4d0cee No.16296248
Libbie worship edition
Resources
>>>/agdg/
>>>/vm/
>#8/agdg/ via irc.rizon.net
>Dev resources: http://8agdg.wikidot.com/resources
Links
>Wiki: http://8agdg.wikidot.com/
>Beginner's guide: >>>/agdg/29080
>Previous thread: https://archive.is/LJKXI
Announcements
>QUARTERLY DEMO DAY SCHEDULED FOR May 5th (05/05)
>Please contribute to the wiki if you can
017e4c No.16296406
>QUARTERLY DEMO DAY SCHEDULED FOR May 5th (05/05)
I hope everyone will have at least something by then.
4d6d4e No.16296447
Friendly reminder that if you aren't programming, or learning how to program, in C++ and/or Assembly, you shouldn't program at all.
4edf94 No.16296451
>>16296447
>in C++
What does C++ better than C?
69cf67 No.16296455
>>16296451
It's like having two C's
4d6d4e No.16296467
>>16296451
Literally everything. The STL, after all, is one of the most efficient and simplest to understand of all libraries of similar caliber.
5d0814 No.16296483
>>16296447
Is csharp good or is that just a meme language unity uses?
095121 No.16296502
The guy in charge of Nim's been talking about proposed changes lately to keep memory safety without the run-time cost of the current reference counted GC (and without going full on crazy like Rust's borrowing). I'm liking the direction he's going: https://nim-lang.org/araq/ownedrefs.html I don't think many anons use Nim (maybe one other?) but I think it's a neat language. On the other end of the spectrum I've been stuck on the idea of doing small NES game for next demo day as an ASM learning exercise. Would be fun to be able to just share a ROM.
>>16296447
>C++ over C
2998eb No.16296520
Finished the cooking tools, the player will use those to bring up the cooking UI
7 total tools, 4 have portable versions to cook away from the city
2998eb No.16296530
Also forgot to post the progress on the city reform earlier
74fec7 No.16296536
>>16296483
C# isn't as bad as it could be, and there's plenty of worse things out there, but it is an interpreted language with a garbage collector; not "ideal" for gamedev. The real question is, will that matter for you? Are you good enough that you can squeeze optimal performance out of C/C++? The answer is likely no. In which case, use whatever as long as it's not java.
69cf67 No.16296546
>>16296536
>C# is an interpreted language
<An interpreted language is a type of programming language for which most of its implementations execute instructions directly and freely, without previously compiling a program into machine-language instructions
No it's not. The code is universally compiled into an intermediate language that works on a virtual machine that exists between the software and hardware levels. It is strongly typed and doesn't anything in common with stuff like Python.
f81e91 No.16296550
>>16296530
This is starting to look really cool
2998eb No.16296570
>>16296550
Thanks, i also made one of the caves, but the video quality is poor, i need to find a better webm converter.
74fec7 No.16296571
>>16296546
>that works on a virtual machine
And therefore, is interpreted. Anything that isn't compiled to machine code is interpreted, anon. And C# sure as fuck doesn't compile to machine code.
69cf67 No.16296577
>>16296571
But your code is still preprocessed into a machine language (albeit a virtual one). The source you write isn't directly interpreted because it is compiled (which then gets interpreted). Actual interpreted language (again, like Python) are magnitudes slower
74fec7 No.16296605
>>16296577
C# compiles into IL which is interpreted. So whatever the fuck you write, is still going to be interpreted, just not "directly". I'm didn't stab him officer, I just threw a knife and it's velocity lodged the knife inside his body.
Macro$haft has been shilling how incredibly not interpreted their language is for ages because interpreted == bad to programmer brains, and there's plenty of dumb fucks who believe them and parrot their bullshit, but fact remains that if your language does not compile to machine language that is interpreted by an actual machine (and not a fucking interpreter even if happen you call yours a virtual machine) it is in fact interpreted.
75c183 No.16296607
69cf67 No.16296613
2998eb No.16296626
>>16296607
Would be like if Arthur Morgan retired and settled down in harvest moon with animu waifus
75c183 No.16296637
>>16296626
wish you luck, from a technical perspective this is the best thing i've seen in these threads in a long time.
2998eb No.16296659
>>16296637
Thanks bro it means a lot
We are releasing the new cooking system on the next demoday, don't miss it.
5d0814 No.16296687
>>16296530
This looks really good bud
>>16296626
>if Arthur Morgan retired and settled down in harvest moon with animu waifus
Very cool, I'll have to check it out on demo day, sounds like a game tailor made for me.
2982f2 No.16296720
>>16296546
>>16296536
C# is JIT'd in every modern implementation. I think it does hotspot JIT-compiling like Java does. It's compiled into bytecode, a bytecode interpreter runs on it first, and sections that are run a lot are JIT-compiled to native machine language, which is executed directly on hardware.
It's a middle ground between interpreted and compiled, and it's disingenuous to simply call it either.
50fcf1 No.16296735
>>16296451
Not having to use malloc.
>>16296502
>I don't think many anons use Nim (maybe one other?)
I tried to use it but I couldn't get the SDL wrapper to install on Linux. Despite that, I like it because the garbage collector is actually suitable for gamedev unlike Golang.
2982f2 No.16296737
>>16296720
I actually just looked it up, and .NET and Mono actually JIT everything on run and never interpret any bytecode. So in no way is it ever appropriate to consider modern C# interpreted at all.
74fec7 No.16296818
>>16296737
What does an interpreter do? Compile non-native code into native code at runtime.
What does a JIT do? Compile non-native code into native code at runtime.
JIT is just a specific type of interpreter that caches some shit and looks ahead to parallelize stuff.
You're all pretty much /pol/acks saying "this is not a nigger, this is a muslim" about a black muslim from africa.
2982f2 No.16296845
>>16296818
>What does an interpreter do? Compile non-native code into native code at runtime.
Abso-fucking-lutely not. That's what a compiler does. An interpreter by definition executes without compiling. There are bytecode-compiling language implementations that don't JIT, like Lua, Ruby, and Python. All of those use bytecode interpreters, which involve no compilation into native code.
> What does a JIT do? Compile non-native code into native code at runtime.
You're right about that.
> JIT is just a specific type of interpreter that caches some shit and looks ahead to parallelize stuff.
Not even close. A JIT complier takes input and compiles it at runtime into machine language immediately before executing it as native code. It does not refer to any sort of caching or parallelization technology. I'm not sure where you got that idea. A JIT compiler usually does the exact same thing that an ordinary compiler does, just with faster less-intensive optimizations so that it can do it during execution.
> You're all pretty much /pol/acks saying "this is not a nigger, this is a muslim" about a black muslim from africa.
I'm a professional programmer. There are distinct differences between interpreters and compilers, even when the lines get blurry with bytecode languages and JIT compilation. This isn't being pedantic, and they are not the same thing. These are also incredibly easy to look up by name, so you have no excuse to be as ignorant as you are.
ed1dd6 No.16296848
>>16296818
>What does an interpreter do? Compile non-native code into native code at runtime.
But that's wrong you fucking retard.
74fec7 No.16296938
>>16296845
>>16296848
Ah I forgot, it actually reads your line of code, turns it into a demonic summoning ritual, then tells the daemon to magick the bytes of your computer into a specific sequence:^)
Good luck executing quite literally anything without turning it into a sequence of machine code instructions at some point. Even every keypress you make while responding to this is turned into machine code sent to your CPU. Everything done on a computer is machine code at some point.
>these are also incredibly easy to look up by name, so you have no excuse to be as ignorant as you are
Oh yeah, let me just look up Interpreter in a search engine and click the first result. Oh boy, it's Wikikepedia, but whatever:
https://en.wikipedia.org/wiki/Interpreter_(computing)
>Early versions of Lisp programming language and Dartmouth BASIC would be examples of the first type. Perl, Python, MATLAB, and Ruby are examples of the second, while UCSD Pascal is an example of the third type. Source programs are compiled ahead of time and stored as machine independent code, which is then linked at run-time and executed by an interpreter and/or compiler (for JIT systems).
>I'm a professional programmer
Nice appeal to authority; So am I, your point being?
7f9efb No.16296954
>>16296570
Just use ffmpeg.
781132 No.16296972
File: cf70e656ef72970⋯.gif (Spoiler Image, 110.1 KB, 115x130, 23:26, 9877f2995da0e50daa8c7171bb….gif)
>>16296248
Of course I make a progress report and the thread is page 14. Working on interaction triggers to get my deathtraps tested, also there is a method for checking if a kinematic body is touching a ceiling. Just gonna despawn the pawn and spawn chunks if a pawn is gibbed.
2982f2 No.16296992
>>16296938
>Ah I forgot, it actually reads your line of code, turns it into a demonic summoning ritual, then tells the daemon to magick the bytes of your computer into a specific sequence:^)
What the fuck are you talking about?
>Good luck executing quite literally anything without turning it into a sequence of machine code instructions at some point.
switch (instruction) {
case OP_ADD:
...
case OP_JMP:
...
}
That's what an interpreter looks like. You don't have to turn your data into machine code to operate upon it. You operate upon it as data, dumbass.
> Even every keypress you make while responding to this is turned into machine code sent to your CPU. Everything done on a computer is machine code at some point.
No, it's binary data. The keyboard interrupt, if you're using an interupt-based interface, is machine code, but the keypress data is data. Do you really think that every piece of memory that your computer interacts with is run through the CPU as code? Do you not understand the difference between code and data?
> Oh yeah, let me just look up Interpreter in a search engine and click the first result. Oh boy, it's Wikikepedia, but whatever
Holy fucking shit, you're retarded, look at what you pasted:
> Source programs are compiled ahead of time and stored as machine independent code, which is then linked at run-time and executed by an interpreter and/or compiler (for JIT systems).
> executed by an interpreter and/or compiler (for JIT systems)
That is specifically differentiating between an interpreter and JIT compiler. Actually read the fucking article and understand what it's saying.
> So am I, your point being?
I really hope you aren't, if you can't get basic concepts like the difference between code and data straight, or if you think that every single piece of data that your software operates upon is literally compiled into native machine language and executed as instructions.
Learn some fucking assembly, man.
095121 No.16297127
>>16296735
>I tried to use it but I couldn't get the SDL wrapper to install on Linux.
Through Nimble? I remember some months ago the wrapper wasn't up to date with 0.19.x changes and wouldn't compile without fixing a few things.
74fec7 No.16297136
>>16296992
>case OP_ADD … case OP_JMP …
Which in the end is run on the CPU as a conditional jump to one or a set of instructions, which are then ran. Hell, what do you think a compiler looks like in the end? The difference is that instead of the … in your code block being "shit this out to the CPU now" it does "shit this out in a binary format in a .EXE blob so it can be shat to the CPU later". And guess what a JIT does.
>hurr it's data
Machine code consists of OPCODE + DATA, wowzers!
>That is specifically differentiating between an interpreter and JIT compiler
It's explicitly listing JIT as a type of interpreter in a list of types of interpreters on a page about interpreters.
>the difference between code and data
Data can not be used on a machine without code, which was my fucking point. Everything is run as a series of instructions. Compiled code does so directly reading from disk. Non-compiled code has to go through other code first. That's called interpretation.
1012ba No.16297142
>>16296845
>>What does an interpreter do? Compile non-native code into native code at runtime.
>
>Abso-fucking-lutely not. That's what a compiler does.
All my what? A compiler turns the code into native code when compiled, the native code is run at runtime, it doesn't get compiled every time you run it.
9ae6fd No.16297253
Is this gay slapfight about the definition of interpreters really happening? The real question is whether the language is garbage, for example javascript, even if thrown into JIT.
80180c No.16297258
I've been reading a few books about physics engines and decided to write my own, I have no idea how to make rigid bodies stack, they keep shaking and bugging.
74fec7 No.16297385
>>16297253
You're right; javascript is inconsistent garbage, java is garbage garbage; and C# is most likely not garbage enough that you should avoid it; That last part being the exact argument I made before people started yelling at me for implying C# is interpreted.
>>16297258
It's been a while since I worked on physics engine stuff back in college, but as far as I remember this was fixed with multiple iterations and a bit of handwaving of small forces. Having objects have a rest state should also help, where they act as static objects like the floor until a large enough force is applied to them that they'd move, which is nice anyway as it speeds your program up as well since resting objects only need to check for collisions with moving objects.
781132 No.16297409
>>16296248
To the poster that was having a rough time with their game mod: keep at it and read until it makes sense. If the docs lead you to a word that has no context to you, look it up and read on extended topics until it makes sense. You probably know all of the potential health advice (eat,sleep,etc.), so do that too. Try to post to forums for help, eventually someone MAY see your efforts and offer advice.
>sorry for the filename, mobile posting
80180c No.16297432
>>16297385
yeah right now I'm trying to set up a resting/sleeping state but I believe the only thing that'll do is make my program pretend like it's working fine, if a stacked object awakes it'll still start shaking like crazy.
a5cf79 No.16297435
>downloaded unity and am working through the tutorials
IM DOING IT IM GOING TO BE BESTEST GAMDEV EVER
69cf67 No.16297482
a5cf79 No.16297508
>>16297482
I DONT HAVE ANY YET.
My idea is to start off with maybe some simple projects just to get my feet wet.
Then make a larger one. Its going to sorta be a joke project but funny enough where I won't get bored but give me valuable experience.
2998eb No.16297968
2982f2 No.16298168
>>16297136
>Hell, what do you think a compiler looks like in the end?
A compiler is the same thing, but instead of executing it, it emits machine code which is later run through the CPU.
> Machine code consists of OPCODE + DATA, wowzers!
Where data, if present, is usually a literal value, a memory address, or a register specification. The data is never compiled into the machine code literally. Is that how you think data processing happens, that the data is literally compiled into new machine instructions at runtime and executed?
Honestly, I'm not sure what you're arguing here. You want to consider a JIT compiled execution an "interpreter", fine, but most people won't agree with you, even you apparently don't agree with yourself ("if your language does not compile to machine language that is interpreted by an actual machine (and not a fucking interpreter even if happen you call yours a virtual machine) it is in fact interpreted", C# is compiled into machine language that is interpreted by an actual machine, that's what JIT compilation is).
781132 No.16298330
>>16296972
Interaction with triggers that track player input while in an area. Kinda bugged because it doesn't close the door but this is good. Tomorrow I will work on fixing this, making the player slow down upon turning to allow for percise turning like in Solatorobo, and making gibbed art as well as blood splashes (hopefully I can find sprites).
cf4285 No.16298383
>>16232757
Você está aqui pra sempre.
74fec7 No.16298560
>>16298168
Let me break out the semilegalese:
>Defining a CODE INTERPRETER as a piece of software that that processes CODE from it's base state to another state
<Whereas CODE is defined as a series of instructions that does not consist of OPCODES for the CPU and requires (further) interpretation
<Whereas OPCODES are instructions directly executable by a CPU
>Defining a PROGRAM as a series of data that can consist of a combination of CODE in one or more LANGUAGES and DATA
<Whereas DATA is defined as binary values stored on a hard drive or in memory
<Whereas LANGUAGES are defined as a series of code instructions capable of being interpreted by a CODE INTERPRETER
>Defining a COMPILED LANGUAGE as LANGUAGE that is interpreted at COMPILE TIME into a combination of OPCODES and DATA
<whereas COMPILE TIME is defined as the moment a programmer creates the final distributable PROGRAM
>Defining an INTERPRETED LANGUAGE as a LANGUAGE that is interpreted into CODE plus DATA at COMPILE TIME
If a part of a program has to be changed from code into anything else at runtime, that is code written in an interpreted language.
This includes scripting languages included in programs written in compiled languages, such as a program written in C with LUA code definitions for enemy behaviors for instance, no matter if that implementation of LUA is LUA or LUAJIT.
c7c464 No.16298571
>>16297508
Definitely start with small stuff first.
Even just building certain mechanics within a test project is very valuable to learn how shit works and build a workflow.
7a00f8 No.16298596
>>16296447
Assembly is easy, it's all opcodes and the compiler can do it for you. Why bother when 90% of the assembly you'll use hasn't changed for decades? Algorithms are more important and assembly only makes spaghetti which is unmaintainable. Even the most dedicated are turning away from C++ now so don't expect its reign to continue for much longer.
360d07 No.16298637
>>16298596
>the compiler can do it for you
The reason you use assembly is to optimize the code using information that the compiler literally can't guarantee or know about because it's not possible to express those garuntees in C. Alignment, data padding, etc. It's just not expressible with just C because of the rules of the language.
>Why bother when 90% of the assembly you'll use hasn't changed for decades?
It changes every few years or so, see:
https://en.wikipedia.org/wiki/Streaming_SIMD_Extensions
>Algorithms are more important
Algorithms are just the first thing you optimize. Assembly is what you do when you can't gain more speed by improving the algorithm. It's just the next step of the optimization process.
84838f No.16298647
>>16296248
Man I wanna fug Libbie
2998eb No.16298735
7a00f8 No.16298903
>>16298637
>not possible to express those guarantees in C
It's possible if you rely on undefined behaviour. Embedded programmers do it by sticking with one compiler at a specific version that has the undefined behaviour they want. Most programs rely on compiler specific extensions anyway so relying on a specific compiler version isn't much of a stretch.
>It changes every few years
Instructions are added but the ones used in programs hardly change, see pic. Unless you want to prevent people on older systems from playing your games (or worse, have people play the same game at different speeds with optimised subroutines chosen at runtime), you'll want to ignore yesteryears newer instructions.
>the next step
I'm not sure what year you're from but nowadays, if the algorithm is optimal and compiler outputs inefficient assembler, it's a compiler issue that should be fixed upstream and backported. Outside of compiler work, there's no reason to dip into assembler. Hardware is no longer the bottleneck, algorithms are. Concurrency, anyone?
360d07 No.16298950
>>16298903
>It's possible if you rely on undefined behaviour.
At this point it's much better to just write the assembly yourself. This isn't an option on software that has to run on multiple platforms and toolchains.
>Instructions are added but the ones used in programs hardly change, see pic.
You aren't looking at the whole picture. It's generally only useful to rewrite the functions that need to be optimized. The procedures that you aren't leaving up to the compiler are going to be using mostly new SIMD instructions to gain this speed over the compiled code. The vast majority of what you will be writing should be SIMD instructions. Otherwise there isn't much of a point.
>it's a compiler issue
We already talked about limitations of the programming language. It's not a compiler issue if the language cant express the garuntees you need.
>Hardware is no longer the bottleneck, algorithms are. Concurrency, anyone?
Just because you can optimize in other ways doesn't invalidate using assembly as an optimization tool. Optimizing code is a process and assembly is generally the final step in that process. There are certain problems that require this level of optimization. For example image and video codec encoding and decoding is generally implemented in assembly. Or the compression and decompression of files. Or, finally, video games… all are places where you would gain a lot from this level of optimization.
7a00f8 No.16298982
>>16298950
>C is not suitable for game development
Agreed.
360d07 No.16298995
>>16298982
I didn't say that. What are you trying to say?
2982f2 No.16299007
>>16298560
So you're considering a compiler to be an interpreter if it's running at a different time? Sure, but most people don't use that definition. The people who write LLVM consider it a compiler, whether you're using it as a library for AOT or JIT. Nobody who writes JIT compilers consider them interpreters.
e02699 No.16299122
7a00f8 No.16299126
>>16298995
>I didn't say that
Oh, you kept mentioning the language limitations of C and topic was C++ so I kind of assumed. It's okay, the resident C enthusiasts know that modern games are not written in C. That's not to say hobbyists can't still write games (or engines) in C if they really want to.
>>16299086
And it just got a whole lot worse. >w<
379f14 No.16299156
>>16299126
>modern games are not written in C.
I remember hearing that Cryptic Studios still writes their games in C.
08b450 No.16299178
Someone post that video about why Mike Acton uses C++
360d07 No.16299179
>>16299126
It was more within the context of translating C into assembly language. Writing fast assembly routines is all about how much context you have about your data, you can't express that context in C or C++, in that way C++ is just as limited, although to be fair both C and C++ are getting slightly better, with alignas()… But generating the best assembly still usually ranges from really hard to impossible within the rules of the language.
>the resident C enthusiasts
You've found me, although because we were talking about writing assembly I ended up criticizing C. I don't think C++ is good to use for modern games, unless you restrict the features you use heavily (basically using it almost like C, no fundamental difference) and it's about the same.. and this is how game developers end up using C++ any way.
The whole point of C++ is to enable the style of programming where you have a massive standard library (STL) and you just hook the smaller parts they give you together and watch them go. Faster to write the code… but not good for games, really, and game developers, don't want the overhead-
so it's not really that different.
a3c19b No.16299380
What the fuck does fighting over languages have to do with gamedev? Under 18s are not allowed in here…
360d07 No.16299409
>>16299380
language shitposting because the only "progress" I have made is changing which errors my validation layer shits out
e02699 No.16299430
>>16299409
Imagine how much you could accomplish if you didn't waste your time with language shitposting.
61503f No.16299903
>>16296546
>>16296571
You're both right! parts of C# are interpreted, parts are compiled into native bytecode
f8c626 No.16299929
>>16299086
They can't all be winners.
61503f No.16300047
who was that anon that made the chef game/art for a kiosk?
80180c No.16300224
>>16298383
>>16298735
O maior erro que Portugal cometeu na sua história foi deixar o brasil tornar-se independente
2998eb No.16300365
>>16300224
Artigo 11 e 13 :^)
1ddf15 No.16300434
For a game like Monster Hunter, what kind of structure is used for the AI boss monsters? Would it be a state machine, a behaviour tree, Goal Oriented Action Planning (GOAP), or something else entirely?
2998eb No.16300471
>>16300434
I bet my balls it's just the simplest of simple state machines, with states like "target behind you" and lots and lots of "if random"
61503f No.16300512
>>16300434
>Goal Oriented Action Planning (GOAP)
If something has a name like "Goal Oriented Action Planning" it's probably bullshit that doesn't benefit anyone
2998eb No.16300543
>>16300512
nah, it's a pretty neat setup for complex AI. Mostly unneeded, since AI is hardly ever complex enough to need this.
It's for things like
AI wants to start fire -> AI needs wood -> AI needs to cut tree -> AI needs to get an axe -> AI picks an axe -> AI cuts a tree -> AI gets a log -> AI starts a fire
It works and it's neat, but, have you ever seen a game where the AI NEEDS to do these things?
08b450 No.16300555
>>16300543
Sounds like more of a game design topic than a technical one. There's numerous ways you could accomplish that effect.
47d044 No.16300691
>>16300555
>Sounds like more of a game design topic than a technical one
OK, here's the technical parts of GOAP.
GOAP is a decedent of STRIPS (Stanford Research Institute Problem Solver) and uses backward chaining with A* to find the shortest distance from the desired state to the current state. It generates the connections from one action to the next automatically, so that the AI programmer only needs to write an action then define the results and the prerequisites. The AI will assemble the actions into a list that accomplishes the goal, so that the programmer doesn't have to. That really is the main benefit of GOAP, it's easy for a single AI programmer to get pretty good behavior without so much work.
While GOAP is faster than STRIPS, it still falls apart when it comes to performance in the grand scheme of game AI. Other planners, such as HTNs (Hierarchical Task Networks), can often do the same thing faster. Modern day implementations of GOAP will off-load many tasks to subsystems. For example an AI may have an attack action, but must have a sword drawn to actually attack. Rather than create a sword draw action that GOAP has to assemble into the action plan, that sub-action would be given to the attack subsystem. The AI is told to attack, so the attack subsystem checks if the sword is drawn. If it's not drawn it draws the sword and attacks, if it is drawn it just attacks. GOAP meanwhile waits for the task to complete. It doesn't care whether the AI had to draw the sword or not.
>There's numerous ways you could accomplish that effect.
I would like to hear them. I'm always open to learning about new AI systems.
2998eb No.16300854
Finishing ingredient xml
5 done!
84 to go!
74fec7 No.16301323
>>16300854
As much as we may fight in our language wars, I am pretty sure all of us are happy knowing that your brother's monstrosity is being replaced by functional system.
2998eb No.16301459
>>16301323
Ha you are talking to the wrong brother.
I was the one who created the hardcoded fuckery, while he creates the new actual code i make the xml following his samples since it requires no thinking at all.
His job is doing the new code, mine is doing everything else, xml and csv files are content, not code.
f367ec No.16302283
>>16300854
Although probably too late but if you use attributes instead of content for content, XML actually looks neater and only use content for nesting multiple tags within
0699b7 No.16302424
>use Unity's FBX Preset importer to save time and mass reimport animations with different settings
>fucking thing renames the animation clip IN the FBX as well
>pic related occurs for all 100 animations
Goddamit. It's still all linked properly but now every animation has the same clip name and I cant tell which one is which anymore.
2998eb No.16303013
36c075 No.16303046
Hey, my shitty isp finally unblocked 8chan
2998eb No.16304274
>>16302283
sounds gay, content can't be arrays if done as you say
61503f No.16304788
>>16303046
If you live in an area that can block your hobbies, just get a vpn
c1bd56 No.16305138
Hey here's a fucking retarded question, would anybody here have need for a voice actor?
I can't guarantee i'm any good but, i exist.
47d044 No.16305256
>>16305138
Upload samples and post a link. If you're unsure what to record act out some dialog or a monologue from a movie.
https://vocaroo.com/?upload
c1bd56 No.16305268
>>16305256
Alright, is there anything you'd specifically like acted out?
47d044 No.16305355
>>16305268
I'm not looking for a voice actor right now, sorry, I'm trying to help you sell yourself better. If I had something specific I wanted acted out I would have given direction for that.
c1bd56 No.16305374
>>16305355
Ah okay, all i have at the moment are various readings from a TGquest and some odd requests.
47d044 No.16305381
>>16305374
Share them, so long as they reflect your current ability. It's better than nothing.
c1bd56 No.16305386
178451 No.16306204
>>16298330
This is what I want the menu in the game to look like:
TEK/MOD: select abilities and toggles that modify them.
EQUIPMENT: each party member can hold 3 items, some have passives.
SQUAD: change active party member or order them to do a command (like Xbox's Brute Force)
CONSOLE: open the terminal text editor, can be used for hacking computers.
Also I want to make a a chamber like the Dead Space stores that add new armor, but instead it breaks down one of your party members into stored matter and they can be added back into the party at any station, like the PC in Pokemon. Any of your guys can die in the platforming/stealth/combat sections if you can't respawn them with an item and can be spotted or targetted, so this way you can reduce the amount of people you have to be concerned about on a mission. Last 2 levels wont have these. 10 levels planned.
I grabbed some more textures, plus blood and energy sprites for graphics (all pulic domain or with credit).
Spent today planning, getting back to this Sunday.
3f9c4e No.16306290
I really hope you guys downloaded the Libbie archive before Mixtape went down. Because I accidentally deleted it with dd…
2ee1a7 No.16306299
>>16306290
I can share my Libbie collection through IPFS tomorrow.
3f9c4e No.16306303
File: 861753c6c9e43c2⋯.png (Spoiler Image, 835.71 KB, 1976x3311, 1976:3311, 861753c6c9e43c2ffb050f7896….png)
>>16306299
You got the lewd folder too?
47d044 No.16306768
>>16305386
Amateurish and flat. You need a lot more practice and confidence in your line delivery. Watch theater and movies with lots of dialog, then try to copy the vocal inflections and tone. Don't copy the voices, if it's a girl don't try to sound like a girl or if it's a guy with a very deep voice don't try to do a deep voice, copy the emotion. Listen back to the recordings and figure out where to improve. This is easier at first when you have something that you're copying and can compare to directly. Try to capture the emotion of lots of different sources. Angry, happy, sad, aloof, focused, direct, maniacal, worried, bored, monotone. Anything and everything.
As a non-voice actor there's only so much advice I can give based on what I've heard and read, and the above amounts to "figure it out for yourself" and "develop a good ear". If you can find someone with actual experience to talk to that would be for the best, and if you find a source from an actual voice actor that contradicts what I've said, go with that other source, not me.
Your recording quality is also sub-par. The audio peaks at multiple moments. Your voice isn't clear and I can hear background noise such as you fidgeting and clicking the mouse in many of the recordings. I can't see you, but I would guess you're too close to your microphone and talking directly into it. Give the mic some distance and talk just to the side of the mic. If there's a gain knob turn that down to the point where when you are at your loudest you're just under the point of peaking.
There's a series online called "how to be an online voice actor" that goes over everything in more detail. I'm not going to hold it against you if you don't go out and buy an XLR microphone, audio interface, and foam sound dampening pads since you aren't doing it for money at the moment but there's still stuff you can do to improve your audio quality and plenty you can do to improve your acting.
Godspeed anon.
d2c721 No.16307663
I can't decide if I should go for choppy or fluid animation for a game of this style. Here's my attempt at fluid using my first non-plagiarized tile and it looks like a gbc game trying to act like a dos game. What do you think? I kinda like it.
Other than that its now possible to load sprites of any size, simplified collision formula to scale with sprite size, floating sprites let me make a menu, and 'object sprites' are now a thing and are rendered onto the map tiles but they do nothing yet. baby steps
1f31ad No.16307789
I'm going to make a multiplayer game, but my friend told me to make a singleplayer and put a multiplayer later.
I'll be doing that in unity, but I'm not convinced it will be easy to implement something like that later. How would you guys do it?
979591 No.16307848
>>16307789
If you make a single player game, you'll probably never finish a multiplayer game.
If you make a multiplayer game, you'll probably never finish a multiplayer game.
97a201 No.16307868
>>16307789
it's kind of a double edged sword
at my current job, a good portion of the game was made with multiplayer in mind, but certain features branch a bit too far apart and after a while some stuff ends up being cut from multiplayer because figuring out how you're gonna make it work for both and implementing it would take too much time
c1bd56 No.16307978
>>16306768
I don't have a volume gain knob sadly, as for emotion i am trying to put it in,guess i'm failing.
1f31ad No.16307981
>>16307868
>>16307848
This is what is making me worry, realizing it is beyond what I can do.
9ae6fd No.16308020
>>16307981
You have to think about these if you really want that multiplayer to work and not become wasted dev-time.
>update data size vs. network speed
>constant speed deterministic simulation vs. dynamic framerate physics
>wait for peer input vs. predict next state
And basically construct the engine around it.
Good luck.
66f696 No.16308086
Haven't been posting any media lately to avoid spoiling the game, but here's a little peek at some of the new stuff.
979591 No.16308097
>>16308086
What kind of genre are you going for? And when's the release date?
ed1dd6 No.16308105
>>16307663
Looks cute, anon. I like it.
66f696 No.16308114
>>16308097
It's a puzzle adventure kinda game. It's still far from finished, so I don't know when it's going to be out…
140c91 No.16308123
>>16308086
spooky gondola you got there but looks very comfy overall.
Will you present it on the next demo day?
52e5a6 No.16308136
66f696 No.16308171
>>16308123
Thanks. Yes, I'm planning to make a demo for 05/05.
7abf24 No.16308186
>>16307981
If you're set on a multiplayer game, my advice would be to code it section by section, first in single-player and then in multiplayer. Makes a single unit of the game (enemy wave spawning, weapon switching, something like that), and get it working for a single player. After that, make it work for multiplayer. Then, and only then, move on to the next game unit.
The benefits are that you get more evenly-spaced out successes to psych yourself up, you're not delaying a huge workload in the future by having 20 systems you need to check and cross-check later on, and every time you work on a system in multiplayer it's still very fresh in your mind from when you got it completed for single-player.
1f31ad No.16308327
>>16308020
>>16308186
I suppose it all boils down to discipline and organization then. It's a hard habit to train, but I'll have to at some point.
Thanks for the insight.
05c009 No.16308373
>>16307663
Try a choppy animation instead, I think it would fit the old style graphics more. The fluid one looks kind of over engineered.
990f8d No.16308460
>>16296467
>The STL, after all, is one of the most efficient and simplest to understand of all libraries of similar caliber.
KEK
E
K
66e257 No.16308857
>>16307663
it looks quiet nice and helps to it looks distinct from ussual rpgmaker trash
d2c721 No.16308913
>>16308857
thanks, but I'm not using rpgmaker..
017e4c No.16308943
Progress was a bit slot the last few days. But slow is better than doing nothing at all and finished the graphics options and character selection. (Obviously still waiting for our artists to finishe the characters.)
2998eb No.16308981
>>16307789
Multiplayer complexity can range from easy as fuck to balls deep hard
What is the game to begin with?
And it's better for it to be a single player running on a local server from the get go if you want to add multiplayer later
1f31ad No.16309188
>>16308981
>And it's better for it to be a single player running on a local server from the get go if you want to add multiplayer later
Fuck, I didn't even think about that. I'm retarded.
It's just a platform game like kirby. Player 1 will have health and will be ranged and player 2 will be like tails, he can die and come back after some time. It's nothing too complex.
40ff96 No.16309339
>>16308460
It's actually true. The basic fact is that you don't need most of what the STL does. The STL is slow because of all the guarantees it needs to provide about consistency, coherency, and iterator invalidation. You're better off sticking with the STL until it's your bottleneck, and then replacing it with a container that only offers what you actually need. There's no reason to hand roll everything off the bat or to write your own before you need it.
9253f4 No.16309456
>>16296248
Libbie hoof worship is best worship, just imagine getting to lick her feet after a long day of programming with her
0b5b79 No.16309466
>>16309456
>footfags
>smelly ones at that
9253f4 No.16309516
Algorithm discussion so the thread doesn't get completely derailed:
What interesting algorithms have you anons found and worked into your game? Anything that strikes you as particularly unique? Anything that you actually had to remember high school calculus class to make?
>>16309466
I never said they had to be smelly, but I certainly wouldn't mind it if I had to gag a little bit at first :^)
69cf67 No.16309542
cab112 No.16310704
>play Spellcross for a while
>fiddle around with the DOSBOX midi configs and the accompied sound CLI program
>after several hours of fiddling around it doesn't work
>look up if some goy uploaded the midi files instead
>find several tools for this game instead made by a czech guy
check it here m8: spellcross.kvalitne.cz/index.html
Gud shit, I don't know yet if it works in wine or not but I will try it soon. I am not going to plan to do any serious modding for this game. Because I have other stuff to deal with for now. I have uploaded the game and the tools at the share threda volafile because only the devil knows how long this site will stay. I have seen enough modding and mapping sites that got lost in time.
a7a518 No.16310736
Silly question, but how would you implement a Roman-like MMO with different castes and slavery and that could be sandbox and "survival related" by exploring, building cities and expanding?
Maybe a limitless skill learning in the same way of Oblivion, Eve Online or Mortal Online.
88db32 No.16310889
>>16308327
>>16309188
I don't even know what kind of game you're planning to do. Local multiplayer doesn't need all the network stuff and deterministic/predicting engine. I intended >>16308020 as a half-joking jab, to ensure you drop out now or get ready for all the work if you really want network multiplayer. Ping-time of ~20ms vs. frame time of 16.6ms for 60 FPS gameplay forces you to think about how you make the engine to keep all clients synchronized.
0ea6bc No.16311034
>>16310704
Neat! I did like that game even though I think it's mediocre at best.
I hope to one day see another game execute its premise decently.
77d896 No.16311214
>>16306204
Made the switch to Godot 3.1, lots of issues in script that weren't issues before. It is really obnoxious.
cab112 No.16311500
>>16311034
I like playing this game more than UFO Alien Invasion mainly the reason for that is that there is a lot more vehicles, tanks and helicopters available. There is however 2 major flaws I could already see in this game, first is how the upgrade system works you can upgrade your experienced units into other units as long it belongs to the same group thats okay, the real trouble starts upgrading your units with items even if its "fairly" cheap compared to a commando unit (85 credits~) for example but if you do that for 5 infantry units it cost you already 50 credits which is fucking a lot and once you did that you cannot get back those items anymore so when you upgrade those units which has those items into a different units they will just simply lose it. The second issues is that while you can group your units in tactical model it is far more troublesome than it is worth it mainly because you don't know if your slower units in your group can even reach the destination target and once they do its nigh impossible to tell where they will stand. So for those that don't know yes 50 credits is fucking a lot in veteran mode, considering that the more injuries your unit suffers the more it will cost to "repair" it and even more so if you want to keep its experience which it will lose when you choose lower then elite recruitment.
So I think this is actually 3 issues I could think of so far, the third one is as I mentioned before how this experience dynamic works. You need as much experience units as you can get as it will determine its firepower against the other side experienced units, on second level those units can reach from 6-8 stars already which can and will decimate your lower ranked units. But on the other hand having to develop also a feature where you have more control over your items so that you can plan your budget much more efficiently was probably already too late for those developers to change it just in time. Considering in the czech version you need to even pay attention that you don't allocate too many points for the research else you will get those over allocated money not returned. But at this point it is just speculation unless there is a czech anon that shows up that knows more about this game development I wouldn't know it better either. I did looked a little bit more what other sort of game Cauldron used to make and so far it looks like spellcross is their first and last turn based strategy game which is a bit of a shame since in my opinion it is a untapped intentional to make this game more complex but not lossing to much of its action aspect either, for example having a more in-depth economical and resource procession aspect would be great as it will give those territory much more different values it could have.
The English version did have some additional fixes and several other changes made to the game, the first thing that is noticeable is that the english intro version is much shorter and showcase the overall situation as less desperate than in the czech version. According to a jewtuber the development of this game was quite troublesome and the producers of it didn't had much faith of this game which would sound a bit plausible why it didn't had more exports in the western European part.
In the other hand this game died peacefully and its corpse hasn't been necromanced to serve as a vassal for cash grabbing such was latest C&C titles nowadays is.
tbh fam I want to make a spellcross clone too once I git gud at SDL 2D C engine development. I got at least some plans how hypothetically the engine and the game could be structured. It is far more complex game then a asteroid game obviously but it is also in one way simpler in features compared to X-COM: UFO Defense.
6806e7 No.16312215
Today's progress:
>functional pause menu
>game over screen
>groundwork for win conditions
Y'all better be making progress, even if it's just a little! At least 1 productive commit per day!
f965aa No.16312311
>>16312215
Never does a day go by when I don't do at least something. I'm sure most anons make far more progress but as far as I'm doing something, I don't feel bad, even though my current expected release date may be upwards to 4-5 years from now.
a7a518 No.16312536
>>16310790
I just want a comfy sandbox simulator without grieffing ridden servers.
873afc No.16312632
You guys motivated me to actually post progress. I'll try to post weekly updates. I'm learning Godot. And what I'm doing is Cathedral board game clone. It's somewhat similar to tetris you put pieces into the 10by10 board. I picked free sprites to make progress. 10 by 10 grid I've implemented via TileMap and pieces I've implemented via multiple Sprites. Not sure if a good idea tho. I've done only white ones. Black ones and Cathedral still to be done.
Fans of the game already developed notation how tiles should be represented. So I'd try to stick with those.
Now I need to figure out how to put the piece on the board.
The next step would be to determine what square are already occupied.
Long term goal is to move to isometric. Than to pre rendered model isometric and after that to the 3d.
The question that bothers me: Is it OK to put it on itch.io under the original name at some point and ask a buck? Probably, noone gonna pay, but potential shekels might motivate me.
69cf67 No.16312648
>>16312632
Just Like Make Game™, anon.
If you build it and post it on itch.io and do PWYW model, you're probably still comfortably in anons' green zone. Patreon is a warning flag, but would be suitable if you're a poorfag and can actually deliver on stuff.
If you can hit 1,000 downloads, you can probably call it a success.
2998eb No.16312660
>Haven't even finished the fucking thing
>Already having to rewrite the very core of it
Gee what a glee
69cf67 No.16313040
So vid related is the game I'm trying to clone. I'm trying to make my own implementation of their wire/logic system and extend it (so it's not a direct copy and the creator of that game doesn't get mad at me). They have on/off switches, as well as AND, OR, NOT and XOR logic gates, which is pretty standard.
What I wanted to do was instead of having a standard on/off state, change it to a RGB spectrum, so wires could carry essentially 3 flags with them. The problem is that I'm not entirely sure how to resolve multiple inputs or handle the output. For example, looking at the other pic, the XOR gate would be triggered by having either (RGB) from the left, or (R) from the top, or (B) from the right (but only one of those sets). In the case that it has exactly one input, what would its output be? Should it repeat/clone the input into the output, or would it make sense to have a colored output gate (green in this image) and only send out that signal?
I understand truth tables and boolean logic just fine, but when you get multiple valid inputs into a gate, it's confusing
cab112 No.16313056
>>16313040
I wonnud how long it wil l taken that a autismo will make a complicated circuit thing of your game map beer posting is beer best posting
69cf67 No.16313076
>>16313056
>Of my game map
Pretty long, considering I have no game yet.
But in Subterra 2, there is a more complicated level to that effect. In the first Subterra, there was no logic gates, and the developer did make a turing-complete byte-adder level out of rocks and conveyor belts and stuff. It was pretty cool
f965aa No.16313957
In Blender, I'm trying to use a semi-transparent texture (not final, it's the first attempt at it). In Edit mode, the model shows up fine as it should, but in Object Mode, it makes certain tris behind it transparent. Pic related. Is there any way to fix this? Sidenote, there's no pattern I see to which tris remain visible. Those from another object do, yes, but so do some from this object at seemingly random.
2dee9c No.16314176
>>16308943
Wow,anon.
That is a really nice cutscene you have there.
2998eb No.16314288
4th day filling xmls and localization csvs
I no longer feel my hands
Please send help
32ddb4 No.16314297
>>16313040
Well what are you trying to do then, specifically? Have the xor operate as a switch to a mux to pass data through or have it operate as a boolean gate that simply activates when conditions are met?
88db32 No.16314315
>>16313957
Transparent polygons are kind of fucked up in the editor. A fully transparent invisible polygon still writes to the Z-buffer, so any polygon drawn later behind it is basically erased there, leaving some random background color or polygon. I think the solution is ordering polygons by Z-depth, but Blender doesn't give a fuck about that.
Actually, this problem is not specific to Blender, I've seen MMD videos with similar issues, like a whole model erased by a half-transparent beam of light.
The problem should be limited to hardware based OpenGL/DirectX rendering. Blender's render engines shouldn't have this, even the old Blender renderer goes for a multi-layered rendering to solve this.
I don't know how to get the editor view unfucked.
873afc No.16314483
>>16313040
As a quick and dirty solution: you can just make colours ignore each other. For example OR gate receives 2 red, one blue and three greens the output would be one green and one red.
Anyway, why do you need XOR for? When it's not needed to be Turing complete.
22f6f9 No.16314700
>>16313957
Switch to perspective mode.
69cf67 No.16314705
>>16314297
>Well what are you trying to do then, specifically?
See that's what I don't know. I feel like I'm trying to merge two separate concepts, and getting lost in the middle
>>16314483
>Why XOR?
Well, the problem extends to other gates too. Like if I had an AND gate instead, I'd still need to muddle through the color combinations and have the same input/output problem. Also, and/or/not is just 3 gates, and xor rounds it out to an even 4 which makes it easier to lay out spritework
I think the cleanest solution for the gates is to iterate each direction, then for each direction treat the RGB value on it as a mask against the input wire RGB values. If the mask matches, it has boolean input in that direction which is then applied to the logic gate; it will then send out whatever RGB color the gate was.
Or maybe I can still color the output gates in a similar way. It's hard to convey the information I want in a simple, clean manner.
f965aa No.16314708
>>16314700
You mean from ortho? Because it already was on perspective. Ortho looks pretty stupid, too.
69cf67 No.16314730
>>16314705
Oh okay this is a lot closer to what I have in mind I think.
>Top: AND gate has two inputs and one output. Both input colors match, output is enabled.
>Middle: AND gate has two inputs and oneput. Both input colors match, output is enabled. If the top wire was pointing out, then it would emit green. If the top was was pointing in, then the gate would be missing an input (3) and would produce no output.
>Bottom: AND gate has three inputs and one output. One input color is incorrect, and it produces no output.
The only thing here is that I would want some way to indicate at a glance whether a gate/wire was active or inactive. I have it animated to fade between black and the color, but even so, the blue is very dark, for example.
32ddb4 No.16314763
>>16314730
I admit I'm a bit lost as to what you're ultimately trying to do, but it seems like the logic you're working with is actually sending 2-bit inputs around (3 colors and no-color), so if you're trying to use base-2 logic to solve base-4 issues, this might be a/the problem.
cab112 No.16314764
>>16314737
I have no idea what the fug is going on.
69cf67 No.16314810
>>16314763
No you're basically on the money for that. I'm taking a 1-bit system (boolean on/off from >>16313076 ) and trying to somehow extend it to 2-bits.
69cf67 No.16314812
>>16314810
>>16314763
Or more specifically
>Have logic gate, detects adj tiles as either input or output
>When it receives a wire signal (RGB flags), compares it to its own RGB flags on the direction it arrives from
>Each input can be either on or off, effectively converting a 2-bit input to a 1 bit value
>It then takes all of its inputs and applies the gate logic to
>If the truth table resolves to true/1, then an output wire signal is created
68697d No.16314829
>>16314737
Do you think this is a motherfucking game?
da7cd6 No.16315722
>>16314737
Get out of here, nigger comedian.
6806e7 No.16315753
>>16314176
Thank you, Anon.
Can you tell me more about that phenomenally handsome fellow in your pic? He looks like the sort of man who'd make the best video games in the world.
77d896 No.16315849
>>16314737
There is an april fool's containment thread, please post garbage in there as it is hard enough to keep /agdg/ threads stable as-is.
2ee1a7 No.16316450
>current year + 4
>not writing your game in PL/I using a homebrewed compiler
git gud
cab112 No.16316463
>>16316450
So where is your game that utilizes this PL/I language with your self made compiler?
2998eb No.16316478
2ee1a7 No.16316487
>>16316463
It doesn't exist but it will be better than every C weenie game and compiler combined.
There are no reserved words in the entire language except for DATE and TIME.
cab112 No.16316506
>>16316487
>but it will be better than every C weenie game and compiler combined.
>It doesn't exist
So well looks like there won't be any benchmark of your imaginary game I suppose.
2ee1a7 No.16316517
>>16316506
But at least it won't suck.
f7e17e No.16316522
>>16316517
It sucks. And it would probably suck even more if it existed.
2ee1a7 No.16316531
>>16316522
You'll never know if it sucks or not because you won't get a chance to try it.
cab112 No.16316537
>>16316517
Well of course it cannot suck because the best performance you can have is code that doesn't get executed in the first place. So it's no wonder your imaginary game is winning every possible imaginary benchmark test and contest available, clearing off the scrubs program with your superior performance.
>>16316531
>because you won't get a chance to try it.
Congratulation Mr. PL/I Anon aka multics ranter, was getting caught part of your plan? C'mon dude, /tech/ is already fucking terrible with all the shitpost going on there, can you at least spare this threda from your unhinged autism shitpost at least?
2ee1a7 No.16316545
>>16316537
The actual UNIX hater fag is too pleb for libbieposting. Speaking of which, I'll dig up more Libo shitposts from my backup drive before uploading the folder to IPFS.
f7e17e No.16316546
>>16304274
He's telling you to do it like this: <type val="HOT"/>
Which makes parsing easier and uses less space.
2998eb No.16316565
>>16316546
>Which makes parsing easier and uses less space.
Maybe it uses less space, but parsing will be just as easy, except my parser is already finished and does everything I need, so redoing what's working is not really what I'd call "easier"
f7e17e No.16316571
>>16316565
>redoing what's working
It's just a tip to optimize when you got some spare time.
You can probably simplify the already existing XML files with regex.
5f5a87 No.16316585
Shinobu game will now have lewd. Working now on advanced loli labia pussy and feet physics.
2998eb No.16316589
>>16316585
I thought those were done years ago
97a201 No.16316592
>>16316585
shinobu is not for lewd
f26346 No.16316597
>>16316585
Perfect, April is a great month to implement lewd features.
2ee1a7 No.16316601
>>16316597
Alwaysfap April?
b27ac2 No.16316608
Shmup anon here,
In between phasing out and various stuff to do and reworking the explosion effect, I'd say it's much stabler.
Instead of having a giant unstable mess, I divided layers and minimize mixing them:
- Light and heat only expand and retract in their own data
- Pressure only affect speed
- Control pixels moving around depending on speed
- Smoke will be generated from control pixel, move according to speed, and have color affected by light and heat
Currently working on the control pixels, since I have less thing on screen I can better see if something is moving weird (currently the case), but I have higher hope than before.
69cf67 No.16316639
Now that I have my wire signal system pretty much concepted, I can cross that off my list and go back to fussing over my actor query system for two months (again)
As a reminder:
>Up to 256x256 tile map
>Actors can occupy zero, one, or more tiles (typically 1-2 as moves)
>Tiles can be occupied by exactly zero or exactly one actor
>Actors are updated in order from bottom right, moving horizontally, up to the top left (N..0)
The issue I was trying to solve over the last few threads were query commands such as GetActorAt(xy). Right now I have an Actor[][] which works in this specific instance since the number of tiles is fairly low, but it doesn't scale. I also maintain an Actor[] (up to 2048 elements) which gets sorted by position every tick for updates.
If I try to have Actors that don't occupy space (eg a conceptual one such as a switch toggle that exists and then disappears, or a spark that travels along wires to verify them) then it kinda breaks down. This makes a simple 2d array, dictionary, and hashmap unsuitable for this problem. I think quadtrees were suggested at some point but that seems like overkill for this.
f7e17e No.16316717
>>16316639
2D arrays are a shit idea.
Just use: [x+y*map_width] in getActor();
Welcome to elementary schooler math.
I'm assuming you're using C# from what you wrote.
>Actors are updated in order from bottom right, moving horizontally, up to the top left (N..0)
for(short y=255;y>-1;y++) {
for(short x=255;x>-1;x++) {
getActor(x,y).update();
}
}
Holy shit, am I supposed to make your game?
>If I try to have Actors that don't occupy space
Just keep your your actor list and your actor map separate. How complicated is that?
f7e17e No.16316722
>>16316717
meant y– and x– obviously
b8e987 No.16316724
>>16316585
That better not be a joke.
>>16316592
Are you for real? She is lewded in the official materials.
f965aa No.16316727
>>16316585
Nice april fools joke, anon.
69cf67 No.16316792
Oh fuck I just realized from this mockup, that while it looks pretty, it's literally just as function as if it were all 1 color. The additional colors offer no mechanical depth
>>16316717
The updating is just fine, I had already figured this as I said. Besides, what you've done is overkill when there's only 2000 elements (at most) in a sorted array.
It's the spatial mapping in an efficient way that's the issue.
2998eb No.16316814
I'd like to ask you all for insight
My brother suggested our alchemy system be some sort of bitshift puzzle
Each substance have some bits turned on or off, different tools would actiave/deactivate bits depending on which one are on/off and maybe some other state, like temperature
The issue I raised is that this seem to simple to make a calculator, just insert the desired bits you want, and the computer would A* it's way to an starting item and state the needed steps to acquire the substance you need
A) Is there any need to care if the alchemy puzzle is easily solved by an online calculator? Because I feel it would ruin game balance and remove all fun of the system
B) If so, what's exactly the key to make something solvable by humans but nearly impossible for computers? (which is also run and evaluated by a computer at the same time)
Having the substance bit combinations being randomly selected on game start would only add a calibration step to the calculator, stealing even more fun out of the system
69cf67 No.16316823
>>16316814
>A) Is there any need to care if the alchemy puzzle is easily solved by an online calculator?
Players will always "cheat" in your game. Best to let them if it's a single player game. However, if they need a calculator, then maybe it's best to think about ways to make it more intuitive. Perhaps a skill/feat activates a UI function that shows them the result?
>B) If so, what's exactly the key to make something solvable by humans but nearly impossible for computers?
A captcha. Don't actually do this.
2998eb No.16316833
>>16316823
>A captcha
First off, computers are already able to fill captchas https://web.archive.org/web/20180817045933/https://please use archive.is2014/4/16/5621538/google-algorithm-can-solve-recaptcha-almost-every-time
Also, where am I meant to add it to solve the issue? How does it help in any way even if it was a good idea?
2998eb No.16316836
2998eb No.16316841
>>16316823
Also, using a calculator is not cheating
You are still playing within the rules of the game, you just took the best possible route effortlessly
People who would never use cheat engine would use calculators (like me for instance)
69cf67 No.16316880
>>16316841
The intent is for them to learn the system, solve it and figure it out, in order to create the best alchemy results, right? Using a calculator, while still within the rules of the game, bypasses this and is not explicitly intended (Eg you didn't design the system around it). I would say it qualifies as cheating.
But again, that's not necessarily a bad thing. It just means you're aware of it ahead of time and could think of ways to encourage them not to do it (as opposed to punishing them for it)
f26346 No.16316960
>>16316823
There are better captchas, like rotating shapes so they are the right side up.
2998eb No.16316976
>>16316880
>(as opposed to punishing them for it)
I was thinking of making a calculator impossible rather than somehow punish the player for using it (as there's no way to detect to begin with)
f8c626 No.16316978
>>16316592
>>16316597
>>16316724
>These obvious non-regulars don't know
97a201 No.16316984
>>16316978
i've been here before u
9253f4 No.16316993
>>16316976
Dumb idea. If someone wants to make a calculator for your game its fundamentally impossible to stop them from doing it. Instead of punishing them for using one, or trying to make one impossible to write, make the player feel smart for figuring out a puzzle and let that be the motivation.
873afc No.16316996
> Figured out how to put pieces on the board.
> 10 by 10 grid implemented
> Orienting piece is working
> Calculation where piece is, probably, also working.
So far so good.
63772e No.16317066
>>16316993
Is it fundametally impossible to stop people from creating custom clients for your online game too?
873afc No.16317096
>>16317066
Yes. But usefully you just need to make it hard enough.
Ideas to try:
- don't use Java/JS/C# they are way too easy to extract
- absolutly, encrypt traffic
- add some key into your app to access your server
If it's not very beneficial to reverse engineere your app it should be enough.
9253f4 No.16317098
>>16317066
Yes, it is. The only reasonable ways to mitigate it seem to be either obfuscating the shit out of your game code and doing shady shit client side, like installing borderline spyware, or by using machine learning to figure out when someone is fabricating data. Neither of these are 100% effective. Don't waste time trying people not to cheat at your game, make your game fun to play and minimize the number of people who will want to cheat in the first place.
2998eb No.16317318
>>16316585
How many will fall for this?
d5b81b No.16317337
>>16316585
>mfw this will never happen
2998eb No.16317342
>>16317098
you can 100% stop cheating on multiplayer games except for botting
auto aim is botting, prove me wrong
9253f4 No.16317456
>>16317342
Right, and how do you plan on doing that exactly? I'm sure you have a new and unique method that nobody has ever tried before.
61503f No.16317464
>>16317456
Install malware that searches for programs with the name "aimbot" in them and terminates itself if it finds them.
2998eb No.16317474
>>16317456
>and how do you plan on doing that exactly?
What am I planning to do?
69cf67 No.16317491
>>16317464
I've named my aimbot calculator.exe, what now?
f965aa No.16317500
>>16317491
Search for programs called calculator.exe and delete system 32 if they're detected.
Should have said calc.exe
88db32 No.16317527
5f5a87 No.16317717
>>16317527
LOL holy shit that does match.
61503f No.16317954
>>16317928
>Whenever your games are ready, I'll be looking forward to playing them.
I'll send you a key in 5 years. I hope you don't die.
36c075 No.16318140
7263be No.16318389
I'm making a dungeon mod for Morrowind. Currently have two areas plotted out, not entirely decorated yet. I really wish the Openmw construction set had grid placement, because I prefer its look and feel immensely over the standard Morrowind construction set.
7263be No.16318727
>>16318389
Found better models and textures for the tileset. They're much needed, because the stair tile for instance had no collision on its walls.
0699b7 No.16319024
Nah but seriously I finished remaking her pelvis+butt+legs+feet. Better deformations now. Also made a parallax corrected cubemap shader for fake realtime reflections. And working on a bath bubble shader.
0699b7 No.16319025
Works for vertical mirrors as well.
a74a87 No.16319035
>>16317954
>I'll send you a key in 5 years. I hope you don't die.
uh… thanks.
>>16318140
Whatever you say man. I still think your efforts are worth more than AAA output.
095121 No.16319040
>>16317928
>I still listen to the song from your demo often.
It's funny how many people mentioned that it was oddly fitting. https://freesound.org/people/imaginaryband/ is who did it. Sadly it looks like he doesn't do free songs anymore. My friend is constantly on me to get back to working on that puzzle game.
4edf94 No.16319079
>>16319025
So the player is a vampire, too?
4edf94 No.16319216
7263be No.16320124
>>16318727
Players really ought to bring a torch to this one. I've got the layout of the first section of the dungeon finished. I'm considering having an underwater part, but the thing is, you can't see shit underwater when it's this dark. Either I'll have to put some lighting in the underwater part, or scrap the water idea entirely.
2998eb No.16320314
873afc No.16320753
>Added pieces' scenes to act as a buttons.
>Added action for mouse clicking
>But unfortunately, _unhandled_input() intercepts all clicks on the screen and not only those on the piece.
How do I know that user clicked on the piece without using Button? I wan't to reuse pieces on the board. Is it a good idea?
41ab9d No.16321033
Too busy for any big hurdles right now but I made it possible to render repeating tiles of any size which helps with making the menus. I still plan on limiting myself to 16x16 tiles or multiples of.
7263be No.16321087
>>16320124
Decided to scrap the water for this section, might have it for section 2 though. Now on to decorating and enemy placement. Because this is the Soul Cairn, most of the physical items are foreign to the realm, brought by necromancers and wizards who seek to make pacts with the Ideal Masters or even reside in the Soul Cairn. Unfortunately for them, they almost always meet their end in the ghostly grasps of the wrathmen.
2998eb No.16321125
wew its finally over, 89 ingredients into xml form, now just gotta wait my brother finish the implementation of the effects
f26346 No.16321594
>>16320753
This is Godot, right? Use events.
d75ae6 No.16321959
whoops forgot my image
>>16311214
I know what I want my demo to be. 3 rooms, one an empty debug room with a sound player computer and an animation computer, the north wall has a transition door like Mega Man Legends which loads the next room. This one will have a room with a pit in the middle, preventing the player from getting to the next exit. There will be a platform on a track near walls with computers attached to them. Activating a computer near the platform moves it along the track, staying in one spot for too long makes the platform tilt the player into a pit, forcing them to climb a ladder. The platform is by walls to obscure the player from a camera and a turret. The player has to get to the other side of the pit to disable the camera and turret then go through a door. The third room will have an RPG battle where the player is auto-spotted and must navigate in the Vandal Hearts inspired grid system. On winning, player will probably just get gibbed and a thanks for playing.
d6a9b2 No.16322765
Does anyone know any comprehensive guides, preferably written ones, that can help me understand how classic FPS movement mechanics work?
I've been looking at the source code of various iD and Valve games but I haven't found any explanations for what I am looking at.
66c9d3 No.16322776
>>16322765
What? You float X distance above the ground, that's all there is to it.
e02699 No.16322846
>>16322765
There is no "classic FPS movement mechanics". Almost every engine does things slightly differently.
If you're looking to reproduce commonly recurring quirks like bunny hopping, don't worry. They'll find you. Those things exist because they're hard to iron out.
abf5dc No.16322945
>>16322846
Keep bunnyhopping, make each level have slopes for ramping, you are a man on hover skates or something like what Lucio has. Except you are better than Lucio in every way.
da7cd6 No.16323020
>>16322765
Have you checked Doom source ports? There are the "Game engine black book: Doom". But I can't recommend it since I haven't red it.
7263be No.16323216
>>16321087
Section 2 is immediately brighter than section 1, as well as larger. Section 2 is a place known as the Forgotten Kalpa Chamber, a remnant of a previous kalpa's Soul Cairn A familiar scamp from Vvardenfell has branched his business out here.
7263be No.16323235
>>16323216
An in-editor shot of the Precincts.
08b450 No.16323297
>>16323235
Is it supposed to look like an evil bald dude with a mustache and/or a cigarette?
7263be No.16323321
>>16323297
No, but now that you saw that, yes.
d6a9b2 No.16323418
>>16322776
>>16322846
I meant specific things like want-direction and actual move-direction, velocity, friction and air-friction and how all of these things fit and work together. Like how that system was probably put in place because how in Doom strafing and moving forward added speed separately which allowed players to straferun diagonally to move faster.
>If you're looking to reproduce commonly recurring quirks like bunny hopping, don't worry. They'll find you. Those things exist because they're hard to iron out.
I figured that was what I was gonna have to do so I guess I'll just continue to iterate on my stuff. That probably answers my question.
>>16323020
I haven't looked into those yet but that book looks interesting. Thanks I am putting that on my list.
4ba251 No.16323459
>>16323418
Probably won't be helpful since it's pretty easy to find but here's some resources.
https://archive.is/JyUH2
https://archive.is/fVyjN
d6a9b2 No.16323915
>>16323459
Thanks, I already knew the second link but the first one was new to me.
873afc No.16323962
>Figured that to make smth clickable you need Area2D which has appropriate signlas.
>Touched my every single piece to make them clickable
>Putting it into Autoload on click and removing from pare
>Adding to board when entering gaming zone
>Returning it to first parent when another piece selected
I do save position in first parent and setting it before returning. But for some reason it end up misplaced by some constant vector. I'll research that tomorrow.
Another strange thing is that when I don't release mouse button and move into the game board it jumps to position where I released mouse button after I select another piece.
1e1c5e No.16325024
>>16322765
Doom movement has been done from hell to back, try something like Titanfall 2 with super powers instead.
1e1c5e No.16325069
Why didn't this thread bump after the post I made?
979591 No.16325219
>>16325069
Due to low quality posts coming from your IP address, I've set you to autosage for all threads.
2998eb No.16325343
>>16325219
wow what a faggot, there's no need for that
>>16325069
don't worry, you can easily fix that by installing 8chan gold to bypass autosage
d6a9b2 No.16325376
>>16325024
>something basic and elemental has been used over and over again
You don't say!
>try something like Titanfall 2 with super powers instead
Never played it, don't care. You do your stuff, I do mine.
1e1c5e No.16325524
>>16325376
Very well, best of luck on your game anon.
d99169 No.16325583
>>16325219
I didn't know that was possible.
0699b7 No.16326090
Finished filling the bathtub. No water plane clipping whatsoever.
47d044 No.16326105
061cff No.16326125
>>16326090
>opaque water as censor
DAMN YOU, ANIME!
0699b7 No.16326132
>>16326125
No the water is a little bit transparent. She will have bubbles to censor her body though.
69cf67 No.16326154
>>16326090
In theory, is it possible to replace Shinobu's model with another rigged model, and degrade your game into a simple, generalized waifu simulator?
0699b7 No.16326162
>>16326154
animations wont match up perfectly, no look at pupil logic, all logic would need to be edited, and wouldnt be compatible with an upcoming clothing system. So not really. Yes superficially in a broken way.
93ce68 No.16326314
>>16325524
Thanks. Sorry if I was being mean other anon but I am still trying to grasp the basics. More complex things are impossible right now.
348744 No.16326492
>>16326132
>She will have bubbles to censor her body though.
061cff No.16326787
Only tangentially gamedev related, but I spent a few hours last night and today writing a Brainfuck interpreter that has a section of memory mapped to a palletized image output.
It's a standard Brainfuck implementation using a 0.5 MB memory chunk of 8-bit values, with one additional symbol (! to be precise) in the language for refreshing the screen as that's currently the slowest thing in the program.
The first 8 values in memory are not part of the "VRAM" yet as to allow code to skip past the it to get to normal memory.
The screen is currently 128x122, half of the actually usable resolution of the NES, with the palette being the NES's palette as well (pic 2). I could theoretically add a full 256 color palette though, just wanted to ape the NES a little for now.
Pic related is the output of
>>>>>++++[->++++[->++++<]<]>>.[->+>+>+>+>+>+>+>+>+>+>+>+>+>+>+>+>+>+>+>+>+>+>+<<<<<<<<<<<<<<<<<<<<<<<!]
It's just a loop to get the value 64 into address 7, then looping over a bunch of visualized addresses and incrementing them once, drawing a frame, and repeating that for each color in the palette.
Maybe I should add a symbol that like the comma returns keyboard input, but returns what it was at the start of the frame instead of blocking to wait on user input. That should make making a game with it a little more viable, or at least non-turnbased games.
2090d5 No.16326897
Where the fuck is the keyframe edit enable option in Godot 3.1? I've been looking for it for an hour.
717bd1 No.16327152
7454b2 No.16327154
Today's progress:
Made my mind up on a screen layout so I can start woking on items and inventory system.
Barebones midi playback with looping and muting (mapped to M).
Sprites can be scaled at will (as well as collision boundaries).
7454b2 No.16327169
>>16327154
oh and I wanted to ask, what's the best strategy for having different music play depending on player coordinates? I don't want to have map areas defines as strict squares ala runescape but checking for every single tile sounds slow and tedious.
f965aa No.16327218
>>16327169
You could add invisible triggers for an area switch at all possible paths to get to another area. Going through it one ways sets current_location to 2, the other way back to 1. Use that value then to change the music.
9253f4 No.16327230
>>16327169
In an editor, draw continuous or equivalently smooth curves over the areas you want to play certain music in. That is, partition the space arbitrarily. Then recursively subdivide the map into a quadtree and if a partition is entirely covered by a given song, assign the node to that song in that location. Then to play a song just do a lookup in the table.
873afc No.16327684
I've made tiny bit of progress today. Still have that bug, that pieces are shown in the wrong place after reparenting, but now it's showing where they should've been in the previous parent. So, that makes tiny bit more sense now.
Hope your progress is doing better, fellow nodevs.
6806e7 No.16327700
>>16327684
Got some progress done today:
>Finished porting to latest Unity beta
>Scene loading logic with animated transitions
>Fixed a bug where performance would get worse each time you switched levels
>Made some particle systems look nicer
14dd54 No.16327723
Anyone know a formula to decelerate in 3D axis without drifting in one direction?
6806e7 No.16327738
14dd54 No.16327767
>>16327738
accelerated/decelerated movements like a car.
9253f4 No.16327786
>>16327767
Cars drift in one direction as they deccelerate, thats how acceleration works
2998eb No.16327886
Recipes, items, ingredients, effects, about everything that benefits a lot from modding is now loaded from XML rather than my retarded proprietary ScriptableObject whose only benefit was easy AnimationCurve editing. I might make a AnimationCurve -> XML editor because how much of an ass is declaring one in xml
6806e7 No.16327894
>>16327767
Just manage the direction vector and speed separately. I don't exactly understand what the problem is.
ce7032 No.16328829
The youtube app on the 3ds absolutely sucks, it crashes all the time for every video now. Anyone has an idea on how I can stream youtube videos on my 3ds without the app? Something like the VLC streaming function.
da7cd6 No.16329214
>>16327767
As I understand, what you need is friction. So it won't drift infinitely. **That's why cars don't drift infinitely after all*
da7cd6 No.16329269
14dd54 No.16329295
>>16329214
>As I understand, what you need is friction.
Ah ok, thanks.
08b450 No.16329904
>incredible desire to make fun little games
>mfw I'm still at the pre-engine stage
>haven't even started on the biggest component, i.e. opengl rendering
>haven't made any progress in weeks due to lack of motivation and mental energy
Should have stuck with browser games
6806e7 No.16329920
>>16329904
Then why build your own engine in the first place?
08b450 No.16329932
>>16329920
Because I hate using other people's shit.
6806e7 No.16329957
>>16329932
I can see why someone might want to run with their own engine, but that seems like a silly reason. Aren't you using libraries?
08b450 No.16329978
>>16329957
I have a laundry list of reasons if I really want to get down to it, that just summarizes it pretty well.
>Aren't you using libraries?
No, while I don't mind using very focused libraries for very complicated things, I have reasons to avoid libraries as well, and they're similar.
08b450 No.16329981
>>16329978
Oh actually I do use STL for loading image files.
08b450 No.16329986
*STB
Handing image files would require me to deal with someone else's shit (the image format) more than using a library (someone else's shit) will.
4ba251 No.16330004
>>16329981
Are you using any libraries for input or window creation?
d686d8 No.16330005
>>16329920
Why doesn't every animation studio switch to using goAnimate?
061cff No.16330012
>>16329986
>not just writing your own image format and making all your assets in that format from scratch
Heh, fucking scriptkiddie
08b450 No.16330040
>>16330004
No. I used SFML in the beginning but began to hate it as well. SDL or whatever didn't sound much better. Not like it was difficult to do it myself anyway because Windows API has great documentation and plenty of guidance online.
>>16330012
I unironically considered it, but I made a BMP parser myself and I don't think I could make anything much better myself anyway, even less something that could rival PNG or JPG in compression.
ad3c71 No.16330045
>>16329986
>Not making his computer from brick of copper and sheet of textolite.
Chinacuck
4ba251 No.16330050
>>16330040
SDL is much better than SFML, way less useless stuff and easier to port.
2998eb No.16331001
>Finished importing my old 89 ingredients to the new system my brother is making
>Time to import the recipe types
>I think there were 60 or so? no prob i can do it!
>There were over 125 recipes
f26346 No.16331018
>>16331001
write an auto-converter
2998eb No.16331065
>>16330040
Did you code Windows API yourself? Damn you are good
Or are you actually using someone's else shit? Or are you going to claim that Windows is not shit? Checkmate atheists
>>16331018
It's converting his dumb hardcoded trash to XML
7263be No.16332463
>>16323235
Some shots of work in progress areas. No clue where to go in the dungeon's progression from here. Probably going to add some more optional areas.
2c7991 No.16332517
>>16332463
Those screenshots remind me of the Sleeper's lair in Gothic.
69cf67 No.16332527
So after playing a lot of Baba is You recently, I think internally that might be a fine way to give tags and commands to my puzzle objects. I'd write it in one of two ways in a flat text file, and when I load a level set, it can generate the logic for that group of tiles, and I wouldn't have to compile it
Yes I know text is shit, but this seems like a fun/flexible way to soft mod the game too
7263be No.16332545
>>16332517
Part of my inspiration is from the depiction of the Soul Cairn in Battlespire. I couldn't find a prefabricated tileset that quite matches it thematically though, and the one I'm using is rather limited when it comes to dungeon interiors. The way I've made it is a bit too red as well.
47d044 No.16332590
>>16332527
>I'd write it in one of two ways in a flat text file, and when I load a level set, it can generate the logic for that group of tiles, and I wouldn't have to compile it
Congratulations. You discovered scripting.
cab112 No.16332599
>>16332517
alter sumpfkraut
23fc5d No.16332614
Good lord, FPGA development is so obscure. I just want an Arduino like platform that isn't fucking obsolete or requires a whole gay proprietary design suite. Not even that, but FPGA's have WAYYYY better uses as Microcontrollers than dedicated processors due to parallel computing and larger bit lengths- all sorts of cool shit you can do that you can't with some dinky AVR chip.
69cf67 No.16332665
>>16332590
Yeah but using a full language just feels like overkill
Fucking
VIDEOGAMES
0699b7 No.16332990
Finally starting to put shit together. Still need to texture her new hair.
abff5f No.16333228
>>16332990
That sharp divide between neck and body, is that visible in game? I know it's Blender being dumb with normal vectors.
1d68f6 No.16333270
>>16332614
pick up an ICE40 they have a free as in freedom toolchain and have a $20 evaluation kit. Im not sure what this has to do with gamedev, but if you make a decent game on fpga please post progress here.
aae94b No.16333295
>>16332990
LET ME SEE HER COOCHIE ANON
DON'T HIDE IT FROM US
81db16 No.16333315
Big think mode: engaged
Should I use a comfy rural skybox or stick to grimdark?
6806e7 No.16333319
>>16333315
Looks rather distorted.
eb2760 No.16333321
>>16333315
I think Dark Souls 2's Majula did a really good cross between those two options.
31c85c No.16333322
>>16333315
what about a night sky filled with stars and a full moon? i think it would be a perfect mixture of comfy and dark.
81db16 No.16333323
>>16333319
it is, just experimenting now
>>16333322
thing is, even the grimdark version is supposed to be kind of a twiligh/evening thing more than the middle of the night
81db16 No.16333326
>>16333321
>Dark Souls 2's Majula
thanks for the tip
5190ef No.16333462
>>16330040
To be honest, what you are doing is pretty cool. But it's okay to do it to learn smth, to optimise part of some bigger system or smth. But doing it because you just don't like using someone else's shit is… unreasonable.
You can do it if you don't care about ever making the game and just enjoying the process.
But we are here about, just like make the game (TM). So I think it's more reasonable to write smth as quick as possible and than rewrite as autistic as you want. Because it's unlikely that you know what you want/need beforehand and you would just waste time doing shit wrong and don't have result.
61503f No.16333616
>>16333462
You should probably go back to reddit, friend.
6806e7 No.16333733
Implemented the logic for the designated target area at the end of a level. If the main character goes down, you have to save her first before you can finish the mission.
Also added an indicator to the HUD that directs you towards your incapacitated character.
ba31b4 No.16333749
>>16332990
show bobs and vegana
Did you sculpt her hair? There's actually anything under the bubles that needs to be censored?
We need more of your dev posts on /agdg/.
2998eb No.16334017
>>16332990
add some asymmetric bubbles to the forearm and knees, hide the fact that the convenient bubbles are so convenient
061cff No.16334118
Had some more insights on my BrainFuck game, uh, thing. I implemented the non-blocking input as I stated in my previous post on the subject (using the symbol $), but realized I'm going to need to make some more changes to what I have.
Currently it is impossible to knowingly skip past the visualized memory ad stop at the other end, aside from just typing > width*height times. That'd be kind of retarded. I think the only realistic way to fix this is to add another two symbols, which skip to the left and right scrwidth+padding times and add some usable memory between each horizontal scanline.
That way, assuming I use { and } as go left/right a scanline, you can set up a loop to add the y value in the extra memory between scanlines and end up at the free memory using just }+>+[[-}+<+>{]<}+>+]
Now my question is, at what point does this stop being a game made with brainfuck? The ! symbol I added previously didn't interact with the basic language, just the external display, and the new $ one is the same as the comma but nonblocking, but { and } would be actually changing how you use the language. Am I just being autistic here? And does anyone even understand what I'm doing in the first place?
0699b7 No.16334122
>>16333228
It wont be. I havent applied a data transfer yet.
>>16333749
Yes it still needs some work. No genitals though. Ill post some more later.
>>16334017
Thats actually just a mockup. Theres way too many bubbles censoring. Im gonna get rid of like half.
2998eb No.16334123
>>16332990
Just the right amount of visible skin
061cff No.16334136
>>16334122
>not using 3 bubbles, one on her crotch and two on her nipples
2998eb No.16334252
I've spent the last 3 hours on this juice, I demand praise.
Speed and color can be easily adjusted
61503f No.16334264
>>16334252
Maybe if you didn't make 38 shitposts it wouldn't have taken 3 hours
1d68f6 No.16334406
>>16334118
I think I understand the general idea, but i couldn't find your post about $ and ! commands. If you feel its important to use "real" brainfuck why not have the graphics and other special features be controlled by outputing nonprinting characters?
061cff No.16334555
>>16334406
Th only other post I made on the topic is >>16326787
There I explain what the ! symbol does, and suggest what I would later implement as $.
>nonprinting characters
Interesting idea, especially since the period has been kind of made useless outside of debugging. I'm not 100% certain which ones are truly nonprinting and which ones are nonprinting-but-useful, though. I know some of them play a sound on windows when output, for instance.
But either way it doesn't 100% address what I was thinking of, although I guess I didn't make that entirely clear. As long as you replace $ with a comma, currently all code written for my games would work in any brainfuck compiler or interpreter and result in the same resulting memory, just not visualized and not realtime. { and } wouldn't, unless you replace them with 256 < or > symbols. Maybe I'm hypocritical for thinking a single character replacement is fine and multiple isn't. I don't know.
And if anything tricking myself into allowing things because it's just return codes is a bit dangerous as I could add for instance a setpixel function or sprite support using a set of memory values yet "it's still all real brainfuck :^)"
If it wasn't clear, I'm mostly having trouble figuring out what limits I want to set for myself and hoping someone opinionated could draw the line for me, but that's kind of an unrealistic thing to hope for given the subject matter and how no one probably gives a shit about it. Especially considering the Snake implementation I plan to write for it is probably about the limit of what you can do with it, anyway.
I don't think there's much of a "brainfucker" community out there, and even if there were they'd probably be plebbitors so I wouldn't want anything to do with them.
1d68f6 No.16334652
>>16334555
I missed the part of your post explaining ! and $ because I was taking a shit and phone posting like a retard. If you want to do everything in brainfuck including graphics routines you could have multiple brainfuck machines with memory maps like your graphics example that all talk over a bus with . and , You could even make some sort of keybuffer that returns a special code for nothing available to avoid needing $, though it might be hard to implement that in particular as a memory mapped brainfuck machine.
b2b846 No.16336397
Would it be possible to write a shader for Unity that skews a texture on the mesh but keeps it seamless? It's easier to show than tell, so pic for reference.
0699b7 No.16336399
>>16336397
Yeah if the texture is seamless.
70b741 No.16337110
Thanks to an anon from another, as well as some thinking, I came up with a potential hack for an ability system that has many different types of targets for an ability. For both sharing and feedback, I wanted to share it here. Keep in mind that this was developed for a non-tile-based, real time strategy or real time tactics, game.
Let's say I have an ability that needs you to target a character to heal and another ability requires you to target a world space position to teleport. This is where the hack comes into play. A "spherecheck" can be used for both abilities! For the fireball ability, just get the closest object from the center of the sphere to throw the fireball at. Use the center of the sphere to get the point the character will teleport to. You can even use the same spherecheck for area of effect abilities.
Downsides about this is you would have to either use an engine that supports physics collider shape casting, for example Unity's Physics.OverlapSphere, or create your own physics shape casting. I think any performance hit for doing it this way is meaningless because the alternative would involve ridiculously cancerous code. You would still have to do some checks, but I don't think there will be as many checks using an alternative.
70b741 No.16337114
>>16337110
Thanks to an anon from another thread*. I need to learn to proofread
e02699 No.16337150
>>16336397
Literally "uv.x+=uv.y*skewfactor;".
6806e7 No.16337167
>>16336397
What >>16337150 wrote would obviously happen inside a vertex shader. You'd modify the texture coordinates in there and let the fragment shader do its thing.
70b741 No.16337639
I added a page for game ideas for beginner game devs to the wiki and linked to it on the resources page. I-If that's ok.
08b450 No.16337830
>>16337639
Should've made a generic ideaguy page instead, and categorize by complexity or type.
061cff No.16337837
>>16334118
>>16334652
Some sleep and a whole lot less sleep deprivation later, I realized what a dumb fuck I was being. { and } are less intrusive on the brainfuck language since they can actually be replaced directly instead of having roughly the same functionality, and besides who gives a shit. Honestly I should probably add some way of moving in x times using loop because that's impossible with current brainfuck and sorely needed for nicely written code. But whatever, I can manage for now.
Added the padding before every scanline instead of just the first, tested my "set up line numbers" code which works wonderfully, and made a hello world script which doesn't require it. It's not the cleanest code since I optimized for symbols instead of making the letters easily repeatable, so they get drawn differently depending on if you're approaching them from the top or bottom pixel. It only saves a few symbols but you don't try this shit without being autistic so >muh cycles. Here's the code, with comments: >>>>>>>>>}
Huu +}+}+>+<}+}+>>+{+{+{+{+!
Eud >>+>+>+}}+<+<+{+}}+}+>+>+!
Ldu >>+>+>+<<{+{+{+{+!
Lud >>>>+}+}+}+}+>+>+!
Odd >>+>+<{+{+{+{+>+>+}+}+}+}+!
>>
Wdu >>+{+{+{+{+>>+}+}+}+}+<+>>+>+{+{+{+{+!
Ouu >>+>+<}+}+}+}+>+>+{+{+{+{+!
Ruu >>}+}+}+}+{{{{>+>+!
Lud >>+}+}+}+}+>+>+!
Ddd >>+>+<{+{+{+{+>+>}+}+}+}
!du >>+{{+{+{+!
Maybe I should focus on getting the interpreter's code clean enough that I can show it to the world without embarrassing myself so you guys can try it instead of just reading me waffle on about unintelligible bullshit. Any interest in that?
794c98 No.16338247
>>16337110
>Using the game's physics engine and collision detection system to get the nearest object
that's some pajeet-tier shit.
I finally ground through the Vulkan tutorial once again using my memelang but the validation layer is screaming at me again about image layouts when I resize the window and redo my swapchain+render state+commands and I don't understand why.
>Images passed to present must be in layout VK_IMAGE_LAYOUT_PRESENT_SRC_KHR or VK_IMAGE_LAYOUT_SHARED_PRESENT_KHR but is in VK_IMAGE_LAYOUT_UNDEFINED. The Vulkan spec states: Each element of pImageIndices must be the index of a presentable image acquired from the swapchain specified by the corresponding element of the pSwapchains array, and the presented image subresource must be in the VK_IMAGE_LAYOUT_PRESENT_SRC_KHR layout at the time the operation is executed on a VkDevice (https://www.khronos.org/registry/vulkan/specs/1.1-extensions/html/vkspec.html#VUID-VkPresentInfoKHR-pImageIndices-01296)
794c98 No.16338277
>>16338247
Never mind, I was confusing myself between the swapchain image index and the "frame in flight" synchronization object index.
b2b846 No.16338281
>>16337150
Thanks. This was exactly what I needed to figure out which nodes to use in the shader graph.
6806e7 No.16338294
>>16337110
>>16338247
>that's some pajeet-tier shit.
Why though?
If you wanted to do it properly yourself, what you'd have to do is build some sort of spatial acceleration structure (quadtree, octree, kd-tree, …) and use that for a nearest neighbor lookup. That's exactly what the physics engine does too, with the same kind of data structure.
Rolling your own solution wouldn't necessarily be better. After all, you'd have to update your tree on top of the physics engine already doing that for its tree too. So you have redundant work right off the bat.
The only way I can see this being worth it is if you make A LOT of lookups per frame. Because your own data structure will only be filled with potential targets, rather than every physics object. Therefore you'll get a much more sparsely populated tree which would make the lookups faster.
But for a single check every now and then, it's not worth it.
61503f No.16338499
>>16338340
It looks rather concise if you can adapt to it.
69cf67 No.16338509
>>16338499
Concision is what you want the compiler to do. You want your code to be readable, immediate, and self explanatory, not open to interpretation. You might as well just name everything two letter names
61503f No.16338515
>>16338509
No, you want languages to be concise too. There are many reasons Java is hated and retarded boilerplate is one of them.
69cf67 No.16338520
>>16338515
The language itself is concise, the standard library is not.
360d07 No.16339166
Thanks to renderdoc I get to visually inspect each peice of junk that goes through my drawcall. No idea why all my data is bad right now… they seem to pass through the shader fine anyway. So I guess all my verticies are messed up? But I don't really know why right now.
7af1fe No.16339390
>>16338281
I use λ for lambda and ƒ for function in my code.
da7cd6 No.16339536
>>16339390
If you never going to show your code to anyone, you could probably get away with it.
But if another human being going to work on your code than I hate you, you fucking pajeet, for wasting people's mental energy on thinking how the fuck they should comply with your retarded code style. Seriously, please, don't do that
9253f4 No.16339708
>>16339536
Programmers should conform to math style because its correct, not the other way around
f2dfbd No.16339779
>>16321959
Gibs work, controller debugged a bit, hud needs work, moving to fix player movement code before I do more with these sprites. The player turning modifies their forward velocity, I need the player to move forward only on input (or collision) and velocity not modified by turn angles.
f2dfbd No.16339780
>>16339166
Keep it up, I like how this looks.
80180c No.16339846
>>16338281
I can never get used to looking at that image, absolutely terrifying
2998eb No.16340785
>>16339708
how do you write λ and ƒ?
I copy pasted from your post
f965aa No.16340827
>>16340785
Alt codes, probably, though the one I found for lambda, 955, doesn't seem to work for me.
I use various Alt-code symbols in my textbox scripts to trigger special functions, as all of it has to be in a widestring.
For example, ¢XX to change the color, using a function where I declare default colors (00 being light blue, 01 being green, 02 being orange and so on)
Another example, ∟for a linebreak.
Technically, Space is also a special character since instead of drawing something like most other characters do, it just adds to the current horizontal position and proceeds to the next character.
6806e7 No.16340835
>>16340785
Do you happen to use a keyboard with US-layout, Anon?
873afc No.16340837
>>16340785
That's the reason I'm so agains it. Though every language has its own code style. Maybe if it some kind of math Scratch than it's OK. But for C like - not so much.
Social interactions interfered with my ability to make progress. But today I finally fixed that stupid bug that items were place in wrong position. It turned out I had signal to remove piece from the board but it wasn't connected for some reason. Thus I wasn't removing piece from wrong parent. Now I'm working on collision detection. I need to allow placing piece only in case if all of its tiles are on the board. Position on child sprites are misplaced by 0.5 of the greed tile for some reason.
2998eb No.16340854
>>16340827
>I use various Alt-code symbols in my textbox scripts to trigger special functions
You are actually insane
>>16340835
Nope, ABNT2. I can't foresee a future where I would use alt codes on a regular basis
f965aa No.16340862
>>16340854
>You are actually insane
How else do you trigger things like that on the fly? Better it be handled by the text printing script itself than any other method I could think of. It's interworked with my regular character checks anyway, since the script has to retrieve each character one by one anyway to get the order in which they're laid out in the texture sheet, so it's just a quick check for a few special characters, and if it's not any of them it does its regular math.
Again, how else would you colorize two words in the middle of a sentence somewhere?
6806e7 No.16340874
>>16340854
>Nope, ABNT2. I can't foresee a future where I would use alt codes on a regular basis
Well, there is a keymap called "EurKey". It's used for US layouts and extends it to make using European symbols really easy. Everything else is normal, but I can press right alt + M once and the next letter I type is from the Greek alphabet. So if I press right alt + M and follow up with a, I get α, b I get β, l I get λ and so forth.
If I hold right alt instead, I get the German Symbols like ä,ë, ö, ü, ß, …
It's quite convenient, if you need those symbols regularly and better than either alt codes or automatic replacement of stuff like %ALPHA% for α.
2998eb No.16340914
>>16340874
that's actually pretty good
to actually type both on russian/japanese/"other languages that don't use english alphabet" and english, and nothing else
061cff No.16341000
>>16340862
What's wrong with just using /c instead of ¢?
f965aa No.16341063
>>16341000
I wanted to have both slashes available within the font itself. I consider them regular enough characters to where they should be there. In fact, pretty much any character that you can easily print with the US keyboard layout, I wanted to be available. Since C++ will not let me use \ regularly, I use | in its place and / as itself.
That, and I think special characters are easier to read. They stand out far easier within several lines of text. Your eye immediately catches them since they don't belong. Hell, just looking at your post, the very first thing I noticed was ¢.
There is one advantage to / that I can see, it's something I might have to try out later - it could be a slight optimization since it would consolidate all special characters to one check, except spaces.
211574 No.16341107
Anyone here has experience with Haxe? Do you happen to know if it's easy to make a 3D game that can compile for both desktop and browser?
873afc No.16341112
>>16341063
What's wrong with good old HTML? It's excessive, but not as autistic as you people would be able to read it no problem.
873afc No.16341118
>>16341063
And when you get back to that part after forgetting everything about it HTML would be dumb clear. I think, being obvious about intent is one of the most important qualities of code. And unlike labdas from previous post - your invention is far from obvious.
7af1fe No.16341281
>>16339536
The text editor changes the way it's displayed. The source file stays the same.
2998eb No.16341316
Anyone ever faced the issue of Visual Studio not getting breakpoints in generic classes?
9253f4 No.16341351
>>16340837
Just program in haskell, math syntax fits right in
16c295 No.16341447
I've made a little bit more progress with my RTS game terrain. Also trying to work out the water animation. It's a pain since I have to use palette cycling rather than conventional frame-by-frame animation.
2998eb No.16341467
>>16341447
looks cool. Can you walk behind cliffs?
6806e7 No.16341470
>>16341316
Have you tried setting a breakpoint for the caller and then stepping in?
2998eb No.16341876
>>16341470
The caller is Unity's ECS system manager or whatever
I might actually put a non generic static method that is only called to get breakpoints to work though
22f6f9 No.16342717
>>16296248
Working on a pong clone in godot now.
93846d No.16343111
What are good resources/places for learning hit detection? Specifically for melee combat.
93846d No.16343142
I reformatted the "Random Anon Game Ideas For Beginners" page into a general game ideas page. There are now sections for beginner, intermediate, and advanced game ideas. Feel free to add your own if you want to that page.
47d044 No.16343466
2998eb No.16343754
>>16342717
Good for you, will be a invaluable experience even if turns out shit
061cff No.16343859
>>16342717
The very first thing I do whenever I download a new engine is to avoid all documentation and implement a basic version of Pong. That's how I measure how good an engine is.
22f6f9 No.16343980
>>16343111
For 3d, look up AABB collision.
9253f4 No.16344058
b27ac2 No.16344387
Shmup anon here.
The most undewhelming explosion shader done, need to add color mixing but at least it's linked to the death event.
I think I've spent too much time on it but at least I get the gist of writing graphic effect and some limitations when going shader only.
Curently: adding weapons, and graphic effects associated to them…
6806e7 No.16344406
>>16344387
>The most undewhelming explosion shader done
Post pics
873afc No.16347358
Cathedral anon here. Implemented collision detection. Now it isn't possible to place one piece on another. Guess I need to start working on turning pieces.
83d6ad No.16348000
Serious question
Have any of you considered picking a random Sega game that has gameplay similar enough to a game you want to make, make a game with that skin, then ask Sega to license it?
This seems like a no brainier way to get a company off the ground if you're starting with nothing making a game in your spare time. Sega is clearly willing to license any of their old IPs to who the fuck ever and their list of old IPs not only spans most every genre but is one of the broadest of any company's.
I guess I'm only an idea guy but my idea isn't for a game, it is for getting a company off the ground.
Fuck I bet you could make any game but with Alex Kidd and they'd fuckin license it why the fuck not they licensed Sonic to a fuckin rom hacker
f965aa No.16348043
>>16348000
Any sort of license is limiting. Suddenly, you can't have fun with it, you have to adhere to strict limitations, essentially meaning you have to be a blind fanboy, spouting Sega Reference after Sega Reference. Want to do something original? Fuck you, go make an After Burner reference, cocksucker.
I've had an idea for a Tails 3D platformer, his flight ability could give an emphasis on verticality that no other 3D platformer series has (Mario coming the closest), without outright breaking it, but god forbid I don't want Sonic's 70 fucking friends to be shoehorned in, or I want more than two stages to not be Green Sand Hill Zone and Chemical Plant. They wouldn't let it happen.
Shitting out a licensed game is a get-rich-quick-scheme and nothing more. Most people here, I imagine, are making their dream games and a license attached to it would bring in the shekels, I'm sure, but it would never live up to the dream game status that anons are aiming for.
83d6ad No.16348343
>>16348043
Damn I was sitting here hoping that one could make whatever game then ask Sega to ok it then it would go by fine. Have you any experience with obscure properties outside of Sonic?
f965aa No.16348368
>>16348343
Anon, I'm only speculating, so maybe Sega's far better than I assume, but consider Mania. Most speculation is that it only had two new zones not because the devs wanted so but because Sega demanded so. But hey, they referenced that fucking popcorn machine and the Segasonic arcade game-exclusive characters, so it's all good, right?
If something is safe for work, I'm sure that would be the only limitation, but if some anon was really fond of making a game with raunchy humor, do you think Sega would allow jokes about Blaze being flat or Rouge having massive fucking tits?
I'm sure a lot of anons here could make variants of their games starring Sega characters, but would have to abandon some mechanics, some humor or something or other along the line. If you went into it with that thought from the getgo, it might be okay, but you have to keep those exact limitations in mind - you'll never be as free from a creative standpoint when working with someone else's IP.
83d6ad No.16348458
>>16348368
I see what you mean but your biggest emphasis comes from censoring Sonic characters which are clearly going to be the most protected class of characters. Sega has video games from the 1970's. Sonic is completely irrelevant to their history of games. Would Sega accept adult jokes in a game? I don't know please fucking try.
f965aa No.16348464
>>16348458
My project in particular would never work as any Sega franchise, for both the gameplay style and the style of humor. Maybe other anons would be up for it.
da7cd6 No.16348630
>>16348458
That could be done. But I can imagine two scenarios.
1. You present your game to them. They actually give you good advice how to make your game better. You do that and they release game for you and give you some percent of sales.
2. You present it to them. They force you to bend your game to their shitty corporate standards and remove all fun from it. Because you have nowhere to go with this. They force you to give up all rights on it, pay you some tiny amount for it and remove your name from development history. When game turn out shit blame you.
So I think doing that is OK if your game is actually 1 to 1 clone. Basically source port and you did it in a month with Unity. You can make some shekels after all. And shekels are good to feed you while you do your dream game. But it's no go if it IS your dream game (TM)
7934bc No.16351247
Hello /agdg/ I have literally looked everywhere for this and I want to find it. This UE4 marketplace asset was free for a month, does anyone have it?
https://www.unrealengine.com/marketplace/en-US/slug/modular-cliffs