I'm kinda wanting to do something different than a straight-path storyteller. So far what I have is sidescroller semi-survival about a goblin creature combining/crafting items to advance and explore the forest using a 1-3 lvl upgrade system, 3 being stronger/better for healing/stuff. Example: wearing tree bark wrappings dipped in a curative to survive a dead lake.The combinations are mostly unexplained and require slight observation. One main mechanic is stocking and preparing your home against harsh seasons, and desperate repair if something fails. Progression is seizing new safer homes like an abandoned overgrown fortress nesting with creatures, safer from rat infestations, etc.

Mainly I need to know (since my first real game) how can I keep a game like this becoming repetitive and maintain good replay value? This is a fairly bigger game backtracking and stuff is my main concern, moving back and forth between so many screens, and also item/challenge pacing.

How do you combine a sidescroller with real exploration? Aren't the levels too simple and too small?

Consider typical roguelike games: apart from procedural generation, which is likely to be a good idea for your forest and your homes, they have large environments and a more natural 2.5D structure (many planar levels or regions, with enough passages between adjacent ones) that's highly suitable for exploration and long-term travel.

Even worse, how do you combine staying home to make repairs with going out to explore?

Maybe explorer is a bad term, it seemed that way in my head, but 'resource gatherer' exploration is probably better. You learn how to efficiently defend against the seasons in different ways using familiar and new materials. The game starts before a hibernation cycle, which is when you're out gathering and preparing, then when it hits hibernation starts. With good defenses you wake after it's all over and assess the damage learning through observation and item hints. If you did badly something will fail and you struggle to fix it before everything goes down. Exploration is to find new items for building/healing/weapons/lore and more meaningful stuff. New seasons bring new creatures and items/growth, and maybe unique weather effects like special rains enhancing certain stuff, just to give areas a kick in value worth a trip. If those ideas sound appeasing. In itself, defending a house maybe seems a little too hollow, so it's gonna have to tie in with drawbacks to something bigger like a story? Idk.

I was actually planning some light procedure generation for goblin camps and stuff, entire levels is probably too much for me though. Camps meaning several sized huts holding 3+ enemies in a small screen.

I don't think there's any problem with the core idea but like everything testing and development is the only way to know for sure. As for repetitiveness, I find that stress and very risky decisions helps mitigate. This to me implies either a real time game with strong time limits or a turn-based game with very strict, limited moves.

I like the aspect of experimenting with different solutions, and changing the efficacy of different items could help with repetition: In one game, tree bark is good, in another, it's bad, and if you have a good but subtle hint system I could see you creating a matrix of solutions which varies extensively from game to game.

If it doesn't work I probably screwed something up heh.

I'm liking the idea of making risky decisions, hopefully without p**sing the player off to much :D

Another addition, perhaps catchable creatures like birds and critters along the way advancing a skill? Birds could be 'broken down' in to feathers/meat/beak for items, or craft the actual bird in to a glider for luring larger prey.. which sounds kinda funny for some reason.

Sounds like a very interesting idea, though I'm not clear on certain details. It's a sidescroller survival game from what I understand. Is the goal of the game solely to survive as the gobelin ? Or is it to have a base strong enough to survive the next winter ? Is there permadeath ? Do you want to make it a roguelite ? Do you lose if you die or if your base is not strong enough when winter hits or both ?

I think you could make a pretty original roguelite, it's a cool idea to have the base be an end in itself and not just a mean to an end like in most other survival/roguelite games. I think it would be a really good angle to explore, making a survival game that is as much about your own survival than it is about your base's survival. You could explain it by saying that you can't survive the winter without a base in decent condition or that the gobelin is a creature that withers and die without a home.

Pacing shouldn't be a problem, it's mostly about fine tuning and balancing the game. You could have increasingly harsh winters requiring the player to constantly upgrade his base, or non/slowly renewable resources forcing the player to always explore further away to keep surviving. And as you explore further, zones are more challenging.

Backtracking is the real problem. It's especially a huge problem with non linear sidescrollers. Look at a game like Kingdom, you end up going back and forth over and over and it's really tiring. They introduce fast travel options to try and fix that issue, but it's still there to some degree. There is no magic solution to backtracking : either you make the travel interesting or you get rid of it altogether.

Fast travel options or a dynamic world that will change and make the journey back different from the outward journey. It's a sidescroller so you can easily make the journey part of the experience by introducing platforming elements and puzzles (which you have already done it seems). It's not just an obligatory step to get from one point of interest to the other. Then all you need is to find ways to make the journey different, either by actually making it different (using verticality) or by modifying it on the way back. That or fast travel.

You could have the gobelin go down a cliff to access a certain zone for example and not be able to climb back up so you would need to use a tunnel that runs below it to return to your base. Or have platforming or puzzles change as you do certain things, for example you cross a bridge that later collapses or have your item be consumed when you go through that deadly lake and you need to find a new solution to cross it again. Just having 2 different solutions for each "puzzle" or obstacles and forcing the player to use both to go back and forth would be enough.

I realized it needs an arching story to be interesting, so what at first were exploration tools (like the lake) may become plot points, but still hold meaning beyond that. At the moment (because it keeps changing) you'll survive to develop the story at any given pace. Along the way you unlock privileges from quests, like a goblin chief clearing tough enemies of value for you, then death would revert any progress gained this way, or something like that. Just a thought. There's also this idea I had about him being spirit-bound to an object you're sheltering if destroyed reverts him back to human again, so the chief would not recognize or follow you, as one example. But that's kinda wacky and maybe pointless. Anyway about death, I can't say for sure until I figure more out, and idk what roguelite is actually :unsure:

Increasing winter deadliness sounds good, should definitely tie that in with mid-late game at least.You're probably right on that, It should have some fast travel and as much content as I can throw in. I've been thinking about the whole backtracking thing, and as much it needs engaging stuff I need reasons to shift focus to them individually in ways the player isn't face-planting their screen in boredom. Basically finding interactions to make them feel they won something or that something cool/unique is happening, that's the toughie though if you get my late-night rambling.

Access points also sounds good, got some stuff to think up for that one. I had a decayed bridge in mind that needed fixing for the story, but recurring repair could be a thing too like you said, and more than one 'round them parts.

One other thing, I really think games like this need at least some types of 'human' factor. Npc's should be deep and quirky, and especially not always stagnant but participant in different ways. It's like adding another dimension to it, unless I birth some pretty unappealing Jar-Jar's. It should be fine though.. I think.

Lastly, if I end up on the art-side of things, it will probably look something like this below. Sketched up a few oddities, which are suppose to be trees and the char. Clearly have no artistic gifts, which sucks, so I might have to hire some people if I even can. Otherwise I'm stuck with this in hopes someone talented will touch it up with a color revamp.

http://i.imgur.com/W697LMa.png

I think that's not a bad idea, obviously adding narrative is never a bad decision, but I don't think you need it, I think you could do something simpler that would still work. A roguelite is a game that shares characteristics with roguelikes without really being a roguelike. Roguelikes are games like Rogue, games that feature perma death, procedurally generated levels and turn based combat. Basically you start the game and you try to make it as far as possible before dying, then you have to start over completely, similar to an arcade game.

Roguelites are games that share the same philosophy but with slightly different mechanics, usually they have perma death and procedurally generated levels, but they abandon the turn based combat and they also might try to mitigate the perma death with progression options where you unlock stuff with each new playthrough that is persistent when you die. That way you don't lose all progress everytime you die.

I think you really should check out some games in that genre, that might help you figure out exactly what you want to do. Kingdom is the best example I think, it's a sidescroller survival roguelite focused on base building and management. Its gameplay revolves around a super simple mechanic, so that is different from what you want to do, but everything else is pretty much what you described. Don't Starve is another good example, it's not a sidescroller, but it's an excellent survival roguelite.

The first thing you should figure out is if you want perma death, personally I think it would fit your game perfectly. Then each playthrough would be about surviving for the longest time (or you could have an endgoal like in FTL, another roguelite). Then figure out the survival mechanic, how you die and lose the game. Then build around that. You should focus on those things first because they are the core of your game and they will shape the rest of it, so if you try to think of specific mechanics first, it's most likely you will have to redo everything because it won't fit.

Regarding the art, I think it's ok for a first try at pixel art, no one is born gifted, art is something that you learn like everything else, I think you could easily improve enough to make the art yourself. Pixel art is fairly simple. I've seen people become really good at it even though they were way worse than you when they started.

Check out dont starve. It has a lot of replayability and you seem to want to create something similar.

Why replayable? Becouse world is highly random and there is a lot of creatures, crafting, feutures and items to explore such as stuff behaving differently in different seasons etc.

Yea I'll start brainstorming the core stuff, rather than throwing random ideas in as patchwork. I wanna make it a type of hybrid between exploration combat and crafting with random encounters without disjointing the gameflow. Will also check out those games, probably plenty for me to learn there. I'm also not that new to pixel art, but is not something I practice so now's the time to learn I suppose. If anyone ever has any idea's I'll definitely check em out and thanks for the posts guys!