Desynced

There are little things and big things that keep me from liking this game. Overall it feels like the idea came first and then they decided to bolt a game onto it without a good understanding of what makes a game feel good. The quality of life features that one would expect are just not there, and greatly hamper the experience. I love the idea of combining combat and RTS elements with an automation game, but for one, the bot controls are just not great. The move command is right click, but the game constantly drops that because right click is also drag to move the camera. Rebinding it to another key makes it very clear that the dropped input is the game misinterpreting your action, but even with a rebound action it's a delay every time you use it and a delay before your bots respond. If it were an intentional design choice that you had to play safely to work around a signal delay (since it's a remote bot), it would be one thing, but it is entirely poor game responsiveness over action and delay. I can even see how it could be a great tool for pushing players to interact with the behavior system since the bots could respond themselves, but there are woes there too. I have worked in game development for 15 years at this point, and was part of a project that was building a toolchain with visual scripting up from scratch. The idea of using visual scripting in an automation game is a really exciting one, but it's just very, very clunky. The system dramatically lacks quality of life, and the features it does have are just the last thing a player would expect. Let me give you an example: the Divide node presents you with "From" input, "Num" input, and "Result" output. Is it From / Num or Num divided From the From register? The help text says "Divides the value in Num from the value in From and stores the floored integer result". So it's From / Num, but it doesn't give you a decimal/floating point/fractional number, and integer. So the most obvious uses of division, finding the current percentage of max (health, battery, etc), requires an extra step of multiplying by 100 beforehand so you can check 0-100. Oh, and each register has a "Data" and "Number", so a Value you can store is like "Iron Ingot : 25", so the Divide node should actually read "Divides the Number of From by the Number in num, and writes that into the Number of "Result"". Want to set a Value? Gotta Set Data and Set Number as separate operations (unless you discover the unhelpfully differently named "Combine Register" node, which ONLY lets you set both rather than have optional inputs. Want to get clever and create a building Behavior that sets the product of each of its Production components? Well, you can use the helpful "Loop Producers", but you can't easily set the product. You have to find Set to Component, which sets one of the component registers based on the slot, but you don't know which register a looped Producer result is in to set it (it sets Number to their production time??). Oh, and both 0 and 1 are register 1, because why not. So what do you do? Output it to a parameter, exit the script, and drag the result to all your production components. You know, intuive-like. You can work with the "store" register and get a lot of functionality out of that, but the primary defining feature is just frustrating to use. It feels like the game is fighting me at every turn, but also not challenging me. You get turrets and walls very early, but enemies don't actually attack your base. They don't patrol, they just hang out near spawn locations. So the "turrets" are actually just weapons you can equip to your bots. You eventually do have to explore to progress, but again, you can just whittle away the enemy and their spawns until they're dead, and they won't repopulate, so that area's safe forever. So not much to engage with there. Speaking of progressing, research is your tech progression, and there are no ways to scale up the speed at which it works until you get overclocking like 15 hours in. Having multiple research buildings doesn't do anything. Your production unlocks are nice, and do serve to move things forward, but really the biggest upgrade was from transporter modules, which let a building move items from nearby buildings. That's right, your bots have to move everything until then, and even after you get that and start building massive blocks with item chains, realize you can tell your buildings to only move things by transporters (the bots will try to help, occupy themselves, and slow things down otherwise), and get things up and running, it's never really for anything. Like I said, you're hard-capped on research rate, so scaling up just fill your surplus faster. Some of the unlocks are great, like new bots, buildings, and the like, but a lot of them are like "ok now you have virus turrets and virus cubes to make", and only give you another bot-automated build to construct. The production chains aren't puzzles, they're just "build this, and the bots will sort it out" or "put a storage here that maybe you can set a bot to fill (or not)" and that's it. Again, I think there are interesting things here if the scripting becomes friendly and straightforward, or the factory expanding had meaningful payoff, or the combat was engaging at all.

Если очень кратко - лучше бы тут были конвейеры. Боты не могу пролетать через друг-друга - ок, но они при этом не хотят облетать друг-друга, стоят пока кто-то не отойдет, если есть курьеры - быстрый бот будет биться ему в спину, но не облетит. Заводы строить можно, но логику тут фиг настроишь, нет приоритетов распределения ресурсов, у меня при большой фабрике самые начальные печи стояли без руды, потому что ее туда просто не доставляли по приколу, прям рядом стоит склад забитый рудой, но ни 1 дрон не может наполнить печи.... А еще тут нельзя забивать мехи с запасом, т.е эти самые запасы боты просто унесут на склад с выходной продукцией, зачем, почему.. Я не отрицаю, что я это не понял, но в обучении об этом не говорится, улучшений я не нашел нужных. При расставление стобов(копированием вместе с основанием) нет зоны распространения, приходится ставить на глаз, а потом переставлять на сколько промахнулся, ну или терять просто так некоторое распространения из-за наслаивания.

Честно, оно своих денег не стоит, и нечего это сравнивать с Факторкой — у них нет ничего общего, не считая стоимости.

我就喜欢这种能把“自动化”做得有脑子的游戏。不是摆几个生产线就完事

Gonna say the same thing so many others have. The problem with the game is the tedious and painful "visual programing" The idea is interesting. The actual gameplay effect is ♥♥♥♥♥♥♥ ♥♥♥♥. Want to make a bot carry more than a single item? Okay you need to craft an internal component (hope you have the mats or you cant even program) Now your down a slot in your bot, oh well right? Now you have to go through the menu and create a program.. "oh its simple just drag and drop." Yeah IF you know what each of the over 100 little programs ACTAULLY DOES. No explanations, no visual queues. Just ♥♥♥♥♥♥♥ *wiki work*. At 50 hours in, ive hit the wall of "fun" because at this point nearly every bot needs a program, and I am sick of searching for guides, looking up wikis, and dealing with the jank. Combat is where the jank lives. You can really tell they just tacked that on at the end. Projectiles can pass through hostiles through buildings. Getting hit will lock your units AI up. ♥♥♥♥♥♥♥ stupid pathfinding. "But you can program around the bad AI" Yeah I see you and ♥♥♥♥ you. The game also struggles on normal systems from what ive read...because Ive had to do EXTENSIVE ♥♥♥♥♥♥♥ research just to PLAY a game. I have a pretty beefy system, all i noticed was my fans running a bit more. Another ♥♥♥♥♥♥♥ notch against the game. bottom line. Its pretty, and the first few hours of gameplay are mechanically sound but once you are outside of the "demo" levels. The game falls apart under its own bad design. Tacked on garbage mechanics designed to be "unique" instead of fun. If your looking for a factory game, just get Shapes. If your looking for an RTS, just get anything else. Its the worst of both worlds.

Factory building, mysteries, limited hand-holding (a negative for some), and easy programmability for your units and buildings! I love just about everything about this game, and have already spent a number of long sessions deeply engaged with it.

I spent a fair amount of time trying to figure out what this game is trying to be, and my conclusion is that it's several "rule-of-cool" ideas in a trench coat. The experience that I had playing it was a let-down. There's so much potential for good gameplay, but a lack of cohesion leaves me always wanting. Most of this comes from the way that the behavior system is integrated with the main gameplay. The main faction you can play, the robots, all come with integrated behavior controllers. This lets you design behaviors for them from the very beginning of the game. Even the runner, a unit with no enhancements and a single inventory slot, can be programmed. This is really exciting! But then, when you try to do much of anything with it, you find out that the systems that the runners use outside the behavior system are difficult to inspect within it. Orders can be given to the logistic network, but you can't read them or figure out which order the runner is fulfilling. Nor can you change the network settings, besides turning it wholesale on and off. The information asymmetry is strange for a programming game--programs fundamentally just move data from point A to point B. From here, things only get worse. Say you have a unit with a factory component equipped, and you'd like to produce a unit. If you're using the UI, you can select a blueprint. If you're using code, you can use the produce unit instruction. So far so good, but then the problems pile up. * If you don't have enough room for all of the blueprint's components, then production blocks, even if those components are needed at different stages of construction. (e.g. constructing the chassis versus adding components to the chassis.) * If you do have enough room for all of the components, then you'll get a unit produced! You'll also have to wait on the component to finish, since the produce command doesn't block. And if you want to know which unit was produced, then you'll need to basically guess. Now, that could be an interesting balance problem. Complex units require buildings with lots of inventory, so you can invest more in a building for more inventory! Except that inventory modules are gated further down the tech tree, and require materials more advanced than the units you're constructing. You could spend time unlocking them and expanding the factory with new materials, but then you'll run into the problem that enemy difficulty is gated on the number of techs you unlock. Have no fear, though! You don't need to do *any* of that. Instead, you'll just use the integrated behavior controller. Create a blueprint that sets the behavior to a construction routine that orders the parts you need using the unit's built-in logistics network. Now the unit can build itself! And this would be clever, but it turns out that behavior controllers subverting mechanics is basically the theme of the game. * There's a signal register on every unit, and a specialized component to read it. But the behavior controller can already read it. You don't need a signal reader. You can actually use the behavior controller to make a very, very powerful signal reader. * There's a radio receiver component, and a specialized component to send a signal. But the behavior controller can already read it, and it can write a faction register, which broadcasts on a radio channel. You don't need a radio transmitter or receiver. * There's a radar component, but it's also a command, so you barely need it. * Unit construction is made more difficult, but you can sidestep inventory requirements using behaviors. I'm 60+ hours in, and most of that time I spent figuring out how each of the instructions function. And I'd still be having a grand old time of doing that, if the game didn't have such cool ideas that it seemed to undermine at every turn. One thing I can say about this game is that the devs do care about it. They're currently working on the behavior system, making many new things possible and giving the behavior system an overhaul in the discoverability department. As I said above, this game is full of potential. At this moment, however, it's not got much else going for it.

I know the reviews on this game aren't great, but I'm not sure what they are missing. I love playing this game. You don't need conveyers, and you don't need a million bots to move things around. You have internal transport attachments that can move things around without you any oversight. This game is tons of fun to figure out how to play it, and you can customize it to your hearts content. You just have to take a second to learn how the game works.

不是纯无脑盖，也不是纯坐牢算数，平衡得还可以

是那种你越研究越有意思的游戏，不只是看结果，更像是在看自己怎么把一个混乱系统慢慢调顺