-
Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
Hobby and Indie Game Dev |OT 2019| Designing, Drawing, Coding, Composing, Everythinging
- Thread starter Popstar
- Start date
- Status
I didn't even know I was showing up in the hashtag searches until you said that and I went to look. So that's how twitter works!
Thanks for this, I'll try it. Created a private account just to post videos and see how they look, lol.
This is a nice thing to hear.
More on the way! When I get something else done.
Thanks guys. And thanks for the earlier advice on Twitter, was stuck in an obsession loop for a while.
This already looks like it's getting fun to control. I've never worked on a 2d platformer, but it looks like you've got the right idea! Do you have a particular game idea you're working toward, or just learning about using the tools and making a platformer?
Oooh, smart, didn't know you could make private accounts! I normally tested stuff by simply posting and deleting the tweet afterwards, but it probably pops up annoying notifications on followers, so your method is much better.
If you come across any tips for improving twitter video' potato quality, please let us know!
Looking for free texture tile set that looks cartoonish like System Shock 1 Interior and Exterior.
Learning, I'm not interested in selling because I'm still a beginner (programming) and today I've decided to make my player "play" on a map and seeing a white wall, white floor makes everything boring :)
I know it is not right selling something I didn't made, so don't worry I don't want to be remembered like these people that "steal" assets and sell their product.
I give you my word about that.
Yeah I'm just asking because I wouldn't want to point you toward copyrighted stuff if you were working on something you want to sell.
When it comes to looking for materials to get yourself inspired while you learn, I would steal whenever you can. Steal mercilessly. I think it helps learning a lot when you can take something apart you really love and understand how it works.
If you google "SS1 Texture Families" there seem to be some collections of textures from System Shock you could download and play around with. A LOT of old games have communities that have taken the game apart over time, converted the textures and models to more common formats (often they're stored in some custom format on the game disc).
I most certainly will! My current strat is to run my game in 1080p and then record a smaller 720p box with ShareX (not the full screen) so things look better on phones. That seems to compress the file smaller than anything else, and gets the crappy-at-first-then-decent results you see on the tweet. That's all I got :)
Thanks, so after I learn programming I guess learning how to create texture is also useful :P
I mostly record at 1080 (with nVidia's own tools), crop the duration as needed, and pass it through that site, which resizes the video itself. It seems to result in better video quality, but it could entirely be placebo, hahah. I also pick 30 FPS, as the default is 24.
In my case, since it's a pixel art game, the difference between low quality and high quality video is the difference between "what the fuck is that" to "oh, its a game!". :P
ugh i'm running into a bug where occasionally i get stuck in walls... must be something to do with my collision checks, especially since I didn't put in my 'shove yourself out of a wall if you get stuck' code yet. I bet my x movement is higher than 2 and my collision checks are only currently checking for x+1 pixel movement right now.
Very likely! Thats one of several collision issues people run into early on. Look forward to skipping through walls because you're moving too fast too!
If you want a place to start for a fix (and this is by no means comprehensive), rather than doing X+1 for your checks, do X+Entity Velocity (in the appropriate direction).
I'm using a 2d manifold to deal with this in 3d. Entity movement is "locked" inside the manifold and can only move on its vertices, edges and faces. This has some really nice properties, like making tunneling impossible regardless of velocity. Probably not the best fit for a 2d platformer though.
Last edited:
That's an interesting concept! Does it integrate well with physics libraries/engines, or is your game one that doesn't rely too heavily on physical interactions?
I spent a few hours yesterday and this morning drawing a sky backdrop for some of the levels. I'm much, much happier with how things are looking now. I'm not sure if I'll need to draw a few more of these for additional levels, as I imagine it would become quite obvious after a while if the same one was used again and again. Maybe cycling around 3 or 4 would do the trick?
I even made a little 'arty' screenshot outside of the level bounds for my own enjoyment.
I was making a shmup for a while but my heart wasn't in it, but I like brawlers a lot more so I was going to try that. I just hate z-level brawlers (which sucks because that's basically all of them). I actually do have a concept with this one but as usual I'm really just trying to get a framework for something where I can just start churning assets haha.
It's nice though, 3-4 years ago getting this far would have taken me about a month, now I'm up to a pretty decent looking character running around with a state machine and enemies in 2-3 days. I'm gonna try and implement hitboxes tonight... that's gonna be a pain.
It is really good to be at the stage though where I can just grab some assets off itchio and rip apart their gml projects to see how they do their thing.
Thanks! I think that'll work... I realised I'm calculating x and y collisions differently i think...? I'll have to wait until I get home to check.
can you make a few that you could apply some basic effects to that will make them look different, or make a few with slightly different colour schemes?
Assuming it works for your game, you might want to look into "Axis-Aligned Bounding Box Collision" (AABB), it's fairly standdard stuff for 2D games and will probably give you a nice foundation to work on for when things get a little more complicated!
What tools are you using to make it?
Ah cool, I'm sure there'll be a wealth of resources and tutorials out there for collision related things in GM!
Re: Colour schemes: I might try messing with the cloud colours a bit but I don't think I'll deviate much from the sky colour. Might just make a couple more images and then do the old "horizontal flip" trick to make it look like there are more
Why not make a non-Z-level brawler? You would have a competitor count of about one (Ninja Warriors Once Again). That sounds like an untapped market to me!
Anybody else here going to be at PAX Australia next week? Swamped with making new business cards, putting together a press kit, emailing media outlets, business business business etc.. I just want to get back to work on the game!
I put together a trailer which was, not exactly fun. Video editing software looks so much like audio, but behaves nothing like it. Getting to grips slowly. Anyhow, this is what I have at the moment. I'll be contracting somebody to do the next one haha.
I put together a trailer which was, not exactly fun. Video editing software looks so much like audio, but behaves nothing like it. Getting to grips slowly. Anyhow, this is what I have at the moment. I'll be contracting somebody to do the next one haha.
wow, that sounds awesome! If i can get away from the booth I'll stick my head in.
Hey guys,
I submitted my game to IGF and I've finished what I think is the first build. I've been working on this for over year now and I'm quite excited as worked on it alone. I start so many projects I don't finish so I'm so glad I was able to finish this up to a point. I guess it's like early access.
This is the trailer fr anyone interested. I dont mind sending ERA members a free copy to test. In fact I'd love to get some feedback from some ERA members interested. There was someone who showed interest above so I'd send you a copy.
it's easy to start something new, difficult to finish.
I submitted my game to IGF and I've finished what I think is the first build. I've been working on this for over year now and I'm quite excited as worked on it alone. I start so many projects I don't finish so I'm so glad I was able to finish this up to a point. I guess it's like early access.
This is the trailer fr anyone interested. I dont mind sending ERA members a free copy to test. In fact I'd love to get some feedback from some ERA members interested. There was someone who showed interest above so I'd send you a copy.
it's easy to start something new, difficult to finish.
I guess it depends on what you mean by heavily relying on physical interactions. As long as the final positional update is handed to the manifold system things should be ok. Another place this system really helps is with debugging. Since I'm dynamically generating a manifold in real time I can also draw it out for a perfect visual representation of reachable/traversable areas. If an unintended or forgotten collider is blocking the path then I'll see it instantly and likewise if I want to place some new objects into the scene I'll immediately see how it affects the traversable area.
Last edited:
woo, i figured out the wall collision. It turned out I was already checking the x+hspd or y+vspd so that wasn't the issue... the issue was I had a 'while' loop to shunt me to the wall as close as possible for so long as I wasn't actually up against the wall, except I was doing that by directly setting the x coord instead of just adding to it.
I still got stuck in a wall doing testing afterwards but it seems to have fixed like 90% of the common occurrences, which is nice
Anyway now I have to figure out this hitbox code.
/edit oh my god i spent 40 minutes debugging trying to figure out why or if the hitboxes were spawning or if they were invisible or what... it turns out because their width is determined based on a var that tracks which way an object is facing, except instead of storing it like I should have been doing it was determining it live based on the hspd of an object... so if the object was sitting still it was at 0, so the hitbox had 0 width... argh...
anyway, i fixed it
I still got stuck in a wall doing testing afterwards but it seems to have fixed like 90% of the common occurrences, which is nice
Anyway now I have to figure out this hitbox code.
/edit oh my god i spent 40 minutes debugging trying to figure out why or if the hitboxes were spawning or if they were invisible or what... it turns out because their width is determined based on a var that tracks which way an object is facing, except instead of storing it like I should have been doing it was determining it live based on the hspd of an object... so if the object was sitting still it was at 0, so the hitbox had 0 width... argh...
anyway, i fixed it
Last edited:
I hope someone is, just realized today that they were gathering dust in my Humble account :p
i have a dumb question about spritesheets for animations where a character's feet or grounding moves as part of the animation set. Say I have an attack string like this:
As you can see, the sprite character moves during the moves, and so should the origin of the character. In the state I'll apply the required level of hspd so the object will correctly move, that's not an issue.
I know what I need to do to get the animation correct, i.e. move the sprite frame by frame in such a way that the origin is correct relative to the character for every frame for the spritesheet. Practically speaking though, what is the best way to do this?
As you can see, the sprite character moves during the moves, and so should the origin of the character. In the state I'll apply the required level of hspd so the object will correctly move, that's not an issue.
I know what I need to do to get the animation correct, i.e. move the sprite frame by frame in such a way that the origin is correct relative to the character for every frame for the spritesheet. Practically speaking though, what is the best way to do this?
i have a dumb question about spritesheets for animations where a character's feet or grounding moves as part of the animation set. Say I have an attack string like this:
As you can see, the sprite character moves during the moves, and so should the origin of the character. In the state I'll apply the required level of hspd so the object will correctly move, that's not an issue.
I know what I need to do to get the animation correct, i.e. move the sprite frame by frame in such a way that the origin is correct relative to the character for every frame for the spritesheet. Practically speaking though, what is the best way to do this?
I have no clue about Game Maker, but what I do in Unity is, I never move the character during the animation, but move the position of the hitboxes relative to the character (this is done by the animation itself, since in Unity at least animations can affect things other than the current sprite). So technically the character object is still at the same point the animation originated, it's just that its sprite looks like they're further forward, and its relevant hitboxes have moved as well. When the animation ends I "teleport" the character forward at the exact same frame that the character returns to its neutral stance.
This is far more flexible and less headache-inducing than trying to actually move the character during the animation. You don't need to know or care how many pixels the character moved between each subsequent frame; you just visually adjust hitboxes to match the sprites (which you would need to do anyway, as attacks hit e.g. at different heights) and you're done.
If this sounds confusing, I can record a video of how I have it set up.
Again, this is probably one of the things where Unity is very different to GameMaker, but the idea is that I'm moving the collider (hitbox) forward with the animation (specifically, I'm changing the offset between the collider and the character object), then when the animation ends, the collider snaps back to the default / idle offset (typically zero) at the same time I teleport the whole character forward, so the collider itself remains in the same place. Since the checks (done by Unity itself) are performed on the collider, it works fine as long as the collider offset snap back happens at the exact same time as the character snap forward.
Hi all,
I have been on a crazy marketing spree for 2 days. 2 of the articles might be of interest to gamedev part of ERA and it would be interesting to know your thoughts.
One about difficulty on the Playstation blog: https://blog.eu.playstation.com/201...v-talks-difficulty-player-chosen-checkpoints/
One about the focus on enemies variation on the Xbox Wire: https://news.xbox.com/en-us/2019/10/01/driven-out-preview/
I have been on a crazy marketing spree for 2 days. 2 of the articles might be of interest to gamedev part of ERA and it would be interesting to know your thoughts.
One about difficulty on the Playstation blog: https://blog.eu.playstation.com/201...v-talks-difficulty-player-chosen-checkpoints/
One about the focus on enemies variation on the Xbox Wire: https://news.xbox.com/en-us/2019/10/01/driven-out-preview/
i have a dumb question about spritesheets for animations where a character's feet or grounding moves as part of the animation set. Say I have an attack string like this:
As you can see, the sprite character moves during the moves, and so should the origin of the character. In the state I'll apply the required level of hspd so the object will correctly move, that's not an issue.
I know what I need to do to get the animation correct, i.e. move the sprite frame by frame in such a way that the origin is correct relative to the character for every frame for the spritesheet. Practically speaking though, what is the best way to do this?
I'm not entirely sure, I actually haven't had to implement such a system just yet! You could attempt to move the hitbox of your character on a per-frame basis, and create new hitboxes for the "attack". That way your character would still be confined by physics, so if they hit a wall the collision detection should prevent them from crossing into it, or if the attack takes them off a ledge they would still fall. I guess it depends how explicity you want to tie the animation to the motion, as it could get rather complicated if you're using physics to drive the character rather than explicit X+Y modification.
Have a little mess-around, experiment a bit! Let us know what you end up with!
Well done! I've seen this around and about a few places over the past few months, saw some people talking about it on Era too!
If you don't mind me asking, what is the process you need to go through in order to get those blog posts up? Im assuming you need to get in contact with certain people?
Nice that you have seen be around! That was the point :-). Sorry I can only discuss the content.
Haha fair enough!
If you'd like a little bit of anecdotal positive feedback, there were several people watching the trailer in the office today, so your marketing looks like it has been paying off!
:-) That makes me very happy. What do they think about the voice over?
Nobody mentioned it I'm afraid, I can ask them if you'd like?
They did say that it seemed interesting though, and it was a good narrative explanation for why the character wouldn't be very skilled as the player starts playing the game for the first time!
No no.. No need to ask. It is just that so many complained about the narrator in the comments on the Playstation youtube page which surprised me since I really like it..
I just released my first ever game and Im quite happy.
To add to this: I just checked and there's a small detail in my actual implementation that's different than how I described it: I actually do the teleport first, then play the animation. So if I have an animation that takes two paces forward, the animation itself (in terms of both sprites and colliders) begins two paces back, and ends at the same position of the idle stance. Then right before I play the animation (exactly the same frame), I teleport my character back two paces. -2 paces (from the sprite and collider) +2 paces (from the teleport) = 0; colliders and sprites begin at the same position they were during idle.
The only thing to keep in mind about this is that if you get somehow interrupted mid-animation (say, get hit by an enemy), if the offset is relatively high, the player will seem to teleport when the animation is interrupted into the hurt animation. The way to solve this is to check the collider offset at that point in the attack animation, and teleport the character that distance in the opposite direction, so that again the overal sprite and collider position remains the same. Again, this needs to be done in the same frame so that the character doesn't seem to "blink" for one frame out of place (and potentially hit things with its collider).
This may seem a bit complicated but it's all done at the systemic level, so that when you add further attacks you just forget about it; you just move the animation sprites to have the last frame match the idle position, and plug in the appropriate offset in the attack information. It's also kind of hard for me to explain without actually showing it, but it's simpler than it sounds, I promise. :D
Ah, too late I had already asked! 😂 They said they liked it, someone said it reminded them of a grandmother telling stories, and it felt a bit like one of the characters from GoT in the first season who tells the ghost stories to Bran. Fitting!
Congratulations! You've done something that many never will, you should be proud! :)
Added a bunch of visual effects and background elements to the game. Let me know if you guys like how it looks.
- Status