Skip to main content

Indie game storeFree gamesFun gamesHorror games
Game developmentAssetsComics
SalesBundles
Jobs
TagsGame Engines

The-Balthazar

62
Posts
132
Followers
A member registered Mar 13, 2022 · View creator page →

Creator of

Recent community posts

Sorry if this is a lot, I've ordered the points in the approx order of importance to the overall experience. I have no idea what you'll be graded on, or how much time you have to budget towards any of this, but I hope this helps.

The level design doesn't give you enough time to actually learn how to handle any of the obstacles the game poses before it makes you deal with everything at the same time or in quick sequence, for example, the game makes you dash jump to a platform with low headroom before it requires you to do even a simple jump. It makes you deal with a wall jump section and a ledge mantle with an instant death spike near the top giving you no space to learn how ledge mantles feel first. It makes you deal with your first enemy with an instant death spike behind you giving you no space to learn effective ways to handle that enemy, and then to top it off after dealing with all that the game rewards you with two off-screen instant death cars. Those give me the vibe that this is supposed to be a hard platformer à la a kaizo or a troll level, but as a stand alone game you have to give people a chance to learn the game feel and get invested enough to get through those. For example, I have no idea the timing on dodging the car from the front because to even try it I have to get through the whole rest of the game to even try it. As a section of game what I saw up to that is probably fine, but as the first section of a game it poses too many barriers to learning the game. As a result, I have no comments on anything after that point because I've not seen it.

I'd recommend easing people into it with a soft tutorial with some generous checkpoints; you could have it so, for example, the first area is enclosed with a wall with a ceiling on the right so you're immediately signalled that you're supposed to be going left, and then left is a ledge that you have to jump up that's just a little shorter than the max jump height, then one just a little higher so you have to do a ledge mantle, followed by enough space that you don't feel in danger of overshooting the landing, then your first hole that you only need a regular jump to get over then a checkpoint and a comfortable amount of flat ground. Then just a hole that's wide enough that you need to dash jump over, then a checkpoint. Then a flat section that with a reasonably generous lead in before a just fly so you have enough room to experiment with fighting and dodging, then a checkpoint. Then a section that's just flat with a spike to jump over, then a checkpoint. Then, with enough spacing before it so it doesn't feel like it's in the way of jumping over the first spike, a wall jump section with enough flat ground at the top that you don't feel in danger of overshooting the landing into another threat, then a checkpoint. You could have something like a ceiling above the outside side wall on this first wall jump section just so you have to actually go between the walls at least once and not just spam jump up the outer wall, but that's not a huge deal. Then, so that you don't have two wall jump sections too close together in quick succession, you could have one or two sections that combine any two of the other obstacles together, with checkpoints between and after. You could also do that slightly cliche, but still helpful thing, where the first time you need an input, there's an in-world signpost telling you what the button is.

Then there's the matter of the instant-death cars; I'll start with they are at very least telegraphed with the beep. You could at this point have the player arrive at that platform to the right of the current start and just continue onwards and they would have theoretically had the opportunity to get most of the skills needed to actually proceed and would have at least some degree of personal investment from the feeling of progression from getting there, and hopefully enough to make it through some difficulty, but, assuming you actually want people to be able to make it past the cars, you really need a section where people can practice the timing on a car coming towards their face with no other obstacles between retries. You could do this by having a checkpoint in between the vertical car and the front-on car and let people suffer it out there, but if you want to be helpful or give them a good introduction you could have a section in between the previously described soft tutorial and the current start which is a flat area with an obvious road texture, have one of those 'monkeys crossing' signs at the start of the road to really literally sign post it which could double as the checkpoint, then, assuming people are holding left this whole time, at the point where you'd need to jump to dodge the car, have a splat mark on the road or some other visual identifier to help people actually learn the timing. If you're feeling generous you could also use the 'monkeys crossing' sign later on as additional indication of them, but if you do that make sure you keep it consistent as to if they count as a checkpoint or not.

The combination of enemies requiring multiple hits, the lack of any kind of knockback on hitting, and the fact that you only attack forwards makes it very hard to avoid damage, and makes combat feel like you're just mushing your faces together until one of you falls over. Adding knockback on hitting enemies would make hits that don't kill feel more impactful, and give you a reasonable chance to actually get through combat unhurt. Similarly if you had enemies hitting you knock you back it would make failure to properly engage them actually interact with the platforming, potentially knocking you off things, and make it so that combat with a fly has some back and forth instead of you just both rubbing faces. If you also had enemies like flies ramming you also cause them to recoil back a bit it would also help sell the interaction and make the whole thing feel more visceral. An aside; the first time I encountered the fly I jumped on top of it, and when that did nothing other than have me ride it for a second or two taking damage until died, I then went back to the game page to see what the attack button was that I'd obviously missed.

The visual shape of the terrain not matching the collision shapes, aside from often making it look like you're floating on nothing, makes it very hard to visually judge edges. Functionally speaking you would be better served with using a flat texture for the terrain, then for visual appeal adding separate decorative objects and/or edgings to break up the shape, like grass tufts, bushes, hanging moss, cracks in the rock and so on, although if you do that you need to be careful to make sure that there's some visual differentiation between things you're supposed to care about like collidable terrain and spikes, and things you're not. It is however good that you have something to stop you just falling off edges that you slowly inch up to. It's subtle; I almost didn't notice it, which is how good mechanics like that should feel. They're the things you only tend to notice if they aren't there. It would be good if it, or something similar like coyote time which might be easier to implement there, also applied to walls you're clinging to, to make wall jumping easier; I resorted to holding one direction and spamming jump to get up walls, even if that is much slower than properly wall jumping, and feels more like a cheese. I'd originally attributed the fact that I'd never missed a jump that it felt like I should have gotten to the fact that the visuals were smaller than the collision, but I realise it was probably that mechanic, which you deserve credit for.

It repeats the jump noise whenever you press the jump button, rather than whenever you actually jump, which makes it more obvious that there's only like 2 noises. I don't know how much you care about not being able to spam the sound, but either way if you vary the pitch of sound effects like that by up to 5% either direction it will greatly improve the variance.

Similarly it makes the dash sound at times when it doesn't dash, such as when you're not holding a direction key and press the button, although on that subject it might feel more natural if it remembered and used the last pressed direction or used your current facing or movement vector rather than requiring one being held. On a personal note, it feels very unintuitive to me to have dash on a direction key and not be directionally affected by that. I've found myself trying to press right to dash right even after a bunch of playing. I'm sure I could get used to it, but something as simple as not having it on a direction key, or having both left and right dash, whether or not it dashed in those directions, would probably go a long way to help with the control feel of that.

You could perform some more advanced movement if you didn't have it cap your horizontal momentum so harshly when you come out of a dash while not on the ground. It would also probably improve the feel of the movement, and maybe add some speedrunning potential. Like making it so you can do a ground-dash straight into a jump to do a longer jump.

These are less important, and probably things you're already aware of, but are points that would help with the overall presentation, and I might as well mention at this point:

There's no visual cohesion; some bits are edited photos, some bits are mspaint drawings. This isn't inherently a problem, but it'd look more cohesive if you picked one and stuck with it; be it edited photos, mspaint doodles, or if you roped in someone from art class to help create art assets.

If you had some kind of background element connecting at least some of the platforms to the the bottom of the screen it'd go some way to making the place feel like somewhere that exists in a world rather than just some floating rocks.

Nothing is animated, even things that would be easy to animate like making the saws spin.

The game instantly snapping you back to the respawn point on death gives you almost no time to process what happened. It's good that it's fast and requires no input to proceed, especially for this sort of game, but it'd be good if it at least dwelled on your death and the cause of it for like a quarter of a second or something

It's a shame I can't see all those images, since imgur completely blocks the UK.

(1 edit)

I don't have enough experience with getting it to run on mobile to give great advice, but if you don't have a file browser to run the .love file from, you should be able to find it in your downloads to run. I'm not sure if there's a way to create a shortcut to a file or something to make that easier. I can look into creating a mobile build though.

You can force a zoom setting in the options, and it might let you bind physical buttons to zoom in and out, but I've yet to implement gestures.

Oh it looks like the love2d was taken off the play store since I last checked. Some phones can be a pain to get them to install APK files, you may have to find a guide for your specific model. Once that's done if you try to open the .love file it should give you the option to run it with love2d.

You can, I just don't provide a build for it: if you get the APK of the engine from https://github.com/love2d/love/releases/tag/11.5 (IIRC the one on the store is out of date; 11.3 or something), then you can open the .love file from the full version with it. I don't provide that for the demo because it's literally just the games source code.

Glad you got it working! I thought it might have been because my launch script loaded the window module early if the config has no values for resolution set. I already changed V48 so that it delays the window module and user config loading until after the other core modules, I'd be interested to know if that would interact with that setting, or if it would have still crashed anyway.

Sorry it took me a few days to get back about this, I have an idea what this could be, and you can bypass it with an edit to the config.

If you open "~/.local/share/SternlyWordedAdventures", if there's no file called "userConfig" (no extension, plain text) create one with:

return {
    window = {
        width = 1920, -- or appropriate values for your system
        height = 1080,
        fullscreen = false,
    },
    volume = {
        music = 1,
        effects = 1,
        master = 1,
    },
}
If there is one already, add the window section to it or add the width height definitions to the existing window section. The file needs to be a valid Lua table.

You'll probably want to go through the options and set everything then restart the game if you did set that bare minimum config.

I'll change how the code I think is at fault functions to hopefully prevent similar issues in the future for future versions.

Is something preventing it from creating a window? What are you trying to run it on?

I was debating adding a button for that this last version, but if you're on windows then your save directory is in %appdata%/SternlyWordedAdventures

Your metaprogression, stat-totals, and other permanent flags like for one-off tutorials are stored in persistentSaveData, if you delete that you should also delete mainSaveData and combatSveData; those are your current character (or you could get them eulogised first), since they could put your metaprogression in a impossible state. If you played pre V40, deleting persistentSaveData will have it try to import the save data at %appdata%/WordTileGame

Deleting userConfig would, as you might imagine, delete your settings.

You could preserve the contents of saveStats and get the majority of the new player experience; you get a predefined single hero option with an empty stats folder (the adventurer with no curses or boons), the halfling unlocks on a non-empty stats folder, and world seed is based on the number of stats entries you have, so if you keep your stats history you'd instead get 3 hero choices with random curses/boons. You'd have a very high chance of rolling mechanically the same as the starter with no unlocks/seen items though, and you could technically get that experience as a full newbie by just going to the arcade first.

Whether or not you delete any stats, if you delete persistentSaveData you should also delete statsIndex; it'll trigger a recount of global stats and reindex them.

I've done a rework of how it does the background of those screens. That should hopefully deal with it. I'll likely be releasing it tomorrow or the day after depending on how much time I get to finish a few other things in this hotfix tomorrow, since I've still to look at the corrupt forest issue.

(1 edit)

Wishing well or village well? I could see that maybe on a wishing well but they aren't supposed to have a blurry background, and the village water well blurry background renders once then stores the result. On my setup it naturally caps the framerate, so it's probably related to an individual setup not capping it.

The forest not completing is probably a missing check or trigger my end. I'll look into it although it's not vital for progression.

This is super cool. It's late so I'll check the video out tomorrow.

Thanks. Hopefully I can get V46 out soon, it's already a larger update than V40 by some metrics.

(1 edit)

On investingating further, it looks like you chose to fight on past the required enemies then quit while still in the dungeon. There was a load order issue in V45 (and possibly also a few versions before that) in that specific situation.

The good news is that crash was fixed along with a big combat load order rework in V46 some months back. The bad news is to save that character before V46 you'll need to do some light save editing, which I can guide you through. Mostly just deleting the last few kill stats so it rolls you back to the chest.

Edit: Or you can email me the save or drop it in the discord and I'll fix it.

Was it a save from a previous version?

It's on the github.

Thanks!

Yeah with how late I started I had to choose 3 of the 5 of netcode, AI, deck construction, audio, and sleep. I nearly made time for the audio; I woke up like 2 hours before the end of the jam, and reached for my microphone, but the battery was dead and I took it as a sign to go back bed instead.

Your tiles, including upcoming tiles, refill at the start of your turn if you still have health, and then effects like ! tiles falling off happen during your turn.

To activate equipment that require manual activation you click/press thier icon.

There are plans to release the OST when my brother is happy with it, although he's currently sitting on a big music rework that should be coming soon™.

For most enemies where defeating them counts as murder, you can make them run away by hurting them enough (enough usually being less than half health). Enemies that count as murder that wont run away only happen if you're already on a murdery path or choose a murdery event option. They'll have if they will run or not listed in their statuses with a little exit icon.

No worries. Thanks! Feel free to report any issues or missing words you find, or anything else.

I've not added Steam keys for Itch to distribute to you yet; since it's not released on Steam yet they'd function just like pre-order keys so it wouldn't let you play it there anyway yet outside of the demo. IIRC the claim button would be through your non-demo download link.

Well that's problematic. Thanks. I'll get a patch out shortly. At least it's only a one time error. It's the warning I added about when the demo ends.

There already is a free download demo. Unless you're after one for a different platform?

Web demo I've tried in the past and it didn't work; the java emulator for the engine can't handle it.

So after looking through the logic of how that could have happened, you must have gotten to a forest, had an event that wasn't the hell opens event, then entered the forest.

I'll add a clause preventing it from happening in subworlds. In the meantime a workaround is to avoid level 4+ nodes in subworlds until the hell opens event has actually happened. Hopefully you can either get through or get to another place where it's safe to trigger it.

For whatever reason it likes to run Proton instead of the Linux native version by default, which is definitely a choice, and based on the fact that that compatdata path includes the Steam AppID it'd load a different folder for the full game when running on Proton. (Demo is 2570560, and full version is 2562130)

Linux native version in and out of Steam and/or the demo should hit the same ~/.local/share/SternlyWordedAdventures folder regardless.

Love to hear it! In theory it should just load from the same folder. If not you can just copy paste the save files over. If you open the parallax viewer on the almanac then tell it to save it will load the save folder for that version without having to go digging for it.

I've just done a re-upload of V40 with a few fixes that were important enough to get out now, but not big enough to do a full hotfix for.

This was enough to track it down and fix it, thanks.

Removing it or reordering the list will affect the RNG for everywhere else, and it's an alphabetised list, so it's now Gyattsville.

Do you have the text of the bluescreen? If it's not rendering correctly you can press C to copy it to clipboard.

I guess that doesn't look good, although it is a real place in Yorkshire.

An easy way to handle this that doesn't require waiting for me to get the fix out or edit save files, is to load the Steam demo, then quit and go back to the itch version.

It's a script that affects migrating old save data, that had a last minute addition specifically for the Steam version that shouldn't have been enabled for the Itch version.

The same crash or a different one?

The button to download purchased games is at the top of the page on itch.

I've found the source of both bugs and fixed them. I'll get a hotfix out shortly.

Not intentional, I'll investigate both.

I'm trying to work out how this could have happened unless you managed to download v37 in between me checking the analytics and seeing no downloads, and finishing uploading and publishing the amended version (that I thought I'd only privately referred to as 37.1).

Can you confirm that the issue persists on the currently available version? And if so could you describe the game state when it occurred?

Other than using the capital city to regenerate the world you don't really. Have you found one of the rare seeds where there's a disconnected land mass?

(1 edit)

This appears to be a conflict between the voiceless rough stone charm and the bronze bonsai. If you're able, unequip one of them until I can get a fix out.

Edit: I have a fix, I'll see if I can easily untangle it from the unfinished new update for a hotfix, or finish the update in the next day or two.

I introduced the Bookworm Adventures speedrunning discord to this, and this was one of their results:

That is the current end yes. More is planned later, but I can't say when that will be. There should be an unlock the first two times you do it though. Depart essentially takes you to a fresh world for a newgame+ on a new seed. The option is technically available the whole time, but costs 1000g if you've not done that fight. Explore is currently unimplemented.

Yeah, those can sometimes throw some super obscure words at you. If you check the stats, it'll tell you what they were on there. I should probably add a postgame screen to them.

Latest update should solve this.