Indie game storeFree gamesFun gamesHorror games
Game developmentAssetsComics
SalesBundles
Jobs
Tags

Shweep

185
Posts
14
Topics
472
Followers
17
Following
A member registered Jul 01, 2018 · View creator page →

Creator of

Recent community posts

Thanks! Haha I get that a lot with SIGNALSIS, happy to provide more of what people want!

Thanks! Still working at it.

Game is now released, you can play and vote for it here! https://shweep.itch.io/echo-gunner

Level 3 is kind of just an empty level with some story bits and music left in. Sadly the game is not entirely complete but the first 2 levels have actual gameplay in them. I'm mostly proud of the music Delaney did and hope it wins an award. The music really did set the whole mood for the game.

didn't quite get what i wanted to when planned but the todo list is slowly shrinking . likely the last update before a post mortem sort of thing!

Not too much the report I've fallen behind on some stuff and still have the last bit of level 1 done, but with the last 5 days or so off I decided to slap all of the tasks on the whiteboard and look at it from a lateral progress angle than a vertical one. I tend to bounce between the two but it's hard to get out of the habit of being very vertical and making everything super linearly.



I also tested the game with some folks and level 1 probably needs work balance/onboarding wise but I think worst case I just have a default easy mode where you merely take a penalty to score if the game ends up being too hard. It took 3 weeks to get a solid foundation and level 1 done but I think I'm just going to have do a little bit of corner cutting to get 2 levels done in a week. It's really a matter of how polished/varied the game will be that's the big question mark. I think I just want to live up to the music though as Cowberry's been doing a really solid job with it.

Probably won't be too crazy! But gotta honor the DSD rule somehow.

(1 edit)

So got the first half of level done enemy placements wise. I had to overhaul how enemies are spawned since basically echo location makes its pretty useful to see enemies behind walls but if you see through walls... well.... it'll look weird if enemies spawned out of thin air. So i had to set up parameters like enemies being activated at certain times rather than created. But I think having them already in the scene is easier to manage, I just have to be careful about loading the ones that you don't even see til the camera orientation actually sees them. One of those things where I try not to engage in premature optimization and just roll with it til it's a problem

 My plan is to get the first boss done then flesh out then subway, then test it tomorrow with local gamedevs at the meetup (maybe using the wiimote). Then get the shop done and whatever extras and then gun for the last 2 levels. The first level is likely going to be the framework for the next 2 and I don't know if I can promise the most spice and variety. 

Ultimately I'll have something but question is if it's going to feel complete or conclude in a way that isn't abrupt. 🤞



So got the title screen done and placeholdered the rest of the game. next thing to do is actually flesh out level 1 with enemies and challenges and figure out how it all ties into the shop/abilities. That's the one big problem area is just the abilities ideas I have, like melee attacks or some kind of piercing round thing idk. 


The story is well... the least thought about but I'm going to wing a very anime rules sacred pact plot that uses the friends premise thing. Not exactly sure who's who in the scenario and the scene text in the devlog is mostly temporary, just figuring out how the beats would play out in a typical level. What's different is the text is happening during the gameplay so I have to time and wait out the text (right now it does it evenly regardless of length). Another thing I'm realizing is that I really want a jazzy art deco setting but art deco is slightly difficult to accomplish with a low poly quake level editor. It's not like brutalism where i can just throw in some cement cubes, there's a lot of fancy details the more i research the movement. So it might not come out as much as I'd like or I'll have to find clever ways to simplify the look.

There's a lot to cover with where I'm at and I can feel the time squeezing even though progress is pretty good all things concerned. The lack of enemy placement is what prevents the first game level from feeling complete, so my first priority is that and then reward/progression. The biggest wrinkle is probably the bosses... and the composer is cooking some bangers so hopefully they can live up to the audio spectacle.

Doing something a lil different, taking a break to walk around the local cemetary since I'm planning to make a graveyard level for the second part of the game. LIkely going to "protoscope" the whole game today to get an idea of what needs to be done.
(2 edits)

So I've modeled a basic enemy with vertex colors painted on as opposed to UVs. That one shader i talked about with making the sprites pop out from the background wasn't wasted. The ones on the left of each pair has a custom shader i made to make the lighting affect the colors much less. Making them pop out more from the background.



This isn't an artstyle choice so much as a readability thing. I might be overthinking it though. As long as the humanoids are the only thing moving and the environment is relatively simple. It shouldn't be too hard to point out what's relevant. Having said that it's also why I've been trying to keep the environment fairly Doom wad-ish. I plan to make anything that's modeled to be interactable to some degree, and the rest will likely be baked into whatever I manage in the quake editor. 
 
Before rigging and animating I tend to just throw the models into the ingame scene because I never know what something looks like until it's ingame. The enemies are a bit generic but I'm liking how they're brining back the Virtua Cop influence, the lack of any complicated detail I think will help with standing out. This model is actually based on the protagonist but just puff the chest up to look more stocky and bam there ya go. I'm probably going to reuse the hell out of this model though.

Next up is rigging and animating these fools

thanks! glad it's useful for people 👍

So birthday today and you know what that means! More jamming. Got some character animations in but had to figure out how she was actually going to aim her gun without breaking her back. Then I need to move on to enemies and hopefully effects and such before the weekend. Canada Day is coming up which means a 3 day weekend (a monday after canada day for some reason). So I should spend it wisely to get the game to a vertical slice where everything is in for the first half of the level, then scope out the rest of the game with some placeholders and see where to go from there.

It's hard to fully know when to lay out foundation and when to polish/art things up because the game content just doesn't get worked on when you're an artist/programmer mucking around with art assets. I think my 2d sprite pivot means ramping up the art production side because i wanted to validate the look as soon as possible rather than assume it'll look good after more placerholdering. So it's a tough balancing act and there's no formula to the process I feel.

Hopefully I can get to beyond this one street sections and get rid of these sprite billboards.

(1 edit)

So in a chaos of directions I... decided to use realtime 3D for the characters for a list of reasons, but that meant concepting, modelling/rigging/animating stuff so I hit the ground running.

For some reason i had it in mind for her to be like Iris from MMX4 but then I realized thick long hair is a huge no no in 3d modeling so I went for an anime pony tail thing. Then I had it in mind to go not just low poly but early PS1 segmented limbs and potentially vertex colors. This is something FF7 was known for doing for enemies, which involves assigning colors to vertexes as opposed to UV textures which became more dominant as 3D became more understood. I see it as an excuse to avoid having to do extra skin weighting and get some nostalgia points. I'll likely use faceless enemies with no textures to cut down on any UVing.


Tiringly enough this means weighing each bone to each body part specifically as I'm using Rigify to speed up the rigging process. There's probably a method to auto binding the individual meshes to only one bone each. I still need to figure out the hair as it's one solid piece but it might be more interesting to make the hair into a poly segment snake and animate it manually via programming since the auto rig doesn't provide custom options.

we in business

Thanks! Yeah it was really cool to get the wiimote working, my pipe dream is to usher in some kind of wiimote renaissance via indies now.

In this update I get the first half of a level mapped out (which is like 2 mins of raw gameplay...). And I talk about some of the difficulty with shaders in regards to keeping the enemies visible but also look like they're part of the environment + transparent objects and echo location becoming a problem. Fun times

Decided to make a big change which was make the game third person. I think generally I just want to a do a mix between Sin and Punishment and Wild Guns. Probably will slow things down but I don't know wasn't as bad as I thought and I feel like the game will be less about twitch reactions and more about adapting to situations. Lots to think and worry about though. 

Heya I'll be documenting my dev process here once again, probably uploading progress videos every once in awhile. I do have less time though. But essentially I went in with railshooters in mind and decided to apply the idea with whatever character interested me or would suit a shooting game well. 

Last year I did a simular log https://itch.io/jam/robot-party-jam-2022/topic/2087398/gigantimus-zx-love-never-... looking back I really spent a long time on environments and characters... so I'll be doing a more 2.5D approach and using simpler environment tools. Also the game being more linear and focused should help things. Or I'll just find a way to make that more complicated who knows.

TRIK!!! Adorable and whimsical, love the hit stun pauses.

Glad to know the story was understandable, it relies on a prior context that was brainstormed in a stream. Thanks for playing! 

(1 edit)

Posted 1.1, the last time I will ever update it unless there's some glaring crash to prevent. Will post a post mortem later on today or tomorrow but also the game might be streamed today so that's exciting. Should get around to playing other peoples games after I rest for a bit.

(1 edit)

1.1 Changes Added since release:

Changes - 

-Minor maintainance fixes
-Flowers now have graphics
-Arcade and vending machine have 3D models
-Averagmus spellchecked
-Characters are no longer T-Posed
-Warning messages for alien pirates attack, just generally less abrupt
-Barrier to escaping into the ocean added
-Anime opening added to the menu as sort of a what got cut DVD bonus
-Credits added

I doubt this will affect judging much, but just being transparent here.

Alright so working on 1.1 for the potential monday or later stream just to fix some bugs

However, the anime intro wasn't made but the song is still in the files, I'd like to at least implement the storyboard I had planned for it and play the music so that all that planning doesn't go to waste. I know it's poor form to add stuff after the deadline but I'll be transparent about marking what was clearly for the jam deadline release and the updated version and hopefully judging is based on the jam release sans the anime intro.

The game is in a rough state and is basically unfinished with a really uhhh anticlimatic cliffhanger I will not follow up on. Obviously the game was crushed by its overt ambition but I will likely do a post mortem going over my goals and uhhhhhh what the takeaway from it is.

fixed... hopefully.

actually scratch that looks like a different error???

Yeah found a fix for it, great error to have happen 30 minutes before the game jam. But download should work now.

submitted

https://shweep.itch.io/gigantimus-zx-love-never-rusts

I want to update a 1.1 later and maybe even a 1.2 if a monday stream for it is possible



So just got off work and it's thursday here's my plan for finishing the game cutting up the environment texturing/modeling into 30 min chunks. Then switch to actually making the dang game 8 pm onwards til 3. I'm doing all of this quickly to make any time for the anime opening which might just be characters standing around who knows.



Decided to model two characters at the same time and texture also, really wanted to at least rig them in the end of the day but time caught up to me. A devlog to show them in the game at least:


Little happy accident happened where leaving the ik on lets you spin the waist, since Gigantimus spine just sorta... collapsed it made it easier to hide the twist in the spring part. Which I think adds to the differences between two characters.

Another devlog going into it:

Speedrunning these devlogs, these are hot off the presses:

New devlog don't have time to type:

Against my better judgement I decided to show some irl footage of my work setup and went over some the woes I've been having. Mainly that maintaining a 9-5 game studio job (remote) while making a game is hard to adjust to. Sure it could not be remote and not be computer based but I feel like since both activities are at the same work station it's a bit hard to seperate the two or want to keep working after more work. I also not been getting enough sleep lately mainly due to the cat waking me up at 6am and me getting very fragmented 4-6 hour sleep at best. So really the week days didn't see much progress beyond some work done on the anime intro. Compared to the holiday I did full time on. Today Cowberry is gonna do the rest of the vocals on my decent microphone and I'll be conditioned to get the actual game part of the game done.

I'll report back after sunday if I managed to make do.

(1 edit)


So today was Victoria day (a work day off holiday for Canadians as far as I know, not really in UK for some reason). So me and Cowberry the composer decided to work together in the same room for some good ol fashioned local game jamming like we used to before the pandemic. I got animations done for Avergmus and got them sort of okay blended together in the movement (some things seem to be missing from the run cycle somehow...). Truth be told there isn't much platforming in this, it's mostly an excuse to bounce between VN/dating +shooting segments. Some things came really well together, I also took a stab at animating a storyboard sequence for the anime opening and Cowberry timed out all the parts reall well as you can see in the devlog.


THE PROBLEM, is I forgot that doing ingame cinematics is a fucking nightmare especially when timing it to music. There are multiples ways of charting it out but since Godot's animation editor isn't as robust as say.. premiere where you can scrub through audio, it's really hard to line things up. I had to export the animation to 60 fps and did simple tests to make sure a cut would line up based on the exact frame, but no dice, the frames seem to be off by 7-8. At 60 fps this can get a little annoying so I'm considering at least trying for 24 fps to narrow it down, even if the output will do 60 fps (and EVEN if godot has a really bad measuring method). That's not even getting into syncing into the audio which is a bit of a headache along with the editing.

This isn't even getting into the animation part which I think I can handle. It's just a matter of making the characters and giving them simple posing and getting the most out of smart camera work. Lot of anime in the 80s tend to just have characters standing or walking in place (due to budgets) but also sometimes it was just impressive to animate a character spin around... which is easy with in-game cinematics and full 3D. Anyway you can see me complain about this in the devlog, and also listen to my cat meow to the music almost on beat (also I had no idea she was scratching the door, because I'm hearing impaired and with headphones).


So this is pretty much the run cycle finished... hopefully the rest of the animations such as jumping and landing will be pretty basic. Most of the animations are basic, I haven't had time to really inject the actual character and personality of Avergmus into the animation, but that usually adds 10+ hours to the animation process (and potentially a lifetime of professional experience).


Against my better judgement, I have about 2 hours of footage of me making the entire run cycle (though some extra time after recording was spent polishing aka 3 hours of tweaking graph handles). It's mostly me fumbling with Blender but hopefully this gives some insight on how 3D game animation is made:





Doing some animation work! I tend to be getting into the groove at the start of midnight and then I realize I have work tomorrow and will probably get 5 hours of sleep. Which is probably why it takes me til midnight to get back into it. Maybe this weekend I can re-calibrate.

Here's a devlog showing off the animation process to give an idea on how I do stuff:

I was going to do animations but instead I found this shader on github (https://github.com/paddy-exe/Godot-RealTimeCaustics) and had to try it out, now the water has caustics you can see through.  Though you won't get underwater uhhhh and you probably won't see it maybe I'll find a use for it.



Maybe I'll have water in a beach cave and have that create caustics?

The problem with caustics is that it basically takes place inside of a cube that acts as a cut out, but if a polygon is angled like in the top middle, it kind of distorts due to the cube being basic (to save on processing I imagine), meaning I'll have to model the cave in a very boxy minecrafty way so it doesn't stretch.

But it seems doable! What's the purpose of the cave? Cause I have ADHD and need a way to justify a fancy toy I found on github. Uhhh potentially a dating spot location, yeah that's it.

I finally got around to modeling and texturing the main character, now 2-3 more to go... I'm probably just gonna double down on the first design and not stray too far with Gigantimus. I used the same shader I've been using since... Train Trip. Which was my first project in Godot and wow only one year ago? Just modified the same shader a lot to get the most mileage out of it. It merely takes two textures and celshades between the two rather than just pure color. You can do some neat stuff like the hair shine disappearing in shadow or eyes glowing in the dark wind waker style. Best of all you get complete control of the colors and can even use it with sprites using Material Overrides.



Managed to test out blinking. Because uhh robots need to liquify their camera eyes too you know.



The dirty secret is to offset the UVs by the percentage of total textures on a sheet. Pretty sure they did this for Megaman Legends.


Since most of the visual novel parts are going to be first person, I imagine there will be wayyyyy more faces to use for animations.

Here's a devlog that goes over some of the rigging stuff and how stuff is in motion. I recently got a new computer and started using it, so apologies if audio is a bit different or something is off. 


(1 edit)


Andddd we're doing some character modeling, because I figured I would switch gears and just get a character in the game to also gain some momentum, over the weekend I'll probably sit down and get some actual content in the game, but for now it's easier to just grind out some character models as I don't want them to be too last minute. Time is running out though and my time for working on this is effectively cut in half. Here's another video log to briefly talk about character modeling.


Edit: Also I decided to change the color of Avergmus' face, which movie is this face paint a reference to? It has lots of robots.

So I've been a bit busy as of late, started a new full time job at a game studio and well, have to adjust to waking up early and being tired at the end of the day (also known as being in your 30s). So I've been trying to do some easier tasks like drafting up some character designs: Averagmus, Singuliera and Damian



I realize doing a full 3D dating sim plus an action shooter means LOTS of characters especially if it takes place in a town so... hopefully I can model and rig these fast enough. I didn't bother with side view as I've gotten used to modeling from the front and going from there (yay jam habits). I made these in about 2 hours roughly so uh these aren't meant to be pretty just something to go off of when it comes time to model/texture them. What's left is Gigantimus but you can expect him to resemble Averagmus' design (totally because of motif and theming and not laziness). Don't know how I feel about the steel jaw, it's hard to convey... but I'll worry about it when it comes time to model. The atenna idea is me really wanting to communicate that the mech is going to be remote control, so expect cheesy radio waves when swapping to the mech.

As for the other characters I'm not entirely sure, I might have to wing it. There's also the alien ship designs but I might get a little lazy with that. I really don't want to leave the characters to the last minute like last time when the sprites + designs were rushed in the last week or so. Very smart of me to multiply the work load by doing 3D for this one. No video update I'm just going to sleep.


So I lied and implemented a few shaders and basic colors to the base scene. However I also realized the groundhog day esque mechanic needed to be further elaborated so that multiple different ways could game over and funnel into the scene that Damian keeps rewinding back to. Now the game properly resets and can remember things that carry over between resets. Here's what I mean: 

Basically from the Damian scene you can branch out and explore the town to prepare properly for the date. You can start the date any time you want but without certain knowledge that Damian aquires from your past failures or triggering certain events the date will eventually reset back to the Damian scene. Every time you do an event or progress the date, time will forward. Enemies will invade the city prompting you to defend it, but if you're in the middle of the date you will just have to let the city take it. The city isn't gauranteed to be destroyed in a single invasion but if you let too many invasions happen while you're in the date then both you and Singuilera will perish. 

So the whole objective of the game is to not only prepare for the date but to make sure the date happens at the right timing. Damian will also let you know when invasions happen the first time you encounter them. I'm wondering if the date could be split into multiple dates as well, like you meet up with Singuilara at certain spots at certain times but that might be complicating it a bit.

I updated with a quick devlog to show how the game works as of now: