge, or a very difficult enemy that must be defeated to pass, is what's called 'sequence breaking', and is a staple of speedruns. It also has a significant tendency to unbalance large sections of the game, because you, as the skilled player, now have in your hands an item that you are not supposed to have yet according to the game's designed path. Until you hit the part where you get double-jump, you can often use your new item to breeze through without any real threat - the aforementioned 'Press X to win' scenario.

In other words, if you're going to allow the player to get items out-of-sequence, make sure those items aren't high-powered weapons/attacks, or any other item that reduces the challenge for the player. Feel free to make it something that changes the gameplay however, even significantly; it means that the player will have familiar ground to try out their new ability if they are actually playing the stage over again for the item. This can work especially well if it's a skill that permits rapid travel - the player can now blaze through the rest of the stage thanks to their foreknowledge of its layout, but the first-time out-of-sequence player will have to be more cautious.

On the other hand, if you don't allow sequence breaking and instead require the player to replay a stage, give them an alternate path a short ways into the level. Make that alternate path challenge the player, even with his new abilities, then either drop the player right next to the level exit after giving them their shiny new item reward, or let them teleport out of the stage once they get it.

In short, there's ways to make backtracking worthwhile, as long as there's new content, whether it be due to a change in the player character or a change in the level itself. Artificially lengthening the game via any other means besides that will earn you the ire of gamers.