itch.io

There are several indicators, but everyone misses them. Don’t know if you got an idea?

Oh right, I caught a glimpse of that at my first coven in mission 1 but I didn't make a special note of it because of all the other things getting absorbed at the time. It's still something that player should have control over while trying to keep the witch king safe and i.e. have the UI option to stop producing dropships even if it's at the cost of half the overall unit production, because once they are accidentally produced it's kind of awkward to clean up.

I used to get a lot of complaints that the game was too easy/pointless, so I guess it’s good I managed to swing the pendulum the other direction at least. Now to perhaps find a middle ground.

Maybe the game is inherently hard to balance. Normally the resource economy dictates the tempo, but here dynamics are different with big payoffs from calling dibs on free covens as early as you can. But maybe I'm wrong and simply mission 3 obliterates me too hard. Mission 2 seems impossible to fail because there was no offensive pressure and I could take my time.

Here you go https://files.catbox.moe/8v8prp.zip

Not my genre for sure.

Visually it looks like a mishmash of different asset packs but it's not that bad on its own. The bigger problem is that it undermines any sense of world building and atmosphere due to lack of consistency. Interesting ideas are being thrown in with the naming but then all interest goes away because of that. I can imagine that the concept of controlling different enough factions while keeping general cohesion is hard as fuck to do without resorting to generic fantasy.

It seems like a very simplified rts at first, but in turn the pacing can really ramp up when taking over covens one after another while having to manage each previously taken, and my APM is too pathetic. Mission 3 is completely kicking my ass even at slower tempo. The gates are bulky as fuck and can appear literally next to me without me noticing. I guess I'll need to find a Juggernaut coven first to prevent them from sweeping everything.

The tutorial wasn't really needed because mission 1 does a good job. What's missing is some info on buildings - I don't know what they do before I actually place them. Same goes for items, I can't see any information on what they do. This might be extended to units too, not sure.

The fog of war is really ugly as it is - not a deal breaker but because of how it works with obstacles you are forced to look at it in close proximity to the witch king. It's easy to mistake summoning circle unit preview for actual units when you are in a hurry. Normally moving cursor to the edges of the screen moves the camera, but that's not the case when selecting units - this is a bit annoying for large covens. Some prompt texts are too small like invasion gates appearing. I didn't realize through the entire mission 2 that mechs have stances, I just thought the ones with guns out were special unmovable scripted units. Should mechs and dropships be summoned both at once? I don't want to keep producing dropships which then I have to send to death to get back my unit slots. When you have a lot of units selected the unit UI collapses into unit type counts, but the unit icons are all over the place. The launcher window cuts some elements at its default resolution.

The skeleton gimmick is clever, though it's annoying to do because of how aggressive witch king is. Same goes to overtaking covens, I need to constantly execute hold position. The music is great, though I'm not sure it matches at the moment.

tldr skill issues

Unfortunately I'm not too good at linux so let's just leave it at that. I'll try to upgrade the build tools and port over crash logging for the future.

Yeah, no idea. It could be some AMD cpu quirk or a cpuset setting, and the problem is so far below abstraction level I would have to try different compiler settings or turn off thread pool completely. Thanks again for testing.

Hey, what was the exact distro again?

Thanks for testing. Seems like it fails to wake up threads from the pool, but no idea why would that happen. I guess it's time to install arch.

Thanks! Yeah, those are things I have to add next if I decide to take the leap. The trailer has a preview of the weapons (with a few more to come). The second minigun is accessed through a breakable wall below. Seems like I broke the casings for it in the last update, whoops.

Sorry for the delay.

I thought it was unity slop, but it's actually some completely different engine, neat.

The whole 3d free movement format is way too bloated, especially if you are just prototyping the game mechanics. This would work much better as a VN-like with less friction. Unless you are going for specific sort of interactivity that would give you an edge, it seems detrimental on all fronts except for some potential immersion (that would require massive amount of polish anyway).

I didn't get in depth with the game and I'm not familiar with the genre, so I could only judge some technicalities, and all problems stem from being to navigate in 3d and being in early development stage so it might be useless. Having several text boxes and a loading screen before entering each rooms sucks especially since there are a lot of rooms to visit. Camera controls are too sensitive. Exiting the swimming pool brings you to gambling room's entrance.

Sounds like the character roster is critical for the game, so it would be good build up the familiarity and parade the characters on screen as often as possible, maybe even use them for the tutorial. Due to that it feels weird that hanging out would cost energy, but let's see how it plays out when it's implemented.

Thanks, but I'm not sure about the metroidvania part, at least as a big metroid fan. The closest I could get is Demon Crest-like due to being more arcade-y and fiddling with already complex movement would backfire.

The boss mostly checks dodge and mid-air attacks (which most players still can't do at that point). I'll be adding minor tweaks to help with that. He ramps up to match your aggression though and going for nearly uninterrupted full health bar minigun combos can be pretty fun.

Alright, the mp system makes sense now so there's more decision making for sure. But holy shit, I just had the worst streak of unlucky hands ever, and no amount of mp could help me. No idea how could I have beaten this first try with no skills.

Current setup works mechanically, but there are probably more interesting ways to make this twist. Right now it feels like fighting against the community cards rather than the npcs, especially since it's asymmetric. If you are going the game-y route and slowly become balatro/solitaire battle then it should be fine.

Oh wait, the mp isn't per whole game but per hand, holy shit, nevermind. I won tricks by using no skills because of being too conservative with it. I'll replay it tomorrow.

It's interesting for sure. Tricks have too much bullshit going on, to the point it makes me instantly care less about what's going on and about using my skills because the community cards can all completely change. It could turn out good after some tuning though. What I hoped to see was trying to bridge psychological aspects into single player and giving some deduction tools to the player.

I like the commitment to being a doom-like and removing head shots.

The basic shooting enemy is completely broken and does no damage, so sadly I can't judge much about the gameplay in the end.

The city level is sprawling way too much. At least add a map if you are doing that. I gave up after searching for the green door for god know how long. It also lacks visual clarity with all the darkness and countless small point lights. The hospital is a step in a good direction though, as it's recognizable from a distance. Some collisions are broken i.e. with stairs (one can be clipped through into the side, another makes you stuck). Level design has a lot of good shit going on though, so I believe you have the most important stuff figured out.

I think being able to shoot the pistol by holding left click is a mistake, although I'd have to retest it with fixed enemies.

How the hell does it run so slow? My gpu might not be an rtx, but below 50 in the tutorial is a bit... Thankfully the visual style still works with lower resolution.

Damn, those might be some of the best controls I've seen in an agdg game (played with a keyboard). Pretty much no fighting with the controls. Did all the bones and the gauntlet no problem.

Butt slam might be a bit too fast, as it looks almost like a teleport. Swimming animation can glitch out by butt slamming in shallow water. It's easy to panic when doing aerial maneuvers, so air dash could use more frames where the character is stopped mid-air and you can aim the dash direction, and the duration could depend on the dash key press duration. It's not really necessary though, at least on keyboard because of how responsive it is.

You totally have to add claw marks on walls when sticking to a wall and fumbling down, shit's going to be hilarious even when failing.

Unless you are adding more mechanics it looks ready for pumping out assets and content.

Funny how the barrier skip you immediately discovered almost makes the entire level doable gunless. Thanks a lot again for streaming, I had a blast.

I'm not quite feeling it. Technically it's all well made but feels overcooked with all the complexity being thrown into the mix, especially level up upgrades, before letting me grasp and enjoy any nuances of the healing combat. I can't quite come to terms with the camera where I can fully see the back line and the front line isn't tiny, maybe the fov could be lower.

Works well and looks nice.

It lacks a bit of visual clarity but it's fine except for some confusing tiles that i.e. look like one sided platform but actually doesn't have collisions at all.

The platforming is good but needs more time for me to get used to. Jumping straight into timed platforming is an instant progression wall. It would be better to simply be able to fuck around with all the new mechanics in a more relaxed setting rather than putting a confusing sequence of platforming right after introducing sliding and platforms. My WSAD hand isn't good enough for platforming this precise and arrow controls aren't comfortable with mouse either so I would need way more time to get good at this game. I would prefer showing cursor XY cursor position during the target jumps as it's less confusing and allows for faster and intuitive angle adjustment. Arrow up doesn't open chests, only W does.

The game is way too quiet, at first I thought it lacks most sound effects until I boosted the system volume x3.

The secrets are really neatly hidden, especially the water one.

I crashed while playing:

-----------------------------------------------------------------------------
Date : 03/08/2025 14:46:21
System.IndexOutOfRangeException
Message : Index was outside the bounds of the array.
StackTrace :    at TangEngine.AudioClip.AudioUpdate(AudioStore buffer)
   at TangEngine.AudioCallback.ProcessAudio(AudioStore buffer)
   at TangEngine.AudioE.BufferCallback(AudioStore buffer)
   at TangEngine.AudioE.Callback(Void* stream, UInt32 framesCount)

But don't tell gedg you don't have a linux build.

Feels really good. Screen shake didn't look as nauseating while playing like it did on stream. I managed to freeze the game when doing over 10k combo. The additional hit stop (also known as lag) is working well in letting me know I'm doing good.

The large bouncing combos are the biggest appeal but also the biggest problem if you would go for gameplay depth instead of dopamine pumping.

I'm not quite sure what the health is for.

Thanks a lot for the video! It will be helpful for me.

I'll try as soon as I can wrap my head around it. As of now I have an almost full revamp ready with better graphics and I'm gradually planning out an improved concept.

Movement reminds me of my game more than I'd like to admit.

Air control is really good, but as opposed to that having only one option of starting from the ground doesn't exactly feels right, especially if you are about to jump into spikes.

Glass not changing the direction is just too funny, great idea.

It doesn't exactly bring out the full charm of bumble bees. It needs some buzzing, funny leg movement and tall flowers bending under its weight (catapult prop?).

Feels like a complete game overall, very polished. Last stage was a neat challenge.

Kino progress. Now you need to start pumping out gameplay.

Physics can sometimes glitch out. You can get stuck while standing next to the body of water next to a cave exit with some wind, until you blow so much of it, it rips you out of there and throws you across 20 meters. Some wind movement doesn't ever gets dampened and stays as perpetual vortex.

Thanks for clarifying, that's true. I biased the system to make player fight mid air too instead of clinging to the ground to prevent that playstyle pitfall. It kind of worked as intended then, and now I'm being dumb because I started expecting to have it both ways.

I'll experiment with dashing while working on the flamethrower, but I doubt I'd like it as a main mechanic. Momentum acts as a reward for skill and confidence, so it's never expected but rather encouraged (outside of the practice stage).

It works fine. The biggest problem for me is visibility. Shadows, buildings, fog, barely shaded enemies, crosshair, window cracks and fortress mode on top become noise to me as I mow down enemies by the health bars alone, and spam sprint to preemptively avoid missiles. It feels like audio could also be better.

At first I didn't even notice the damage taken indicator because of all the things going on. Sometimes I get teleported or turned around instantly, I'm not sure what's that about.

Thanks. The setting is completely separate, but they share some themes while branching out to do different things with them. I see them as the opposite of each other.

Does the horizontal on-ground recoil feel slow or did the tutorial fail? It's a bit gimmicky but it adds to shooting direction decision making and works around mech animation restrictions. There's also a variant of it with holding down direction, but that's not tutorialized yet.

Missing a parry can be really bad, though that's mostly the case when you are playing aggressively. I'll do some more testing with the bosses - optimally the snake boss gets constantly harassed, but the acrobatics involved are insane. Theoretically taunting could cover for that in two ways. Gameplay can actually be far from methodical, as high momentum and aggression is viable without any limits - this is why the style meter works this way.

How's your demo btw? I hoped to see it this DD.

It has some potential. Mechanics are simple and it's not a matter of gameplay tutorial, but the power mechanic scaling needs an in-world explanation to make it perfect, something like power grid load. Other flavor texts would be nice too, like selling power could be skimping out on lighting, and you could even make it affect the background art to let you directly see the result of your actions. With minimal effort additional flavor text could push it from doing semi-abstract actions to actual immersive storytelling.

I don't think the hover tooltips should have a delay. Time is critical when you first boot up the game and quickly try to learn all the resources and buildings. It doesn't bother after that initial stage though. Buildings could have their names always visible.

The game could scale really well with new mechanics since you can get away with anything as long as it's intuitive (anything business or industry related). It doesn't feel like there's much complexity right now so a greedy strategy works just fine.

I couldn't change the game mode in the options so I assumed only one was implemented, but it actually is random after booting it up again.

It really needs attack and hit sounds. It's like all the juice went into the pretty particles and bloom to compensate.

There's a skill with regen icon and ice aura description.

The game takes a bit to get going, but once it does the waves are fun, although it's still not that challenging. I like having a bossfight at the end, but the summoner is too simple as it is. I expected being gatekept and build-shamed by it.

The bomb character is kind of meh when it comes to the basics, but it eventually does get the job done with bomb spamming. Animation bugs out when you press both mouse buttons.

Thanks. Levels have hand placed casing absorbents to avoid blocking the path. I'll make a dedicated shield holding animation for the boss to make the vulnerability visible.

Thanks a lot for the video, it will be plenty useful. There are some subtle things to learn, like some ledge spots not being obvious.

On windows I can only see a white screen and the console. Can't call doscript due to an error. As it is, it would be cool to have vsync on and key repeat.

Works fine this time. Can't tell much about shaders, but the floor had a nice subtle specular if I'm seeing it right.

I had quite a bit of trouble building on ubuntu. It lacks -ldl flag to build properly.

TARGET_LINK_LIBRARIES( bengine dl )

I'll come back to it later to try building it.

Thanks, I'm glad you liked it more this time.

I fucked up big time with the controller, I hope you didn't play with it all the way through with it because the kick doesn't even work properly as I forgot to rebind stuff. After some testing I decided to change movement to the left stick (at least until I find a better way for up/down input effects).

I need to be careful with bullet mode (and other recoil movements), as it can make the controls feel unresponsive in a jarring way when it activates by accident. There's a lot of leeway with preserving momentum and you can simply aim down for some time, but I think the practice level misrepresents it a bit - optimally you enter and exit out of it quickly to mix up with other moves giving you time to think and rest.

Something like that. Level is subdivided into pillars for simplicity and performance.

Yeah, what I meant by that is that I wish mouse controls were back, and keyboard is 100% viable but not exactly fun to control as it is.

Made it to wave 26. I might be full of shit seeing how far I got this time, but keyboard controls feel worse than mouse controls form last time. There is still some weird movement in certain angles, but as long as you don't do anything too crazy and only press keys for very short time it works pretty precisely (though I would prefer doing smooth flowing motions rather than constantly correcting my course).

Same problem with the font as below.

Does odin work with absolute paths to the point of embedding them into code, or something?

Works perfectly fine. The AI is decent, and seems like the background is subtle enough to not give me motion sickness.

Nice minecraft reference.

Thanks for the info. What does it look like at 1440x1080 on your end? There didn't seem to be anything wrong when I tried it, although it snaps left on fullscreen (well, my screen doesn't even support it).

Is there an error_log.txt for the crash? What system? I only know one other case where that happened.