Springsong; progress, review, and bug reports

Welp, after playing for 42+ hours here is my town thus far. This is the sixth or seventh town I had to restart do to numerous bugs. I’ll try to cover them below, particularly some I’ve not seen posted.

Springsong, front at the gate.
Click to view

Showing the Inn and various shops.

Rear view

At night

A farm

-In order to get this far I had to do some serious save scumming. Pretty much every building I made I had to save before putting it together
-To successfully construct buildings, I had to make sure I didn’t have the doors or windows for it build. I’d design the building, wait for the workers to finish it (without doors or windows) and then have the carpenter craft the embellishments. Furniture was added in after the whole building was done.


-The town has a total of about 17 citizens at this point. However, I’m starting to see the PC hitching and the yellow pathfinding bar constantly. Keep in mind that every house has furniture; 1 bed or 2, dining table, chairs, rugs, and some have internal lighting.
-Crashes occur every 5 to 10 minutes. Large slab constructions are impossible without pausing the game. In fact I have to keep the game paused now any time I go into build mode.
-In the above picture, note the little girl sitting by the fire in the town hall. Notice there are two empty beds on the top left hand side of the screenshot. She fell asleep on the floor right there instead of looking for one of those beds. Bed seeking priority should be increased. Town should have more beds than the number of citizens.
-I’m unable to place lamps, pennants, pictures, and signs on the outside of my buildings without their floor plan turning red (as if showing an error)
-Buggy construction after building 20+


-Note the floating villager.
-Note all the trash that I can’t clean up because for some reason the villagers don’t consider them Tools. Please allow us to either delete stuff like this (the hoe’s on the ground) or fix their designation.

-The goblins have made the game impossible to play and more importantly to test. You can’t test or see the goblin issues until you’ve built up a town about this size (there is another thread showing the same problem).

Click to view

-This is not an issue of pathfinding, as there are clear paths to the village (both before and after I erected the wall).
-This is an issue of a badly designed goblin spawning system. It seems that it spawns a massive gaggle of goblins as soon as a villager uncovers an unexplored area?


-I’ve actually had the goblins spawn on top of my villagers even though the area was already discovered:


Summary thus far:
-I know a lot of folks like the scaffolding and how cool it is. I’m starting to think it’s a bit unnecessary. Perhaps optimize it some more to not take up so much pathfinding and AI routing resources?
-I don’t see the need to put a ladder on every window, even those within reach. I put my windows two blocks high.
-I think the goblins need to be turned off until the pathfinding and AI is much further ahead. Additionally, until we can set patrol sectors for our footmen, goblins are getting in the way of any progression.
-Goblins and any spawned enemy should always follow the single encounter rule, and not be one massive gaggle, regardless of the size/equity of the village. Currently a massive gaggle of goblins is spawned (size based on total village asset worth?) and then trickled at the player. This doesn’t work, and almost always ends up with the player getting all their villagers killed piecemeal. If you’re throwing small bands of about 3 to 4 goblins at the player, the system shouldn’t spawn any more than that at their camp. The system should not spawn more goblins at that spawn point until that warband is destroyed. The current setup is creating massive pathfinding bottlenecks, lag, and crashes.
-When villagers get hungry, they drop what ever they are carrying to run back to town. I’ve had villagers half-way across the map bringing back stone, suddenly drop it, run all the way to town to eat, and then run all the way back to pick up the stone they were carrying… I think eating should be at the personal level and not always synchronized with all other villagers. This is going to be a problem later on, especially once you have a city going and its entire population drops what they are doing on the ground and runs to the food depot. This problem is exacerbated by the pathfinder having issues with multiple food bins.
-Workers won’t build scaffolding when doing slab constructions.

I’m afraid the game in its current incarnation is going to have some serious scalability issues later on down the road. If these issues aren’t squashed now, the programmer is going to have a nighmare of work later on:
-Housing construction needs to be simplified on the backend (not the visual end). I think the game’s scope has gone beyond the need of scaffolding supports. The details are cool… but seriously Devs, it’s out of the scope of the game at this point. You could save so much on CPU cycles and free up memory resources by reducing or simplying this feature. You have pathing to and from resources, up and down scaffolding, slab construction, in and around furniture, in and around the game world… And then you’ll have additional pathing implemented in the future… dungeon monsters, AI Villages, and so on.
-Pathfinding and house/furniture navigation needs to be solid before additional units and features are added.
-Footman should function identical to the Trapper. You set up a Yellow area where the footmen patrol. Remove their patrol of storage areas as a trade off. Currently Footman are patrolling storage areas, so this should be an easy implementation.

You guys have a solid foundation and what could be a great game. I just hate to see this game fall into development hell like every other game of its type has done thus far.

This saved game can be downloaded here: [http://www.mediafire.com/download/a6rfypi6c26xoii/saved_games.zip]


I believe many of these issues, such as the item designations and footmen zoning have been discussed extensively before… though I am not sure because I don’t religiously scan the forum all the time :stuck_out_tongue: I do like the focus on scalability you’re taking though. New features are awesome, but I personally have enough similar issues with goblins, AI or even crashing to play the game for long as-is.

Goblins are a very big issue once the town grows larger, but for testing all non-combat game modes I believe the peaceful game mode works just fine. I think it’s agreed that the current spawning model isn’t ideal, especially when villager count eventually runs even higher, and the goblin horde grows exponentially with it. I’ve seen posts regarding it before and have addressed it myself as well.

I too believe scaffolding is very problematic at this point. A simple start I would like to see is villagers being able to build with a slightly higher range (maybe even as a learnable skill for a class) and thus require less scaffolding. Currently anything a block above requires scaffolding, which for some things, as you noted, like ground-level windows, is a bit ludicrous. I mean it makes more realistic sense for the worker to be only able to work at eye level efficiently, but I’d hardly call realism a goal in a game like this. I honestly believe it’s a system that’s very much unfinished and unpolished though, as with many other parts of the game, as it still doesn’t even support many building styles by itself, such as multistory or slab building where manual ladder-placing is needed. As such the AI is likely not optimized 100% and is probably bound to change either way.

You see, that’s the problem though. I have this sinking suspicion that it started with the idea of Minecraft; a single player or group of players building things and requiring supports and scaffolding. Think 7 Days to Die and how that game requires scaffolding and supports for multistory construction.

The game evolved from that into something closer to The Settlers. Except that it’s still trying to cling onto the micromanagement of a single player (or small group) exercise. Templates help that more or less and I think once that feature is solidified it will help greatly when the player reaches the City level.

Unfortunately, internal villager usage is still stuck at the local “small group” level. This is going to create a major problem when you’re running a city that requires multiple construction jobs, farms, stone/lumber gathering, and military management.

Keep in mind they also plan on having adventure and RPG modes of play. Where by you’ll have a village hero or group of heroes that you can send out to dungeons, POI’s, and exploring. The good thing is that villager automation is pretty robust, and it’s what has attracted me to the game. I.e. I don’t have to micromanage them to bring supplies to the job site.

Some suggestions:
-Building construction by villagers (not the player build mode) should be simplified. In fact I’d almost suggest going the RTS route where villagers hammer away at the building artwork as it slowly rises. Player still has to design the building however. This would cut out any of the pathing such as on the roof, picking up all the excess wood and roof stone and so on. Villagers would still need to bring wood/stone to the job site.
-Buildings should be treated as a single obstacle. The programming is already there for this, as buildings are treated as one object. Unfortunately villagers/enemies aren’t treating it that way. The only time a villager should go inside a building is if the Player assigns them to that building.
-To help with pathfinding, enemies should build their own ladders as well as destroy buildings and walls blocking their path.
-Take a look at games with similar tones. Rollercoaster Tycoon 3 and The Settlers are good examples. In RT3 park guests will eat while going to and from rides. Although they sometimes do sit if a park bench is close.

Most of your observations are correct. There are lots of optimizations needed in the game, and there are several persistent issues plaguing the gameplay that makes it less fun than it could be.

Currently, we can basically ignore any issues with Goblins attacking and such due to the fact that the spawning system is clearly not at all final. Yes, they do need to fix that bug where they just form a horde like that, but I wouldn’t really want them spending much more time on it than that. Just make Goblins and other enemies not crash the game (or spawn on people’s heads) and call it good for now. Once things like dungeons and such are added, revisit Goblins and such.

We will have to disagree on this point. Scaffolding being removed would be almost a deal breaker for me and several people I have introduced to the game. It is one of their favorite features, and it helps the game stand out and gives it character. A traditional rts model as you suggested doesn’t fit the atmosphere of the game, and would break my heart personally to see implemented. Scaffolding may have some issues, and it may be troublesome, but without it, Stonehearth just doesn’t feel like Stonehearth. The loss of personality and flavor would outweigh the benefits.

This is the core of all the issues you seem to have with the game right now, and I agree. The perpetual pathfinding bugs are what keeps me from playing more than an hour or so, and I never get any buildings done beyond the first. Pahtfinding is somehow tied to job assignment, if I remember correctly, so if the pathfinder breaks the entire game grinds to a halt. There are several threads about various issues with it, so I do believe that it is a main focus of the team right now. Hopefully they can fix it, or if necessary, rewrite it, without sacrificing things like scaffolding. As I said, this is the biggest issue for me right now, and it has killed every attempt to make anything in the game for me.

Other than that, those are all really good observations. The area patrol of the footman is a good one, and being able to place things within reach would be handy. Your town looks awesome, and I am jealous that none of mine have ever gotten to that size. :stuck_out_tongue:

I don’t think I understand fully why the removal of scaffolding from the background would be an issue.
Is it a visual thing? Because you know scaffolding can still be put up as a just a visual effect.

Here’s the thing right; When you start and only have two or three buildings, watching your little villagers walk up and down the scaffolding laying out block by block, roof shingle by roof shingle, yeah sure I agree it’s cool to watch the building come to life like a construction site simulator…

Lets fast forward to a thriving city… Now you’re spending hours working on your castle, equipping your heroes, running a dungeon, or inspecting footman/knights/cavalry to make sure they all have your latest Mythical armor… Are you really going to be concerned about house #158 and all the workers laying out each ladder, platform, and shingle anymore?

I’ll use my city as an example. Right now, aside from laying the floor plan and designing the house (I don’t use templates since they are buggy for me), I don’t pay attention to the construction any more. In fact I haven’t since the third house or so. The only time I look back is if the pathfinder is going crazy and the bar is on the yellow.

That’s what I mean, it’s a cool novelty at first. This can be perfectly represented by having the scaffolding just be a visual effect as the workers construct the building. My concern is that I don’t believe we’re there in terms of resources needed to build an entire city block without it having a detrimental effect on the gameplay as a whole.

My other concern is based on my gaming experience. I just haven’t seen it done (Maxis with The Sims series, or Chris Sawyer with Roller Coaster Tycoon, Transportation Tycoon). Path finding is a major challenge. It isn’t just “a small concern which will be taken care of by the time Beta comes”.

Another alternative would be a path finding handler based on overall path finding requirements. So as more demands are placed on the CPU for path finding calculation, less important path finding jobs are simplified (housing construction being one of them). The larger your city and the more simultaneous activities are going on, the more simplified things become.

I just have a nagging feeling that the team may have set themselves up for more than they can handle.

Your concerns about the pathfinder are valid, and I share them. I disagree on the scaffolding, and after a long day of work, Ill do my best to convey my reasons.

Stonehearth is, at its core, a city builder. Everything you do revolves around building your town, providing supplies, making people comfortable, etc. Later on in development it will be fleshed out with rpg elements, have a storyline, but will still be a city builder. One of the unique things about it is that you design your own buildings, similar to games like the Sims and such, which lets you add a deeper level of customization to your city. All of this comes together to make a flavorful and colorful game.

Scaffolding, as it is currently implemented, is an immersive and different take on the method you use to construct your buildings. It takes advantage of the voxel nature of the game to provide that little extra bit of coolness factor that sets it apart from other games. I do not feel I am in the wrong to say you would be hard pressed to find anyone in support of removing scaffolding in its current format in favor of a Warcraft, Starcraft, Majesty, etc. style of construction. The game itself doesn’t want to convey that type of construction feeling. So, I say that scaffolding should stay because it is wonderfully flavorful element of the game that gives both immersion and uniqueness to Stonehearth.

Pathfinding, on the other hand, is where both you and I agree on many points. I want to be very clear on this: Pathfinding, as it stands, is pretty borked. I cannot, as of the current revision, build more than one building without it breaking and having people idle about the place. Placing furniture in a building before it is done traps people in it. Putting a workbench inside a building creates some kind of issue, though I would have to test it more to nail it down. The Pathfinding, in my experience, is not working right. I am not worried about house 158, I am worried about house 2.

In my experience, a ladder up a cliff breaks Pathfinding far quicker than Scaffolding does. The developers even pointed this out. So, Scaffolding as a cause of such issues is relatively low on my list of things that need to be fixed about Pathfinding. We are years away from a thriving city of several hundred inhabitants, as depressing as that might sound. So yes, Pathfinding will have to be fixed before we can reach that point, and it will. I just do not agree that removing Scaffolding is currently the way to go.


I too share your concern on pathfinding and especially it’s scalability at later more populated points in the game. I’m not going to say it’ll be easy but it’s the most critical part of the entire game. I’m sure there’s plenty of hairpulling over this every day at Radiant and even though I’m not a programmer, I understand the complexity has got to be daunting. I’m being honest here when I say, I think there’s a reason that noone has attempted this sort of building method before. I haven’t tested this but does limiting the number of actors working on a house simultaneously cut out some of the borkiness? Is it simply a matter of too many cooks? Perhaps that’s a direction, instead of removing scaffolding and the like, since that’s part of the core aesthetic that makes Stonehearth Stonehearth.

To many villagers working on a building can be part of the issue.

Think of all the events going on at the same time:

Goblin pathing, monitoring, spawning.
Carpenter crafting player queued items. Gathering materials needed.
Weaver doing the same as above.
Trapper setting traps and inspecting hunt zones.
Footman patrolling and scouting.
Farmer planting, harvesting, delivering food.
Workers collecting/harvesting.
Workers gathering needed materials for construction.
Workers plunking down materials at job site.
Workers plunking down materials at Slab job sites.
All of the above routing in and around town, map obstacles, scenery obstacles, cliff faces, ladders, scaffolding and roofs.
All of the above multiplied by how ever many villagers and goblins there are.

And then you’re going to add additional units; Masons, Blacksmiths, Engineers, Knights, Heroes, and entire villages.

…and that’s why most RTS games take a very simple approach. Settlers is the closest I’ve seen to this game, and even that one uses a simple approach to building construction.

Thedrgnrbrn mentioned The Sims series. Having looked under the hood of that game (the first, second, and some of the Sims 3), it has a dynamic handler that only runs the simulation based on either the Lot or the location the player is at. It dynamically stores lots (houses, clubs, or PoI’s) until the player loads them and only keeps a few loaded. In other words even if you have an entire city or village in The Sims 3, only a few are loaded, with the current one the player is at taking priority. Additionally, house construction is done completely in build mode, the sims themselves do not construct the buildings.

Second, there is absolutely no gameplay application for the Villagers to use CPU/Path finding resources to construct a building when it can be simulated visually.
-Villagers can’t stop mid construction to make major changes. I doubt the developer will allow that anyhow (I.e. stop midway on the roof to change its direction).
-So villagers path around a building site, but they don’t around the Slab sites? I built my tower and the villagers did it perfectly without issue and without scaffolding.