/agdg/ amateur game dev general

bitches and whores

I made a few new sprites for Moonman mod.

Fullscreen HUD is neat now. An actual proper spriteguy took pity on me and made it pretty.

Isn't the autism logo a blue puzzle piece? The /agdg/ logo is just a square made of letters, which in this case is rendered fractally by some user and was recolored to have the SNES button colours by me.

I can't stand the grind and futility of posting on social media… The only people that ever see my content for the minute it's actually visible under the #gamedev hashtag are goddamn bots.

And I only made 2 damn tweets…

Any suggestions on how to maintain social presence without one's soul withering and dying in the process? Some of you guys must have tumblrs and shit.

Anyway,
I hope at least some of you might enjoy reading my blog posts
beelzebox.net/2017/04/12/heart-on-a-platter-blood-and-gore-part-1/

It's shit and you know it. And a gud alternative would be something Gahnoo+Social based it respect your 4 freedums :——DD

Sounds like a cool idea when you already have an audience and want to create a platform for them to communicate with you. Currently, when I'm known only by the beloved /agdg/ community, I feel that if I want to (((promote))), I need to bite the bullet and start accounts on all your usual "LIKE US AND FOLLOW" cancer sites where all the possible audience is. Hell, even tetrachrome dev has a tumblr from what I saw.

It is the reality of the situation, you'll probably never get much of audience by only advertising on here, because we're just too small. I'd say it would probably be better to even be on reddit at least your potentially audience will be better.

Make a unity tutorial nobody has ever seen before and shill your game in the beginning, I have an addmefast bot, I'll help you rank your vid if you need it.

look, it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue, and it's blue

it will be bigger not better

I just tag #gamedev and #agdg
Also follow a few blogs that look cool and they might follow you back … the trick is to get them to reblog your silly/cool shit so their follows see it

Shilling for each other should be a good start. I followed just now. Let's create a clique! especially that most people under the #AGDG tag are most likely cuckchan users

I guess it's like my wife says… slow and steady. Maybe one of those people that generate traffic at beelzebox.net is a human… That would be some sort of consolation.

Sounds both interesting and shady as all hell. Are those addmefast generated views/visits even people who want to play vidya?

I am making vids for the blog but they are not tutorials. Might make one someday. Perhaps a trailer would work? Or is that too… obnoxious?

wypok pe el

Are there any benefits of having good audience over having large one?

>wypok pe el

artykuł w GWnie
występ w dzień dobry tefauę

Addmefast is mostly bots but it helps ranking your video so actual people can see it higher on the list.

Small audiences need less filtering to get good feedback. Most games with small audiences are good, but as their audience grows the creator tries to accommodate more for the public at large's need for casual gameplay and the wishes of SJWs. Examples are Dwarf Fortress (to some degree), Minecraft, pretty much any interesting new game that gets 2 AAA sequels that Holla Forums tends to hate, YandereSimulator, etc.
Examples of small good audiences are for Aurora, SS13 (I think, never played it myself), Iji, games like that.

>sugerowanie że byś odmówił darmowej ekspozycji jakby Cie Michnik zaprosił

Rejecting the use of normalfag sites is completely pointless, it is just virtue signaling.
Use all media outlets available, just don't behave like a whore while doing it.

better feedback and not having to sort trough the same five or six questions a thousand times each
some OC

I agree for the most part. I can see the potential in all of them. Seems that as long as you are getting to potential customers, it's worth it. Then again, there's the debate you guys are having, regarding the quality of the audience. Sigh…

makes good points. I really don't want blizz fans and koreans to force me to make Beelzebox a papamole D3 clone and it's not out of hate. I just don't share their taste.
But on the other hand, I'm sure there are people among them who want to play a game like the original Diablo- so I want to be noticed by blizz fans and koreans. What a dillema!

And regarding mr. Michnik, I actually gave his newspaper interviews on two separate occasions(nothing vidya related) and in all honesty It's not an experience I'd want to relive.

do you come here often or is there a way for me to contact you should I want to use this method?

sounds like a nightmare

I'm in every /agdg/ thread, just make the video and post it.here. Make sure it's unlisted then make it public once you have a hundred thumbs up.

This is some halfassed sprite I made for a game project. This should be good enough for the project but I'd like to know what I can do to better my drawing? I'm not too good on linearts and I'm pretty sure there's more I can do colorwise.

I'm pretty sure the vid will be posted here first for you guys to dissect.

Did you boost your own project like that?

...

Fundamentals, >>>/loomis/, color theory, etc

Getting better feedback is a good reason but as you audience grows you could always create list of fans that give sound advice and stick to them.
All other drawbacks you mentioned come from creators that are willing to bend over to the will of the audience for the sake of greater profit. Would you change your game for casuals?

For example if your game would get big boost in audience thanks to some meme exposition you either change your game to accommodate them and lose better fans or keep to your own rules and lose some/most new audience but you would still have some additional fans that like your game the way it is. To me it seems like a win situation for both attitudes.

This question of better or worse audience to me sounds more like a question whether the creator has right personality to handle large audience or not.

MSPaint-tier.

I would recommend an old trick I learned a while ago, make a regular drawing, then put a reticule over it, try to "convert" it to the nearest color in pixels.

I can't really start my own project yet, but I have seen people ranking videos like this. The goal is to give the video a proper title and rank it high for the keywords that it uses, that way someone might click on it if your video seems interesting, that's why I told you to combine it with a tutorial, or something impressive (like that damage system you showed us in some previous thread). You bait a few people and send them to a tumblr and then you bait the tumblr and send them to the youtube video to create some kind of feedback.

Who gives a fuck? I'm sure as hell not making my game so normalfags will shower me in praise or money. I'm doing it for the self-satisfaction that comes from making something I can be proud of. Putting my work out there so I can receive criticism to improve is a hell of a lot different to putting it out there to receive backpatting/dicksucking fans like that attention whore cocksucker YanDev.

If you're making vidya so you can get pats on the back you need to head directly back to reddit/4cuck/whatever shithole you came from.

I wanted it to say "your chances of making a game are as real as me!" but i needed more text to cover the original message

Is floodfilling itself considered a pathfinding method, or just a useful technique?

Should my pathfinding shit be static or instance based? The latter would let me apply settings and state and simplify the PF calls to a common interface as well.

Also should it operate on any array of values, or is it worth making my own object to hold a grid?

So is anyone here developing something in unity?

Floodfill is not a pathfinding algorithm per se. In the context of this particular problem, it can be used to determine whether point B can be reached from point A, rather than the actual path. However, since it's not much faster than breadth-first search (which isn't fast to begin with), this is usually something you would do only once when your map is loaded, and cache the results. I posted two algorithms for that in the last thread. They allow you to find out whether the destination can be reached in the first place, before you do the pathfinding. Since unreachable nodes are the worst case (entire map needs to be analyzed by your algorithm), this is generally a good idea to do. However, it becomes massively more complicated, if your maps are in any way dynamic.
Either works, since the actual meat of the pathfinding are a) the algorithms and b) the data structures. If you want to perform multiple queries at once, it would be smart to have a pathfinding class that is instantiated. That way you each entity of your game that wants to calculate a path can create a new instance with its own data, so they don't get in each other's ways.
Not entirely sure what you mean here.

Not anymore, but I do have quite a bit of experience with it (primarily its API). Any questions?

Or if he hires someone that can handle it for him. Both seeing the recent post-kickstarter PR fiascos and how yandev mishandles fan communication by refusing to hire a secretary makes one want to do it better
I really hope I can use my wife's enthusiasm to make her my social media person without it blowing in my face. I have to pave the way tho

This was supposed to be my next blog post (I even uploaded unlisted videos to youtube) but perhaps presenting it in a more "bombastic" way will be the way to go?

quite a few of us, yes

Do you guys ever do collaborative work?
One person does sprites/models, someone else does code, some else does music and SFX?

Anyone getting the Newsground Madness Series Vibes from this?

Well lets say I have Grid which contains a T [] and a Rectangle (for bounding box info to make the array have 2D semantics)

I could either have it require a Grid, making it so I have bounding information and do things like subregions and shit easily, or I can have it require a T [] which makes it more generic but require the bounds/stride

This plus explaining your code, you might end up getting that whole indie horror guys crowd who want tutorial on how to redo it. Try to copy vid related a little and , also post on the unity3d forums.

Thanks, I guess the only way to get better is to practice, I don't know why I expected anything more.
Funnily enough I'm not a drawfag(this is the first actual drawing I've made, and I was shit at art) but I really want to improve at this now that I've gotten my hand going in it.

another thing i used to do before the uber-specific art style i stole prevented this was take pictures, usually from gelbooru, and scale them down in Gimp with Nearest Neighbor, AKA the Interpolation being set to none. it keeps the pixels sharp.
then just touch it up.

theres little reason to really write code generically unless you are gonna use the code generically.

yeah I have been getting that recently. I am still a bit torn whether it's a good thing or bad. As I said some threads earlier, the art style comes from my webcomic that started way back in 2005. While I might have been influenced by my early visits to Newgrounds back then, I hope the style is distinctive enough not to be called a rip-off.

I am told to make promotional artwork like wallpapers and shit, hope I still got the skills to make new strips

good, everyone has to find their niche. Now how to find people who enjoy the same niche, outside Holla Forums? the answer is not one I like

Good plan, thanks for the advice. What I am sure of is that it will require quite a lot of time to achieve, so the blog post will go live and the tutorial video will be done later, referencing it or lying that it's by popular demand I am making it.

My brother makes music, my friend is working on parts of the code, I have an ez time working with them.When you treat a project as a hobby and have relatively reliable people to work with, it's very fun. Start applying deadlines or work with strangers/volunteers and it's all going to crash and burn. and I will have to start setting deadlines, especially for myself, to complete Beelzebox

sorry for reusing content, I will go back to work this Easter

I fucking hate UMG so fucking much. Why is everything so fucking hard in it? I just want to get the fucking position of a widget but no, always returns (0, 0) even though it damn well isn't.

Its probably failing to find the widget and returning a default value in the null case.

Relax user. It's probably quite simple and you'll feel like a retard later on, once you figure it out.
What exactly are you trying to do?

I blame breath of the wild and Holla Forums

Don't feel too bad, I left my game for six months before coming back to work on it.

Wheres screamsinternally.jpg, I need it to represent my lack of skill

Yeah found the solution pretty quick, though I still hate UMG for being so annoying: getting the position of the widget as a canvas slot will return the canvas in that widget, so if the widget is a child to another (which it likely is unless it's your main UI widget), it will return the position inside itself. This does make sense, but the node needed for the actual positions really aren't the kind you'll find without a search engine or looking at (a lot of) Slate code. What you actually need for the absolute position is Self->Get Cached Geometry->Local To Viewport->Viewport Position.

Wish there were a "Get Absolute Viewport Position" node. Maybe I should just dive into Slate and avoid the BP mess.

At least im getting some brillant ideas from botw, best openworld game i played since morrowind

user what about Skyrim tell us about Skyrim

...

The thing about Tumblr, nobody's going to deny the majority of its userbase is shitty, shitty, shitty, shitty.
But it's reeeeeally easy to both filter out the nonsense and follow only the stuff you care for. The only thing I follow on Tumblr are food blogs, cat blogs, and retro vidya.
I already follow Tetrachrome dev, if any of you have anything I'd be happy to follow it.

Twitter is more annoying, because of all the dumbass suggestions that it keeps giving. No, I'm not fucking interested in your dumb trending hashtag suggested-followers shit. Fuck off already.

I remember this shit, straight off after registering to Twitter to fight this very thing.

well, the content I'll post on the social media will be most likely things I show off here first anyway, but giving each other shout outs from time to time for our non-Holla Forums followers If I'll I have any would be nice.

I've got the typical "follow us on…" footer on beelzebox.net/.

oh and that twitter suggestion thing is just too awful, it's nothing but drivel I despise.

@-moz-document domain('tumblr.com') {.alt.takeover-container { display:none !important ;}.radar_post { display:none !important ;}.radar { display:none !important ;}.fullscreen_post_bg { display:none !important ;}.radar_header { display:none !important ;}.right_column{ display:none !important ;} .is_reblog { display:none !important ;}}

Is anyone surprised that MoM user uses Tumblr?

I wanted somewhere to post that wasnt one of the other trashy art sites. I wish they was a good Art Site.

You could always post on Deviant Art and then when you gitgud just move on to Artstation. There's good artists who start on DA and after a while they move to AS or use both.
Look at this guy:
balaskas.deviantart.com/
balaskas.artstation.com/

I dont my MoM stuff is worthy of being in a site like that!

That's when the gitgud part comes in, mate.

burger

Your MoM stuff might not be worthy of being on artstation, but deviantart is not worthy of your MoM stuff.

...

...

I made the cube rotate in vulkan mode a few days ago.

Anyone know anything about GBA romhacking?

I took HxD and slapped it against Castlevania:COTM and managed to carefully look at offsets and corrupt one 8x8 part of a sprite that I planned, but how would you manage things like inserting new graphics, etc?

Would I have to fuck with the ROM and basically build a memory map and make sure offsets aren't fucky?

...

I recall hearing about that being one of the reasons why GBA romhacks aren't that common. Graphics data chunked up very finely and scrambled all over.

By inserting new graphics, do you mean completely new content, or just editing what's already there? Because you could probably find or build a tool for the former to turn it into an easier-to-process form, but the latter is going to get wacky real fuckin' fast.

Well I assume that for any rom. You insert new data at an address, you push all the subsequent data along, and anything that was programmed to a particular offset gets garbled

If social media isn't your thing it's better to accept that and do your blog and have it actually be worth reading. What you can do though is link your updates on a twitter to have it somewhere publicly. Use pictures since your style will surely catch peoples eyes.

Opinions on making a text game?

As long as it has actual gameplay( ie: fail states), doesn't cost a million dollars, and isn't just a book for a computer I'm fine with it.

can I use unity for it?

why would you use unity for making a text adventure?

because is easy enough and the assets store.

why do you need assets in a text-based game?

transitions and effects.

that seems like an awful waste

>Brain says hey wouldn't it be great to work on something more interesting like let's think about that all the time instead
AAAAAAAAAAAAAAAAA

What's your game all about ?

chicks with dicks

Well I almost managed to compile Bullet Physics properly, but I get some error in the example files when it compiles. It gives me an incorrect conversion from int to wchar_t, it can't compile from 16 to 32 bits or whatever, I'm fucking tired, I can't believe that this thing was engineered and coded by professional, it feels like such a hackjob and yet it is used by the top game engine and some scientific simulation. I get the feeling that once it's going to be installed it's going to be a clusterfuck and I'm going to hate everything about it.

How does one import materials into Godot? I've imported the model I need, but the object file for the model doesn't contain the materials. I've consulted the docs but they're pretty shitty.

Just a puzzle game to try and discipline myself into getting something done and out.

Pic related.

Figured it out with enough digging. Apparently you can't import objects into godot and have them maintain their materials. You have to import the object using some collada exporter or something.

I can help.

...

What program can I use to produce music? I only have access to a craptop so it has to be light.

FL Studio does okay for me.
it's not anything resource intensive, i don't think.

easy, just use any tracker like Milkytracker or Sunvox (though SunVox is quite a bit more powerful than most trackers)

or Tracktion or Audacity if you don't like trackers

I'd mention Renoise, but they require a higher resolution monitor (with mine just barely hitting the minimum requirements.

I'm writing a simple scripting language, compiler and virtual machine for game dialogue trees with external state. It's intended for visual novels but can probably be used for other types of games. Also writing a visual novel module for Unity that makes use of it.

Does anyone have any non-obvious suggestions for what a language like this might need to include?

I said light

OpenMPT.

outside of the regular dialog options, a lot of flexibility could be added with some of these features.

I would say that what you're doing thus far is definitely unnecessary. Your variables are dynamic, your if-else system is reliant on labels which require additional lines to continue a script, and you precompile the script into bytecode but if it's just for text, what the fuck for? For a simple script to check user input for gender it's a somewhat long script. Imagine writing an entire game using this scripting system.

You could bind Python or Lua to a C/C++ program to power a game, and it allows you to use a good deal of modules. And they both treat variables dynamically as well. What you've created seems to be an Assembly language hybrid, and I just can't see why you would go so far to create such an awkward script system?

Anyway, if nothing else I would recommend that you implement procedures/functions (whatever you want to call them), and strong variable typing. I never understood the love affair people have with dynamic type variables, a variable changing its type is an inherently bad thing and non-programmers should definitely not be fucking with scripting languages for the same reason.

Consider that Python, Lua and Javascript (all which can be implemented into virtually any program) have so many data structures and paradigms for implementing things, and how elegant code for those languages can be written.

By contrast, your system has so much overhead. Right now you have to recursively search the program once, and that's not so bad, but imagine if you had four or five elif checks.

Have you written your own compiler?

...

how about you add multiplayer support. your not finished.

Is there some music program that is deligned as a drawing program?
You run it, do a few random clicks in a "canvas", and you get sounds?

I wanna make video games

would an easy first project be… a card game? easter egg hunt? deer shoot?

one of those "match 3" things that your mom plays on facebook.

variables aren't dynamic, they're just integers (in this screenshot 'actors' were also variables, but are not anymore and are not mutable). it's meant to allow very simple flags, like checking if a player has either done something inside or outside of the visual novel layer. the program runs sequentially and does not mind if labels or 'asks' fall through, on line 13 the jump instruction is redundant.

i didn't want to use language binding or marshalling because i wanted something that could be easily written and modified by artists and writers who have minimal programming experience. the scope is intentionally very small.

i'm not sure what you mean about recursively searched outside of compilation. it's just jumping to different bytes in memory with an instruction pointer and executing the byte.

you can already easily loop, but i think that signifiying questions that have been 'asked' and 'answered' would be very useful.

chosing a dialogue option can change a varable because it controls flow, so you could compare that variable later. there is also an external variable system that i am working on, my hope is that the user will just have to type import local global and export global local. i'm not sure how globals will be indexed as the individual programs do not know about the state of the virtual machine until they are run, the easiest way i can think of is just to index globals by strings and hope they exist.

that was never a goal of the project, you can ship a game without multiplayer

You just earned my respect in 45 characters
this comes from someone who has given you shit for being an enginefag

Okay, i just remembered a 2014 half-life mod that I abandoned, then didn't try again for a whole year.

2/3 completed, sorry. I've finished everything I started since july 2015, but I only did two projects in that time. My projects take a while.

so your not complete.

...

Its a single player game, I know you are just trolling me, but I decided that it was too ambitious to start learning the WSAPI on top of all the other things that needed to be done.

Been practicing pixel art.

Are you being ironic?

No, I'm using a software called paint.net; for the last 30 minutes i tried to draw a hot woman (pixel by pixel) using the pencil tool, after a couple of tries I took a capture of it ,saved it on my computer, got here and made a post showing my progress.

Can't see how that's me being ironic.

it looks like something you'd see in ResistJam, user.

Well fuck dude, he just started.

here's his participation award.

your pixelart is shit.
git gud

B-but is supposed to be aesthetically ugly!

Here's a better one

Good shit, thank you user.

That's actually pretty informative for game art in general, especially the third one about color. Do you have something similar about making game music?

...

Sorry, nothing specifically on game music.

...

...

...

Gave me a hearty kek over here.

Reconsider this. There is nothing with worse usability than a buggy and limited language and VNs aren't the kind of thing that need their own lang. Why aren't you writing a VN module for an existing scripting language?

This was my entire friday night. Was it worth it?

Disappointed simplescreenrecorder didn't record the terminal cursor. Then again, I had to switch to xterm from urxvt to not get it to fuck up more

Speebot demo debut
It's finally here: drive.google.com/open?id=0BzIMurAb1QxDLWNQd2tBVjVuZHM
Let me know what you think of the game, and please report any bugs you find here or at kircode.com/contact.php
Enjoy!

Downloading now, faggit.

Aaaand it crashed immediately.

I have been waitng for this but i have work until late night today

Does it crash on launch or after the intro or when you play the first level? Have you properly unzipped all the files into a single directory?

It crashes as soon as it's launched. The game's window appears, but nothing ever fills it. It's all extracted into one folder, named the same as the zip file. I tried re-extracting it, just to make sure, but no dice.

Runs for me. Holy cow that's a lot of resolution options.

…mate?
immediately shuts down after this.
files are in the same directory, tried running as admin.

Here's the error message, if it helps.
Problem signature: Problem Event Name: BEX Application Name: Speebot.exe Application Version: 0.0.0.0 Application Timestamp: 58f2450d Fault Module Name: StackHash_0a9e Fault Module Version: 0.0.0.0 Fault Module Timestamp: 00000000 Exception Offset: 00000000 Exception Code: c0000005 Exception Data: 00000008 OS Version: 6.1.7601.2.1.0.256.1 Locale ID: 1033 Additional Information 1: 0a9e Additional Information 2: 0a9e372d3b4ad19135b953a78882e789 Additional Information 3: 0a9e Additional Information 4: 0a9e372d3b4ad19135b953a78882e789

I suspect it's because I'm running a shitty laptop from 2009. That said, I would expect a toaster to run a game that looks like that. I was particularly looking forward to it, because it reminds me of Mega Man Legends a bit.

What GPU do you guys have? The engine requires at least OpenGL 3.3 (2010)

pic related.

- I can use my controller for menu input but it seems to default to keyboard for play. I'd recommend defaulting to whatever input the player is using to move through menus with if you can't switch on the fly (which would be more idea).
- Doesn't seem to support analog stick on my controller (PS4).
- Personal preference but I'd like to be able to skip the zoom into level start up animation by pressing something.
- Circle/B for the drone view is a bit weird. Maybe instead use a shoulder button? I think it would let you use X/A for confirmation and O/B for back/cancel then.
- Ideally pushing cancel/back on the controller after bringing up the menu should cancel out of it. "Resume" should be highlighted by default too, so the player doesn't need to input a direction to resume play.
Still playing so will give more feedback as I go.

I'll try running it in Wine 2.5-staging.

Quickly became furious with jumping to platforms on the sides of things. 9/10 it's alright, I'll probably play more a little later on.

Holy fuck user this is really good

Seems that's the problem, then. I've got a generic on-board Intel laptop potato chip.

The game runs really smoothly on Wine, even if a Linux binary would be ideal.
Are you planning on selling this eventually?

Seems like a quirk with newer GPU models, I may have fixed it, if not - there should be a bit more detailed error message - could you try redownloading the game (same url - drive.google.com/open?id=0BzIMurAb1QxDLWNQd2tBVjVuZHM) and running it again?

Noted, thanks, will look into improvements a bit later. Regarding the analog stick - does the Gamepad Movement Mode option have any effect?

Might make that spot in 1-14 a bit easier if it's too frustrating.

Glad you like it.

2.1 is 2006 tech… I thought if it's at least 3.1 (2009) I could maybe try making a downgraded version for it, but OpenGL 2.1 is so old that I'd have to rewrite the whole rendering system. Sorry, user.

Good to know, and yeah.

tried a redownload, still the same error , mate.
do work on it though, from what i've seen, it looks like fun.

- 1-11 is a bit of a spike compared to levels around it.
- Maybe add an NPC at the start of 1-13 to tell remind players power ups only last the stage they're picked up. I like how NPC text blurbs aren't over the top or trying to be too goofy too.
- Ah I missed that gamepad option. It's probably simpler to just accept both d-pad and analog by default with a sensible deadzone.
- Analog stick is still binary input so feels kinda clumsy.

Getting the same shader error as well, Win7 x64 and R9 270x GPU

this is incredible. i love the light hearted art design, and the controls feel so fluid.
keep it up, this is amazing work.

Are you sure its the same message? This time it should print the name of the shader in the error. Maybe it's Google Drive fucking with file versions, try this link: drive.google.com/open?id=0BzIMurAb1QxDLWNQd2tBVjVuZHM

Thanks again for the feedback.
Analog stick is not binary input. There's a sensitivity slider in the options that controls the deadzone, I suggest you lower it if it feels sticky.

Thank you!

Wrong link, try this one please: drive.google.com/open?id=0BzIMurAb1QxDYy0tUmhKQjFYems

my bad, has the name as well

Didn't have too much time to dick around with it, but I played for several levels. There's a lot done damned well, but the audio stands out as just not good. From the little things like some sound effects having too much low end (clicking any button in the menu is fuckin' heavy), to clashing sounds, mostly SFX with music, to the general composition and choice of instruments in the music, to the mix and master, or lack thereof. It feels like shovelware audio.

I'll mull on the other details once I get more time, but everything else questionable seems to be either easily alterable later, or just a matter of polish.

Fuuuuuuck
yeeeeeeeah
niggaaaaaaaaa

Will give a shot, been super-looking forward to this.

Thanks, will look into it…

Thanks, will try to improve that…

do you plan on open sourcing YUME any time?

I personally really liked the sounds, although navigating the menu with the "next" and "back" buttons was kind of confusing.

What I would look into is to use a shadow that is directly under the character to make it easier to understand where you are actually positioned when you jump.

For me it just werks.

t. gahnoo/loonix user

By binary I mean I can only seem to move at full speed, no matter how far I tilt the analog stick (as long as it's over the dead zone), making any fine adjustments with the analog stick rather hard.

It's a lot of content for a demo build, you might want to consider splitting some of the levels up to different themes/BGM. I thought the audio was okay but the music begins to grate after a dozen+ levels.

On level design: I think a few too many crystals were 'you've done the stage, now jump down for a crystal and redo it'. A bit more variance in crystals might be fun, like some are on the level completion route but tucked behind a side and more obvious in drone view. That's just my personal opinion though.

Fun stuff though! I hope Mark didn't steal your cake.

Every 10 levels has a different song, user. Themes probably change per world, so 2-1 will look different.
Currently on 1-17 with all crystals so far, really having fun.

Oh my bad, I probably missed that since I was tabbing back and forth typing shit. Just finished with all crystals.

Got this problem as well.

Tried this too.

Don't cry because it over, smile because it happened.

One minor problem I found is that if you hit a directionless spring after a directional spring, you appear to have pretty much zero control in the air. This means that if you try to go fast in 1-5 by landing on the spring after the first jump, you can't steer backwards had enough to save yourself. That's not very Spee. Time trails when?

Can you try the new version? Same url: drive.google.com/open?id=0BzIMurAb1QxDLWNQd2tBVjVuZHM

I don't plan on it, but I can't say for sure yet.

Thanks, noted. The next and back buttons appear automatically when the current resolution can't fit all the buttons. I've thought about using shadow blobs for the characters, but I don't think I want to change the way the shadows are now.

I'll make a native linux build eventually too, but this is great to hear!

Thanks. I'll experiment with analog sticks a bit more later. There is a different song for each 10 levels as the user below you said. Regarding the crystal placement - I had the same thought when designing the levels, so I avoided that pattern after the early part of the game. Crystals eventually become more of a puzzle element when new mechanics are introduced later in the game.

Glad you liked it. Yeah, I'm aware of that, but some later levels in the game use that as a mechanic, so I guess you can call that a feature… Might add a speedrun mode later if I think of a fun way to do it.

It works!

works on R9 270x now

Great, let me know if you get any performance issues or visual glitches.

runs pretty good for me, at least i havent noticed anything glitching out.
though i do find it weird that the mouse cursor always stays on screen during the gameplay, where there isnt any use for it.

It should be invisible, that's a bug right there. Thanks.

wew

maybe this will help: the cursor is only visible in windowed mode, borderless and fullscreen show the cursor during gameplay.

fuck, meant to write that the cursor is only HIDDEN during windowed mode, shown in other modes

Is reaching a spot like this intended?

Ideas:

...

I'm sure this will help, thanks.

This particular spot is useless, so I might block it off, but I've deliberately left some tricky shortcuts in other levels for speedruns.

...

Thanks, good points. Noted.

Does the robot dance when finishing a level?

Also, you should be able to pick Speebot's gender.
Picking female just adds a generic and ridiculous curly blonde wig to the top of the model

He jumps and waves his arms around

Let's see, what else…

One small bug I found is that my controller option to user D-Pad or Left Stick has gotten flipped, I now select D-Pad for Left Stick and vice versa.

Guess I'll second this. I've found it difficult to judge Spee's position in the air, in particular along the north/south, x axis. Even though I'm personally willing to accept that as a part of the game's inherent difficulty, I just want it to be noted that I do notice as well and it is a source of frustration.

Otherwise, I'm really enjoying the demo. Good work.

Oh, also you can set all the inputs to use the same key. Should fix that

also i noticed I can't bind anything to mouse buttons I would like to bind jump to left mouse button so my right hand has something to do.

I rebound movement to arrow keys and jump to z.

Thanks, all noted.

Maybe those should be the default controls?

I'm gonna stream game development for a bit this time on youtube since it seems to work better with my shitty internet youtube.com/watch?v=s3klDxnl9rQ

Rebind menu in windowed mode cuts off some options

What do you mean? The Back/Next buttons don't work?

Those worked fine for me. I'm getting an annoying stutter. My fps is usually ~170, but every 5-10 seconds I'll lose 20-30 frames, and it messes up the feeling of the controls. Also, do you have a thread on /agdg/? Feel like we should do this kind of debug posting there.

Is it not possible to have multiple inputs mapped to the same control? Like on a controller you shouldn't have to switch pick just dpad or leftstick. Simple games often have wasd and arrow keys mapped to movement simultaneously as well.

Having the same issue. Your GPU probably only support opengl2, and not 3.1 or whatever is being used.

Oh I see. I'm on the graphics menu and there's like 14 options or so, and they get smartly sorted to the next page based on screen size… It just feels weird to not have all the keybinds display at once, you know?

Also consider having the back/next buttons on the top corners or bottom corners, maybe relabel them to prev/next, and then have a back/accept button? I dunno, just an idea.

Yeah, I'm the same user as but my IP/ID changed. I updated my drivers to see if it'd get me anywhere, and it now gives that error message, instead of just crashing immediately. Guess I'll be building a new computer sooner rather than later.

No need to apologize. If you don't mind me asking, did you draw on any particular sources for inspiration on the visual elements of your game?

nah man a pink wig

So it's also pretty easy to dick around with your save data

Stutter is bad. Can you check task manager to see if it's eating up a lot of resources? I don't have a thread but I should create one next time, didn't expect this much valuable feedback, sorry for derailing.

Since keys are remappable there's only one key per action. The dpad-analog switch actually exists for a reason - if someone's analog stick is loose it will dispatch input even though it's not touched. In other words, a guy might be trying to play using d-pad, but the character would move unexpectedly, which is actually the result of a loose analog stick. There's a deadzone slider in the options for this problem, but this kind of behavior might not be obvious to an average player if it does happen. I'll see if I can think of something else for that.

I'll find a way to make the Back/Next menu obviously different from the regular items.

Old school Runescape, but it quickly turned into my own art style as I introduced new stuff like dynamic shadows, reflections, etc.

Yeah, I didn't bother adding any obfuscation to the file format.

I use WASD and J, which is my usual setup for platformers when using keyboard. K/L for shooting/extra.
I did try controller input for this game, but even at .25 deadzone it still felt rather annoying for some reason, so around 1-11 I switched over to keyboard and enjoyed it a lot more.

Are you going to make a thread for your demo SpeeBotDev? If not, I vote we make the next thread SpeeBot edition.

These are some high quality PDFs, thanks

The third pic in page four should say Hue + Value / Saturation.

Value =/= Saturation though.

is there an easy way to turn any class/data type in c++ into text that i can save to a file?

I'll make a general Holla Forums thread when I'm closer to completion, and I will probably make a dedicated thread at /agdg/ next time I build a demo.
Thanks for all the feedback guys, got a lot of good suggestions today.

The slash implies the bar on the right I think.

Look up Serialization

A pink bow.

I want to try my hand at making a galaxy (in UE4).

I've made a black hole, good to go there.

I can make star particles

But where I start to suffer is both a material for nebula in the galaxy, and for arranging it in a spiral manner (probably involves some fancy math beyond my current knowledge)

Anyone have any advice?

A feminine dick

Look up for math equations on spirals, you will eventually find one that can represent pic related. I am sorry, but i dont remember the exact equation's name and it's form.

it was called the involute of the circle (or the archimedes sprial, cant remember) searching for one of those should yield your answer

In Blender the effect is done by having several cloudy-type textures as the cloudy part of the galaxy, and each texture is on a plane with transparency, and each plane is arranged on a curve modifier that allows you to move all the planes by changing the curve.

Basically lots of overlapping textures (like a flamethrower's fire or a cloud) and have the distance to the center modify its color and size. Then a big blooming thing in the middle.

Also the planes are all billboarded, aka pointing to the camera to keep the illusion its a cloud.

I agree, we need a drop shadow. It's just something that comes so naturally you don't even have to really think about it when playing.

Thanks anons, I'll get started and see what I can come up with :^)

Fun fact, Yooka Laylee could not decide between what shadows to have so they just added both

Would it make sense to have some kind of node at intersections between aisles and stuff?

Or I guess the don't need the radius check, they just need both NN's to resolve to the same node?

I did not expect that. It has a much cleaner look than Runescape did.

I do not have nearly enough anime girls with interrogative marks above their heads for this decision.

can you zip all these pdfs for me, senpai?

anonfiles.cc/file/6902b936da812640e25509cc1536ebeb

...

I mean I guess it was "productive"

Well that's fucking lucky

thanks, but I mean all of your pdfs. If you die, horribly, tragically, I want to pass on your legacy.

Please don't disrespect my sensei

Saitama would be able to defeat Goku easily
That's the entire point of his character

I ported my game engine to linux.

It depends on how we look at the character.
Going just by feats,Goku has done more powerful attacks and taken more powerful hits than that Saitama. It's mainly just because after so many years there's been so much power creep and bullshit hacks in Dragon Ball that Goku is of corse going to be absurdly powerful.
If we go by the purpose the characters serve in each of their respective stories then Saitama would defeat Goku in a single punch because he's supposed to be the character than can always take what his opponent is dishing out and retaliate with enough to bring them down. Goku, meanwhile, is supposed to be someone who gets stronger over time and overcomes a challenge that he previously wouldn't have been able to. Goku is a person trying to reach the top. Saitama is the guy who's already there. It's just the top we see in OPM never manifests itself as high as the highest we've seen in Dragon Ball.

Will this game be Ａ　Ｅ　Ｓ　Ｔ　Ｈ　Ｅ　Ｔ　Ｉ　Ｃ？

They're just test images, but I've considered making the game art "out there" because I can't draw for shit.

If you have a neat idea, I can do the game art because I can draw for shit. Wacom Cintiq and everything. Let me know what the game is about and what sets it apart, if you please.

I'm the type of person that comes up with stuff as engine development continues, so I don't have _ANY_ ideas.
Sorry.

Basically, if you've seen this, you'll see that it has a circle shadow for underneath Yooka-Laylee as well as a dynamically-generated accurate shadow, shown at different times.

The reason being that if you're jumping on platforms, you want something to show you exactly where you can land. Seeing a 3-D world on a 2-D plane can lead to not knowing where you are on the Z-axis, which is why FPS jumping is so troublesome. This solves the problem and is a nice throwback to the N64 era when most shadows were simple circles.

Then I suggest you e-mail me anyway, I got a couple simple ideas that would make a pretty penny if a programmer like you helped out. Do you work in Unity?

Not at all.

Still, I'd appreciate your help and insight.
What programming language(s) do you use?

C++, Python, (occasionally)Ruby.

anonfiles.cc/file/4860d559aaea217c6375a37909faa7b1
anonfiles.cc/file/1f7f2d227a7df961a7b2a4e9c6c15edd
anonfiles.cc/file/f870ff66d82baa780f72654cd11f48bf
I'm not the one that compiled these, I'm just like you. This is new material, and there's another one that's 400 MB.with the ones I posted and ones too large to be posted.

That's actually incredibly helpful for making stuff in Unity and Blender 3D, respectively.

If you could make a floor modifier in Blender 3D that actually stops the entire bone from going through the floor even when in an IK chain, it would be incredibly popular and helpful.

That's just the tip of the iceberg when it comes to useful things you can do with your knowledge. So what do you say?

I'm somewhat scared of people relying on me and it's 2 AM.

Here's my e-mail address. Just keep it in case the thread 404's.

I know the panic when someone suddenly relies on you, but the rewards are worth the risk. Send me an e-mail when you're curious about what we could do together. You'd be relying on me for art assets, so the feeling would be mutual. But together we could get things done much faster than having to be an expert in each other's skills.

OK if it doesn't show the e-mail just message me on my website, Brocartoon.com

I get why they made the decision, I'm more wondering if they thought that it looked good.

I might just be drunk, but I'm finding the lack of a shadow beneath speebot to be a major issue with the platforming
Either than that, it's a solid game and I want to give you money for it

You have my thanks user

hmm sadly it seems there is no easy solution at all this is pretty much a pain in the ass
I guess I need to create my own save/map/campaign serialization

I'm sure there's plenty of C++ serialisation libraries, did they not fit your needs?

I generally don't like using other peoples code but I guess I probably should.
sage because I'm posting to much

barely finished last week project so don't expect any fun
volafile.io/r/q3JhiM
only tested the linux build but the windows one should work just as bad

you know i was thinking of some kind of library that would somehow magically do everything for me but i realized thats not realistic, however this serialization library is actually really fucking useful and saves me a lot of time and effort so thanks for posting.

You're really not.

MoM user here, with another weekly showcase; this time I am showing off the Mage Class, His Exclusive Weapon and Special Abilities!

Any Questions are per usual are Always welcome!

So if you render your monster sprites, is there some sort of a easy way to automatically give those sprite the correct direction numbers? I made a 5 framed rocket render and kurwa was it annoying to name those properly so that it looks correct in-game.

Now when you say 5 frames, do you mean you have 5 sprites, with the left/right sprites being mirrored?

Post pics of how it currently looks in Slade for me will be helpul!

Well this projectile sprites looks correctly in-game but I had to do it per hand, so I am wondering if there is a way to automate this process.

They is a quicker way then doing it all by hand; if you renamed somthing like MISL then added MISL^, it will add A, B, C, etc in the correct order. Alternatively, you can add & to have it do numbers. so for example, if you had MISLA, then added & at the end, you would get MISLA1, MISLA2, etc.

However, you cant really get it more simplified then that unfortunately!

That is the current result of doing this: PPN*A^A%% . So Slade3 doesn't really provides a decent way to automate this process in the same way it has 7 different presets for the sprite offset welp.

I must ask why your sprites are not just PPNSA1-5, and why its PPNSA1A0?

The other A and 0 is not necessary to my knowledge?

the A0 thing happened with the %% so and the description doesn't shows one that decrease the number instead of increasing it so I thought I had to use the %% for that. Also I based on the sprite naming on the other rocket sprites since I have no clue how I was supposed to name it. It doesn't really help me any further that Blender doesn't have some sort of a native support for 2D game asset naming.

Just use PPNSA^ Buddy, and you should be fine and dandy!

why do people mainly post art here and not code?
are you guys ashamed of your shit code?

Isn't game play footage a testament to the code?

isn't your comment a testament to stupidity?

// nod a comment :DDD

I'm gonna stream game development again youtube.com/watch?v=tKe-GBs2pOw

what kind of mic are you using? I've been looking for an alright one

Because no one wants to look at your code.

blue yeti

because people in general appreciate visual depictions as opposed to text depictions. Even people who like to code would probably rather see a webm of gameplay as opposed to just blocks of code. People usually only post code if they're stuck on something.

However, if you've got some really good looking code you're confident in, feel free to post it.

Well they is also a risk of some one stealing the code too, unfortunately.

honestly I don't consider that a big risk. Unless you're developing a new state-of-the-art algorithm or engine, chances are your code is probably unoriginal (for example, if you're doing something like a sort, or programming game AI, there's basically already theoretical and practical solutions for these issues, so you aren't making something original when you do these things). Not to mention a lot of people in this thread are amateurs, so their code is likely inefficient or based heavily on tutorials elsewhere.

there simply isn't anything you've done someone else has done better and for free

Its still a risk however. I mean, Obviously they cant do much with the code without the assets either of course.

while(CodeIsPostedHere == true){
>Yes D:<
}

[hackerart]private static void GetThing (Thing thing) { return thing; }/[hackerart]

it really isn't nobody gives a shit about your shitty code
any problem that is worth solving has already been solved and the solution posted online for free
I could make an entire RTS game open source and no one is gonna want to look at the code.

I hope by your/you've you mean that in a general sense and not specifically me. I was speaking in a general sense when I said it could possibly happen

I meant people in this thread

In fact, posting art/design/music is far more risky, because someone with more experience or talent could easily steal it and use it on a project they might end up finishing before you even get close to finish yours.

yeah im sure someone wants to steal yandere-dev-tier code

I think I've made a mistake

Into Miracles?

Then why are people begging me to release my shit code and garbage art every time something goes wrong?

Yooo just had a great idea

Tldr; >put sign >it werks

BUENO
cute blimpo

some people might think they want your code but once they realize there's thousands of lines of code that make your game work they will quickly lose interest.

THis may sound completely buttfucking retarded to you guys but I know a lot of C, C++, and some C#, along with Javascript and other shitty languages like that
And while I know these languages, I don't know the first thing about practically applying that knowledge. Where would I start in regard to that?

how is this possible when no one on /agdg/ can program?
its just no-devs, idea guys and artists here

codeeval.com/

oh my, those digits

Sick quads but, I'm not gonna go fronting and say I'm badass enough to take this challenge. Would this really help me learn to apply code I've learned practically?

well you asked for something to do with your (theoretical) knowledge, that site offers you a variety of challenges where you can apply your wisdom about algorithms and shit
make sure to post your sub-par scores here once you are done so we can laugh help you

Well fuck, guess I can give it a shot.

I have several ideas…

1) Procedurally generated, 2D infinite Grand RTS, almost like Warcraft II meets Minecraft.

2) Top-down 2D hacking RPG, which unlike Watch Dogs involves realistic hacking and device trees.

3) A 3D survival game where you play as intelligent rodent in a procedurally generated home. You find supplies, build nests and avoid predators.

4) A standalone version of the "Cat And Mouse" gametype from multiplayer Starcraft, like how DOTA was based off a Warcraft 3 gametype.

5) Side-scrolling platformer ARPG that's a blatant exaggeration of the Mary Sue and various esoteric wish-fulfillment tropes.

Whatchall' think?

Make em

Great

do the rodent

thanks for signing up with my email harvester, expect spam soon fellow nodev

Also, has there been a TBS where each race is a different mythology? I see lots of generic fantasy ones, though

delet this

TBS?

Turn-based strategy. Like Civilization, or any of those hex-based resource games with cities.

I guess the "proper" term is 4x

third berson shooder :DDD
turn based strat. probably

...

Oh, like Heroes of Might and Magic 3!

I keep forgetting that is a genre considering its generally dwarfed by RTS games.

Okay, I stopped fucking with social media for a second and sat down to work with generating items.

how would that even work
unless you take out basebuilding? but even then….

I think he means theoretically infinite, not actually infinite, like minecraft 1.6. Now there is a wall, but in the past it continued.

Oh no, there's basebuilding. You build towns and cities where you gather resources and train troops, and either ally or attack other, procedurally generated factions. Alot of the most unique units and structures are enabled through upgrades which you get by friending/killing factions, doing quests or completing dungeons.

You can name each town and quickly zoom to it from a drop-down menu

Decided to practice some C# by attempting to make a SS13 esque game using monogame.
Haven't used or even heard of monogame before so I expected it to be similar to HTML5 game engines like Phaser, but as I've come to find out it's surprisingly barebones.

Took me a shameful amount of days just to figure out how the fuck to draw a colored square on the screen rather than a sprite/texture.
It also doesn't help that half the information you find on monogame/XNA is usually wrong yet has a huge amount of people posting thank yous and "this solved my problem".

Either way that's my rant, have a screenshot of me perfecting the art of drawing a randomized sprite on the screen (that changes on mouseclick but you can't really see that).

https:[email protected]/* *//greenman/rts.git I don't feel like counting but the main file is over 4000 lines long

What would prevent player from snowballing out of control?

Yeah, the world limit is something like 32 million from the center. Unless you are fatally autistic you will never get there. It apparently took a minecraft player 90 IRL days straight to get to the world limit.

I think that there is a guy doing a series of walking towards the farlands. I cannot recall his name, but I think he hasn't actually got there yet.

are there any good low polly tutorials and is there a workflow that allows you to sculpt them first?

i will now steal this and use it for nefarious purposes

Some biomes require special upgrades to inhabit, and the generated faction can use the unique upgrades, which means if the player needs to adapt their tactics when dealing with them.

Resource scarcity and biome-dependent faction friendliness slow the game down abit.

I had a neat idea for multiplayer, where the player can build an embassy, fill it with units and resources and have a friend connect to their world and play alongside them.

doesn't bother me

that doesn't really make any sense

It is a good idea overall, but you need to delign it all before starting.

Grid based or freedom of movement?

I wouldn't recommend just using a library if you actually want to make a game. monogame is just like you said, very barebones.

I know Monogame pretty well (actually XNA, which it's based off of)

Let me know if you have any questions. Not so good with shaders, though.

i don't think most tutorials have parrots

500 bitcoins, please.

Farlands or Bust, but that guy doesn't really go as fast as possible I think.

...

What would you recommend using instead of/in addition to monogame?
I'm pretty beginner when it comes to C# gamedev (and with gamedev in general), so any suggestion is appreciated.

I would recommend using a full game engine. If you want to use C#, then Unity is probably your best bet. If you just want an easy game engine for 2D (and you don't mind scripting languages), then Godot or Gamemaker is probably better.

Is that site actually a botnet though?

What is the best program to create lowpoly models?

Should I use one geared towards minecraft models?

Complete newbie, first timer, etc., here. I've decided I want to make a fetishy /d/-appropriate text-based rpg-type thing in the vein of Corruption of Champions and Nimin, but I don't know where to start. I figured it'd be, after writing some initial design documentation, to find an engine or particular language, but I have no idea what to use. For those unfamiliar with the games, while they're text-based, game navigation is done through a gui. Both games are done in Flash, but I don't know if that's because Flash is good for that kind of project, or if it's because that's what the devs knew already. I may end up adding some more complex graphical elements to the game, later on, so that's also a factor. I do know basic programming, but I've never worked on anything big.

just use renpy i think

can ren'py make a CoC style game? I thought it was for VNs and shit.

I'll take a look, but I'm unsure that it will do everything I need. There's gonna be a lot of stat-tracking and turn-based combat. Also, though this is far from a concrete plan, I was entertaining the idea of having a simple 2D kart-racing minigame as a part-joke part-actual feature, but that would be well down the line, and I realize it's not exactly reasonable to be thinking about doing something like that this early in making it.

Oh, and it's Python, which I don't have much fondness for.

oh in that case just use godot or love2d

I'm back.
Did you miss me?

...

Yes, it got kind of lonely without as many almostyesdevs.

Welcome back, mate.

I had to rework a lot of my existing code and procrastinate a little.

What is a state machine?

A machine that has states.

You've got a loooooooong road ahead of you, user.
Google "Finite State Machine for video game development". Get all the knowledge you can

I've worked on something like this before.
I was using Javascript.

There's a few reasons for this.

Since it's something so simple, no matter what you decide to use, I'd suggest pure code. This is going to be your best occasion to learn it since it's not too complicated.

For fellow Unity3D fags, I feel bored as fuck and decided to share a simple script if anyone wants to use it. It's mostly for shortening your references to components on an object efficiently and safely.

hastebin.com/agihegavum.cpp

It saves you from having to expose components of an object in the inspector to get reference, and saves you from writing shit like
private Renderer _myRenderer = null;public Renderer MyRenderer { get{ if (_myRenderer == null) _myRenderer = GetComponent(); return _myRenderer; }}

For usage, an example of how you can use it is like
private ComponentCache _myRenderer = new ComponentCache();public Renderer MyRenderer {get{return _myRenderer.Get(this);}}
or you can look at the source and use it however else you feel.

Here's a useful resource
gameprogrammingpatterns.com/contents.html
The whole book is free online, it's best for C++ programmers really

Thanks user.

No problem, my bad for saging
It's the one good resource for game programming I can find really, there aren't any good tutorials for things like structuring game code, or common practices.

i post my code and nobody responds to me at all

You free the wrong thing in rc_free_vbo_cache.

yeah, that's been fixed since i took the screenshot. maybe its helpful posting code since everyone cant wait to call you a faggot for messing up

whenever I post code its always an untested first iteration so it probably has a bug

I'm still having issues with pathfinding… So let's say I forget about A* and just keep it to floodfill, it looks like this

But my problem happens during my while loop. Basically, where/when do I add the node to the visited list?

If I add it and the check for visited before the neighbors are pushed onto the stack, then the neighbors can be added multiple times. If I do it during the neighbor check, then the starting node gets ignored

hastebin.com/igaponabid.cs

They take control of the donated units and resources as their own separate faction, and they have access to all the upgrades you've unlocked. When they disconnect, their stuff disappears and you get a chunk of what it's worth back.

Grid based, although freedom of movement for some things would be neat.

That was one thing that bugged me when watching footage, especially later on during that toybox demo

Added bonus: A mockup i made a few years back, with no units, blocky map design and one biome.

I'm thinking a little ahead of time, so I don't have too many examples for this, sorry.
I just want to make a little sequence for an NPC. Just a little fireball to spawn from his hand while he aims it at you. I'm using Unity by the way.

Problem is, I'm not sure how to wait until he's sticking his hands out to actually fire the ball. If you spawn the fireball upon entering the Attack State, he'll fire it regardless of whether not he's actually aiming at you.

Sure, you could split the attack states into two:
But I don't know of any games that actually do this + What would trigger the "attacking" state? A timer?

As far as I know, there's no way to tell what point you're at in an animation in Unity. What's the best way to go about this? Am I going for the wrong goal here? Should I be more worried about the timing than the animation precision? Or should I be more concerned about distinguishing the two states?

docs.unity3d.com/Manual/animeditor-AnimationEvents.html
I don't know if there is a fancy way to do this but this is what I'm using.

It works! It's smooth and works!

Please, change the text box to a dark, stony blue instead and either make the text bigger or less verbose. People don't play FPS's to read.

Also make the hands do more of a Psychonauts-type hand motion when firing magic out of them. Look at how The Heavy wields his fists in first-person view.

I don't know what I'm going to do with this one yet, probably an rpg

I animating hard as shit? Even for basic fighting ones? It's the one thing I've never done for game design and it's coming up.

depends on how your brain works

All I know is games have significantly less animations than meshes and textures. It's also probably one of the most recycled assets.

Animations are what makes the difference between a game that holds up and one that looks clunky and unplayable.
It's what you should put most of your visual efforts into, imho.

Well the game I'm planning on doing will be a MMO, where the client basically just displays stuff while the server determines things like atmosphere, health and collision.
I don't know if unity or game maker supports that, but I'm 90% sure gamemaker doesn't.

lol

first of all, don't make an MMO. You just mentioned you're new to game dev. There's a reason you probably can't think of any good indie MMOs, because MMOs are some of the hardest and require the most work. There's no way you're going to make a decent one as one guy with no experience.

Secondly, gamemaker supports online features. You can set up servers and have multiple clients communicate with it. It is entirely possible to make a multiplayer game in game maker.

I'm working on something in love2d, yet I noticed that each new love2d version requires more in terms of openGL and hardware requirements, where 0.9 required openGL 1.1, and 0.10 now requires 2.1.
Is it worth to develop on an older version of love to support more player's cards? There are some features from 0.10 that I'd like to use.

Even older integrated shitcards should support 2.1. Anything higher than that and you will run into trouble though.

No.

Nice eye cancer inducing fontsize and color scheme.
One of my professors literally died of eyecancer from staring at small font for years while he worked as embedded systems dev in the car industry.

streaming development again today I'm doing more serialization youtube.com/watch?v=bBZ0qfGzB1Q

is that atom?

sublime

Know a good IDE for it?

What would you need an IDE for when working with javascript? You write into notepad and open it in your browser, there is no compiling involved or anything.

It would be handy to have syntactical errors and such highlighted for convenience. I'm prone to making simple typos that break everything.

Every. Single. Time. Why is it that babby devs are always choosing MMOs for their first game, i mean come on, even idea guying should have some limits. As someone who is new to deving like you, it might probably even take a while to make something like pong, so how do you expect to create a working MMO title, something that even a team of pros often take years to finish. In the end you'll just dick around with the level editor for a few hours, import some low poly shit tier model from blendswap, use the standard third person controller to run around with it and never touch it again afterwards.

Do yourself a favour and start with a scrub tier space shooter or some shit, gain some base knowledge and keep building upon that.

probably because an MMO is such a grand concept, that there's a billion ways to do it, and they're /ideaguy/ing really hard.

Frankly just use Notepad++ and autocomplete, you can't misspell things if you don't have to spell them.

any editor with syntax highlighting is good enough

I want to again say I don't know much about anything regarding things like tools.

fuck. Is there a point where you start to slowly get 'it', knowing how to better structure things over time? Or is it a learned method?

programming requires a certain mind set and of course practice.
can you post some code to elaborate what you are struggling with.
i won't laugh too much

Sooner or later that should happen. A large portion of the issue you're talking about can (note: "can" and not "has to") stem from the paradigm of the language you're working with. Unless you're working with something exotic, you will be working with either of these:

You can solve any problem with any of these, but the nature of the solution will vary greatly. What language are you working with?

What are you programming in?
Eventually you'll get there, for now try studying some source code and keep practicing the language you're working with.

Both, you're already on the right path.
Write, compare, learn, optimize, repeat.

Just C++, as it sounded like a good place to start.

This code was just a part of a solution I came up with from a text-book of chapter2 quiz tutorial on learncpp. Don't know if it's worth anything but bully bait to mention it, as I may just be stupid slow to absorb information.

didn't know how to properly pass new values from functions as each operation was made.

Learning programming with C++ is asking for pain. Using C++ at all is asking for pain, but that's a different issue. If you want to make games, please start with a language like Lua or Python, you will get something presentable much quicker and save yourself a lot of frustration.

Take your time with C++, it's definitely not a beginners language.
Make sure to cross-reference learncpp with other sites, and I'd suggest picking up C++ without fear, it was a great help for me.
Couldn't post with the PDF, so here a.safe.moe/38Mkk.zip

Python and pygame is slow and Lua doesn't have case statements. Neither are good for games, but are good for beginning languages.

From the image he uploaded, it looks as if he's doing okay

Don't be retarded. Starting with C++ is fine, python is cancer and do you really want to start learning programming with a language that is 1 indexed?

Watch vid related and don't fall for the OOP meme.

you could shorten it something like pic related, writing every case out manually is pretty silly, especially should you do something larger
pretty rusty with my code but should work

I've never found Lua being 1-indexed that big of a deal. You already have to shift mental gears switching between language syntax and features, I don't know why it trips up so many people.

this is wrong starting with C++ is fine and starting with a weak typed language like python will massively fuck with your understand of whats going on
Python is great if you wanna get something done quickly but if you actually want to learn and understand a C or C based language is better.

No case statement = unsuited to games?

Declaring something cancer isn't an actual argument, m8. Indexing barely matters in the grand scheme either.

Have you ever seen a dev in /agdg/ who finished a C++ game? Serious question, I've been in these threads for three to four years now and for the hundred C++ projects I don't remember a single finished one that wasn't
The finished games were usually in Unity or Game Maker.

And what were Unity and GM written in?
c++heckmate m8

>tfw i can't tell whether this is an ironic shitpost or sincere

Creating games is a part of why I want to learn programming. Making other applications alongside trying to understand computers more also sounds kinda interesting. Don't know where the pain could be, though.

thanks. Will stick with it.

It was a dumb solution, even though I was trying to find a way not to write every case. Couldn't think of a easy answer that included the solution being printed with it.

Your code appears to work. A bit more complex than I'm currently used to, but I can kinda understand it and see where it's going. cool.

>he cant tell ironic shitposts from shitposts and normal posts
back to reddo you gooo

HypnoRain, by SpeebotDev? Red Sky? That's two games from this specific /agdg/ finished in the past year or so. Plus any Unreal games are also C++.

Speebot and Hypnorain are written in Haxe, not C/C++. at least thats what i remember
C/C++ still best languages

C++ is the devil's plaything.
It even says so in the new thread!

Honestly, he should just download Unity and start with Javascript or C#.
I had few coding lessons in school, but truthfully i learned the most by just opening Unity, starting a project, and just randomly adding gamelogic.
I started with having something spawn at runtime, then moving when i press a button, then rotating through mouse axis, then having a light go on and off through raycast/hitbox and so on, having some mumbo jumbo mix that wasn't even a real game. It was fun, adding features and testing it out, always trying to expand the project further through new features, and afterwards i knew much more about coding. I just asked myself "I wonder how x works in games?", did research it and implemented it, learning game code and general code at the same time, just make sure it's not something too big but something small that you can keep building upon, step by step. Not only that, you get quick achievements and something to look at, with barely any blockades or frustration that you could get otherwise, which is important to keep going at the beginning too.

yes, quite simply yes.
Show me your finite state machine in Lua and tell me how that goes.

No problem, with time it will just click, and dont be afraid to ask questions

Aight boys I'm having a weird issue.

I have an entity manager class that just holds onto references to my game entities and an entity id. Im doing collisions by looping through the entity manager map for an object then looping through the same map again to compare against all objects that are active (my Entity Manager has an active list and an inactive list).

_entmgr.loadEntity(&_player_girl,1,1); _entmgr.loadEntity(&_block, 2, 1);

However when I run my code here It doesn't seem to be getting to the second entity in the list.

Any thoughts?

void Game::handleCollisions(){ for (auto& this_entity : *_entmgr.getEntityList(1)) { // Get a rough idea of what kind of entity this is then short circuit logic to ignore conflicts bool this_check_move = (*this_entity.second).getMoveable(); if (this_check_move == true) { // Get bounding box and then go into the loop again to do collision checking sf::FloatRect this_box = (*this_entity.second).getSprite()->getGlobalBounds(); for (auto& other_entity : *_entmgr.getEntityList(1)) { std::cout

To clarify here, I have some objects that are moveable and are not. In this example the girl is movable whereas the block is not and they are both in the active list

Thing is, I'm not new to the concept of MMO.
I've been in the private server scene for years and have dev'ed for several projects, so I'm familiar with how networking, packets and all that jazz works, it's just the client end that I'm not too familiar with. I've always been interested in that part but never had the time (or at the time, interest) to do it.

So now I'm here, trying to learn how to make client end.

Yes I know it's not easy to create a network protocol, so my plan is to keep this local until I got a firm grasp of how to print all the shit I want to the screen, and will be using this to practice various things like perlin noise map generation before I even think about touching the networking part. And even then I don't expect it to immediately be perfect and a fun playing experience, I expect it to be years before it's even beyond "you can see other people move on your screen, so magical!".