
Dive into a factory-building game with the focus on just that – building huge space factories! Design sprawling multi-level factories and satisfying production lines without limits. Tackle progressively challenging automation goals at your own pace, without the threat of enemies.
Modeled estimate, not a reported figure. How we estimate →
Media

About
shapez 2 is a 3D factory-building game where you automate the extraction, production and delivery of geometric shapes. Each shape is its own set of building blocks that your factories can take apart, stack, paint and put back together in near-infinite different ways.


Deliver specific shapes to unlock new technologies, expanding your factory’s capabilities and scale to make even more unique shapes. Grow a single conveyor belt into a massive multi-station space factory across three interconnected layers. Keep it efficient and optimise every platform, or expand into the infinity of space and connect a million buildings by space train – you decide how you engineer your factory.

shapez 2 is the purest factory simulation game imaginable: all buildings are free, resources never run out and there are no enemies or time limits. Effortlessly delete, redesign and rebuild your spaghetti belts as much as you want, without penalties. You’re in complete control of this sandbox, so you can focus on how to craft the next shape.


Whether you’re new to the genre, a casual or an automation veteran, shapez 2 is for everyone. Pick between three difficulty levels with completely unique logistic puzzles each or take on the experimental Hexagonal mode. There are many reasons to play shapez 2 again and again.


If you’re looking for a list of what shapez 2 has to offer, these are the features you should know about:
Multi-Layer 3D Factories: Design and optimize your factories across three building layers for both buildings and platforms. Connect everything by space belt or train!
Multiple Difficulty Levels: One main game mode with multiple difficulty levels for both new and veteran players, as well as the experimental Hexagonal mode if you’re up for the challenge
Challenging Shape Mechanics: Discover shapes types with unique properties and logic
Open Production Buildings: Build stunning space factories with animated open buildings, so you can always see exactly what’s going on or going wrong
Research System: Unlock new buildings, mechanics & upgrades that open up new ways to build your factory
Blueprint Library: Save, load, export and share factory design blueprints with anyone
Easy controls: Effortlessly cut, copy & paste parts of your factory and undo & redo up to 50 operations with intuitive keyboard controls
Genres
Stats
All-time low is the lowest price we've recorded for this game since we started tracking it.
Languages: English, French, German, Spanish - Spain, Korean, Turkish, Polish, Russian, Thai, Japanese, Portuguese - Brazil, Simplified Chinese +1 more
Updates
Build ids and update times come straight from Steam's PICS — the same source SteamDB reads. Patch history grows as we record new builds.
Engagement
Hours played at review time, across 611 reviews with recorded playtime.
Reviews
Themes across 609 recent reviews and how positive each mention is. A keyword signal, not full sentiment analysis.
positive critical· bar length = how often the theme comes up
Reviews
While a huge fan of factory building games this one seems to undercut itself with the endless supply of everything. There is no real reason for optimization, and often you scrap things for the next design. While the game has unlocks they are quite limited and often, due the fact everything is endless, feel underwhelming. It is no bad game at all, but there are better choices out there.
Where did all the time go? The purest factory builder, no abstractions of "copper" and "bolts", just geometric shapes and colors. My only complaints are that even with the Supporter Edition soundtracks and Shapez 1 soundtrack the music repeats a little much, but it's a great soundtrack. Technically sound as long as you don't stray from the default DX11 renderer. My only real technical issue is having problems zooming to a platform lying under another platform.
I have chosen to be a bit obtuse with "negative review" here as I'd describe my thoughts as: "I liked my time with this, but there are definitely issues that have now irked me enough that I'll let this one rest for a good while". If I talked about something in that tone to a friend, then I don't think that's a positive review at least. Insert comment about wanting a middling option for a review here. Shapez 2 is an automation game distilled. You don't have xenos gnawing at your concrete plant, and you don't have to concern yourself with annoying terrain getting in the way of your concrete plant blueprint. Hell that concrete plant doesn't exist, because all you produce is abstract shapes that have no use in building, they are just made to be thrown into the void for rewards. Every building is free, and you build in an endless expanse populated only by scarce resource nodes (that you can pave over in case you don't feel like you need another source of quarter-circles or green) I don't dismiss this approach on the face of it, as seems to be the main criticism. Again, it's factory-building distilled. If a belt isn't fully saturated in Factorio I can just dismiss that as some butterfly-effect from something not caused by me, Venus is in retrograde and oil pump 5623 has dropped enough in efficiency so that this beselheim-plate line can only sustain at 99%. But in Shapez if a belt has gaps, I know that a design I've made is wonky and needs a seeing-over (Still haven't figured out why my splitter-utilizing designs end up with an ever so occasional stutter, even on the fifth revision). It's interesting to be in a world of theoretical perfection where everything does exactly like I asks. But those people complaining weren't entirely wrong. When everything does as I ask, I have as many resources as my computer can handle, and nothing is allowed to happen without me saying so, we've just reduced ourselves back to programming (perhaps even better than programming since I'm more sure about my factories doing as I tell them than the compiler, splitter problems notwithstanding). Much like programming, you pretty quickly come up with functions, things that come up over and over again, and just call upon them when needed. In my blueprint library that got every shape up to at least level 50 I have about 40 blueprints/functions. Out of those I'd say that half were actually useful, and the other half could have been recreated when needed, but stuck around for neatness sake. Save for setting up the starting platform, I would not need to go to the factory scale to beat normal-hard again. To put it into programming context, save for setting up the main function, I wouldn't need to write anything other than function calls to complete a project. I had fun creating some of those blueprints. Figuring out something small that could swap two quadrants of a shape at (almost) 100% throughput was great (Make anything machine wasn't, but I have one), but now I've built it, and if I had built it 1% better I would never need to look at it again. I'd just paste ten down and move on to the next part of the shape being made. Could it be even smaller? Maybe. Do I care for such a tiny boost in available space? No since space is infinite and my budget for platforms (the only limiting resource) has never swung at the scale that saving ten would even move the needle. I've already done the fun part of a playthrough, now if I boot another one up, it's just routine execution. I've yet to feel that blueprints have spoiled the fun in any other automation game, since none other I've tried has been 100% blueprint-able Another interesting problem that the lack of resource scarcity causes is that logistics are really unimportant. Until you can chuck entire trains into the void, anything entering it can only do so at one speed, and it can only take twelve tripple-layered space-belts. If you aren't building your shapes in the Andromeda galaxy, having twelve belts go all the way to the voiding hole won't bankrupt you. The ability to chuck trains into the void comes quite late, so for a good part of a regular playthrough you are hard-capped in how many shapes you can consume. I won't drag a belt across a continent in Factorio for several reasons, but I constantly do it in Shapes. People have also complained about the ephemeral nature of your factory. Unless you are producing the static and always useful "operator shapes" (and even then two of those randomize), then you are at most producing a shape for half an hour until it's completely useless. I don't have much of a problem with this, but then I can paste a new factory to do the next shape in a minute or two. I suppose it's either frustration or boredom. I had my money's worth, but I think there are better automation games, and most IDEs are free to download if you like the programming-like aspect of the control that Shapez 2 offers. It sat right with me for a while, and then I started to think about my time with it and frustrations kept mounting. Maybe you don't have the same brain-worms I have.
对用笨办法来设计自动化的人才(我)来说,就是个挂机游戏嘛。虽然效率低得令人发指,但只要我只要摸鱼时挂得够久,什么条件不能满足啊(哭)。看了看世界成就,这个游戏好评那么多,完成成就的人却那么少,难道大家都不挂机嘛(沉思)
I decided to try this game, curious to see what this was about, with some reservations because I've never played anything like it... and it's magic. My biggest concern was that it was going to be unintuitive, confusing and janky. The trailer is accurate: easy to get into, a challenge to master. The controls become intuitive very quickly. Understanding the geometry of how to put things together is a joy and a welcome challenge. A truly fantastic find in a genre I never anticipated I would enjoy this much. Massive kudos to the devs and a massive thank you from me personally for expanding my idea of what fun looks like. Keep building this awesome game!
没点工程师思维还真搞不定这个游戏, 每当你觉得这就是最佳版本的时候, 总能在一个小时后想到更优化版本,玩起来贼带劲 很像编程
很有意思的自动化游戏,和一代相比升级成了3D工厂,也有新的设备,玩法增加不少。28小时通关主线,目前还在折腾万能机中。 总体还是和一代一样,通关前和通关后是两个游戏。主线任务比较简单,相当于是各种形状和设备的教程。由于加入了支线任务,本作的游玩流程更多是建了再拆、拆了再建,解锁新设备搭出产线→交付形状达到目标数量→把有用的部分保存到蓝图→拆掉整个产线→开始造下一个形状,基本都是这样的循环。 相比于轻松的主线任务,通关后做万能机就头疼了,因为有新的形状和颜色,比一代的万能机复杂很多。不过那些本来就是通关后的额外玩法,不用强求,光是通关前的主线内容就很不错了。 遗憾的是,随着游戏更新,部分教程没有及时调整,导致有一些错漏。比如,游戏内教程说晶体形状只能通过交换器操作,不能用切割机,但实际上在不切到晶体的情况下,用切割机和半破机都是可以的。再比如游戏内教程没有说白色+白色可以合成黑色,以及白色+黑色可以合成灰色(未上色),需要自行摸索或者刷社区才能知道。
La boîte de Pandore a été ouverte. Shapez 2 est probablement l'opus parfait pour s'initier au genre des jeux d'automatisation. Un peu moins si vous êtes du genre terre-à-terre. Heureusement pour moi, j'ai un certain pourcentage d'autisme de la perfection des rendements et de la logistique en intraveineuse dans le sang, donc ça ne m'a absolument pas dérangé ! Ici, pas de fer ou de charbon à extraire pour alimenter une machine douteuse qui produira autre chose. Ici, il n'y a que des formes et des couleurs. Et c'est tout. Vous pouvez découper les formes en quarts, les faire tourner, échanger leurs positions est-ouest, les peindre et les assembler. Voilà. Grosso modo, c'est ça le jeu. En boucle. Le but est d'alimenter les formes demandées par le vortex (ou les postes d'échange), et vous disposerez pour cela de nombreux outils. C'est là que le jeu est magnifique. Tout est agnostique. Les méthodes et techniques d'usinage que vous apprendrez ici vous serviront dans des jeux d'automatisation bien plus complexes et concrets. Calculer les ratios de machines. Comprendre les méthodes d'acheminement. Découvrir les manifolds. Gérer les fluides. Transporter efficacement des ressources d'un point A à un point B situé à l'autre bout de la carte. Et bien d'autres choses encore. C'est le tutoriel ultime. De la théorie pure. Il n'y a pas de pression, pas de limite de temps. On peut copier-coller des ensembles entiers. L'ergonomie est tellement bien pensée qu'on a parfois l'impression d'utiliser un logiciel professionnel. Effacer des pans entiers de son usine n'est jamais un frein. Tout est fait pour encourager l'expérimentation. Tester une idée ne coûte rien, à part un peu de temps... et au pire un Ctrl+Z. Le jeu vous donne des objectifs, mais soyons honnêtes : le véritable objectif est celui que vous vous fixez vous-même. Après avoir passé beaucoup de temps sur le subreddit de Shapez, je peux vous assurer que les joueurs ne manquent pas d'imagination : - Vous voulez une machine capable de tout faire automatiquement ? Apprenez à construire une Make Anything Machine (MAM) avec des portes logiques, et vous n'aurez presque plus besoin de jouer. - Vous êtes un perfectionniste de l'optimisation ? Concevez des plateformes les plus compactes possibles et chassez le moindre trou d'air. Une véritable œuvre d'art. - Vous voulez obtenir un flux parfaitement équilibré sur toute une chaîne de production ? Découvrez de vraies techniques industrielles comme les buffers, les vannes de décompression et la gestion des flux. - Vous voulez transformer votre usine en Amazon ? (Mon petit kink sur ce jeu.) Construisez d'immenses réseaux ferroviaires, créez des lignes de métro avec correspondances et approvisionnez vos usines de la manière la plus élégante possible. Il y a énormément de choses à faire, énormément de possibilités et de façons d'accomplir une même tâche. Le jeu parvient à être incroyablement créatif avec si peu de mécaniques. C'est presque hypnotisant de dézoomer et d'observer toute la chaîne de production fonctionner toute seule Je suis développeur dans la vraie vie. L'IA facilite énormément mon travail aujourd'hui, parfois au détriment du plaisir de coder qui m'a donné envie d'exercer ce métier. Réfléchir à des machines et organiser un réseau logistique comme on organiserait une base de code, ça fait un bien fou. Ça fait du bien de se casser à nouveau la tête sur des problèmes de mathématiques, de flux et de logistique. Le jeu m'a surtout donné envie d'essayer d'autres titres du genre... mais en version concrète cette fois-ci. Comme Factorio ou Satisfactory avec les copains. J'ai hâte de passer de la théorie à la pratique. Jouez à Shapez 2. C'est un incontournable du genre. Et maintenant, je comprends pourquoi.
Extremely satisfying game, easy to learn and hard to master. Deceptively simple looking but very challenging on the harder difficulty. Game mechanics flows really smoothly and adds to the satisfying game loop. Buy at your own risk, you may very well getting addicted like me.
This game feels like cooking up tiny colored pills, swallowing them, and drifting through a cloudy dream in search of life’s true meaning.
Related
Related
Trends
Modeled estimate over time, reconstructed from Steam's monthly review histogram.
Engagement
DAU/MAU are modeled from peak concurrent players (≈×8 and ×28) — Steam publishes no active-user counts, so treat these as rough estimates, not reported figures.
Audience
Audience
Games most often reviewed by the same players — an audience-affinity signal from our review sample, not full ownership data.
Estimated from the language of this game's reviews — a proxy for where players are, not official Steam demographics.
Related
| Cruelty Squad | $15M | 22.7K | 96% | $19.99 |
| Shovel Knight: Treasure Trove | $15M | 16.9K | 96% | $39.99 |
| LEGO® Marvel™ Super Heroes | $15M | 30.4K | 96% | $19.99 |
| Warhammer 40,000: Darktide | $15M | 106.7K | 72% | $9.99 |
| Squad 44 | $15.1M | 25.4K | 75% | $19.99 |
| No, I'm not a Human | $14.9M | 29.7K | 93% | $14.99 |