/agdg/ + /vm/ ~ Amateur Game Development and Modding

Is there any way to transform a Vector3 into a Vector3 I can use to point to a direction?

I have the Velocity of my Rigidbody, but I only need something that will point in the direction of my velocity like "Vector3(0,1,0)".

In unity.

A unit vector is basically a scaled down vector that is 1u in length, look for a Normalize method, that should work

That did work, I'm a dummy.
Thanks user

How important is it that my collider follows my character precisely? If I don't rotate it, is it going to be a problem?

I dunno what to do on this one

isn't your collider parented to your character?
any change in position that happens to the collider due to physics should also be happening to the character

This is how I "do" things with my game, and I don't know if it's obtuse or not

Where is lolidev?

>>>/agdg/29622

Well no, I found that it's better to have the rotation of my collider and the rotation of my model's armature to be separated.

This way I can rotate my model for aesthetic purposes only, but only rotate my collider in case the game needs it.

I just don't understand why if the collider isn't rotated like in pic related my CapsulecastHit.distance is exactly the same as when my character is walking on flat ground, but if I rotate the collider the distance becomes much longer.

Pic related works, but I dunno, it bothers me a lot

Thanks. So the game is still being developed?

Everytime I get my momentum started life throws like 5 things at me that I procrastinate hard on until the last minute, in which I can't ignore them anymore, and then all my momentum is lost.

...

>>>/hgg/129906

This is his last post, so yes.

I think it means that you have a subset of tiles that are considered walls, and your character will get pushed away from them if you collide with one. Not sure, though, that code is only obtuse because I can't inspect the classes that those objects belong to.

But what the fuck do I know, I'm not some master programming expert.

I check the gamestate helper class to get the tile at (x,y), and if that tile does not have a certain behavior (WALL_SOLID) [eg, it's empty], then I tell the actor to move down (gravity)

From the variable names, I'd assume it's just a constant downward force to simulate gravity.

Holy shit I've actually gotten to the point where I have to decide on content that goes into my game so I can design around it, rather than just make random ideas and implement them

Hacky as fuck but I can create other actors with arbitrary movement flags. They can't collide or do anything yet, but it's getting there

Kind of unrelated but I asked my artist to send over some of the art and this was in the zip

That's a might fine konkey dong reaction image you got there, friendo.

Feel free to save the full res

I want to extensively update the wiki page for my game but I have a few questions about how I should do it

why am i learning philosphy in a fucking IT/Math faculty

images and webms have to be posted somewhere and then just give a link to them in the page

What kind of philosophy?

i don't know, i don't understand the shit
argumentative stuff

...

having a proof of concept and an idea of where its going to go helps.

tbh it's impossible to make something with a bunch of anons

Keyreal, does Speebot have achievements? I'm shilling it to some friends but some of them are achievement whores.

I wonder how accurate this is. Would be pretty shit, but I assume that SOMEONE bought the damn game.

Well the game isn't out so I doubt it's accurate

There's no buy button yet, user.

Well I suck cocks, then.

Progress on entities/objects, got directional sprite support done. Will need to figure out if there's a convenient way to batch render animations with eight cameras for sprites later.

Is pre-ordering on Steam something Valve has to enable now? I've noticed more games that you can't buy before they unlock.

Fairly sure it's the dev's choice

Also I think you should go for really low poly models like in quake, sprites just look weird imho

I don't think that's really feasible, since it's 2.5D. Also how come Steam links are blocked on the wiki? I can't add Speebot's Steam link to it.

Answering a question from previous thread: even though Speebot is on Steam, it's DRM free. However that might or might not change if I want to integrate Steam Workshop later, depending on how their library works.
Elaboration: currently the only way to share custom maps is to copy files and send them around - you put the map file in a folder and the game recognizes it immediately. If the game gets popular enough, it might be worth it to integrate Steam Workshop, which would let players browse and upload and rate maps inside the game (kind of like Mario Maker I think?). I'm not sure if integrating the Workshop would automatically add Steam's DRM to my game, so I can't make a promise that the game will always stay DRM-free.

On another note: can anybody suggest what lets players/streamers/reviewers to contact that might be interested in niche indie games? Perhaps it's a good idea to make a new wiki page with a list?

I've not used it but I know of gamesight.io which lets you set categories/similar games and find people that stream those genres/games and contact them. I'm sure there are other places that do similar things if you search for "indie game influencers" or something. If you don't want to tie sharing levels to Steam you could maybe let people upload levels to your server from in game and run the database yourself.

can't you use the beta option to have a drm free version that doesn't support the workshop?
i'm constantly switching versions in ck2 to try and get certain mods to work

What in the utter fuck. How long has this been the case?

pretty sure you can still work offline
if i try to sign in i have the option to skip it

I can also skip it

vinesauce.com/
They have a bunch of streamers, everybody's pretty chill and (I think) none of them are streamer whores.

twitch.tv/jerma985
Same as above

youtube.com/user/PreRecRLM
Part of the RLM guys

youtube.com/channel/UC4_bwov47DseacR1-ttTdOg
You probably know Boogie

twitch.tv/dingdongvg
Ding Dong makes games and plays stuff with OneyNG on youtube

youtube.com/channel/UCWqr2tH3dPshNhPjV5h1xRw
Superbunnyhop has a whole series where he plays games that are sent to his inbox and chooses the best

I also have this very old list of youtube channels, some might be dead, but it could be useful
docs.google.com/spreadsheets/u/1/d/1lD9Qkf-mts48vUAyTOtZTIg2-gjT5WDxmz_sg28eIGQ/htmlview

Northernlion, definitely. He's got a video series where he looks at indie games and he's pretty popular too.

Is your artwork DEEPER, user?

(heiled)
2deep4me fam

...

I'm glad it wasn't this month after all, because I would have had to rush to get any of the things I wanted in for the demo.

Thanks.

Have you tried tetradev's prencil collider?

And I don't have a PC because after spending 600€ on upgrade components, I decided to cheap out on the 10 bucks mount for my CPU cooler.

...

But I do have one. It's just WAY too shit to work on.
Like, it literally can't handle the engine.

user, pls, wtf do you think you are making here, the next Hallway Dooty: Post Op? honestly though, good luck mate and FFS, go spend that 10 shekels

Anyways, I have TileBehavior and ActorBehavior flag enums. The world is basically a bunch of tile behaviors, and each actor has both a tile behavior and actor behavior. Thus, a rock will look at both the native tile and BitwiseOR the actor at that tile to see if it should eg roll or fall. I still need to implement object tracking so an object can occupy two tiles while it's in transit (some rocks in video related still overlap because of this)

I bought a refurb lenovo thinkpad (one of the good models) as a portable devbox / craptop. Seems to work pretty well. But then, I'm only using 2-5% processor and 20k ram with my game so far

Shit, forgot to record with the debug flags drawn. The lower pyramid tiles are all sloped so the rocks will roll. The upper passage is straight

Funny you should say that, because some time ago I literally put a PC together with used parts and they're exactly what I target as minimum system requirements. It'll be a challenge, but I think I'm up for it.

I cut my teeth on a Commodore 64. Everything post Pentium IIs honestly feels like a super-computer when I stop and really reflect on it.

We need more devs like you two. Anyone targeting Linux and avoiding Winblows?

A great way to also avoid 98% of your potential playerbase.

A Linux exclusive would be suicide. Besides, as far as I am concerned, exclusives are one of the cancers of the games industry and I won't contribute to that practice. I won't spend years of my life working on my game and then not let people play it.

I target Windows Vista or newer (engine limitation) and Ubuntu 16.04 or newer (since it's the most recent LTS) and SteamOS, because why the hell not. Of course I won't go out of my way to break it on other distros, but as far as official support is concerned, that's good enough for me.

Oh boy, I can't wait to put my new beefy rig together and I can't wait for the forced "up"grade to Windows 10 either.

Yeah, I know.

I mean, I get it, if we target Linux we are targeting a very small minority of desktop computers. But if we continue to cross platform support Windows, there will never be any impetus for an exodus.

Well, you would have a point if Microsoft weren't putting in overhours to piss people off left and right. And they're getting worse, too.
At the moment Linux needs to build up its library and I'm doing my part.

Lad, do you not remember the anit-trust days in the 90s? This is, sadly, not even peak-M$ hate.
Same.

Okay, but what if your worst available laptop is shit that can barely run Source games?
1.8ghz Pentium single core
4GB RAM
Intel HD 2000
I'm not planning on making the next Crysis, but I definitely know that what I'm gonna make is gonna require a dedicated GPU and dual-core CPU at the very least.

All my shit will mainly target Linux and other Unix-like platforms with Windows as a lower priority. I'm not bending over backwards to support it.

Holy shit man, that's damn near a relic by this point. What OS is on that thing, ME?

I find this a reasonable compromise.

...

As far as reviewers go, I said it in the last thread and I'll say it again - TechRaptor and NicheGamer are the main ones that come to mind.

Why I am not surprised at this point? I also like it how he talks about modern rendering features yet I am stuck with .md3 which is a ancient format and there is far better options out there like .dae or .iqm (intermediate quake model format) or whatever. But nope its .md3, .md2 and .dmd (a version of .md2). Fucking hell Spring devs got their shit together and added assimp support (being able to use .dae models)

Worse, Windows 8.

Does spending 2 months de-MIT'ing your code count as progress?

But why?

Bruh

Cool, I guess.
Why does this keep happening? Pic unrelated.

burnout

Game Development seriously is nothing but smoke and mirrors.

...

No one is working on one? Are VNs the red headed step child of agdg? If that's the case I won't clutter up your thread with my shitty ideas.

I don't think visual novels have even been discussed here. It's not like you'll be bullied for it but it just doesn't seem to come up

From what I've heard the renpy engine is actually pretty malleable in terms of what you can do and the kind of effects you can get out of it. While I'm no professional artist I could at the very least put together concept art for things I want and have someone a few notches of skill above me turn my concepts into actual art. The rest if designing the story and writing it out which isn't the hardest thing to do but by no means the easiest either.

I always tell myself stories to pass time but I never tell them to anyone else. I figured I may as well give that a try at least once.

I haven't seen anybody working on VNs here, but I don't think that anyone would object to you posting progress on one. It's true that they only sometimes actually count as video games, but there's really nowhere better on 8ch to put dev progress for one. However,
Ideas Guys need not apply. If you don't learn to Just Like Make Game, you'll spend the rest of your life wondering why the game you want doesn't exist. Or visual novel, in this case.

I'm currently doing the tutorial for Renpy so hopefully I can manage to do something with my shite ideas rather than keep them bottled up in my head or posting them here to little effect.

So far the coding doesn't seem terribly complex but I've barely scratched the surface so I'll see how simple it stays. Still it feels better to be doing something rather than nothing which is all I've done lately.

Hello fellow aggydaggys, ready for the hottest new jam? Why not join #allyjam, sponsored by the good people at the ADL.
jams.gamejolt.io/allyjam

But will it be as fun as Resistjam?

I cannot wait until the day I release my game, complete with a feeble female player character in skimpy, translucent clothing to the outrage of the entire gaming press and their SJW army, only for them to realize they can do nothing to me, because I'm self-employed and can speak Japanese fluently (passed JLPT N2 back in 2013, could pass JLPT N1 by 2014), giving me a SJW-free market to sell my game to even if they somehow manage to ruin my reputation in the West.

Cool, I can support this game jam!

I made a new map but my computer renders my videos like shit so you can barely see the rain that is supposed to be in the video. Started sprite work again, forward is a bitch and a half.

...

Funnily enough I'm the guy that asked him that.

The thing doesn't work. Not with the poor info he gave out.

If anything them bitching about your game would only help it in the west see Hatred.

Trying to resolve a Unity error. I'm working on transitions between maps. I have a mapmanager script that holds all of the per-map variables my game uses and a gamemanager script that holds everything that's used on every map. The gamemanager contains a reference to the mapmanager. Every time the mapmanager needs to be referenced by any script, it's done through the gamemanager so I only need to set one variable when I load a new level.

The mapmanager has a function that (among other things) sets the reference in the battlemanager to whatever the map's own instance of the mapmanager is. This function is called during both the Awake() and OnLevelWasLoaded() functions.

In the level I'm trying to contain, I have an empty containing the mapmanager script and a few other objects and enemies. The game is tile-based and enemy AI script needs to have a reference to whichever tile it starts on due to the way its movement pattern(s) works. However, when a new level is loaded, the enemy AI script thinks the mapmanager reference in the gamemanager script is null and it's giving an error when I try to reference it to set the starting tile. The line of code that sets the starting tile is in the Start() function. I'm not sure how to resolve this since I thought the mapmanager reference should've already been sit first since it's called in that script's Awake() function and I've also set it to run before the AI script in the script execution order. Anyone have any idea what's going on or how to make sure everything runs in the right order?

It's really a shame because developing for and in a GNU/Linux environment is reallllly nice. On the flipside doing the same for Windows smells like shit from miles away.

Why can't you develop on linux using cross platform libraries like SDL?

Is walking on a 30 degrees surface too much?
I feel like he should be able to do it but it doesn't look very natural.

Should I maybe not make his body rotate as much?

At that point what you're doing is more akin to hiking than walking, and body movements for hiking are very different from normal walking.

But animating those differences probably isn't worth your time. It doesn't look like you're going for a realistic art style, so unrealistic movements like that aren't going to be off-putting. Tons and tons of games don't make any animation differences for characters going up steep hills versus flat ground. I mean, work on it if it's something you're interested in, but it's a minor issue at worst.

This little guy you've got looks pretty top heavy, so maybe have a different animation for uphill where he positions his weight in front of him to assist the climb…. Maybe scrambling on all fours would look better.

30 degrees is ok, the problem is that your character's body is not balancing itself upward. Even 45 degrees would be ok in this cartoony style, if he straightened himself.

Yeah you're probably right.
I'll add these animations to the list of stuff to be done

I think if you try to slow him down you'll just make the game feel way clunkier. It'll look more realistic sure, but it prob wont be as much fun.

He already slows down a little bit depending on the slope, I didn't program it in, I think it's just a side effect for something else, but I think it should stay there.

I might just try to speed up the running animation depending on the slope and see how that looks.

I also need to work on something that gradually slows him down as he's walking up a slope that is >= 35 degrees and then makes him drop down and slide without player control.
Lotsa shit to do

since you already have the code i would suggest to tilt him for a few frames then reseting, or to tilt him away from downwards instead

I'm not sure I understand what you mean

Maybe if I post a random image Keyreal will notice me.

rotate the model retardo,

I already am cunt

Hey GMOTA user if you are still here, your decorate code works, thanks for that m8.

Well don't give me too much credit, I had help too. But I'm glad I could help you figure that shit out, because it was driving me nuts too.

Yeah it was fucking annoying, I tried to rip the Samsara ACS code for that and it dindu work and now I had help from somebody who tried to rewrite this ACS code so that it works better and I had the same issues, with your or the code from somebody else whatever, it works pretty fine even in Multiplayer mode. So my only issues with that would be figuring a way so the Rocket Launcher pickup becomes a Grenade Launcher one and then a Rocket Launcher item upon pickup because that's how it works in Moonman. If that doesn't work its not a big deal at least its not a bunch of crates anymore.

Oh your poor soul, that was forged in the flames of ijon and only he knows how to unravel its mysteries.

alright, inbetween lectures i finally managed to get the formations to work nearly as i want them to work
before i had the issue that if i had different unit types in the same group they wouldn't get distributed equally among groups of the same type
example:
i want to make a square (4 lines)
each line can have infantry/heavy infantry/phalanx, each of those is set to 25% of their type
and i have 3 of each, so 9 units total
well this should be split 3,2,2,2
problem is that every type of unit has to try to be in every group, regardless of how big that group is
so instead i get 3,3,3,0

NOW
units get distributed equally. only difference on the user side is that the formations need an additional line to know the percentage of a combination of units that a group can have, which is typically gonna be the same as the unit percentages, but you never know
basically a formation group in a file is like this now

0 //x
-0.6 //y
0.14 //max units = 0.14 of all the HI+P+I groups
heavyInfantry:0.14|phalanx:0.14|infantry:0.14 //additionally the group only has 0.14 of each unit of their respective type. alternatively, you could do something like have twice as much light infantry as the other units, or something like that
true //horizontal or vertical
1 //unit distance

in addition to the bugfix, i also got the files to be able to let you decide how a unit type gets place in a group
heavyInfantry:0.14|phalanx:0.14|infantry:0.14 means that heavy infantry is in the center, phalanx on its sides and infantry on the outermost part of the line

at this point the only thing i still want to work on the formations is to get them to push the units closer to the center if there's no groups there, but i haven't figured that out yet

"oh fuck"

**This page is making my coincidence detector go berserk. 15 parenthesis each side.

There is anyway absolute no point in catering to those retarded SJW niggers because they don't play games at all, all can they do is bitching at developers to change this thing and then change it again because they don't like this change either despite demanding for it.

Your free speech ends where my feelings begins goy.

In case I need to record at which times something happens, how is that done?

I thought I just needed to pass Time.deltaTime, but apparently that's always the same value.
Should I make a separate variable called "TimePasses" and then add Time.deltatime to it every frame?

(Unity)

That's where you're wrong, user.

All three of those failed, proving his point that there is no fucking point to catering to them because they DO NOT PLAY VIDEO GAMES.

Hey there 8/v/agdg/. Anyone have decent Mobile Inverted Pendulum code examples? I just got an EduMIP kit and a Beaglebone Blue and I'll be assembling them into a little two-wheeled robot.

The robotics lab at UCSD which is heavily involved in the development of both devices has classes, but as far as I know they aren't offering a MOOC at this time, so unless I learn otherwise I may be on my own getting this setup rolling.

user, you don't get it. You can't call a game a failure if the dev wanted people not to buy it.

Got to read a bunch about classes and got this length of input function working.

Everything works great but it looks like unity doesn't read some of my inputs sometime?
Does anybody else have this problem?
There's a whole lot of values in the Input manager that I don't know how to use, maybe it's that?

Ah, apparently you shouldn't read for input in FixedUpdate, I guess I'm gonna make a script to check for inputs and make a bunch of booleans that I can then reference from my main script

Rather than rotating the whole body, the feet should be rotated to fit the slope.

wew
archive.is/3Z83L
archive.is/xkACF
what a fucking idiot

I tried, it looks really weird because of the cartoony nature of the model

I'm now thinking that I should rotate him a little bit following the ground, but maybe also give him a max rotation.
In Mario 64, Mario can run up 50 degrees slope, but he only rotates himself a little bit, and doesn't actually follow the steepness of the slope.

Lel. Do you think we're all making him just a little uncomfortable user?

He's probably talking about using vim and makefiles with a tiling WM.

Does anybody have any tutorials or guides on what are the best settings for the input manager?
Can't find much on google

Either that, or just read the docs.
docs.unity3d.com/ScriptReference/Time.html

GET BACK TO WORK, POST PROGRESS

Works perfectly for me, and I just implemented it a few days ago.

What you're looking for is "script execution order". It's in project settings.

Look at "center of mass"; you're not following that subconscious rule we all have for walking/running/etc, and thus it looks unnatural.

Look at the other vars for "Time".

wrong thread/board

yup, nice self correction

Just hover over each variable for the tooltip.
It's pretty self-explanatory after u read it.

Note that it should be enough to just do all sequentially dependent startup stuff in Start (easier to sync ordering w/script execution order), with proper script execution order.
Also look at the docs for execution order:
docs.unity3d.com/Manual/ExecutionOrder.html

If not, then u can always do a coroutine if one is null, and let it load; check for it next frame, etc.

side clarification note: the reason I asked u to look at execution order is to have you notice that awake and start are in different intervals; which can cause issues w/manually ordering script execution order for dependent startup/awake stuff.
f.e. if you have script1 executed before script2, it will execute its own awake, and start "first" (on the interval), but script2 will execute its own awake before script1's start. i.e. all scripts execute awake, then all scripts execute start, etc.
Thus, it may lead to the same symptoms of null references you're running into, and is a contributing reason as to why I suggested you leave all sequentially dependent startup stuff in the monobehaviour Start function due to added ease of manually ordering said dependent stuff.

As an amateur gamedev also, this is the best thread I've found for decent programmers thus far. Have any better suggestions user?

It's live.
store.steampowered.com/app/731320/Speebot/

ayo fam, that's a bit pricy for a platformer
5 euros or bust

I think it's a fair price considering it's got 200 levels, unlockables and a level editor. Seems like the kind of game you can easily sink over 50 hours into

i've got over 400 hours in mount and blade and i only payed 1 jewro for that

Well, good luck but honestly normalfags don't think that way. They don't look at the amount of levels or gameplay length. They see an indie game and they expect it to be somewhere around 5-10$ like all the other indie games. I too think that 15$ is an alright price, but normalfags will throw those 15$ at a discounted AAA cinematic QTE shitshow rather than an indie game. 10$ is the magic price tbh.

I was going to say it seems expensive when cuphead is recoomended below it at only a few quid more.
But steam also recommends Yooka laylee at £35 which makes your game a fucking bargin by comparison.

Im not telling you how to run your business. And with video games its easier to lower the price over time or use sales to promote the game. But under £7 people tend to be really forgiving and positive.

...

And I bet you bought it years after release during a Steam flash sale.

Mate, it's certainly a solid game but 50 hours is a tall order for most people. Most of the audience will, at best, beat all the levels and won't care to aim for 100%. For those folks, the level editor and collectibles add no value, and when you think about it like that, 10$ seems more reasonable than 15$.

Cuphead, think of it what you will, set itself apart in the eyes of the normalfag gaming community, and for a small studio, that's hard to do.
Cuphead's artstyle makes it interesting to even the most casual of gamers. Speebot, while a very solid game, will probably look in the eyes of most people as "just another platformer".

Thank you /agdg/ for all the support and feedback over the last 20 months. I really appreciate it.

Price is a bit steep, I'd also agree with ten dollars.

I bought it anyway though.

Congratulations, I hope you make some good money with it.

Yes you showed that those Devs catered to those SJW. And now who is the winner of this?
I assume it is Law breakers which if I recall it correctly is not played by many for whose reason I don't know yet. Fucking pink lip sticked nigger
I haven't bothered looking into this one much
Yeah poorly made animation and a womyn who is autistic, It's a success alright. Probably shit like that happened because they spent too much on marketing department and poor choice of talented "people".

kek

Ok.

Congratulations! Now get back to work on your next game!

Any tips for managing your game data?

I just mocked up a data structure for an item, and pic related is what I came up with. Some item types will need even more properties, I dread the thought of having to manage a list of most likely hundreds of these things. I can give some of the properties a default value so they won't have to be defined every time, but that's where my ideas run out.

Starting on some physics stuff. Have gravity and rough collision resolutions against walls and sectors that are above knee/step height. Still need to fix that draw order bug in the ceiling fugg.

Is this seriously where we're at right now? Bribing people with Nintendo systems to get them to pray to the God of Identity Politics?

For you, Speebot user. I'm a shit drawfag but damn it, I tried.

Congratulations, mate. I'll buy it to support you.
Hope things go well. Good luck!

...

Do you guys have any input on scripting cutscenes? Is there some kind of structure I should make for it, or should I just write it out as simple as "A happens, then B happens, then C happens,"?

A cutscene is like a mini movie, so look up screenplay writing.

Why is boost so fucking popular?

I made a decently shitty engine using C++ and SFML, is it a bad idea to attempt to use it to realize a game idea?

Boost is dead. Boost was always a mistake. C++11 and newer does it better. Before C++11 boost was still perhaps the worst way to do it (anything). If you haven't looked through github.com/nothings/stb you should. He also has some videos and goes into some detail about why boost (and the boost philosophy) is/was a mistake.

The wikipedia page makes it sound nice, at least.

I knew it was a mistake going in. It's just I wanted to mess with html websockets and the best library I could find for writing a server in C++ depended on boost. Ever since then it has been one headache after another. Now the problems have piled up so high I'm going to just rewrite everything in libevent and just gritting my teeth through the handshake part.

Depends how shitty it is.
If it's not too shitty, and you're making a 2D game, might be OK.
If it's your first game engine ever, it might not be OK.
Ideally you want to throw away enough shitty engines you've made and start over from scratch enough times that you figure out how to put a not-shitty engine together that is good enough to use to realize your game idea(s).

Cutscenes are just a set of commands that are either executed in sequence or in parallel.
In other words, code.
But each command, whether executed with other commands in parallel or one at a time in sequence, takes place over many frames/game-engine loops, so you have to keep track of where you are as the game loops over and over.
(Example of execution in parallel -> move camera slowly while displaying text and moving an actor. But there's also times when you'll want camera movement and/or actor movement to wait until something else happens/completes first).

So, as a scriptwriter (not to say you HAVE to use a scripting language), you're probably going to want to be able to specify as an argument when a set of commands should execute in sequence, waiting on previous command to complete before continuing, and when you want a set of commands to all execute at once.
Certain modules in your engine will need to provide a variable or two that indicates when they've completed whatever command they're executing (is the textbox still open? is the actor still moving towards its target location?).
Then it's basically a matter of lockstep, where your game engine keeps looping constantly, but only certain functions are called during the loop depending on the current state of the cutscene.

Hopefully the above aimless ramble gives you some ideas on how to go about things.
Cutscenes are not the most trivial programming problem to solve, but they aren't too terribly difficult, either.
Just spend some time looking at cutscenes in other games and thinking about how they function, if you're been programming long enough you'll probably be able to work out a design for the scripting engine.

Hey, literally me right now. And boy is it tough, even if you have an idea of what you want. The optimizations are the worst part, since you have to do it elegantly, not feature creep, and not make the code obtuse.

Can you programmers explain in laymens terms what is so appealing about making your own engine instead of just using an existing one?

I thought you wanted to make game, not engine.

Foremost, it's a matter of pride. You can say "I made this from the ground up". It's also highly educational, including not just the use of timing, math, logic, shaders, and programming itself, but how and why they work too.

In theory, once you know a lot and move into a more robust engine (say, Unity to pick one at random), you should have a very good idea about all of its concepts out of the box.

why do people climb mountains? specially already conquested mountains. it is not to be the best, it is not to be the first. it is to see if they can rise above the challenge and to say i am here too.

The implied question was not

The implied question was

Or rather, why do programmers seem to value the former higher than the latter?

Fuck, this was a buggy mess getting here. It's starting to almost be interactive.

I originally did it because it was a combination of "I want to learn how these things work so I can utilize it the best I can" and "I want to build something and know each and every part". As I learned I started to realize how much power you have if you build your own engine. Anyone who tells you the gap between consoles and PC's is closing due to diminishing returns is full of shit. Professionals in the industry design their entire game to run as fast as it can and they're all on consoles, where as developers exclusively on PC are mostly just using the PCs power to make up for their shortcomings. If you use an engine you're subject to bloat which slows down your game considerably. This is because they're designed for general use, where as something more specialized can be considerably faster.

the journey and sometimes, the end product can net you vast amount of knowledge.

Also the fact that the end product is literally yours, there's no license fee or anything to deal with unless you dealt with middleware garbage.

There is this odd concept that came along that draws a line between this artificial idea of a "game engine" and just programming a game. Very few people here are writing a game engine from scratch. We are gluing together a few libraries that provide very high level functions: physics, rendering & scene graph, 3D sound, and data import/export. The rest is just game logic - the same as everyone else who is using Unreal or Unity. It's not hard. It's not a lot of coding. It adds flexibility. It provides a much deeper understanding of what is going on and why when something breaks. There is much less flailing and reading through tutorials to figure out the esoteric reasons why shit is exploding in your Unity or Unreal project.

Very very few people are writing a rendering engine, and if they are, it's not a general purpose engine but something that interests them and is limited in scope to their application. Much smaller, much simpler, and entirely feasible for one person.

The whole concept of Game Engine for smaller-scope games is unusual and new. I think it's actually pretty great for quickly making a game that has very similar gameplay to existing games: platformers, tileset RPGs, 3D first person walking sims, and more. It's allowed many more people to enter game making and do some cool stuff.

Want to use a sphere tracing algorithm to render? Want to build a no-load solar system explorer based on perlin noise and procedural generation? Want to have complete control of game timing and multithreading? Not many game engines are going to let you do this.

1. Learning the structure of a wheel and how to build a wheel is a proper justification for re-inventing the wheel.
Game development tends to be so difficult that you really are better off taking the time to understand the wheel than to try and grab a pre-made engine and try to make a game with only a superficial understanding of what's going on and how to put a game together using that specific tool and its user interface.
2. Don't have to pay % of my game sales to someone so that I can remain ignorant about how my game works beneath the surface
3. Many of the pre-made engines out there don't support Linux and other platforms I want to support.
4. I might want to eventually open-source my game.

If you understand and can freely manipulate all of the code running in the engine, you can tune game mechanics to work exactly how you want them to…
Whereas if you only have a superficial understanding of what an engine is doing underneath the surface, even if you create lots of new game mechanics that you know will be fun when implemented properly, you may not be able to implement your game mechanics properly.
As gamers, we all want lots of fun gameplay mechanics.
Thus, as developers, we want our mechanics to not only be numerous, but also meet our quality standards for fun.
Engine deving is a time-consuming approach that helps ensure that quality…When done correctly.

All of that said, there are cases where I believe using a pre-made engine is incredibly beneficial.
RPG Maker is probably one of the most famous and obvious examples of saving mountains of programming time to put a simple game together.

But when you're dealing with something as technically challenging as a 3D game?
user, forget the saved time thanks to Unity/Unreal- you're going to run into a thousand brick walls making the game content if you know nothing about the different ways textures and vertex colors can be used in a shader, what a pole target or inverse kinematics is, etc. etc.
For such a difficult challenge as making a 3D game, you're really best off taking the time to learn the ins and outs of 3D.

Is there a place that has the limitations of unity listed as far as 3D assets are concerned?

It really would be a shame if I started making assets and then had to shred them or cut them down in size and scope because I didn't know the limitations.

That cutscene description is more or less what I've got in progress, so I think I'm on the right track. Thanks, user.

Okay, am I getting botted with upboats or something?

Probably. Also, right at the bottom
red fucking alert the jews are onto you

Isn't that good in theory? Eg I'm a good goy with a nice looking product to exploit?

With jews you ____

unlikely since they're all using the same bot. they're no more aware of his existence than any of the other likes.

Here's a quick rundown, written by a plebbitor but pretty straightforward. The short version is they're a marketing company that astroturfs for indie games that people buy and let rot in their steam library forever.
worthlessbums.com/blog/2016/08/16/should-you-associate-yourself-with-black-shell-media/

Whole day wasted on copying useless shit, it doesn't add anything to a gameplay and even potentially limits it the feature was a speed limit on a character, actually. Is this what the whole "falling for a meme" meme is about?

Lose.

But since theyre painting me as a target, that means there is "something" that might appeal to a wide audience, that my game has innate potential to get popular if I finish it, right?

Don't worry I know how to politely refuse people

Godot 3 worth a shit now?

Yes. And that's exactly why they want to bury their claws into you and burn any of your potential competition into the ground before it even gets started, or make your product popular at the cost of turning you into a soulless corporate whore. With Jews you lose.

Exactly. And if I know this now, I can be coy and diplomatic and extort as much use from them before I politely tell them "no thank you"

(checked)
Jewing the Jew, user. That's the way to win!

Ah, the chink approach, yeah it should be fine.

codegolf.stackexchange.com/questions/11880/build-a-working-game-of-tetris-in-conways-game-of-life

Jesus fuck

Always aim for something that is slightly outside of what you think you can do. Also, watch videos where people break down the tricks that games utilize. With games, many parts are scripted, a trick, a hack, or otherwise. Always think about the easiest way to do something.

For those doing 3d games, you will likely need pathing at some point. An easy search to implement is depth-limited or iterative deepening depth-first search. Generate nodes in whatever way seems easiest. You can place higher value hinting nodes to correct poorly generated nodes on maps or allow the AI to do something harder like go up stairs.

easiest solution for pathfinding that i've found is to just use steering behaviour
and if you want to improve it, you can add A* or whatever on top of it
gamedevelopment.tutsplus.com/tutorials/understanding-steering-behaviors-collision-avoidance--gamedev-7777

You will have different methods depending on what the entity is doing. If navigating the environment while in combat or tracking the player, then avoidance works well. Typically you would use it in wide open spaces that just have rocks, trees, and other things the entity needs to avoid.
I would avoid doing an A* implementation alone. It isn't enough. For instance, consider a large map that has a chasm. The entity tries to get to player, but there is no way a trivial method can get them there, so then they go for A*, and the path requires a search that covers most of the map. You quickly figure out that you need something like iterative deepening A*. Then you have 100+ entities and realize the open/closed nodes require lots of memory allocation, so then you need to address that. I've went down that road. A simple depth-limited or iterative deepening depth-first search works well for most cases without issues. If longer paths need to be done, then you fall back to some sort of hierarchical pathfinding where you instruct the AI to path to landmarks.

Okay, so let's say I have about 2,000 live objects in my game. Would it be reasonable to give each one a list of function points (about 8-12 each) that defines their behavior, as on tick, tile entry/exit, etc?

I guess a better approach would be to attach an ID to each object, then lookup the behavior, but I guess my real question is using delegates/method pointers this way, is that normal?

like doing one big A*, then a smaller A* in the sectors of the bigger one?

Offhand, my guess would be that you would nominally build some kind of event tracker system, make several instantiations of it (one for on tick events, one for on exit events, etc.), and fill it with events that have the object and the event specifications. That way you could simply loop through the event tracker list each time a certain kind of event needs to be run (each time the timer ticks, for instance, you loop through the On-Tick Event System events and run those events on their objects). That way you would avoid having to loop through every object on every time any event could be fired for any possible events it might need.

Maybe this is why user

If you can determine how sectors are connected or know how they are connected ahead of time, you can create a graph representing it and store the paths between sectors. Then your search only needs to determine if it can reach a sector.

My problem is that I don't know how to structure the object interactions. Suppose I have some objects like rocks, gems, keys, these are all affected by gravity and have common behavior. But some are collectible, some hurt monsters when they fall, how do you manage that? What if one is fragile and breaks when something is dropped from above?

My solution was to just use bitflags to give actors certain properties as well as giving them a typeID. Lists of behaviors per event are attached to the typeID which is looked up. Example:

hastebin.com/xupuveqaxe.cs

So a "rock" type might only have a few behaviors attached to it, such as the CheckGravity. The player would be a bit more involved, which could check for pushing objects around, etc

Non-gaming programmer lurker here. Not sure what half of that is doing (why are there two "head" slots and one "torso" slot in equip even though the equipslot is "head" only? why are "id" and "name" different text fields? why isn't "usefunc 'equip'" derived from the "equip_body" type?), but I get the impression that you're putting too much information into this single data structure. Separating the type-unique data fields (equipslot, equipped, sprite.equip) into its own EquipItem data structure will probably make it feel less chaotic.

Also, I'm not sure what you're doing with your sprite data, but it smells bad to me because you have multiple sprites used in completely different contexts saved as separate subentities of a single "sprite" field. You could make a new ItemSprite structure that maps your item identifier with its item sprite at draw time, and a separate EquipSprite structure that maps the equipment id with its equip sprite. Alternatively, you could keep the item sprite field in your current generic item field (assuming every item is supposed to have an inventory sprite) and put the equip sprites into the EquipItem data structure.

I have no idea if any of that helps, or is even possible with how your project is structured. Good luck! [spoiler]:^)

By the way, what syntax are you using? It looks a lot nicer than JSON.[/spoiler]

Is this JSON?

Either way, give it a field called "baseType" or something similar. You can build a tree on the fly and have children of children of children soft-inherit default values if you set it up correctly.

Nice blog, the guy seems like a turbolib but the info is alright.

worthlessbums.com/blog/2014/12/29/mistakes-problems-and-solutions-part-1/

Well, collecting a collectible, breaking on impact with the ground, and hurting monsters when they fall on them would all just be collision events. Gravity would, I assume, be an on-tick event. A rock would constantly be throwing on-tick gravity events into the on-tick queue until it hits the ground. When it hits the ground, it throws a collision event into the collision queue, and I assume that would do nothing. Breaking a tile under a rock would start throwing on-tick gravity events into everything stacked up on top of it. If that rock on top of the tile falls and hits an enemy, it throws a collision event and the rock is destroyed and the enemy damaged.

Whether something is "collectible" or "fragile" would presumably be flags in the object types themselves, or, as you said, you could have an object id and then just a table of some sort with every flag for that id. I mean, I don't know how everything you have right now is structured, and I haven't put a great deal of thought into this event queue, but that's just what's popping into my mind at 4:30 AM.

Guess I can't open source it then. Or maybe I'll OS my engine and have my game-specific code be (((proprietary))). Either way, I want to hit that sweet $11.95 USD price tag

You could have an Entity class that has a virtual functions for each event and inherit and override them to handle the specific behaviors.

I had an Actor class acting as the base behavior, but I didn't want to override every object to give it the same falling behavior, or have a hierarchy nightmare; I'm actually in the process of converting it to what I described

Then you would implement the falling behavior in the base class.

Read through both of these if you do not understand how to use virtual functions and polymorphism
docs.microsoft.com/en-us/dotnet/csharp/language-reference/keywords/virtual
docs.microsoft.com/en-us/dotnet/csharp/programming-guide/classes-and-structs/polymorphism

I understand how polymorphism works, but it fails in a lot of cases for gamedev, and in particular the problem I am trying to solve. This is why many popular engines such as Godot or Unity use a component/entity pattern

You can be open source and charge for the binaries. Most people don't know how to compile things or are too lazy to do it themselves. Aseprite uses that model, and it's a development tool, so it probably has more technically competent users than most games.

None of your events seem like anything special. How does it fail? Are there events you do not know of beforehand? That is the reason a component entity system would be used.

The Aurora Engine for Neverwinter Nights used a C scripting language and entities had a large number of functions that could be implemented for certain behaviors, which were called by the underlying engine. It worked well enough.

Can somebody tell me why if a variable named time equals 11.90, and a variable named speed equals 0.10, that floor(time + speed) would equal 11? Does floor act on the variables within itself before adding them?

The solution to your problem is easily solved. You give objects properties and operate based on those. Why does an object cause falling damage? Perhaps the damage it does is based on its weight. A 20 pound object would do less damage than a 500 pound object. You would not need flags if you have selected good properties. Consider what information can be gleaned from indicating the type of material an object is made of. The only thing you need flags for are states. Cloth may be flammable, but if it is wet, then it is not, but perhaps it loses its wetness state after a time when placed in fire, and then becomes flammable.

Floating point numbers cannot represent all whole numbers. Also, arithmetic operations on floating point numbers will introduce error.

What do

Got it. Thanks, user.

Understand biology on a fundamental level and attempt to envision new animals which could only exist if evolved under the limitations of unusual biomes which do not exist on earth.

Please no.

You sir, are subject to floating-point error. Using exploringbinary.com/floating-point-converter/ you can see that 11.9 is not exactly equal to 11.9 in floating point, and neither is 0.1. Because of this, the approximation of 11.9 plus 0.1 is not quite equal to 12. Are you sure you need floor() and not round()?

You're as edgy as a fucking sphere.

Make edgy looking monsters but make them goofy in some way

We managed to do that for a few bosses, and they legitimately look good, we just can't seem to do it for the basic enemy type that just runs towards you, everything just looks out of place

well what are you fighting in this game? That might help us figure out a monster design for you

You're not fighting anything in particular, that's where we're kind of stumped.
Bosses and enemies don't really follow anything in particular, so far we just now the main character and the archetypes we need to fill for the enemies.

And the simplest one, an enemy that charges at you when he sees you (literally a goomba) is what we can't figure out.

You are full of shit user. Nethack follows a similar approach. Go make a game with more complex object interactions.

no thank you

Well I was talking thematically, but after doing some backtracking in the thread, you're the guy working on that cute little drill faced guy, so how about making the basic goomba dudes based off some kinda tool? Like the gotcha wrenches of Wrecking crew or something like that?

We've already used that idea for friendly characters unfortunately

So what's the theme of your enemies atm?

We don't have one.
We got a main placeholder baddie but even that's something we haven't put much thought into

So let me guess: you guys have decided to make a game. You don't know anything about the storyline, or the characters, or the plots, or their motivations. You just know that there are levels to explore, objects to collect, and bosses to fight. And now you're hitting a creative brick wall.

We know about the storyline, we know about the characters, we know about the gameplay and we have a bunch of ideas for levels and themes.

We just don't know what to do with the enemies. We've got a few deisgns, but since these need to be the most basic enemies that you will keep finding over and over they need to be done well and not look retarded.

Enjoy your scrum meetings faggot.

These are inherently contradictory statements. In order for a story to exist, conflict must occur. In order for conflict to occur, some idea of what will oppose the antagonist must exist. You don't have that. All you have is a googly-eyed bat king.
Is there anything to your storyline beyond "The $macguffins, important and valuable items in the world of $world, have been scattered to the $macguffins.length corners of the globe, and it's up to $hero and his trusty sidekicks $companions to retrieve them from the clutches of the evil $enemies"? I think without something more specific than that, coming up with enemy ideas would be incredibly tough.

Xenoblade, a game you would think put primary focus on that, did it at the end, after they had much of the world and gameplay finished.

Nigger it's a fucking platformer, we're not writing War and Peace. The "story" exists exclusively to give you a reason to play through levels and collect things, it's the least important thing we need.

what will oppose the protagonist*

I've never played Xenoblade, so I can't really say. However, I wonder how much of that "gameplay" were bosses that already had some semblance of a story in the minds of their creators.
As a counterpoint, though, The Banner Saga had their story first before they'd finished their gameplay. It doesn't really matter as long as there's a cohesive whole.
I would be interested in how other games with excellent stories (e.g. Shadow of the Colossus, Bastion, Undertale) were developed.

Which is why you're going to have to try ten times as hard to stand out in the flood of indie platformers on Steam.
a) If it's so unimportant, why did you say you already had a storyline and not share it?
b) DOOM had a storyline. The storyline was "There are demons. They are evil. Rip and tear until it is done." This wasn't War and Peace either, but it gave a compelling reason to play the game. As far as I can tell from what you've posted here, you don't even have an equivalent for "demons" in your game, let alone a storyline.
c) Literally any detail on your current plans would help with brainstorming, but you refuse to share and I don't know why.

I apologize for being abrasive. I'll be going to sleep soon.

You may want to make a list of enemies you liked in other games. Find a good mix of enemy types that would work well for a level and alter their behavior and design. Several games have had an enemy similar to Dark Souls Asylum Demon. A fat enemy that jumps up, lands, does aoe damage, and then does large swipes or breath damage in front.

Oh yeah, the 3D platformer indie market is booming, there's been 2 whole releases this year

This seems like a good idea

I think for you doing gameplay first then story would be a lot easier. Design some fun environments you want to play in, then put in enemies that make sense in that environment. Doing a snow level? How about snowmen and sharp snowflakes (you could give them dyed hair). Doing an industrial level? How about an oil sludge monster, a hammer or a nailgun monster. Theming like this makes it much easier to come up with monster ideas and then you've already got a tone/style to have your big baddies match.

Getting a lot of scam email after releasing Speebot. Here are the most common scams that were attempted so far:
- Reviewers/streamers contacting me with a link to their Youtube page. They are impersonators, because the About page of their Youtube page has no mention of the e-mail address they're using, and they disappear when I ask for proof.
- Game news site impersonators using the same method as above. They all use gmail instead of an address with the site's domain.
- The most elaborate ones actually create their own websites and register domains, then act as game journalists. The websites however use cheap shitty Wordpress themes, and the content can be found on 10+ other websites of similar nature by simply copying and pasting paragraphs into Google.
- Most scammers ask for multiple keys for the "review team" or "giveaway for subscribers"
I wonder how many developers fall for these tricks if it's commercially viable to register tens of websites and create fake game news sites. All of the scammed keys end up on g2a and kinguin, I assume. I'm positive haven't given any keys to scammers yet.

That's what I'm trying to do at the moment.
Doesn't seem to be working great but maybe we're just going through a slump today

It had a storyline. They simplified what they originally came up with. E2M2 was supposed to be the first level of the game if I recall correctly.

It can be frustrating trying to force ideas. I'd recommend working on a broad list over a few days or a week and adding to it whenever something pops into your head. Nothing wrong with having just some placeholder enemies to playtest with until you figure out what you want.

Sounds like a fucking pain to deal with. On the bright side at least they're finding out about your game from somewhere.

There are some legitimate Youtubers among them too, though.
Here's a handicapped man playing Speebot using his chin.

That's…
uh

Something

They will keep coming. At least they are easy to distinguish from real ones. Avoid opportunities that will have a publisher sell your game in other markets. If you want to make money that way, then draw up your own contract. I'd recommend getting an up front payment on that too.

There's cases where layering may be difficult without the ability to put a separate sprite to another bodypart. Also the arms are rendered as separate objects so they can turn and aim.

It's easier to remember an ID like that than a name because the name may be inconsistent between items or just be long or weird. And if I want to change a name, I'd have to also change every reference to the item.

Both good ideas. I already have separate data for tiles too. It made sense to me to separate properties by state/type, i.e. when something refers to the item's properties, it gets the sprites from item.sprite, when you use it you get all the effects from item.used, when you hold it everything is at item.held, etc. I'll experiment with different ways of categorizing stuff.

It's my own language I'm experimenting with, don't have proper syntax highlighting for it yet. It's basically javascript where the colons are replaced with whitespace. I originally didn't have commas either but it looked a little confusing when you have a lot of properties horizontally.

I find this kind of stuff super interesting. Makes me wonder if there's anything I could do to help people who are disabled in some way to play my game.

This guy's minions would be batlings, small blobby bats with tiny wings on their back and no arms. They tend to hop around but when they see you they flap their wings and make a beeline for you to bash into you.
They kinda look like their boss' head with small goomba feet and a derpy :3 mouth.

Make sure it can work with a mouse and a single click. That allows people to play with a mouth stick like in that image or use a touch screen. Having secondary interfaces is a good idea. If you do those things, prepare your feels. Some people do not have many options for enjoyment in life. Your game might be the thing that keeps them going. I got an email from someone like that. It is difficult to read it without tearing up.

If I have a way to check when my last key press and key release for jumping was, how can I make a decent "The longer you hold the more you jump" function?

I was thinking to change the force I apply depending on if the button was released first, but that would mean that I'd need to wait until the player has pressed and released the jump button to apply the upwards force.

Would applying the first force and then depending on when the button is released, applying a downwards force to push the player back down to the ground work well?

There's a bunch of ways but the way I've done it in the past:
>Holding jump does nothing but if the player releases jump and their Y velocity is negative then heavily reduce their Y velocity (like 75% reduction)

I found this is much simpler than trying to add velocity/force while holding jump and should be easier to extend to other things like bouncing off surfaces or some other reason that the player's Y velocity is a negative value.

if (keyPressTime > keyReleaseTime && keyPressTime+jumpHeight > currentTime){
//apply force upwards
}
else{
//apply gravity
}

Help I'm having too much fun making a stumpy angry cannon

Generally this is done by linearly increasing player height as long as the jump button is held, then letting gravity do its work once it is released (or if the button is held for too long).
See pic related.

I'll try this, seems smarter than what I was going to try.

That is probably the case for NuDoom and the comic based on the older Doom, if I recall it correctly Doom 1 storyline was about the UAC had some experiments which went horribly wrong which opened the hells portal and thus fucking UAC shit up. It was far from being gory game considering the UAC Marine (The Player) was send out and to fix this mess with any means. One of the inconsistencies is that the Player starts out with a pistol instead of SMG, A Shotgun or a Assault Rifle since he was a Marine not a Security Man but that is another issue for another topic. Also Brutal Doom popularized the term "Rip and Tear" which is one of the things NuDoom is based on.

Anyway retail Doom 1/2 didn't had a large focus on storyline since there is no way to interact with it, the player gets to read like 2-3 texts if I recall it correctly that describes the event a bit and that's it. There is no such things like books, notes or survivors at all in this game which is further proof that the Storyline is more of a "side thing".

This looks like one of those things with which you have to fiddle with numbers until it works right

Doomguy was sent to Mars as punishment after assaulting a senior officer when said officer told him to fire on civvies. He goes along with the assault team when they go try secure a base on one of the moons after they get overrun, but since he's still a criminal technically he only gets a handgun.
Source: the manual.

Yeah nah this looks weird and floaty

Right, I forget this one that would explain the pistol start, thanks for clearing it up.
Did the manual mentioned what kind of enemy the assault team had to face?

Don't you mean when people rediscovered the Doom comic with it's amazing dialogue? Pretty sure that was before Brutal Doom but my memory is hazy.

This user is right, it came from the comic

As in did they know beforehand? No. But it was demons.Surprising, I know.

Brutal doom popularized it.

Should've added this part as well, I guess. It's relevant.

No, it didn't.
The only thing Brutal Doom popularized was having bad taste, as well as the shitty context takedowns, which nuDoom did borrow from. The Brutal Doom dev consequently started throwing a tantrum about how they're "stealing" from him.

Stop the meme

Thanks for the images m8.

Those tricks probably work on some junior intern working on social media for a big company.
If they didnt work they wouldnt keep trying…

Im not sure if it was you or the other devanon who was contacted by someone offering $100 for 10,000 game keys.

...

...

...

You've taken this too far, opinions discarded

You know, props to carmack for the great engine and pushing the boundaries of what they could do with the complex architecture Doom is capable of, but the first actually good vanilla FPS Id made was Quake

I don't recall how widespread the mouse was around Doom's release, but if a significant chunk of your prospective userbase doesn't have some input peripheral then you don't design your game around it.

Apple had been pushing it since 84.
Windows 95, which (as far as I know) was the first Windows OS strongly based around the use of a mouse was less than a year away.

It's not even about basing the game around the use of the mouse, but adding the option in case you had a mouse would've taken them fuckall time and made the game future proof. There's no reason why you couldn't have both keyboard controls AND mouse look.

Also this method ended up working as intended, thanks user

Also I think my space key might be slightly broken

no auto aim you retard. No vertical aim. Pitching the camera up/down made some of the Doom rendering engine assumptions not work as well. Mouse drivers required DOS memory, which could cause issues. This is back in i386, 486DX and DX2 days. Shit wasn't figured out. Vastly superior systems and OSes such as Amiga and Atari ST existed but IBM PC world was full of mouth breathing retards. So we got what we got. Be thankful they didn't screw it up any worse than they did.

I'm not saying I don't like the game, I'm just saying that it's not nearly as good as people make it out to be.

I have no idea what your sprite code is doing, but as long as it works and you think you can handle it, you probably shouldn't change it. I'd still recommend separating the different sprite contexts into different data structures.
That's what I thought was going on, but I felt like I should ask. I'd just use a find/replace when I needed to change something, but eh.
You possibly might want to use an integer ID instead, or find a way to convert your string IDs to ints at compile time, because those are much easier to read/compare than strings.
You should be able to do this just fine. My brain is sludge atm, so I can't give you a great writeup, but all you realistically have to do is assert that an Item is an EquipItem, then grab item.equip as normal. You'd probably do this by making EquipItem a subclass of Item with its own unique properties, but there are a few other ways you could go about this.
That's cool user, I like how it looks. I find the colons in JSON annoying and superfluous (same with Python tbh).
I'm not sure how easy this would be, but you might want to try using the newline \n as a field delimiter if the comma isn't present. That'll get rid of a bunch of superfluous commas (e.g. at the end of subfields) without significantly changing the syntax.

DAMN

Gotta love the public domain and attribution licences.

Doom 1 storyline as you described it is basically identical to NewDOOM, which is what I was initially referring to anyway.
My point is that the "storyline" can be barebones and basic, as long as it gives the player a compelling reason to play. If you have no antagonist at all there is no game.
This is why trash-tier walking sims like Gone Home and Sunset get panned. There is literally no in-game antagonist at all, which means they are not a game. It doesn't help that their writing is shit-tier as well.

Storytime m8.

Rather, it wasn't as good as people make it out to be (although at the time we also didn't know better, so it was great nonetheless).
But since it was open-sourced and patched for years, it now is in fact pretty damn amazing. And I say that as a guy who didn't play it as
a kid (only played a little bit of Wolfenstein, never got my hands on a floppy with Doom) and only checked it out a year or two ago. So
no nostalgiafagging either. It simply has nice gamefeel and design.

>If you have no antagonist at all there is no game
Who is the antagonist in Tetris?

Block throwing commies

The tetrominos, of course. You spend the whole game eliminating them, and their presence is what ends your game.
I dare you to name a single game that's not considered trash and doesn't have an antagonist.

...

The antagonists are the enemies you fight along the way. Just because there is no single BBEG does not mean there is no conflict in the journey or nobody who opposes you.

You what. There is no bad guy other than when you don't live up to the moral system in place. You can honestly go through the entire game without killing a fucking thing aside from getting them sweet valor points.

Enemies attack you. You fight them when necessary. Those enemies are antagonists. Just because there is no single opponent coordinating the opposition does not mean there is no opposition.
I should admit that I'm going entirely off the TvTropes page, since I've never played this game before. I still thing there's a difference between "no Big Bad" and "no antagonist".
You can do the same thing in Undertale, but even casting aside Flowey, other characters like Toriel, Undyne, and Asriel still oppose you in a True Pacifist run. I fail to see how "not killing anything" means there is no antagonist.

If anything that may or may not stand in the way of your progress can be considered an antagonist, would you say that puzzles in RPG Maker games are antagonists as well?

...

This conversation started because Speebot doesn't have (much of) a story, didn't it?
By this logic the water and porcupine bots would be antagonists. But they're not the reason you're going through the game.>>13573829

It was great up until Duke Nukem 3D and Quake was released. As someone who played it on a 486 near its release, it had an atmosphere that was unlike any game at the time. It would be difficult to describe to someone who did not live through that era. Playing games like Descent, Terminal Velocity, and Freespace were equally impressive. Any game that pushed things forward and offered something that you hadn't experienced up to that point was thrilling. It isn't nostalgia. It is a feeling you never get back, and one that nobody can experience in the future. It wasn't strictly about pushing technology forward. Crusader: No Remorse and Diablo provided a similar feeling.

Pareto's Principle?

Congratulations Speebot user. Hopefully you inspire more people here to dev

Thats it i quit, bunch of parasites

I guess that means there's roughly 200 people on this site, 20 of which post.

Now that's the kind of autism I like. Computers are incredible; you can build one out of anything as long as you have some a fundamental building block that acts as a bit. From there it's just a matter of constructing logic gates to build the CPU, RAM, and whatever else you'd need in a Von Neumann architecture.

When people use computers today, they only see them as magic boxes with screens, and that's kinda sad when you realize all the work that goes into building one. Especially after all the abstraction that it takes to go from a simple computer to the immediately user friendly ones we normally use.

Without your post I wouldn't have paid attention to his, and I really like the idea.

Keyreal already released Hypnorain 2 years ago, so if he's responsible for half the releases there must be two yesdevs here. One of them is Ika I guess, but who's the other?

how do you see useless cripples who can't even work a keyboard as human?

I'll bite the bullet and say yes. Just because the antagonist doesn't have what we would call "agency" or "goals" doesn't mean it's not present. The antagonist in e.g. disaster movies is not a physical person, it's the environment. The plot of those movies is a result of how the protagonist overcomes those obstacles. Man vs. Nature is one of the fundamental forms of conflict, and environmental puzzles would be a form of that.
Compare e.g. The Witness to Gone Home/Sunset. They both have a non-zero emphasis on exploring and interacting with the environment, but The Witness has actual challenges that must be overcome in order to further explore, while Gone Home/Sunset have none of that.
I am, as should be obvious, using a wide definition of "antagonist" to mean "that which opposes the protagonist's progress and plans." This definition may be wide, but it's not completely wild and fits far more types of stories (e.g. Man vs Nature) than a narrower definition would.

Actually, this started because the drill platformer user claimed he knew what the storyline was while having literally no enemies formulated at all, not even in the high-level sense of "Evil demons! Rip and tear!" in Doom. I complained because without some starting point to build off of it's basically impossible to make generic enemies.
Even in the case of Speebot, though, there are both environmental obstacles and enemies, both of which I would consider antagonists.

3/10 made me reply

...

Also, I feel I should add:
The antagonists are not the reason you're going through the game except in the case of "stop the bad guy"; even then, the reason you want to stop the bad guy is generally because they interfered with your implicit goal of "keeping the status quo that I like" (e.g. Bowser kidnaps Peach, you don't necessarily want to oppose Bowser, you just want Peach to be free. Many Mario games like Super Paper Mario and Bowser's Inside Story play around with this).
Often, the reason you're going through the game is normally because you have some sort of goal to complete that has nothing to do with the antagonist (e.g. in Undertale your goal is to reach the surface, in Bastion your goal is to rebuild the Bastion and attempt to restore your country, in Enter the Gungeon your goal is to acquire The Gun That Can Kill The Past). The antagonist is not necessarily required to give the protagonist their reason to play the game.

No problem. It's stuff like building a computer out of arbitrary shit that makes me want to make a game that somehow teaches the player to do the same (kinda like a zachtronics game), or at least helps them get an appreciation for it. Not really sure how I would approach that though, for now I think I'll just work on smaller projects.

Is this going to be a 3D Yume Nikki spinoff? Please say yes. Either way that model looks pretty good, nice aesthetic.

I've decided to keep track of the development of my game with videos
I'll be uploading this to youtube, I'm making a total war clone with mythological units in C++ with irrlicht.
In case anyone is interested here is the link to the first video youtube.com/watch?v=QLbOfS8oHbM
I'll be uploading more programming to this channel whenever i do some and i get time to upload it.

I left you a like.

More editing, showing off the game and less staring at code could make this really interesting to watch

I don't know about him, but it looks like I'm probably going to be headed in that general direction (3D take on Yume Nikki)…Although I'm not looking to have my game associated with YN in any way (I have to sell my game and make a living, so original IP and more gameplay than what is seen in YN).
I have some more original/innovative ideas, but they have stricter requirements (need to make X character models, a more thoroughly thought-out and explicitly expressed story, etc…) that make them much more difficult to develop when as a one-man dev team.
Whereas something like 3D Yume Nikki can tell a more abstract story through world narrative, doesn't really have strict requirements (other than that the more worlds to explore and Effects to collect, the merrier).
It's a much more flexible project that helps significantly reduce risk during the production process.
My game isn't on the wiki yet, but lurk around long enough and it'll pop up some day…The player character looks pretty nice.

...

user there is no game
I just started all i got is a menu
like what kind of editing?

Just removing the bullshit. Condense it down to 10 minutes, don't show parts of code that aren't connected to something that you can't show off.

You barely talking over a 40 minute video with no visuals and almost no explanation is extremely hard to watch and follow.
That is if you're wanting people to follow these videos, if this is just for your own reference then nevermind.

That model looks cool as hell. What are you using to make it?

I agree with , you currently have a lot of dead air. You should probably have a 10 minute cut of the important parts, and an uncut version with a link in the ten-minute cut's annotations and/or description.

most of that is just me thinking
I guess it shouldn't be to hard to just cut out the nothing
Someone recommend me some good editing software for just basic cutting

blender, the latest version even has improvements on video editing

Not a Yume Nikki game.

Blender!

Blender!

I've been perfecting my controller script's jump for a few days now, and thought I'd give my 2cents.

I just poll for input in update, and set a boolean for "isJumping" to true, or false accordingly.
I then utilize my grounded check to determine different states (state machine related, mechanics related).
After the jump key is released, while not grounded, I just set my boolean "isFalling" to true.
This allows a lot of control, and exposes states for animation.

There's a few important aspects of making a good jump.

Also in unity to visualize the "arc" your character is making draw a Debug.DrawLine w/a time duration; in the direction of your current velocity.

There's software out there that keeps track of all the code you change, and then replays it in a video without all the dead air. Although it might only exist on some guy's 20 year old computer.

Somebody get this boy some clothes, he's gonna freeze out there.

...

I was trying to go for that sort of protrusions where the shoulderblades are poking out of the back a little bit, but now I can't unsee it and I'm laughing.

...

...

those are some nice back boobs, I'm getting a boner

looks great imo, it has quite the RO vibe.
Also, the back shading looks like protrusions of the shoulder bones once you're at the proper zoomed out distance (as it's meant).

...

Treasure and cherish them, Anons. You don't know what you have.

I've done that but the problem seems to be pretty much the same, I don't know why.

I made a blog post about how I made YUME and Speebot, if anybody is interested.
kircode.com/post/how-i-wrote-my-own-3d-game-engine-and-shipped-a-game-with-it-in-20-months

Did you anons figured out a way that the items doesn't get picked up when a certain ammo type is full? I tried to make a check for the Rifle round so when the player has 120 of it, he shouldn't be able to pick it up anymore. I tried to do a A_JumpIfInventory for that but it doesn't seem to work well the weapon item itself is still there but then the fake sprites will be gone. ghostbin.com/paste/fu9qn

Would you believe I noticed that exact same quirk about an hour ago? Sadly I can't think of a way around this, so I'm just gonna let players pick up guns even if they have full ammo.

I noticed this issue already when I implemented it. Honestly I would rather try to figure a way so that it has the weapon stay as a default behavior and just spawn a ammo item along it, I will just make the weapon item respawn as fast as possible and the weapons themselves won't give any ammo anymore. It would sadly remove the proper sv_weaponstay functionality completely but there so much I can figure out. The code for it should be at Moonman git repo as soon I changed it successfully.

Some people are just made for dev

alright another video this time with editing youtube.com/watch?v=bNqocS4eess
I kinda have the feeling that it might be a little harder to follow but i guess if people really want to they can just pause the video to read the code.

Welp looks like my attempt doesn't work, I tried making it spawn 2 items the weapon itself and a ammo, the issue? Both won't respawn. God fucking damnit I just wanted a proper weapon pickups per class but noo I have this mess instead. Fuck it I just make the weapon respawn fast and increase the ammo count to "offset" it somehow. Just fuck my decorate code up fam.

Good read, thanks for posting.
I also started on a 3D game project where I'm making the engine and all of the assets right around January 2016, but from February 2017 to July 2017 development completely stalled as I tried to help out my sick grandmother, then lost more time in September attending her funeral, so unfortunately it's probably going to be at least another year of development before my game gets anywhere…
(From here on out there shouldn't be any more non-gamedev related delays, so I'm hoping I can release somewhere around December 2018~August 2019, if you take out the time lost in 2017 August 2019 is pretty much exactly three years of development).

One of my ACS coder managed to fix this issue finally, he had to use ACS script for that and made another attempt, so far my additional changes to the decorate code is working. I will try to finish the whole decorate code and I will gib you the link so you can use it in your mod too. When I am done with the decorate code I need to test it in multiplayer mode.

You're an inspiration, man

You're a fucking trooper, dude.

So how does separating your draw and update logic work exactly? Most engines allow a variable timestep and drop frames, yet still function (albeit choppily) if it gets strained. I imagine yours would be the same since you'd be visually limited as a player

I am sorry for your loss. Take it easy, there's no rush.

It's kind of like this: gafferongames.com/post/fix_your_timestep/ , but a bit more involved. There's a mechanism for interpolation that's built into the renderer, and there's a mechanism for skipping render frames and calling multiple logic ticks at once when the engine can't keep up, and there's also a mechanism for turning that off during loading screens. I might write a separate post about this one day.

DONKEY DONG DOUNTRY

PROGRESS!
Obviously it doesn't look right but I'm getting somewhere.

Wait, that's Blender? It looks really different from when I used it ages ago. How did you make the UI look… usable? How did you change the menu fonts from the default font?

He made the model in Blender silly. That's a model viewer he's using to view it, presumably part of Yume (his engine).

No that's not blender, that's keyreal's model viewer.

Oh mb, didn't realize that.

That model viewer is pretty nice

You're a fucking inspiration, mate. Gonna get some /loomis/ and /mu/ gains to start making my game.

This kills the workflow.

Alright, I need some inspiration.

I still don't know what I want the goal of each level to be. I know I want a placeable exit that is locked in some manner. Some levels would have it be unlocked. I don't know if you should collect keys, some point item (eg gems), destroy certain objects, or what. I named Fire N Ice and Jetpack and Loderunner as my main inspirations, but I don't know if I'll have enemies or not. (I really should though).

Mostly my concern is getting around. Movement is forced into strict tile steps. If there's a 1-tile gap, I can't think of a really good way to get across it other than plug it up rocks (a long process) or some other annoying mechanic

While I can't really tell you what would work best (it's your game, after all), I can give you another game recommendation.
Fire N Ice is known outside of the US as Solomon's Key 2. Embed related is the original Solomon's Key. It uses a similar block creation mechanic as 2, but it's not just diagonal. Every room has an exit and a key. As you can guess, you need the key before you can leave.
Not sure if it'll help you any, but hell, maybe you'll think of something.

Tab closed
Just kidding, thanks user

Nice

Well I added some different gem colors. I'm having a hard time thinking of cool puzzle elements and keeping it as a platformer (since enemies are way harder to avoid and become more action-y than puzzle-y)… Thinking of just making it a straight puzzle game. But then it's harder to get the comfy blocks goin

Some of the enemies in Fire N Ice (the ones that could move) were more like timed obstacles than anything. Can try something like that. You don't have to make them killable, either, just have them be a moving obstacle with a clear pattern.

For getting around, you could add different ground types:

I was dicking around with particles, but not very seriously. Basically just grabbing pieces of the texture and drawing them randomly. It would be kind of cool for destroyed blocks to have some breaky chunks that fly around and disappear

Alright I three choices here. OpenGL engine, Godot 2.x, or Tesseract. I have made decent amount of progress on all of them, so what should I choose for the long run?

I want to put a taunt command in my game but I'm too scared that it might end up like Devil May Cry.

Depends on what kind of game you want to make. Tell us a little about it.

Are you saying that's a bad thing?

Had a genuine chuckle at this, please keep it like this

It's an FPS. I know Godot 2.x's 3D is shit, but if I can hold out until 3.0 becomes stable it'll be way better.

Oh that's the plan. Those are shadow clones summoned by the actual edgelord, and I wanted them to yelp like sissies because I know it's going to catch people off guard.
I like surprising people.

glad this user mentioned it, that's hilarious, and it also adds a lot of character.

keep at it, definitely surprised me, and got a good laugh out of me.

...

There's a few other things I've been keeping quiet about as well, but there's still a few things I've enjoyed too damn much to keep a complete secret from everyone, like webm related

...

Someone else can make the new thread, I have a headache and feel like shit

Ah fuck it, new bread

looks like you skipped over all the regular pron and went strait to scat.