Features of games that had to be abandoned - developer analysis

At the end of 2017, we told stories about sunk costs in game development.

More than a dozen developers talked about the price they had to pay for ignoring the “red flags” and how they continued to work on bad functions or whole projects that were doomed to failure, simply because they invested too much time in them. money or energy.

Shortly after the release of the article, Axiom Verge developer Tom Khepp suggested that we look at the problem from a different angle: what about good games or features that were abandoned in vain?
')
This error, like the non-refundable error, is very simple to make. Game development requires juggling with a variety of incompatible but interdependent elements, so it is difficult to track which of them have lost their meaning and which should become priorities. Many of the respondents noticed that it is much more difficult to assess the need to abandon some "features" of games.

Many elements are cut off much earlier than the end of the game, when the game itself and its systems only take on their form, and this cut-off is rarely an independent event that does not affect, both positively and negatively, other parts of the project. If we continue the analogy with juggling, then if you allow one ball to fall, then the rest becomes much easier to control. On the other hand, when he falls, he can knock other balls off the path.

Veteran developer Noah Falshtein remarked: “The problem with repentance when you give up functions is that often the best parts of the game turn out to be unexpected for developers, or at least they have an unexpected impact. Much of what a designer can consider fantastic is simply adequate. Game design still remains both art and science, so this success is quite difficult to predict. ”

Nevertheless, as we shall see, much can be learned from thought experiments. Even the fact that, for example, the complexity of predicting the importance of the decisions made in design and the functions removed can affect the consciousness of the developer.

Says the leading designer of quests CD Projekt RED Mateus Tomashkevich: “We cut out one quest from The Witcher 3 . Players had to infiltrate the Wild Hunt in order to eliminate one of its generals, Quarantine. In the process of completing the quest, players could learn more about the motives behind the actions of the Wild Hunt and talk more with its representatives. At first glance, it looks cool and advantageous for the game, and sometimes I was tormented by doubts - was it worth it at all to get rid of this quest? But he created a lot of design problems. ”


In the case of this quest, developers had to think about the player not to kill everyone on the spot. Also had to take into account the need to create new motion animation in heavy armor. In addition, there were a lot of other problems that could not be eliminated after several iterations of the design of the quest.

Radical change of direction

Tom Happ told about one more specific example of a slaughtered game, which, perhaps, was not worth canceling. He worked as an engineer for Petroglyph Games during the development of the canceled MOBA End of Nations , which, under the direction of the publisher Trion Worlds, went through a long development cycle through several design changes, after which it finally acquired the MMORTS format. Hepp says that the game finally took shape at the end of 2012, and that it was the best RTS of those in which he had to play.

“I could not wait for her release to play with other people around the world. But when we finished all our work, and a week or two remained, we all wondered: “She should have been released on day X, why does Trion slow down?” We moved to a new office, expecting to switch to EON support with parallel work over other games. On the day of the move, we discovered that the game was canceled. ”

Worse, Petroglyph had to go on massive layoffs to cope with financial failure as a result of this unexpected change. Meanwhile, Trion transferred the development to its own departments and turned End of Nations into MOBA, cutting out almost all the RTS systems, over which developers had been working on over the years, only to cancel this version for the next year.


The suddenness of these events and the lack of warnings from Trion about dissatisfaction with the appearance of the game had an effect on Hepp’s psychological state. “I still feel trauma (recently there was a five-year anniversary of these events). We often hear about injuries when talking about violence with the use of weapons or rape, but I believe that such dismissals should also be considered injuries, and in fact they did not just fire me. It seemed to me that the whole world had changed in one evening. ”

Question Producer / Designer Michael Kelly drew lessons from an incorrectly chosen development direction change, which he witnessed while working on AAA projects at 2K Games.

“Changing direction in the development of a video game is like the turn of an ocean liner. Having a huge team and / or several studios, decisions cannot be made instantly. Having accumulated experience, I decided that it was necessary to determine the scale of the game in the early stages and revise them as much as possible, down to the minimum core of the game, if necessary. Functions that will be included in the finished product, you need enough time to polish; you should not instead create a bunch of optional appendages, dissipating the efforts of the team. "

As an example, he cites The Bureau: XCOM Declassified . Its development was carried out in different places by different teams with different visions, and each of them left its mark on the design and code. When the new team began to refactor the old work for the new direction of development, it did not have the resources to start over. "We had workflows for graphics and code that could not be removed, despite the fact that the features for which they were needed were no longer in the game."

When you make a serious turn in the direction of game development, Kelly explains, you not only create inefficient workflows, but also harm the inertia of movement — many functions need to be redone, and some tools that are suitable for the old job do not fit the new one. “And you have less money and time to complete. Or to justify the investment, you need to sell 30% more copies than in the original project. If you had released the game in the original design, you would at the same time have the opportunity to extend the revenues to the desired level. ” (As a career advice, Kelly adds that it’s better to have one more game released in the resume than a space in a year or two without a game, and it’s better to have a decade’s experience with several games than zero or one, as was the case with the unfortunate working over Duke Nukem Forever during the entire development period.)

In many projects, it does not even matter whether the new direction can be better than the old one. Sometimes the turns can be successful, and the finished games are distinguished by a bright personality and a holistic design, but much more often the developers find that decisions once made come back to interfere with them: “I think that many developers who have spent several years changing direction, ask themselves: “What would happen if we just got rid of non-working features, doubled the investment in the basic game cycle and released the game, and did not spend another year (and a lot of effort) trying to transform it into a new design, success which is also not guaranteed by anything? ""

External factors

There are many reasons for cutting features out of the game during development. Some of them are related to quality (or lack thereof) or compatibility of features with the project, but many factors leading to cuts are actually external. For example, Starcraft: Ghost was canceled for the most part because for several years Blizzard was spending all its resources on World of Warcraft .

The Medal of Honor: Airborne development team spent more than a year due to the EA requirement to move from Quake III to Renderware 4.0 (only to switch to Unreal 3 later after 16 months of torment with raw Renderware 4.0). Dead Space 3 also had problems caused by changes in the EA requirements: most of the depth and complexity of the narrative, for which the game was often criticized, was cut out in favor of microtransactions and marketing tasks; at the same time, the promising remake of Tomb Raider 10th Anniversary , developed by Core Design, was canceled despite an impressive demo , because the publisher Eidos / SCi changed his mind and passed on the remake of Crystal Dynamics, which the Tomb Raider Legend had just completed.

Unravel creative director Martin Sakhlin talks about one case where right before the launch of the game at the request of the marketing department, the management method he was working on was completely cut out. “Perhaps marketing was right,” he says. "I do not know, but it seems to me that the whole game was simply gutted."

Much less often, quality features or whole games fall victim to pessimistic business strategies. An experienced artist and illustrator, Mark Ferrari once told me about a canceled X-Men game that he worked on in the 1990s. The team had a tiny budget and a schedule, but Ferrari and colleagues managed to figure out how to create a good game using tricks with a palette. For example, one graphic file could display an alley with trees and clouds in one palette, and a city with bridges and a highway in another palette.

“We sent the game to publishers, expecting an enthusiastic call and congratulations on a terrific product for such little money and little time,” recalls Ferrari. “But the only thing we received was a very angry call to the head of our department from the head of the project in that big company with the words:“ We didn’t ask for this from you, and you very much let us down ”.”

The client company was angry, because if it had released this really good 8-bit game for a TV controller, it would have made the rest of the line of games already sold in stores too weak against its background and unsuitable for the market. The game was too good for release.

Hepp recalls a less serious example of a feature cut from a game for economic and marketing reasons. "I was the technical artist of the Tiger Woods PGA Tour 2008 and 2009 games." To replace helmet-like polygonal hairstyles, in the 2009 version one of his colleagues tried to insert a dynamic hair system from the Battlefield franchise.

“He managed to create two or three hairstyles that worked with this hair system, and they looked much better. But the management refused to use them, unless all the hair variants were replaced with this new system. ”


Hair from Tiger Woods PGA Tour 2009 in action

The development team did not have the resources to implement such a transition in one year, so they returned to their existing hairstyles. In the last game of the series, Rory McIlroy PGA Tour 15 , the hair in the form of a polygonal helmet is still used, most likely for the same reason.

Contractual obligations

The entire career path co-founder and CEO of Fun Bits Chris Millar is littered with wonderful multiplayer modes, which had to be abandoned. After leaving Blizzard in 2002, Millar walked from game to game with amazing multiplayer modes, which almost without exception were cut in favor of mediocre single-player gameplay.

The first game was Goblin Commander , an early attempt to implement a real-time strategy on consoles, on which Chris worked at Jaleco. “We had a [online] multiplayer running on PC and Xbox. At that time, platform manufacturers adhered to the rule: "Everyone should have everything the same, or everyone should have everything unique." But if we had released the multiplayer mode on the Xbox, then the PlayStation and Gamecube would not let us pass the certification, because we had a feature that did not fit their platform. ”

To be released on all three platforms, the developers had no choice: they had to cut out the multiplayer. Goblin Commander was released on all platforms with only split-screen modes for one and two players, which disappointed critics and, according to Millar, prevented the game from gaining its worthy success.

Then he moved to Lionhead, where he first helped complete the development of Fable , and then in a team of 80 people began work on a hybrid of the God simulator / RTS Black and White 2 . It was an ambitious game, and Millar appeared just in time. He was fascinated by the city's construction modes and multiplayer. He describes the first as a “wildly cool” relaxed gameplay in the Farmville style, the second - as the most interesting game mode in the “monster against the city” and “god against god” formats. However, the realities of contractual obligations, budget constraints and time constraints forced the management to take a hard decision.


Faced with the possibility of canceling the game in case of a deadline failure, the developers threw out the free city planning mode and cut out the multiplayer, which, despite its interestingness and uniqueness, could not remain the only game mode.

Having clarified the lesson, Millar drew attention to PSN and decided that since all the buyers of games on PSN already have a connection to the network (otherwise they would not be able to download the game), he can first of all go in for multiplayer. This led to the creation of the critically acclaimed cult hit Fat Princess , an action-RTS of 32 players, which at the last minute, at the request of Sony, turned out to be a limited single-player mode.

However, Millar had to meet once again with his curse to cancel the multiplayer mode in the PS4 sequel. His prototype was originally developed only for multiplayer, but then took the form of a cooperative campaign with a planned multiplayer mode in the form of an add-on (which was already ready for release), because Sony chose a different direction. After the release, the game failed miserably, because fans and critics did not appreciate this new direction. Millar is not angry at Sony and calls the company a “wonderful partner,” but the decision to shift the focus from multiplayer to campaign mode has borne fruit - according to Millar, “the company refused to negotiate a new contract. And we had to make reductions. ”

For Millar, who is again trying to create a multiplayer mode in Squids from Space , it became clear from this experience that it is crucial to focus on the most interesting aspect for the audience even before experimenting with other elements and game modes. For him and for the companies in which he worked, it was an online multiplayer. If you implement the basic functions correctly, it gives a chance to reach the target audience. After this, Fun Bits wants to use this basic functionality in other directions, but only after the company has a solid foundation for the game, interesting and attractive for the gaming community.

No cuts without damage

Although it is difficult to predict in advance which functions create or destroy the game, and which could take the game design to a new level, it is important to have the courage and refuse what you, as a team, are not sure about, even if it will disappoint or frighten someone part of the audience of the game.

Bruce Shelley as an example of this leads a major meeting of the company during the development of Age of Empires III . It was devoted to a complex system of formations. “The voices were divided as follows: about 51% for preservation, 49% for abandoning them, and we had to make a choice. To make a decision, the majority was not enough, we needed at least 80%, not 50 or 51. Despite the fact that half the team liked this feature, we had to give it up. ”

It is also important to remember that no reduction takes place without damage and additional work. “Refusing functions, one has to make many changes in other places. That is, yes, of course, you can win a little production time required to complete. But we will have to allocate time to adapt the rest of the content for changes, ”says Tomashkevich.

Martin Sakhlin emphasizes that when you give up on features in the development process, they are not completely forgotten. “It's quite interesting to look back and evaluate the cut functions, imagining what they could be. Sometimes we find something useful for the future. ”