How LAI adapted Mind Mould for the Asian market

asia_for-ppt

The Asian market is doubtlessly one of the largest, most lucrative and fastest growing markets in the mobile gaming space. Market research firm Niko Partners predicts the Chinese market alone will reach $8.3 billion in 2017. [1] China is also well-known for its complex mobile gaming marketplace.

greater-china_game market

At LAI, our 24 years of focused video game localization experience gives us an advantage over other publishers of knowing how to navigate through the complex Asian market landscape and its cultural expectations, from both a B2B and a B2C perspective.

As a flagship project, we worked with indie game developer Sillywalk on their mobile game “Mind Mould“, performing all necessary localization steps to ensure it would be received well in the global market, and subsequently publishing the game on iOS. Currently, Mind Mould is available in 12 languages, with soft-launch campaigns in Singapore, Taiwan, Canada, Australia and Japan. The following article is a case study on how we adapted Mind Mould for China. The points mentioned are also applicable to the Asia market, in general.

Mind Mould ( 脑力魔矩 in Simplified Chinese) was designed by a European development team. It is a combination of tetris and tangram, a shape-filling puzzle game with a new game play design. It has unique and simple game rules, addictive tapping, rotating and dragging, and an exotic art aesthetic. It is a F2P casual game, available for download from the iOS AppStore.

Below is a list of the areas we focused on as we localized and culturalized Mind Mould for the Chinese market.

Areas LAI focused on - MindMould Games

We now describe each of these areas in detail below, the challenges we faced when localizing Mind Mould for China, and the solutions we developed.

Part 1:  Text Localization

text localization

Text localization for video games starts with translation, but also involves many additional considerations. In general, well-localized text must meet the following requirements:

1.Translations must appear natural to gamers in the target market. This requires that the localizers who do the actual work must be native speakers of the target language, avid gamers, and experienced translators. 

2. The game’s UI poses additional challenges. The text must be displayed in the correct font, and text strings must be the appropriate length so they properly fit in the available screen space. For Mind Mould, we compared several different Chinese fonts recommended by our native-language localization team of experts, to determine which would best integrate into the game. 

If you are localizing a game for Asian markets (China/Japan/Korea), here are some examples of fonts to consider:

Japanese: HGP or HGS or HG創英角ポップ体 (TrueType)

Traditional Chinese: 方正超粗黑繁体 (TrueType)

Simplified Chinese: 迷你简方叠体.ttf or 汉仪菱心体简 (TrueType)

Korean: 서울한강 장체BL (Open Type)

3. There must be functional equivalency to the local market. Due to cultural differences, certain jokes/humor/puns/idioms can’t be literally translated. Transcreation – the concept of rewriting text during translation to make it more culturally suitable – must be used to create new text in the target language that performs the same function of the source but is appropriate for the native market.

A common practice to localize text strings is for the developer to send over the text in an Excel file to the localization team. Experienced developers will also include related graphics and comments to help the linguists better understand the context. If the budget and schedule allow, the linguists will also prefer playing the game to get a better understanding of the UI and game narrative. This is the best-case scenario.

However, when translating UI strings, it is usually the case that the linguists do not have enough information or context to know exactly what the text means.  In such cases, the linguists must make assumptions based on their experience and understanding of the game. This often works fine but it is by no means foolproof. Mistakes will inevitably occur because the linguists will only receive fragmented information or content without any meaningful context, and won’t be able to see the big picture of how the text fits in the game.

There are a few factors to consider:

games

1) Most developers fail to realize that linguists can’t be as knowledgeable about the game as they themselves are. If developers don’t explain the usage of a particular string, linguists won’t know exactly how or where the string will be used. This can lead to inappropriate assumptions by the linguists, and thus the translation will be inappropriate for the context.

languages

2) Most developers don’t have enough knowledge of foreign languages to be aware of the differences in grammar usage.  A shortcut to save development time or code space that is frequently used is to build sentences by programmatically combining phrases or sentence fragments. Although this may work in English, it will often fail and produce awkward or grammatically incorrect sentences in the target language.

In the worst case, developers may seek to save time (and money) by using free onlinemachine translation tools (like Google Translate) to translate the text. This can be a recipe for disaster if the goal is to achieve correct and appropriate translations.

MindMould Set

For example, in Mind Mould, a “set” consists of 4 levels of the game. On the home screen, there are buttons for “Set 1” “Set 2” “Set 3” and “Set 4”. To save time, the developer initially replaced “set” with its Simplified Chinese Google translation “设置” (which means “to set up”). Google Translate has no knowledge of the context in which a particular phrase is used, so not only was the meaning incorrect, the translation was also grammatically incorrect.

The solution to these translation issues is to do linguistic testing after the localized game is built. During this phase, these types of errors can be easily identified and corrected. The testers can also suggest better translations to the localization team if they see fit. In our case, being able to see all translations in context within the gameallowed us to adjust and optimize those areas that we felt could be improved.

Part 2: Game Lore

Sprock

Twip

(Two Mouldian pets)

Our developers really liked the idea of finding something related to Chinese mythology. We brainstormed a few options based on the art style of the game, including Nuwa restoring the world and Kuafu shooting the nine sons of the sun to save the Earth. After a few rounds of discussion, we decided to create a universe called Mouldia (魔迪拉, pronounced “Mo Di La”). We modernized the story of Nuwa, and localized it for Asian markets by adding some cute pets (Mouldia’s citizens) that gamers could collect and wake up. We internationalized the game to minimize the changes between localized versions by keeping the naming general – referring to the central character as “The Goddess” instead of the more culture-specific “Nuwa”.

Here is the narrative we created:

“A long time ago, the old world of Mouldia was peaceful. It was a wonderful place for all creatures to live in harmony. One day a fight between the gods Water and Fire began. It turned Mouldia into a chaotic world where only ruined pieces of steel and iron remained. The savior goddess happened to visit and tried to use unique pieces of stone to save beautiful Mouldia. But the citizens of Mouldia remained asleep, waiting to be awakened by the goddess.”

English Cutscene video:

https://www.youtube.com/watch?v=7Il7FXq75tE

Chinese Cutscene video:

http://v.youku.com/v_show/id_XMTM1NTQ4NTQwNA==.html

The player can restore the old world of Mouldia step by step by solving the puzzles, motivated by pets that will be unlocked along the way.  By establishing some general goals and achievements, players can be expected to return to the game, keeping the retention rate high.

MindMould-Asian logo

Part 3: Visual Art

When we first came to evaluate this game as Mind Mould’s publisher, we really loved the art style of the game. We decided to change as little as possible for localization, so as to keep it authentic and exotic. The areas we eventually decided to change included the following:

dragon skin

As talented as ever, our European artists did awesome work creating the pets and dragon skin background. However, the goddess figure was a higher level cultural concept that was more difficult to grasp when creating the Chinese version. We discovered the culture gap was difficult to close even after sending over dozens of referential pictures and written descriptions to the artist.  In the end, we asked our own art localization specialist to help complete the design.

Below you can see the evolution of the design concept.

Phase I  Conception

European Artist: 

Conception-European

LAI Art Specialist:

Conception-LAI Artist

Phase II  Development (Goddess with Background)

European Artist: 

Development-European

LAI Art Specialist:

Development-LAI

Phase III: Post-Development (Marketing Visual Work)

Western Market:

Marketing - European

Asian Market:

Marketing-Asian

Part 4: Cutscene and Audio Art

In order to explain the storyline and mission of the game with an exciting animation, we decided to create a cutscene (English and Chinese) and play it right after the game starts to load. To minimize the size of the media, our award-winning native Chinese writer narrowed the script down to 5 sentences for voiceover. In the meantime, we synched the music with the script and the video to make sure they mixed well together. The tricky part was to pick the right talent and specific style for the voiceover. Luckily, we had the luxury of having a native speaker on board to be the judge, giving some guidance during the recording. We contacted our studios in Los Angeles and Shanghai, and decided that our Shanghai studio offered the best choice of voice talent because of the higher quality of the talent pool. As I explained to our LA studio producer and PMs (who are non-Mandarin speakers), unlike picking English-speaking talents (where slight regional accents are sometimes acceptable), the Mandarin voiceover talents have to have absolutely no regional accent. Usually talents from Northern China (Beijing or Northeast China) have a better natural performance since Mandarin is based on the Northern Chinese dialect. The audience will expect a clear and accurate standard Mandarin performance in any media product. Otherwise, it will be viewed as a bad quality product or unprofessional.

Though Shanghai is a southern city, it attracts talented young migrants from all over China. Something unrelated but worth pointing out is that Shanghai has a very foreigner-friendly business environment – it is a very international city and very open-minded. LA doesn’t have the same talent landscape even though they may have more experienced directors, best industry practice and cutting-edge equipment.

e

During a Beijing localization conference, we showed a small video clip of our draft cutscene to the audience. It was the almost ready-to-release version: the actress did a great job of narrating; sound mix and video effects worked well. Most importantly, the native Chinese linguists on our team didn’t see anything wrong with this version. And as expected, the audience applauded for the demo. But what we didn’t know until after the event, was one of our audience members wrote us a very encouraging and helpful email pointing out a commonly-mistaken pronunciation for a Chinese character: 疮（chuang1）which was misread as 苍 （cang1） in the term 满目疮痍. It turned out this audience member was a professional audio auditor specializing in pronunciation issues, so we were fortunate that he was in the audience. We were happy that we were able to fix the issue before we published the game.

Offline events

Offline Events:  We hosted and attended events featuring Mind Mould to warm up the market and get user’s feedback. These events included game or localization industry conferences, design or localization classes in universities, and salon events. We are planning other events in the future to meet up with our global gamers and fans.

Part 5: Game Design

Based on the local gamers’ payment habits and affordability, we adjusted the game economy according to answers we received on a questionnaire completed by our Chinese testers.

Social features are a good way to organically grow your community, while keeping up the retention rate. We decided to use the localsocial media (Weibo/WeChat) since Facebook and Twitter are not an option in China. Be aware that Weibo and WeChat APIs are very China-centric (almost all the service content is in Chinese), so they are typically not very friendly to foreign developers (they are also very strict about only giving permission to businesses with official licenses, sometimes excluding non-Chinese companies).

Other features we decided to add are news and a daily Gotcha (spinning wheel). Gamers can easily find them in the main menu. We hope gamers will be encouraged to always come back and enjoy their free gifts.

Anecdote 2: Adding more guidance for gamers

Too difficult to play?

postcard

Friends and family are always our first test bunnies and can be relied on to give us the most honest  feedback. One of our friends really loved playing Mind Mould (she played for many uninterrupted hours according to our analytics). She told us our game was awesome but “inhumanly” challenging to play. Of course, we loved to hear the first part of her feedback but we were confused about the latter part. So we asked her to show us how she played. It turned out that she forgot to use a very important feature of the game — the “mirroring” function, which enables a player to flip a piece by 180 degrees by double clicking.  We could understand how the game would become very difficult and frustrating (especially at higher levels) without knowing about this feature!  We were actually in awe because she still managed to play for so many hours without using one of the main tools of the game. Of course, that would be really hard!

After receiving more feedback (either by watching other people play our game or from the gamers’ testing reports), we noticed that even players who finished the tutorial still did not learn enough to adopt all the rules of the game – even though the rules are quite simple. I once heard a gamer who loves to play casual games comment, “I love to play casual games with a relaxed mindset, so don’t expect me to act like I’m in a classroom to learn all the rules – especially when I’m just getting started and still deciding if your game is worth my while”. Therefore, to enhance the gamer experience, we decided to add more tip windows on the first 4 levels to remind the gamer to use the main game’s functions which they might not have learned from the tutorial.

Part 6: Testing

linguistic qa

Local testing played an important role for us in helping to improve the user experience of the game. We conducted several rounds of functional, linguistic, UI and UX focused testing in China.

Our internal testing team (made up of native speakers) did most of the functional, linguistic, and UI testing. We wanted to share the test results with our European development team, so to get a large number of testers who were also able to rewrite their report, questionnaire and UI feedback in English, we worked with a local university and found students and teachers from their English college to conduct this part of testing. A gamer experience test was conducted later on with the help from an outsourcing agency based in Beijing.

After we finished internal and student testing, we worked with a professional testing company located in Beijing in order to get real local gamers to do an A/B testing. We collected enough data and feedback to run analytics and fine-tune the levels and monetization specifics.

Part 7:  Soft-launching

audience

Prior to soft-launching, we engaged a professional ASO team who is part of our network to fine-tune our keywords in the target languages for the iOS AppStore.

Before the huge release in China, we picked Taiwan and Singapore as our soft-launch markets because of the smaller market size and similar culture and language preferences (Simplified Chinese is used in Singapore, Traditional Chinese in Taiwan).

We ran online marketing campaigns via ad networks, our own social media and video channels as well as reaching out to youtubers/streamers.

Youtuber’s review:  https://www.youtube.com/watch?v=1fywGrYlI9o&t=4s

We produced an iTunes podcast interviewing the developers and blogged the transcript. So far it is one of the most popular episodes of our LocaLAIse This! series.

[1] http://venturebeat.com/2016/09/06/niko-chinas-mobile-game-market-will-hit-8-3-billion-in-2017/