RAWG Discover

Games discovery powered by machine learning 🤖

RAWG brings a discovery-rich list similar to popular content discovery platforms, that is powered by the world's largest games database. 🎮 Browse an endless flow of games. 👀 Like what you see? The neural network will find more games for you.
🤖
Reviews
  • Антон Орлов
    Антон ОрловDeveloper, orels1.tips
    Pros: 

    A great spin on the default genre-based recommendation system. Based on how much we judge games by the visuals - it works quite well

    Cons: 

    The NN RAWG built still needs training, you can get some weird results atm

    This new feature is a great addition to an already extensive featureset of RAWG.io. I've been using it for a while, and this one is for sure increasing the appeal of the service over all and gives me a way to catch up with new games in a timely manner 👌

    Антон Орлов has used this product for one year.
  • Pros: 

    Finally, an actual one place to recommend game across various platform that doesn't sucks.

    Cons: 

    Still missing a lot of game platform/console such as Symbian, JavaOS, MSX, FM Towns, PC-98.

    RAWG Discover is easy to navigate, get the job done in simple way. They also recommend you game from itch.io when usually other game DB won't have them.

    Christian Elbrianno Yoga has used this product for one year.
Discussion
Hello again, product hunters! It's been over a year since we launched our beta for the first time ob PH, generating around 800 of upvotes. And this Friday is the second birthday of RAWG! So this is a great opportunity to share our progress with this wonderful community. We are out of beta today, and we are proud to announce ML-powered recommendations on RAWG. We launched a neural network to help you find a new favorite game in a few seconds. RAWG is turning into a discovery service similar to popular content discovery platforms, that is powered by the world's largest games database, and lets you browse an endless flow of games. If you like what you see, our neural network will find more games you will enjoy in a single click. Machine learning recommendations work differently than those based on metadata or collaborative filtering. If you liked a game like Super Mario Galaxy, you'd probably be looking for other cheerful colorful games and not just 3D platformers. If you enjoyed a sports game like EA Sports NHL, chances are you are going to like more sports games or management simulators. If you know what you are looking for, RAWG will help you find your next favorite game in less than 20 seconds. I hope that you'll enjoy using RAWG as much as we enjoyed making it. ♥
Hey guys, this is super cool and I've been waiting for such a tool for a long time! Though I only play on NS and mobile, watching games on other platforms really makes me seriously considering buying a console ;) From my own experience, one thing that game industry isn't doing well, especially on console, is that it's hard for players to try before paying. When I'm about to buying a game, I usually go to youtube watching some reviews to make sure the game complies with my expectation. In that case, if you guys can add some related reviews (either text/video) below the game, it would be fantastic! But admittedly this might be just my own thought, not for all users.
@shiyuan_niu The cool thing is we already do that! We don't aggregate reviews from around the web but rawgers can share reviews and comments on each game page. We also aggregate YouTube videos and Twitch streams for popular games and you can browse and watch them right from the RAWG page. Check GTA5 for reference https://rawg.io/games/grand-thef...
@sergey_ulankin Coooool~ Usually I view the list I just walk through and if any picture interest me i'll click and tap into it. Didn't notice this before, but super awesome! Now I have the new shopping list :)
I have long had problems with Steam's UX and their recommender experience. This is great! How did you guys solve the cold start problem initially? How did Steam agree to sharing user data with 3rd party platform like you guys?
@shengyu_chen Hi! Steam actually has an open API (https://steamcommunity.com/dev), so we use it. Everything happens with user consent. In order for us to receive any data from Steam the user first must make it open and discoverable. It's a little bit confusing and buried in the settings. To make the process easier we made short tutorial videos for our community: https://www.youtube.com/channel/... After the import is finished players can choose to switch their privacy settings back, although that'll prevent future syncs, which we also do to add new user games automatically. Hope that helps!
@accujazz Thanks for the fast reply on this. Seems that the youtube channel linked isn't working. When I said the cold start problem, I meant how did you guys start originally for the recommender originally. What did the v1 version of this product look like? I have been thinking about building something similar for some recommender for niche TV genres but what got me stuck is determining the v1 version. I want to know more about your guys' journey and how you guys get here. That'd be pretty interesting.
@shengyu_chen May be, checking out our Medium profile, especially old posts can help: https://medium.com/rawg/archive
@accujazz I will check them out for sure but the account connection with my PSN and Steam account doesn't seem very smooth. Think I waited for 5 mins and the account connection still says "importing"
@shengyu_chen Sorry for that, there was a lot of syncs simultaneously.